HookThe experiment that proved most people cannot read
In the 1920s the Cambridge critic I.A. Richards ran an experiment that still haunts every unseen paper. He handed his students poems with the authors, titles and dates removed, and asked them to judge. The results, published as Practical Criticism in 1929, were humbling: highly educated readers confidently praised doggerel, dismissed real merit, misjudged tone completely, and dated poems by centuries. Stripped of the name on the spine, they could not read. Richards had proved that close reading is a trained skill, not a reflex — and that most people who assume they have it do not.
Paper 2 Section B is Richards's experiment with a mark scheme attached. You are handed a prose extract you have never seen, from inside your shared context — the war and its aftermath, or Britain from 1945 to the present — and asked to analyse its method, infer what its author is really doing, place it in its moment, and then connect it to the set texts you have studied. It is the most demanding stretch on either paper, because you cannot revise the passage; you can only revise the skill. What follows is that skill, broken into the order you should actually perform it under the clock.
Cold readThe ten-minute annotated read
AO2 under time pressure is a procedure, not a talent, and the procedure has an order. First pass: read the whole extract once for gist and shape — what kind of scene is this, who is it with, and where does it turn? Resist the urge to underline on this pass; you are learning the passage's overall movement so you do not build a clever argument on a misread tone. Second pass: annotate deliberately — mark the narrative voice and tense, one or two shifts in tone, any pattern that repeats, and the single anomaly that does not fit, because the anomaly is usually where the meaning lives.
Only then, before you write a word, force yourself to a provisional thesis about the passage in one sentence: ‘This extract sets official optimism against private dislocation and lets the gap carry the meaning.’ That sentence is your spine. Everything you write must feed it or refine it.
The failure this guards against is feature-hunting — sprinting through the extract collecting a simile here and an alliteration there with no reading to hold them. A list of devices is a Level 2 answer however many devices it contains. The worked example beside this block shows the procedure run on an original demonstration passage, so you can watch a cold read become a thesis rather than a checklist.
Read this original demonstration extract, written in the manner of a postwar English novel — nothing here is a set text: The new flats went up faster than anyone could name them, and Mrs Okafor watched from the ninth floor as the last of the terraces came down. Below, a council man in a yellow coat measured the air where her front step had been. She had carried her mother's chair up nine flights herself; the lift was promised for spring. From up here the city looked planned, hopeful, and very far away. She put the kettle on the new electric ring and waited for it to be home. A cold read in ten minutes might land on this: the extract sets civic optimism against private dislocation, and lets the gap between them do the work. The third-person narration stays close to Mrs Okafor — her ninth-floor viewpoint, her memory of ‘her mother's chair’, her waiting — so a sweeping public project is filtered through one displaced person. Watch who owns the verbs of authority: the flats ‘went up faster than anyone could name them’; a ‘council man … measured the air where her front step had been’. Naming and measuring belong to the state; carrying and waiting belong to her. The most telling choice is the last sentence — she ‘waited for it to be home’, which withholds the very thing the new flat was built to guarantee. The register is plain, almost administrative, and that flatness is the point: the prose borrows the tone of the planners in order to expose what planning leaves out. Nothing in that reading needed the author's name. It needed attention to voice, verb, and the withheld word ‘home’.
InferenceRead the author, not just the scene
AO1 here is an ‘informed personal response’ built on inference — reading the implicit meaning and the authorial stance the passage never states outright. Start by separating the narrator from the author. The narrator tells the scene; the author arranges it so that you feel a particular way about it. The gap between those two is where stance lives.
Ask three questions of any unseen. Where does the passage's sympathy sit — with whom are we made to feel? What is the text's attitude to its own material: admiring, elegiac, ironic, sceptical, complicit? And what is withheld — the word not said, the reaction not shown, the judgement left for you to supply? In the demonstration passage, sympathy sits squarely with Mrs Okafor, and the stance is quietly sceptical of the planners' optimism: the author never editorialises against the tower block, but by letting ‘home’ be the one thing the flat cannot yet provide, the passage makes its argument by omission.
This is the objective that lifts a mechanical analysis into a reading. Two candidates can spot the same close third-person narration; the one who infers why the author chose to trap us inside one bewildered viewpoint — to make a civic triumph feel like a private loss — is the one writing an informed personal response rather than a technical inventory.
ContextDate the voice from the inside
AO3 on an unseen is a detective problem: you get no byline, so you must place the period from internal evidence alone — diction, technology, social assumptions, the reference points a character takes for granted. This is a skill you can drill.
For the war-and-aftermath context, the tells are the vocabulary of rank and regiment, censorship and letters home, the euphemisms of ‘nerves’ and ‘shell shock’, mourning ritual, and the flat disillusioned irony of the early 1920s. For the 1945-to-present context, the markers are consumer goods and rationing (which ran to 1954), postwar rehousing and high-rise estates, Commonwealth migration, the pill and the permissive 1960s, decimalisation in 1971, and — a hard lower bound — computers and mobile phones, which place an extract firmly in recent decades. The demonstration passage's electric ring, ninth-floor flat, promised lift and migrant tenant plant it in the postwar rehousing of the 1950s and 1960s without a single date on the page.
