amazon.co.ukTextual Variations and Representations
A* revision for AQA 7702 Paper 1 Section A and OCR H470/01: analyse spoken, written and electronic texts, build representation claims with systematic methods, and compare two texts for top-band AO1/AO3 marks.
🎯 Learning Objectives
- Apply systematic methods from all language levels (lexis/semantics, grammar, pragmatics, discourse, phonology/prosodics, graphology) to unseen spoken, written and electronic texts (AO1).
- Analyse how mode, register, genre, audience and purpose shape meanings in a text, evidencing each with a linguistic choice (AO1/AO3).
- Construct precise representation claims - 'the text represents X as...' - and advance them through every paragraph (AO1/AO3).
- Select relevant contextual factors (mode, era, audience, production) rather than listing context indiscriminately (AO3).
- Compare two texts through integrated points of similarity and difference instead of parallel text-by-text analysis (AQA Q3 / OCR H470/01 Section C).
- Meet top-band descriptors (AQA Level 5; OCR 'confident' band) within exam timing: roughly 25 minutes per AQA text question and 20 for the comparison.
📋 Key Definitions and Core Concepts
Mode: The channel through which language is produced and received (spoken, written, electronic); it constrains conventions of both production and interpretation.
Register: Variety according to use, analysed as field (subject matter), tenor (relationship between participants) and mode (channel) - Halliday & Hasan (1976).
Genre: A text type with shared conventions of form, structure and style (e.g. news article, phone-in); genres can be blended or hybrid, and deviations from convention are analytically significant.
Representation: The version of reality a text constructs about an issue, event, person, group or place, created through linguistic and multimodal choices rather than neutral description.
Multimodality: The co-occurrence of semiotic modes - words, image, colour, layout, typography, sound - within a single text (Kress & van Leeuwen, 2006).
Prosodics: Vocal effects beyond individual segmental sounds: intonation, stress, pitch, tempo and volume; central to speech analysis alongside non-fluency features.
Turn-taking: The organisation of alternating speakers in conversation, managed through cues such as adjacency pairs, overlap, interruption and back-channelling.
Presupposition: A pragmatic assumption built into wording and treated as true without evidence ('yet another pointless delay' presumes earlier delays).
Attribution of agency: Who or what the grammar casts as the doer of an action; passives and nominalisation can conceal, transfer or share agency.
Synthetic personalisation: Fairclough's (1989) term for a mass-audience text addressing each member as an individual ('you'), simulating intimacy at scale.
📌 Key Points
AQA Paper 1 Section A = 70 marks: Q1 Text A 25, Q2 Text B 25, Q3 comparison 20; Section B (children's language development) = 30 marks.
OCR H470/01 = 80 marks: Language under the microscope (2 x 10), Writing about a topical language issue (24), Comparing and contrasting texts (36).
Register = field + tenor + mode (Halliday & Hasan 1976): evidence each element with a linguistic choice.
A representation statement is a thesis: every paragraph must advance it - description without representation caps the mark.
Comparison must be integrated: point of comparison, then Text A evidence, then Text B evidence, then an explicit link.
AQA pairs texts from different modes and eras (June 2023: BBC News online, 2019 + The Bystander, 1928) - mode and context are the fastest AO3 route.
Modes: spoken, written and electronic
Every unseen text sits on a mode continuum, and AQA deliberately pairs texts from different modes and eras - June 2023 gave a BBC News online article (2019) alongside a magazine article from The Bystander (1928) - so mode is often the quickest route into AO3 context marks. OCR tests the same knowledge through Language under the microscope (H470/01 Section A) and Comparing and contrasting texts (Section C). The AQA specification frames the language levels as the systematic methods underpinning all analysis, so mode determines which levels carry the weight: prosodics for speech, graphology for print and electronic layout, discourse structure everywhere.
