Week 5 Writing Studio: Turn Telling Into a Scene, Then Audit the Machine

Part 1 — The Big Picture

This week you learned that a story needs a point, and that a reader feels that point through concrete sensory detail rather than named feelings. In this Writing Studio you practise the core move of narrative writing on a fresh paragraph about a first 5K race: you sort its sentences into telling, showing, narrative summary, and purple prose, then you write a scene of your own, from your essay draft or your own memory, that shows one moment in full. You review your scene against a checklist, ask a chatbot for targeted feedback without letting it rewrite you, and catch the mistakes a chatbot makes when it "improves" a story. The skill carries far beyond this course: every cover letter, incident report, and personal statement is stronger when one concrete moment does the work a string of adjectives cannot.

Time budget: Plan for about 90 minutes.

Part 2 — The Task

Guiding question: Can you turn a moment you lived through from a sentence that names a feeling into a short scene that lets a reader feel it, without adding adjectives or details that did not happen?

Workshop data: First-5K paragraph (illustrative, six sentences)

Source: An illustrative six-sentence paragraph written for this Writing Studio and attributed to no one; it is not a real student's work, and it is printed in full in the Part 4 marking table.

• Rows: 6 sentences, one per row of the Part 4 marking table (that table has 6 rows)
• Topic: the morning of a writer's first 5K race
• Form: a short personal narrative paragraph that mixes telling, narrative summary, purple prose, and one showing sentence

Part 3 — Materials & Procedure

Tools and materials

• A word processor; a citation tool or style guide used critically (the MLA Handbook, the Purdue OWL, or a citation generator whose output is checked against the MLA rules); and one approved chatbot used as a writing coach
• Your browser's find command (Ctrl+F on Windows, Command+F on a Mac) or a careful word-for-word comparison

Ways of working

• draft-and-revise-a-paragraph-against-a-rubric
• peer-review-with-a-checklist
• source-integration-and-citation-check

Procedure

1. Read the six-sentence 5K paragraph in the Part 4 marking table twice, once straight through and once with the comment tool or a pencil in hand. Check: You can say in one phrase what moment the paragraph is about.
2. Mark each sentence in the table as T (telling), S (showing), cut (purple prose), or summary (narrative summary), and write the word or phrase that gave it away in the last column. Tool: A word processor; a citation tool or style guide used critically (the MLA Handbook, the Purdue OWL, or a citation generator whose output is checked against the MLA rules); and one approved chatbot used as a writing coach. Check: Every row of the marking table has a mark and a reason.
3. Choose one moment from your Week 5 essay draft, or another moment you lived through, that you first wrote as telling. Write it as a scene of 5-8 sentences that reaches at least two senses, keeps every detail true to what happened, and names no feeling outright. Tool: A word processor; a citation tool or style guide used critically (the MLA Handbook, the Purdue OWL, or a citation generator whose output is checked against the MLA rules); and one approved chatbot used as a writing coach. Check: A scene paragraph in which a reader can point to at least two sensory details.
4. Review your scene with the checklist table in Part 4, or trade with a classmate and use the same checklist on each other's scene. Then revise and keep both versions. Tool: A word processor; a citation tool or style guide used critically (the MLA Handbook, the Purdue OWL, or a citation generator whose output is checked against the MLA rules); and one approved chatbot used as a writing coach. Check: The checklist table is filled in and at least one change is visible between the two versions.
5. Paste your revised scene into one approved chatbot with the prompt in Part 6, read its feedback, and act on one piece of it in your own words. Tool: A word processor; a citation tool or style guide used critically (the MLA Handbook, the Purdue OWL, or a citation generator whose output is checked against the MLA rules); and one approved chatbot used as a writing coach. Check: One change in your scene that you can trace to a specific piece of feedback.
6. Fill in the critique table in Part 4 for the chatbot's answer to the second half of the prompt, using the find command or a word-for-word comparison to check every quotation against the six printed sentences. Tool: Your browser's find command (Ctrl+F on Windows, Command+F on a Mac) or a careful word-for-word comparison. Check: Each failure you report names what the tool said and your correction.

Fallback route

When to use it: Use this route if you have no word processor or the comment tool does not work.

Copy the six sentences by hand or into any plain text editor, mark them on paper, write your scene and its revision by hand, and complete every table, every analysis question, the coach moment, and the critique step. Submit a photo or scan of the handwritten pages, or a typed transcript of them, for equal credit.

