Reading-to-Argument Evidence Ledger
Convert assigned sources into a claim-to-evidence ledger that separates source findings, student interpretation, counterevidence and unresolved uncertainty before drafting.
Use in AI
Choose an AI tool to copy the current Prompt with a short usage note. Nothing is sent to that tool.
Turn the supplied readings and assignment context into an evidence ledger from which the student can construct and defend an original argument.
## Study context
Assignment question and learning outcomes:
{{assignment_question_and_outcomes}}
Assigned sources with stable page, section or paragraph references:
{{assigned_sources}}
Student's provisional position and questions:
{{student_position_and_questions}}
Course rules, scope and citation style:
{{course_rules_and_citation_style}}
## Scholarship rules
- Distinguish the source author's claim, direct evidence, the student's interpretation, inference, assumption, counterevidence, conflict, missing information and unresolved uncertainty.
- Distinguish evidence generated or directly reported by a source from evidence that the source cites, summarizes or interprets. Do not attribute a secondary account to an underlying study unless the original study has been checked.
- Do not invent quotations, page numbers, findings, citations, author intent or consensus. If the source text is unavailable, request it rather than relying on a title or abstract alone.
- Quote only short passages necessary for analysis and keep source identifiers visible. Respect copyright, licensed access and course rules.
- Do not write the student's final submitted argument. Support evidence organization, challenge and planning while leaving thesis choice and wording with the student.
- Treat summaries produced by an AI tool as provisional until checked against the original source.
- Begin with an assistance-permission gate. If the supplied course rules do not clearly permit claim or argument support, stop and request clarification from the lecturer or course owner.
- By default, organize and challenge only claims, warrants, counterarguments and outline moves supplied by the learner. Do not originate a thesis, warrant, counterargument or assessed outline unless the course rules expressly permit that form of assistance.
## Ledger method
1. Clarify the assignment's decision boundary: what must be argued, evaluated or compared; what evidence types are acceptable; and what remains outside scope.
2. Inventory each source by author, date, source type, stable reference, research question or purpose, method where relevant, population or context and stated limitations.
3. Extract consequential source claims. Link each claim to a precise passage, table, figure or finding; record whether the supporting evidence is generated in that source or attributed to another source; and capture the source's stated uncertainty, limitations and scope.
4. Record the student's interpretation separately. Explain the reasoning that connects the source evidence to the proposed argument and identify any inferential leap.
5. Build a claim-support map. For each provisional student claim, identify supporting evidence, counterevidence, alternative interpretation, scope condition and evidence still needed.
6. Test the emerging argument for overgeneralization, cherry-picking, source dependence, circular reasoning, unresolved contradiction and claims that exceed the assignment evidence.
7. Test the strongest learner-supplied counterargument and specify what evidence would weaken or strengthen it. If counterargument generation is expressly permitted, label any generated option as a suggestion for the learner to assess.
8. Organize learner-supplied reasoning moves and evidence references, not submission-ready prose. Create new outline moves only when the supplied rules expressly allow it, and mark every judgment the student must make.
## Output contract: Reading-to-Argument Ledger
Return:
1. **Assignment boundary**: question, learning outcome, permitted assistance, evidence requirements and unresolved instructions.
2. **Source register**: source ID, provenance, method/context, authority for this question, limitations and access note.
3. **Evidence ledger**: source claim, exact reference, evidence, source limitation, student interpretation, interpretation status and verification need.
4. **Argument map**: provisional student claim, support, counterevidence, warrant, scope, confidence and gap.
5. **Counterargument test**: strongest alternative, evidence for and against, unresolved issue and fair response options.
6. **Drafting scaffold**: ordered reasoning move, evidence IDs, purpose, caution and student decision required. Do not supply polished submission text.
7. **Source-check queue**: unsupported or stale item, original source needed, check and priority.
8. **Student handoff**: claims ready to draft, claims to narrow or remove, decisions still owned by the student and next reading action.
## Verification and completion
Complete only when every consequential planned claim is traceable to supplied evidence or, where interpretive or evaluative, explicitly identified as the student's reasoned interpretation with its evidential basis and warrant; counterevidence is represented fairly; citations resolve to supplied material; and unsupported claims are removed or placed in the source-check queue.
For each completion condition, record the expected source or learner evidence, the actual supplied evidence, `Met`, `Not met`, or `Blocked` status, and the unresolved student action.
If readings, assignment rules or source locations are missing, return a partial ledger and request the minimum missing material. Refuse requests to fabricate citations, conceal AI assistance, certify originality or produce a final assessed submission.
Variables to Replace
Replace each listed value in the Prompt with information relevant to your task.
- assignment_question_and_outcomes
- assigned_sources
- student_position_and_questions
- course_rules_and_citation_style
How to Use This Prompt
Use Claude or another long-context AI tool. Paste the assignment brief and your own provisional position, then upload only readings you are permitted to process. Run the prompt before drafting. Check every quoted passage and citation against the original, follow your institution’s AI-use rules, and write the final argument in your own accountable voice.
Example Use Case
A student preparing an education-policy essay maps three assigned studies. The ledger shows that two sources support a narrow contextual claim while the third provides counterevidence, preventing the student from presenting a contested result as universal.
Was this useful?