Analysing spoken and multimodal texts: reading transcripts and speech-like texts through discourse (turn-taking, adjacency pairs), pragmatics, prosody and the features of spontaneous speech, and multimodal texts through graphology, with mode read into the analysis (AO1, AO2, AO3).

What this dot point is asking

Component 01 is titled "non-fiction and spoken texts" because spoken and multimodal texts sit alongside written non-fiction in the anthology and the unseen material: transcripts of conversation, speeches, broadcast extracts, and digital or multimodal texts. Analysing them well needs the tools of spoken-language and multimodal study, discourse, pragmatics, prosody, the features of spontaneous speech, and graphology, with mode read into the meaning. This dot point covers those tools and the discipline of treating mode as part of the effect.

The answer

The key move with spoken and multimodal texts is to recognise what mode lets a text do and to analyse the resources specific to that mode, rather than reading everything as if it were crafted prose. A transcript negotiates meaning live; a speech projects a scripted persona to a listening audience; a digital text blends modes. Three clusters of tools cover the territory.

Interaction: discourse and pragmatics

A transcript of conversation is structured interaction, and discourse analysis reads its architecture. Turn-taking and who controls the floor reveal power; interruptions and overlaps show competition or collaboration; adjacency pairs (question-answer, greeting-greeting) show how talk is organised; topic management shows who steers. Alongside, pragmatics reads the relationship: politeness and face-work, the implicatures behind indirectness, the speech acts performed. Together these analyse how speakers build a relationship and negotiate power in real time.

Spontaneous speech features

Spontaneous speech carries features that written texts edit out, and they mean something. Fillers ("um", "er", "you know") and false starts mark planning in real time; pauses (often timed in a transcript) signal hesitation, emphasis or turn-management; repairs (self-correction) show the speaker monitoring their talk; non-fluency can mark spontaneity, nervousness or thought. Read these not as errors but as evidence of the spoken mode at work, and connect them to the speaker's stance or the relationship.

Planned speech and the scripted-spoken hybrid

A political speech or sermon is a hybrid: written to be spoken. It uses rhetoric designed for the ear (anaphora, tricolon, parallelism that build aurally), direct address, and prosodic effects (stress, pause, rhythm) that a transcript or stage direction may mark. Analyse it as crafted for delivery: the patterning that would look heavy on the page lands as rhythm aloud, and the persona is projected to a listening, often public, audience.

Multimodal and digital texts

Multimodal texts combine modes: writing with image, layout and typography (graphology), and often a spoken-written blend (a social post's informal, speech-like register in written form). Read graphological features (layout, font, colour, image) for purpose, and read the mode blend: how a digital text imports spoken immediacy and interactivity into the written mode. Always connect the visual or multimodal feature to meaning, not just description.

Examples in context

The texts vary by series, so the moves below are illustrative.

Interaction read as power. "The transcript stages a struggle for the floor: speaker A's repeated overlaps and latched turns seize topics before speaker B can finish, while B's lengthening pauses and rising fillers mark a speaker losing ground. The discourse itself, who holds and who yields the floor, dramatises a power imbalance that no single word states, and because this is spontaneous talk the imbalance is negotiated live, turn by turn." Discourse read as live power.

A scripted-spoken speech. "The address is built for the ear: the tricolon mounts through three parallel clauses to a stressed final beat, and the inclusive 'we' gathers the listening crowd into a single body. On the page the repetition might seem laboured, but delivered aloud it is rhythm and build, the spoken mode's way of carrying a public audience by sound as much as sense." Scripted-spoken rhetoric read for delivery.

Try this

Q1. Why must a transcript be analysed as interaction, not prose? [2 marks]

• Cue. A transcript is structured live talk; its meaning is in the turns, overlaps, repairs and face-work, not only in the words, so reading it as prose misses the spoken mode.

Q2. What do spontaneous speech features (fillers, pauses, repairs) signal? [2 marks]

• Cue. Real-time planning, hesitation, emphasis, turn-management or thought; they are meaningful evidence of the spoken mode, not errors.

Q3. Analyse how a spoken text builds a relationship between speakers, and compare with a written text. [32 marks]

• What the marker wants. Discourse and pragmatic analysis of the interaction named precisely (AO1) and read for effect (AO2), with mode read as context (AO3), woven into comparison with the written text (AO4).

A note on spoken texts

This guide is AI-written and not individually human-reviewed. The spoken and multimodal texts vary by series and transcription conventions differ, so confirm them against the current OCR H474/01 materials and the EMC anthology. The tools of discourse, pragmatics, prosody and graphology, with mode read into the meaning, transfer across spoken and multimodal texts.