4.2 Discourse and Argument

What discourse and argument is asking

On the combined AQL paper, AL remains about 75 multiple-choice questions within roughly 125 AQL items, no dictionary, on passages that mirror tertiary reading. AL is reported separately from QL; MAT is a different NBT domain. The official AL Relations subdomain includes how parts of a text relate across sentences and paragraphs. This OpenExamPrep section is independent teaching of discourse and argument: how paragraphs develop, how introductions and conclusions frame a case, how claims relate to evidence, and how superordinate ideas sit above subordinate ones. Name the job of a stretch: does it introduce, support, limit, or conclude? Does this sentence sit above the others, or under them?

Superordinate versus subordinate

A superordinate idea is the controlling claim of a stretch: the statement the rest of the stretch exists to develop. Subordinate ideas are reasons, examples, limits, and data under that claim. Do not confuse first sentence or a vivid percentage with superordinate. A deletion test: if you can drop a sentence and the stretch still has the same point, that sentence was subordinate.

Worked three-paragraph sequence (original OpenExamPrep teaching)

Paragraph A (introduction). Several faculties now pilot 24-hour library opening in exam weeks. Administrators defend overnight hours as access for students who work during the day, yet librarians ask whether the extra shift changes who uses the library or only when they use it. The practical question is whether overnight opening is a timing intervention, a resource intervention, or both.

Paragraph B (evidence). Gate counts at one campus rose 9 percent between 22:00 and 02:00 in the first exam week of the pilot; daytime counts fell 3 percent. A short survey of 120 overnight users found that 58 percent had a paid job that ended before 17:00. Laptop-loan records showed no similar overnight spike in device borrowing.

Paragraph C (conclusion). A 24-hour policy therefore mainly shifts the clock for already-enrolled, time-poor students; it is not, on this evidence, a device-access programme. Faculties that want overnight opening should staff the desk, not stock extra laptops from this dataset. The controlling close is that timing access is not the same intervention as resource access.

Paragraph-to-paragraph development

Development is not a list of new topics. Paragraph B does not change subject to “surveys in general” or “laptop design.” It narrows A's question with measurements: a clock-shift in the gate counts, a working-student majority among overnight users, and a non-finding on device loans. Paragraph C does not invent a new controversy about building security; it answers A's question using B's contrast. The rail is A's practical question: B feeds it; C returns an answer on that same rail.

Introductions and conclusions

An introduction typically frames: it names the issue, sometimes concedes a popular view, and ends with a question, thesis, or aim. Paragraph A concedes the access defence, then ends on a classification question (timing versus resource). That question is superordinate for the extract as a whole.

A conclusion typically does not add a new dataset. It answers, qualifies, or recommends on the basis of what was already argued. Paragraph C recommends staffing rather than extra laptops because B already showed no loan spike. If C had opened with an unpublished second campus's figures, it would be smuggling new evidence into the closing slot—and an item could fairly say the final paragraph fails as a conclusion.

Claims and evidence

A claim is a statement the writer wants accepted. Evidence is the material offered to make that statement believable: a count, a contrast, a quoted finding, a defined case.

In paragraph B, “gate counts rose 9 percent between 22:00 and 02:00” is evidence. “Timing access is not the same intervention as resource access” (paragraph C) is a claim that uses that evidence plus the laptop non-finding. The 58 percent figure is also evidence; it supports the idea that overnight users are time-poor workers, which in turn supports reading the pilot as a clock shift.

AL traps on claim and evidence:

• Treating a number as the main claim of the whole extract. Numbers rarely sit at the top of the hierarchy unless the passage is a results paragraph.
• Treating a recommendation as if it were a finding. Should and need mark claims about what to do, not measurements.
• Missing the scope. Evidence about one campus does not automatically license a claim about all campuses; paragraph C stays inside “on this evidence.”

Inside paragraph B, the implied superordinate is a patterned change in when students entered, not a device-loan surge. The +9 percent, daytime drop, 58 percent job share, and laptop non-finding are subordinate. If a stem asks which idea is superordinate, pick the statement the other sentences exist to specify. “Should staff the desk” is a policy claim under the still-higher claim that timing and resource access are different interventions.

Exam traps on discourse and argument

• Topic tracking instead of development. “Paragraph B is about surveys” names a topic. The discourse job is: B supplies contrasting evidence for A's question.
• Last sentence as always the main idea. Conclusions often are superordinate for the extract, but body paragraphs often put the main idea early and then illustrate.
• Evidence upgraded to thesis. A 9 percent overnight rise is not, by itself, “libraries must stay open every night of the year.”
• Introduction as decoration. The intro's question is the rail the later paragraphs run on. If C ignores A's question, the argument is incoherent.

Independent OpenExamPrep method: label each paragraph I / E / C (introduce, evidence, conclude), then draw a superordinate bracket over the extract's controlling question. Options that break the bracket are wrong. You still have no dictionary; the argument's structure is on the page.