Week 2 — Writing Studio / Workshop · "Say It, Then Judge It"

Course: English Composition (ENGL 1A) · Silver Oak University (fictional sample) · Prof. Lindgren
Objective: Objective 2 — read critically; summarize accurately and respond analytically · SLO A (compose reasoned, audience-aware prose)
Worth 50 points · Writing Studios group = 15% of the grade · Studio 2
Format: a hands-on summary + response workshop — you'll write one accurate summary paragraph and one separate response paragraph on a real text, review them against a checklist, get a chatbot's coaching, and then catch the chatbot's mistakes when it summarizes the same text.

This is the course's signature weekly component. Every instructional week has one Writing Studio — a short, practical workshop on the week's craft move. All studio resources are links to external sites; there is nothing to buy or download. The habit every studio builds: draft → review → get feedback → judge the feedback.
The text: Chimamanda Ngozi Adichie, "The Danger of a Single Story" (TED, transcript at ted.com). Read or watch it before you start. 🔗 https://www.ted.com/talks/chimamanda_ngozi_adichie_the_danger_of_a_single_story/transcript

---

Part 1 — The Big Picture

This week you learned the two jobs of a critical reader — summarize a text fairly (neutral, comprehensive, your own words) and respond to it with reasons — and that the whole skill is keeping them apart. This studio makes both real: you'll write one accurate summary paragraph and one separate response paragraph on Adichie's talk, then drill the boundary between them. This is the exact move under every analysis and argument paper you'll write this term — so getting it clean here pays off all semester.

Read the text first (~19 min to watch, ~15 to read): "The Danger of a Single Story," Chimamanda Ngozi Adichie 🔗 https://www.ted.com/talks/chimamanda_ngozi_adichie_the_danger_of_a_single_story/transcript . Read it with a pen: underline the sentence closest to her main point, bracket two or three examples, jot one question in the margin.

---

Part 2 — The Drafting Exercise (write this)

Write two short paragraphs on Adichie's talk. Keep them clearly separate (label them).

1. Paragraph 1 — an accurate SUMMARY (~80–120 words). Say what the talk says: her central claim + her major support. Neutral (no opinion), comprehensive (not one cherry-picked example), in your own words (no copied sentences — you don't need to quote her at all).
2. Paragraph 2 — an analytical RESPONSE (~120–160 words). Say what you think: do you find her argument convincing? Agree, disagree, or complicate it, with reasons that point to her claim, evidence, or reasoning. Not "I liked it" — a claim about the talk.

Write both now, in a word processor. Don't polish yet — that comes after the review.

---

Part 3 — Claim & Support Map (fill this in)

Before you review, map the talk so your summary is comprehensive and your response is grounded. Fill every cell in your own words (no quotations needed):

Question	Your answer (paraphrase)
What is Adichie's central claim (the one point she most wants you to accept)?	______
Support #1 she uses for it (an example, story, or reason)	______
Support #2	______
Support #3 (if any)	______
Which support do you find strongest, and why?	______
One question or objection you'd raise about the argument	______

The point: the top rows feed your summary; the bottom two feed your response. Filling this in is your invention step.

---

Part 4 — Self-Review & Peer-Review (apply the checklist)

Run both paragraphs through this checklist — first on your own draft, then trade with a classmate (or reread as if you were Adichie herself for the summary, and a skeptic for the response). Mark ✓ or ✗ and jot one fix:

Check	Summary	Response
No opinion leaked into the summary (no powerful, moving, sadly, weak — judgment lives only in the response)	☐	—
Comprehensive — the summary names the claim and major support, not one detail	☐	—
Own words — no copied/uncited sentences from the talk	☐	☐
The response takes a clear position (agree / disagree / complicate), not "I liked it"	—	☐
The response gives reasons tied to the claim, evidence, or reasoning	—	☐
The two paragraphs don't bleed into each other (summary reports; response judges)	☐	☐

Then revise the weaker paragraph based on what the checklist surfaced. The most common fix: an opinion word leaked into the summary (move it to the response), or the response is really a recap (add a reason). Keep both the before and the after — this is the "revise, don't just edit" move from Week 1.

---

Part 5 — Writing-Coach Moment (required — the BYOAI step)

Bring in your approved chatbot (Gemini, Claude, or ChatGPT) as a coach, not an author.

1. Paste your summary paragraph and ask: "You are my writing coach. Is anything in this 'summary' actually my opinion rather than what Adichie says? Point to specific words. Is it comprehensive — does it cover her main claim and major support? Do NOT rewrite it for me, and do NOT add any quotations."
2. Read its feedback and decide what to act on. Make one improvement in your own words — most often, cutting a judgment word that belongs in the response.

The coach is a mirror, not a ghostwriter. You're using it to see whether your summary is truly neutral — then you make the change.

---

Part 6 — AI-Critique Moment (required — catch the tool's mistakes)

Now flip roles and be the editor who judges the tool — against the real transcript.