| Mode | Typical features | Exam anchors |
|---|---|---|
| Spoken | Spontaneous planning; prosodics (intonation, stress, tempo, volume); non-fluency (fillers, false starts, repair); turn-taking (adjacency pairs, overlap, back-channels); deixis tied to shared context. | Transcripts and phone-ins; AQA's transcription key marks micropauses (.) and timed pauses (2.0) - analyse their placement, don't just name them. |
| Written | Planned and edited; permanent; more complex and often more standard syntax; graphological conventions (headlines, columns, paragraphs); explicit cohesive devices. | Historical and print texts; permanence licenses long-range cohesion, dense modification and periodic sentence structures. |
| Electronic | Hybrid: speech-like informality with written permanence; multimodal (layout, image, hyperlink, typography); interactive; searchable and archivable. | Online articles, email, social media; affordances (immediacy, scrolling, comments) shape lexis and discourse structure. |
- Mode is not a binary. Blended texts - scripted speech, live commentary, email, social media - show spoken-like features in a written channel. AQA texts often exemplify this, so evaluate the blend rather than asserting 'spoken = informal, written = formal' as a law; the historical magazine and the online article in June 2023 both traded on carefully planned structure.
- Electronic affordances drive language: online articles use short paragraphs, bolded standfirsts and headline succinctness for scrolling readers; hyperlinks shift cohesion outside the sentence; audience comments make reception visible. Permanence and searchability also mean online texts are designed for future readers, not just the first audience - a headline is simultaneously a print standfirst and a search-engine string.
- Speech analysis needs prosodics and pragmatics, not just a word-by-word reading: adjacency pairs (question-answer, greeting-greeting) organise exchanges; overlap can be supportive or intrusive; repair and reformulation show speakers monitoring their own meaning-making. Back-channels ('mm', 'yeah') signal listenership and do interactional work - Fishman (1983) showed minimal responses doing the gendered 'housework' of conversation (see the Language and Gender topic).
- Deixis anchors spoken language in shared time and place ('this', 'here', 'now'), whereas written texts must build their own context because writer and reader are separated - a core AO3 point when explaining why speech looks 'incomplete' next to prose.
Register, genre, audience and purpose
Register is Halliday & Hasan's (1976) triad: field (subject matter and its semantic fields), tenor (the relationship between addresser and addressee - formality, power, distance) and mode (the channel). Register analysis is the bridge between context (AO3) and language detail (AO1): each element should be evidenced with a specific linguistic choice, not asserted in the abstract.
- Genre: text types carry conventions - news articles (headline, standfirst, inverted-pyramid structure, reported speech), magazine features (direct reader address, lifestyle lexis), phone-ins (host framing, caller turns). Hybrid genres (advertorial, blog post, podcast script) mix conventions; a top answer spots the deviation from convention, not just compliance with it - a news article sliding into advertorial conventions ('why not try...') is significant because it signals a covert promotional purpose.
- Audience: distinguish primary from secondary audiences, and specialist from non-specialist - OCR's directed-writing task explicitly targets a 'reasonably well-educated, non-specialist audience', which constrains jargon density and syntactic complexity. Audience shapes pronoun choice, sentence length and assumed knowledge; Fairclough's (1989) synthetic personalisation ('you') simulates intimacy with mass audiences. A print article also serves a secondary online audience of searchers - headings and standfirsts work double duty for skimmers and search engines alike.
- Purpose: overt purposes (inform, entertain) often coexist with covert ones (persuade, promote, align). Representation claims usually hang on the covert purpose - a travel article that represents a place as 'unspoilt' is probably selling something.
- Deployment: open your answer with a two- or three-sentence GAP summary, then weave genre, audience and purpose through each analytical paragraph. A standalone paragraph announcing 'this text is aimed at teenagers' earns nothing on its own - the mark comes from showing how that audience is written into the language.
A worked tenor chain, hypothetical but methodologically exact: the same writer produces an email to a client ('I wondered whether you might have time to look at this') and a message to a colleague ('seen this?'). The tenor shift - professional distance to peer familiarity - drives the politeness hedge, the modal softening and the full clause in the first, and the ellipsis, dropped subject and imperative tone in the second. That is the standard of AO1/AO3 integration examiners reward: every register observation traceable to a linguistic choice, every choice explained by a contextual factor.
Building representations: the systematic method
Representation is the version of reality a text constructs about an issue, event, person or group. AQA Questions 01 and 02 (25 marks each) command analyse how, so the representation claim is your thesis and the language levels are your evidence engine.
- Read twice. First reading: identify what is represented (issue, person, group, place) and summarise genre, audience and purpose. Second reading: collect quotations against each level.