This route earns the same credit as the primary route.

Part 4 — The Scaffold

Mark each sentence of the illustrative 5K paragraph so that its telling, showing, summary, and purple prose become visible before you write your own scene.

Student scaffold table 1

Sentence	Text	Your mark (T, S, cut, or summary)	The word or phrase that gave it away
1	The morning of my first 5K race was a really big deal for me.
2	We got up early, ate breakfast, and drove to the park.
3	I was extremely nervous and excited at the starting line.
4	The magnificent, glorious crowd roared like a thunderous ocean of endless support.
5	At the second mile my shoelace came untied, and I could hear my own breathing over the music in my earbuds.
6	Finishing the race was amazing and taught me a lot about myself.

Review your own scene with this checklist before and after you revise, so each change has a reason.

Student scaffold table 2

Check	Yes or no	Your evidence from the scene
The scene names no feeling outright (no word such as nervous, happy, or scared).
The scene reaches at least two senses.
Every detail really happened; nothing was added to sound vivid.
No phrase piles on adjectives or grand comparisons.
The scene is spent on the moment that carries the point, not on the trip there.
A reader could say what the moment means without being told.

Record what the chatbot got wrong so that each correction is checked against the printed paragraph or your own draft, not against the tool's confidence.

Student scaffold table 3

What the chatbot said	Failure shape	Your correction and how you checked it

Part 5 — Analysis Questions

1. Which sentences of the 5K paragraph did you mark T, and which word in each one names a feeling or a judgment instead of showing it?
2. The paragraph has 6 sentences. After you mark the 3 telling sentences, the 1 purple sentence, and the 1 narrative-summary sentence, how many sentences are left that show? Name them by number.
3. Sentence 4 is full of adjectives. Explain why it still shows nothing a reader could picture, and rewrite it as one concrete detail a spectator at a race might really see or hear.
4. Choose one option and answer it in 3-5 sentences. (a) Compare your scene's first and revised versions: which change let a reader feel the moment, and what did you see, hear, or feel that you added? (b) Describe one place in your Week 5 essay where telling still does a useful job, and explain why you would keep it.
5. Sentence 2 compresses a whole morning into one sentence. Is that a flaw in this paragraph, or useful narrative summary? Say what the paragraph would lose if every sentence were a full scene.

Part 6 — AI-Critique Moment

Choose one equal-credit critique route. Route A: ask an approved AI tool to critique your completed work, then check its response against your own evidence. Route B is the no-AI equivalent printed below.

Approved tools: Claude, Gemini, ChatGPT, Grok, Microsoft Copilot.

Prompt to paste:

I've pasted my own revised scene paragraph above. Do two separate things. First, give me feedback on my paragraph: point to the one sentence where a reader would most want more sensory detail, and to any sentence that still names a feeling instead of showing it. Do not rewrite my paragraph. Second, here is a short illustrative paragraph about a first 5K race: "The morning of my first 5K race was a really big deal for me. We got up early, ate breakfast, and drove to the park. I was extremely nervous and excited at the starting line. The magnificent, glorious crowd roared like a thunderous ocean of endless support. At the second mile my shoelace came untied, and I could hear my own breathing over the music in my earbuds. Finishing the race was amazing and taught me a lot about myself." Improve it to make it more vivid, quote its strongest line, explain how the improved version works on a reader, name one published writing guide that recommends showing instead of telling with an MLA works-cited entry for it, and fix any grammar errors in my own paragraph.

Test the response for these likely failures:

• The feedback opens with "What a vivid, engaging paragraph!" and never points to a single sentence of yours or names a change.
• Although told not to, it rewrites your scene in a smooth, generic voice, for example turning your plain "my hands were cold" into "an icy chill crept through my trembling fingers".
• Its "strongest line" is a sentence the 5K paragraph never contains, or sentence 5 with words changed, such as "my shoelace snapped" for "my shoelace came untied".
• It names a writing guide, an author, or an article that supposedly recommends showing instead of telling, with a confident title and year you cannot find.
• It hands back a polished works-cited entry for that guide, or for the illustrative 5K paragraph, and presents it as correct.
• Its "grammar fix" to your paragraph changes what happened, for example turning "I almost fell" into "I fell", or changes a correct past tense into the present.
• Its explanation of how the paragraph works only retells the race in order (they woke up, drove, ran, finished) with no word about which details make a reader feel it.