Two warnings. Do not try to name the author — there are no marks for a lucky guess and real cost in a wrong one. Place the period and its concerns, not the person. And obey the same rule as every other section: context earns only when it is tied to a method or a meaning, so ‘the plain register belongs to an era of official optimism about planning’ earns, while ‘this was written after the war’ on its own does not.
ConnectionsThe pivot: from this passage to your shelf
AO4 — ‘explore connections across literary texts’ — is where Section B is won or lost, and where the discipline changes. Once you have read the unseen, you must bridge to the set texts you have studied within the same shared context, and you must connect the concern, not the plot. Nobody is impressed that two texts both feature a soldier or a tower block; the marks are for connected treatments — how each handles a shared preoccupation, and where they diverge.
Learn the connective sentence as a shape: ‘The extract's treatment of X connects to how [studied text] handles X, though where the extract does A, the studied text does B.’ That single move compares method and attitude at once, which is what AO4 actually credits. The demonstration passage's concern with displacement and belonging connects naturally to the modern period's writing about migration, class and home; an unseen aftermath passage about a fixed clock and an unspoken grief connects to the war poetry's handling of memory and silence — Owen and Rosenberg are the public tradition behind it.
Guard against the two failure modes. The first analyses the unseen beautifully and forgets to connect at all, which caps the script low however good the reading. The second abandons the unseen after a paragraph and retells the set texts, which caps it just as hard. The unseen extract is the spine of the whole answer; your studied texts enter as comparison, never as substitute, and you should be returning to the passage's actual words in every paragraph.
DebateHow the period gets represented — and argued over
AO5 asks you to weigh ‘different interpretations’, and on this paper the interpretations are of the period itself. Every era carries competing stories, and a strong script shows the unseen and the studied texts taking a side in that argument. The First World War is remembered through a disillusionment story of futility and waste — but revisionist historians such as Dan Todman (The Great War: Myth and Memory, 2005) and Brian Bond argue that this ‘myth’ hardened decades later, and that many combatants found the war meaningful; Sandra Gilbert reads the same period through gender, arguing it unsettled the balance between the sexes.
The modern context divides just as sharply. There is the optimistic ‘New Jerusalem’ story of the postwar settlement — the NHS, full employment, the planned city — and against it the story of everyone the settlement left out: the migrants later betrayed in the Windrush scandal, women, the deindustrialised North. Postcolonial reading after Edward Said and the postmodern distrust of grand narratives that Lyotard named in 1979 give you ready lenses for the exclusion story.
The move that earns is to name the competing readings and then show your texts complicating one. In the demonstration passage, the ‘planned, hopeful’ city is exactly the New Jerusalem story — and the withheld word ‘home’ is exactly the passage's quiet dissent from it. Naming both, and showing the text tilt between them, is adjudication; listing critics is not.
ThesisOne connective argument, not two essays stapled together
AO1 closes the loop by asking for a coherent, integrated argument, and the commonest structural failure on the whole A-level is to write half an essay on the unseen, then a separate half on the set texts, and staple the two together with a linking sentence. That produces two competent halves and one mediocre band, because the connective thinking the paper is built to test never actually happens.
A Level 5 script runs one thesis about the shared context and tests it across the unseen and the studied texts inside the same paragraphs. The integrated paragraph has a rhythm: a claim about the period → evidence from the unseen extract → a connective turn to a studied text → an adjudication that says which handles the concern more powerfully, and why. You are never away from the unseen for long, and you are never merely summarising a set text; the two are always in conversation.
State your thesis in the opening — ‘Both the extract and the period's writing present belonging as something the state promises and withholds’ — keep the unseen as your anchor throughout, and return to that thesis at the close, changed by the journey. Do that, and the connection stops being a paragraph you remembered to add and becomes the argument the whole answer is made of.
VocabularyKey terms the mark scheme pays for
TrapsMisconceptions that cost marks
ExamWhat examiners want
Section B is marked out of 25 across the same five levels of response, but its centre of gravity is AO4 — connection. A dazzling reading of the unseen that never connects, or connects only in a bolted-on final paragraph, is capped below the top bands; the highest level belongs to scripts in which the unseen extract and the studied texts argue with each other throughout, paragraph by paragraph, rather than in sequence.
Get the order of operations right under the clock. Spend the opening minutes annotating the passage and forming a one-sentence thesis before you write, because you cannot revise the extract — only the skill you bring to it. Keep the unseen as the spine: refer back to its actual words in every paragraph, and let the studied texts enter as comparison, never as replacement. The most common Level 2 script analyses the passage competently and then drifts into retelling its set texts; the most common Level 5 script never once stops holding the two side by side.
Quote the unseen precisely — it is in front of you — and quote studied texts from memory but exactly; an embedded, accurate three-word phrase outscores a vague paraphrase every time. Above all, read tone before technique: if you misjudge whether the passage is grieving, ironic or celebrating, every clever point built on top of that misreading is wrong — which is precisely the failure I.A. Richards documented in 1929, and precisely what this section exists to test.