- Plan 4-6 paragraphs by representational angle, not by language level. Example: 'Text A represents the strike as chaotic' - so paragraph on lexis (a semantic field of disruption), grammar (active voice and transitive verbs casting strikers as agents of chaos), pragmatics (a presupposition that disruption was expected), discourse (an anecdote placed early for foregrounding), graphology (an image selecting disorderly framing).
- Anchor every point in the chain quotation, then terminology, then effect, then context. An effect claimed without a quotation is assertion; a quotation without effect is description.
- Conclude by returning to the representation thesis and the text's purpose - close the loop the introduction opened.
A micro-example of the chain in action, using a hypothetical headline: 'Flood chaos: thousands stranded'. The lexis ('chaos', 'stranded') builds a semantic field of helplessness; the grammar gives agency to the flood and withholds it from authorities (nobody in the headline does anything to help); the presupposition is that helplessness is the story's baseline; and the graphology - the colon, the bold weight, the front-page placement - gives the claim authority before a reader engages with any evidence. Four levels, one sentence, one representation claim: that density is what a 25-mark paragraph should look like.
- Level menu: lexis/semantics (connotation, loaded or emotive lexis, semantic fields, modification); grammar (sentence type and function, agency and transitivity, deontic and epistemic modality, nominalisation - 'the delay' hides who delayed); pragmatics (presupposition, implicature, synthetic personalisation); discourse (structure, foregrounding, narrative voice, cohesion); graphology and phonology - used only where purposeful.
- Spoken representation: prosodics carry attitude - stress foregrounding blame ('I never agreed to that'), tempo and volume signaling urgency, rising intonation on declaratives softening a statement into a question - and paralinguistic annotations in the transcription key (laughter, overlapping turns) show how participants co-construct the representation in real time.
- AO3 context types: production context (who owns, publishes or funds the text - e.g. a broadcaster's editorial policy), reception context (the primary and secondary audience), and the historical moment (1920s print culture vs 2019 online news). Select the two or three factors that actually explain the language; listing every possible context wastes marks.
- Band capture: AQA Level 5 rewards conceptualised, perceptive analysis with consistent application of methods and sophisticated expression; OCR's top band asks for confident use of linguistic concepts and methods. Description of features, however accurate, sits below it.
- Timing: roughly 25 minutes per AQA text question and about 20 for Q3 - plan for a maximum of six paragraphs and write in full analytical sentences from the first line.
Comparing two texts: method and mark capture
AQA Question 03 (20 marks) asks you to 'explore the similarities and differences in the ways that Text A and Text B use language'. OCR H470/01 Section C (36 marks) specifies the same skills in two bullets: explore connections and variations between the texts and consider how contextual factors contribute to the construction of meaning. Both boards penalise parallel text-by-text analysis: comparison must be integrated.
- Identify the shared representational focus - the texts usually represent the same or closely related subject matter from different positions, modes or eras.
- Generate 3-5 points of comparison spanning mode, register, audience and purpose and the language levels.
- Structure each paragraph around a point of comparison: claim, then evidence from Text A, then evidence from Text B, then an explicit link ('whereas', 'similarly', 'both', 'in contrast').
- Attribute each difference to contextual factors - mode (online vs print), era (contemporary vs historical), audience (mass vs niche). This is where the AO3 marks sit: differences explained by context, not merely observed.
- Do not recycle your Q1/Q2 material: Q3 rewards new comparison points - pick different features, or different angles on the same features.
- Compare mode-specific features only where both texts permit them: prosodics cannot be compared across two written texts, but discourse structure, register and audience design can be compared across any pair. If one text is spoken and one written, the strongest comparison often rests on what each mode cannot do that the other exploits.
- Similarities matter as much as differences: a shared convention between a 1928 magazine and a 2019 news article is usually significant - evidence of enduring genre norms or shared purpose across contexts.
- Comparative language should be pervasive: every paragraph needs at least one explicit connective or comparative construction, and a conceptualised opening sentence naming the similarity or difference being developed.
Timing and preparation differ by board and deserve separate rehearsal. The OCR Section C rubric (June 2023) budgets 15 minutes for reading and preparation and about 50 for writing - so build a preparation grid before writing a word: rows for each point of comparison, columns for Text B and Text C evidence. AQA gives no explicit prep time across Section A's 70 marks, so discipline the reading phase yourself: two texts, roughly 25 minutes per analytical question, and Q3 written from freshly generated comparison points rather than leftovers. In both cases the examiner's criterion is the same - a conceptualised overview naming the key similarity and difference, followed by sustained, integrated, contextually explained analysis.