1. Ask the same chatbot: "Summarize Adichie's 'The Danger of a Single Story' in about 4 sentences, and include one short direct quotation from it."
2. Open the real transcript (linked above) and audit the AI's output for the two classic failures:
   - Opinion-leak — it editorializes ("In this powerful, moving talk, Adichie brilliantly shows…"). Those judgment words make it a response, not a summary. Catch them.
   - Fabricated quotation / detail — check the "quotation" word for word against the transcript. Chatbots routinely invent plausible lines, mis-word real ones, or attribute a claim to her she didn't make. If you can't find it verbatim, it's fabricated — that's the catch.
3. Write 3–4 sentences naming at least one opinion-leak and reporting whether the quotation was real (found verbatim), mis-worded, or invented. Quote the transcript exactly if you cite the correct version.

The habit all term: the tool drafts, you verify — every quotation, against the real text. This is the single most dangerous AI habit in writing: it will hand you a confident, perfectly formatted quotation that does not exist. Catching that is a core skill of this course, and you're building the reflex now, on a short talk, before it matters on a research paper. (A fabricated quotation or source is an integrity violation whether a human or an AI produced it.)

---

Part 7 — What to Submit

Submit a single document (or text entry) with: your summary paragraph and response paragraph (labeled); your completed claim & support map (Part 3); your checklist marks + the revised paragraph (Part 4); a one-line note on the coach feedback you acted on (Part 5); and your Part 6 AI-critique paragraph (opinion-leak caught + quotation verdict). Due Sunday, Sep 13, 11:59 p.m. (50 points).

---

Instructor answer key & model — REMOVE BEFORE PUBLISHING TO STUDENTS

Students write their own summary and response, so exact wording varies. The models below grade the fair representation and the reasoned evaluation, not specific words. Every model statement is a paraphrase of Adichie's ideas in the instructor's own words; no line is presented as a direct quotation of the talk. If a student quotes, the line must match the real transcript exactly (a summary needs none) — verify it.

Model — Summary paragraph (paraphrase, ~100 words):

Adichie argues that when we know only one story about a person, group, or place, that single story hardens into a stereotype and denies people their full, complex humanity. She builds the case from her own life: as a child she read mostly British and American books and so wrote stories full of foreign characters until she found African literature; she had imagined her family's houseboy, Fide, only as poor; her American college roommate assumed she could not speak English or use a stove; and she herself once carried a single story of Mexico absorbed from the news. She adds that power decides whose story is told, and that many stories can restore dignity.

Model — Response paragraph (paraphrase, ~150 words):

Adichie's argument is persuasive, and what makes it land is that her examples show the "single story" operating on educated, well-meaning people, not just on the careless — her roommate is kind and curious and still defaults to a flattened picture, which makes the problem feel ordinary rather than rare. Her decision to indict her own assumptions about Fide and about Mexico is also disarming: she isn't lecturing from above. That said, I'd complicate one thing. Personal narrative is powerful for showing that the problem is real and deeply felt, but it can't by itself establish how widespread it is; a skeptic could fairly say the talk persuades emotionally more than empirically. The argument would be even stronger paired with some broader evidence that single stories shape how whole audiences think — though the vivid, honest storytelling is arguably the right tool for her purpose and audience.

What the models show (the grading targets):
- Summary: neutral (no judgment words), comprehensive (claim + several supports), and in the student's own words — the author would recognize it as fair.
- Response: a clear position (here, "persuasive but complicate"), reasons tied to the text (the roommate example; the limits of anecdotal evidence), and a steel-manned counterpoint — not a rating.
- Claim & support map (Part 3): correctly separates her claim from her support, and the bottom rows (strongest support, an objection) feed the response.
- Revision (Part 4): full credit requires a substantive boundary fix — e.g., moving a leaked "powerful" out of the summary, or adding a real reason to a recap-y response — not just fixed typos.
- AI-critique (Part 6): full credit for naming a specific opinion-leak and a correct verdict on the AI's quotation (real / mis-worded / invented), verified against the transcript.

Grading rubric — 50 points

Criterion	Full	Partial	None
Summary paragraph — neutral, comprehensive, own words (12)	12	6–10	0–5
Response paragraph — clear position + reasons about the text (not a rating) (12)	12	6–10	0–5
Claim & support map — separates claim from support; feeds both paragraphs (8)	8	4–6	0–3
Self-/peer-review + revision — checklist applied and the weaker paragraph revised (boundary re-seen, not just edited) (12)	12	6–10	0–5
Coach + AI-critique — acted on real coach feedback and caught a specific AI failure (opinion-leak and/or a quotation verified against the transcript) (6)	6	3	0–2

Quality gate (self-checked) — citation-integrity + correct-conventions: PASS. The one real text in this studio (Adichie's "The Danger of a Single Story") is used factually and by link; the link was verified to resolve to the TED page with a free transcript. No quotation from the talk appears in this studio or its key — every model statement is a paraphrase in the instructor's own words, so there is nothing to mis-quote or misattribute, and the AI-critique step explicitly trains students to verify any AI-supplied quotation against the real transcript. The summary/response distinction the rubric rewards matches the Week-2 lecture and quiz. No student paragraph is asserted as "the" answer — the key grades fair representation and reasoned response, not specific words.