Write 3-4 sentences that name one thing the tool did well and at least two failures you caught from the list above, each with your correction checked against the six printed sentences or your own draft.

Equivalent route (no AI required)

No AI account is required for this Writing Studio. This route assesses the same objective, Objective 4 (composing in the narrative mode with concrete sensory detail, a scene aimed at the point, and no purple prose), takes comparable effort of about the same 90 minutes, uses the same rubric, and earns the same credit. For the coach moment, read your scene aloud to a friend, a family member, or yourself, and answer the review checklist in Part 4 in writing as the listener would; act on one point in your own words. For the critique moment, critique this flawed feedback-and-revision, written for the course to contain mistakes and printed here in full: "What a vivid, engaging paragraph! I made it more powerful: An icy chill crept through my trembling fingers as destiny called my name at the starting line. The paragraph's strongest line is: 'At the second mile my shoelace snapped, and I could hear my heart pounding.' A well-known writing guide recommends exactly this technique. Works cited: show dont tell guide. writing experts. Grammar fix: At the second mile I fell. Analysis: The writer gets up early, drives to the park, runs the race, and finishes it." Find and correct at least four of its seven problems (hollow praise, voice-erasing rewrite, misworded quotation, invented source, invented works-cited entry, meaning-changing fix, and summary passed off as analysis), and write 3-4 sentences naming what you corrected and why. Submit that marked-up critique in place of the chatbot share link and report.

Part 7 — What to Submit

• The completed marking table for all six sentences
• Your first and revised scene, with the change visible
• The completed review checklist
• Written answers to all five analysis questions
• The chatbot critique table with your 3-4 sentence report and the conversation share link, or the equivalent route's marked-up critique of the printed flawed feedback

Deliverable:

A Writing Studio packet on the illustrative 5K paragraph and a scene of your own: the marked six-sentence table, a first and revised scene, the completed review checklist, answers to five analysis questions, and either a chatbot critique table with a short report and share link or, on the equivalent route, a marked-up critique of the printed flawed feedback. Submit in any accessible file type or as a link; a recorded response may be submitted as audio or video with a transcript, or as text.

Rubric

The rubric below shows how each submitted part is evaluated; bands are full, partial, or none.

Workshop evaluation criteria

Criterion	Full	Partial	None
Sentence marking and diagnosis	Every sentence carries a fitting mark with the word or phrase that gave it away; the three telling sentences, the purple sentence, the summary sentence, and the one showing sentence are all identified.	Most sentences are marked, but the purple sentence is called showing or a telling sentence is missed.	No marking, or marks with no reasons.
Scene revised from telling to showing	A 5-8 sentence scene from the student's own experience reaches at least two senses, names no feeling outright, keeps every detail true, and spends its detail on one moment.	The scene shows in places but still names a feeling, reaches only one sense, or slips into piled adjectives.	No scene, or a paragraph of named feelings and adjectives with no sensory detail.
Checklist review and visible revision	The checklist is completed with evidence and both versions show a change traced to it.	The checklist is completed but no change is visible, or a change is shown without the checklist.	No checklist and no revision.
Feedback acted on in your own words	One change is traced to a specific piece of chatbot feedback, or to the read-aloud reader's checklist answers on the equivalent route, and is written in the student's own words.	Feedback is reported but the change is unclear or copied from the reviewer.	No feedback step is shown.
AI slip caught and corrected	The critique, on either route with equal credit, names at least two failure shapes and corrects each against the printed paragraph or the student's own draft, with a 3-4 sentence report.	One failure is named and corrected, or failures are named without corrections.	No critique, or the tool's output is pasted without examination.

Quality Gate

• Every quotation from the 5K paragraph matches one of its six printed sentences word for word.
• The scene names no feeling outright and reaches at least two senses.
• Every detail in the scene really happened; none was added by a chatbot or invented to sound vivid.
• No source, study, quotation, or works-cited entry is invented anywhere in the submission, and none supplied by a chatbot is kept without being found.
• The revision is visible: both versions are submitted, with the change traced to the checklist or to feedback.
• The critique names at least two failures and corrects each against the printed paragraph or the student's own draft.