💡 Worked Example
Exam-style: Analyse how Text A uses language to create meanings and representations. [25 marks] (AQA 7702/1, June 2023, Question 01 - Text A: a BBC News online article, March 2019)
Mark-scheme logic: Mark-scheme logic (AO1 + AO3, Level 5 requirements):<br><br><strong>AO1 - conceptualised, systematic analysis:</strong> an opening thesis naming what Text A represents and through which genre/audience/purpose; then 4-6 paragraphs organised by representational angle, each applying methods from at least two language levels - lexis/semantics (e.g. connotative or emotive lexis building a semantic field), grammar (agency, transitivity, modality, nominalisation showing who acts and who is acted upon), pragmatics (presupposition, synthetic personalisation addressing the reader as 'you'), discourse (headline/standfirst structure, foregrounding, anecdote placement) and graphology (image, typography, hyperlink placement for the scrolling online reader). Every claim must be anchored in a precise quotation with accurate terminology and fluent written expression.<br><br><strong>AO3 - relevant, woven context:</strong> the electronic mode of a 2019 online news article (immediacy, scannable short paragraphs, hyperlinks, searchable permanence, primary plus secondary search-driven audiences); news genre conventions (inverted pyramid, attributed sources); audience and purpose shaping choice. Context must be used to <em>explain</em> language, not listed.<br><br><strong>What the top band adds:</strong> perceptive interpretation of the representation (what the text makes salient and why), consistent close analysis rather than feature-spotting, and a conclusion that returns to the representation thesis. The same logic applies to Question 02 on Text B.<br><br>For Question 03 (20 marks, comparison): integrated paragraphs organised by points of similarity and difference - e.g. shared covert purpose despite different modes; bolder, shorter syntax in the online text; historical lexis and longer syntax in the 1928 magazine - with differences explicitly attributed to mode, era and audience, and comparative connectives throughout.
❓ Practice Questions & Mark Schemes
Giving careful consideration to the context of the text, identify and analyse patterns of lexical and semantic use in Text A. [10 marks] (OCR H470/01, Section A, Question 1(a) - June 2023 wording)
Answer: Method: read the whole text first, then work the lexis/semantics level systematically with context in mind throughout.<br><br>Model hits: (1) <strong>Semantic field:</strong> identify a dominant field and its contribution to the text's subject matter (field in the Halliday & Hasan sense). (2) <strong>Connotation and evaluation:</strong> quoted examples of connotative or loaded lexis, with the attitude each constructs. (3) <strong>Modification:</strong> adjective/noun pairing that intensifies or softens - e.g. evaluative adjectives foregrounding the writer's stance. (4) <strong>Register lexis:</strong> jargon or formality choices tied to tenor - who the writer assumes is reading. (5) <strong>Context integration:</strong> each pattern tied to the text's genre, audience or purpose - e.g. headline lexis compressed for scrolling readers, or formal lexis licensing authority. In a 10-mark answer, three or four well-developed patterns with precise quotations beat a list of eight named features.
Analyse how Text B uses language to create meanings and representations. [25 marks] (AQA 7702/1, June 2023, Question 02 - Text B: a magazine article from The Bystander, April 1928)
Answer: Mark-scheme logic (AO1 + AO3):<br><br><strong>Opening:</strong> thesis - what Text B represents (e.g. an event or social group of the 1920s) and via which genre, audience and purpose (a niche middle-class magazine readership; an overtly entertaining but covertly editorialising purpose).<br><br><strong>Body (by representational angle):</strong> (1) Lexis: period lexis and semantics - dated or elevated vocabulary, honorifics, semantic fields constructing the group or event. (2) Grammar: longer, subordinated sentences typical of planned written prose; agency - who is cast as responsible; modality expressing the writer's certainty or obligation. (3) Pragmatics: presuppositions carrying period assumptions (e.g. about class or gender roles - see the Language and Gender topic); reader address (synthetic personalisation or overt 'the reader' address). (4) Discourse: structure - feature-led openings and delayed headline gratification vs online news; anecdote and narrative voice positioning the group represented. (5) Graphology: layout conventions of 1928 print (columns, captioned images, small advertisements) reinforcing the publication's identity.<br><br><strong>AO3:</strong> historical context (1920s print culture, social attitudes), mode (permanent, edited print), audience (niche, literate, middle-class) - each used to explain a linguistic choice, never listed.<br><br><strong>Top band:</strong> perceptive claims about how the representation is constructed and why the period context produces it, sustained throughout with precise quotation.
