You are a fly on the wall in a café where three historians are arguing about the same historical question — but they’re talking past each other because they fundamentally disagree about what history is and what it’s for.
Your job: use AI to write that conversation. Then evaluate whether AI actually understands the difference between a Marxist historian and a postcolonial one, or whether it’s just producing labels.
The dialogue itself doesn’t really matter. What matters is the process of getting AI to represent genuinely distinct intellectual positions — and your ability to recognize when it fails.
Good choices are questions where the answer depends on what you think matters most about history — questions about interpretation, not facts. For example:
Avoid questions with obvious answers. The whole point is that three reasonable historians can reach different conclusions from the same evidence.
Think about the historians and approaches you’ve read this semester. Pick three that would genuinely interpret your question differently. Some options (you’re not limited to these):
| Approach | What they care about | Course connection |
|---|---|---|
| Marxist / economic structuralist | Class struggle, modes of production, who controls resources | Marx, Communist Manifesto + 18th Brumaire |
| Nationalist historian | How the event shaped or expressed national identity | Hobsbawm, Anderson |
| Social / “history from below” | What ordinary people experienced and did | Thompson, Rowbotham |
| Cultural historian | Rituals, symbols, mentalities, everyday life as evidence | Geertz, Davis |
| Postcolonial critic | Who benefits from this version of the story; what’s silenced | Said, Trouillot, Guha |
| Great man / event-driven | Key individuals and decisions were decisive | The “old-fashioned” approach that everyone else argues against |
| Environmental / structural | Geography, climate, long-term forces | Braudel, Cronon |
Before you prompt anything, write a short description of each character. Include:
If you can’t answer these questions yourself, AI will produce mush. The quality of the dialogue depends entirely on the quality of your character descriptions.
Give AI the full context before you paste in your characters — the more you frame the task, the better the result.
Prompt:
I’m writing a dialogue between three historians who disagree about [YOUR QUESTION]. Each represents a different historiographical approach. The goal is to show how different methodological assumptions lead to different interpretations of the same evidence. Here are my three characters:
[PASTE YOUR CHARACTER DESCRIPTIONS]
Write a 1000-word conversation in which all three historians debate this question at a café table. Each character should argue from their specific methodological position — not just assert conclusions, but challenge the assumptions behind the other characters’ arguments. Make them argue, not just take turns speaking.
This is the most important part. Read the dialogue AI produced and ask yourself whether it understood what you asked, or whether it just played dress-up with labels.
Questions to guide your evaluation:
Write 3–5 sentences evaluating the dialogue. Be specific: quote a line that seems like a genuine representation of an approach, and quote a line that seems like a caricature. Explain why you see each that way.
Use your evaluation to fix what’s too thin or stereotypical. Some prompts that help:
Make [character]’s argument more specifically grounded in [TYPE OF EVIDENCE — economic data / archival records / cultural practices / etc.]. She should directly challenge [other character]’s assumption that [X].
The [Marxist / postcolonial / cultural] historian is being too vague. Make their argument specifically about how [concrete mechanism] shaped the conditions that led to [your event].
Ask me three questions, one at a time, to help me sharpen what the [approach] historian’s core claim about this event would actually be.
Keep revising until each character has a position you could defend in class — something you could connect to a specific course reading.
I grade this primarily on your evaluation (Step 3) and the quality of your character descriptions (Step 1) — not on how “good” the AI dialogue turned out, although that puts me in a better mood for grading. A polished dialogue you accepted uncritically is worth less than a mediocre one with a sharp, specific evaluation.
What matters most:
What matters less: whether the final dialogue is beautifully written, whether you picked “interesting” historians, or length beyond the minimum.
Post both to the discussion board:
Before class, read two other students’ dialogues and come ready to say which historian in their conversation made the most convincing argument, and why.
And make sure the formatting is readable. One giant block of unformatted text earns a zero.