Explore the similarities and differences in the ways these two texts use language: Text A is a company's email newsletter to customers; Text B is a transcript of the same company's customer-service live chat. [20 marks] (AQA 7702/1 Question 03 style)
Answer: Method: integrated comparison - each paragraph organised by a point of comparison, not by text.<br><br>Model hits: (1) <strong>Mode:</strong> both are electronically produced, but the email is written-mode (planned, permanent, sent to many) while the chat is speech-like (rapid turns, ellipsis, non-fluency) - the key similarity/difference pair to open with. (2) <strong>Tenor:</strong> both maintain professional politeness, but the email uses synthetic personalisation ('Dear valued customer', 'you') to simulate individual address, whereas the chat achieves genuine interpersonal address through personal pronouns and immediate response. (3) <strong>Structure:</strong> the email uses paragraphs, headings and hyperlink cohesion; the chat uses turn-taking, adjacency pairs and back-channels. (4) <strong>Shared purpose:</strong> both retain the brand voice - consistent evaluative lexis - showing that register choices persist across modes. (5) <strong>Context attribution:</strong> differences explained by affordances (permanence vs ephemerality, one-to-many vs one-to-one) rather than observed and dropped.<br><br>Comparative connectives throughout; a conceptualised overview; new comparison points rather than recycled single-text analysis - that is what separates Q3-level answers from parallel rewrites of Q1/Q2.
Explain how electronic mode shapes the language of online texts. Illustrate with examples from two different online genres. [10 marks] (OCR ExamBuilder-style synthesis question)
Answer: Model hits: (1) <strong>Affordances:</strong> electronic texts are produced for screens - short paragraphs, bolded standfirsts, headline succinctness and hyperlink cohesion serve scrolling, skimming readers. (2) <strong>Interactivity:</strong> audience comments, likes and shares make reception visible, so writers design language for engagement - second-person address, questions, direct evaluation. (3) <strong>Multimodality:</strong> image, layout and typography carry meaning alongside words (Kress & van Leeuwen, 2006) - e.g. a captioned photo anchoring a news claim. (4) <strong>Genre illustration 1 - online news:</strong> inverted pyramid compressed into standfirst plus bolded subheads; attribution phrases ('according to...') licensing claims. (5) <strong>Genre illustration 2 - social media post:</strong> hashtags aggregating topics, emoji as evaluative gloss, brevity economy. A 10-mark answer needs the explanatory framework first, then two compact illustrations - not two disconnected text analyses.
Analyse how language in this radio phone-in transcript creates meanings and representations. [25 marks] (AQA 7702/1 Section A style - spoken text)
Answer: Mark-scheme logic (AO1 + AO3):<br><br><strong>Opening:</strong> thesis - who/what is represented (e.g. the caller's regional identity, the issue they raise) and through which genre, audience and purpose (broadcast phone-in; mass regional audience; entertain plus inform).<br><br><strong>Body:</strong> (1) Prosodics: intonation, stress and tempo marking attitude - rising intonation signaling tentativeness or challenge; analyse the transcription key's micropauses (.) and timed pauses (2.0) where they carry meaning (hesitation before a controversial claim). (2) Turn-taking: the host's framing questions, interruptions and overlap - who controls the topic (link to the Language and Power topic); adjacency pairs and back-channels building rapport. (3) Lexis/semantics: dialect or non-standard forms representing regional identity; evaluative lexis constructing the issue. (4) Pragmatics: implicature and hedging; deixis tied to the shared broadcast context. (5) Discourse: how the call is structured - opening framing, anecdote, host's closing summary - and what that foregrounds.<br><br><strong>AO3:</strong> the spoken, broadcast mode (ephemeral, co-constructed, multi-party), the regional audience, and the blend of entertainment and information driving choices - woven through, not listed.<br><br><strong>Top band:</strong> perceptive analysis of how identity and issue are co-constructed in real time, with prosodic and interactional detail anchored to transcription evidence.