Skip to main content

Framing & Political Discourse

What this framework does

Analyzes how political texts construct persuasive realities by activating semantic frames, positioning actors, and tapping into moral systems. The analytical prompt operationalizes Lakoff's frame semantics and Entman's framing functions to reveal the cognitive architecture of political persuasion.

Political discourse is often about constructing the terms of debate. When a politician frames taxation as "relief" rather than "revenue," or immigration as "invasion" rather than "migration," they're activating entire knowledge structures that determine what questions get asked, who's positioned as hero or villain, and what solutions become thinkable or unthinkable.

The analytical approach

Each analysis works through four discrete tasks:

  1. Dominant Frame Identification — names 3–5 major frames operating in the text, maps their FrameNet semantic roles (who is Agent, Patient, Villain, Beneficiary), and applies Entman's four functions to each: how does the text define the problem, assign blame, make moral judgments, and propose solutions?
  2. Source-Target Mapping — selects the most influential conceptual metaphors and unpacks their structure: what does the metaphor make seem natural, and what does it conceal or distort?
  3. Agenda-Setting & Frame Competition — examines which questions the text puts “on the table” and which it systematically excludes, how bridging language redirects attention between issues, and how dominant frames contest or delegitimize alternatives
  4. Contrastive Framing — pairs each dominant frame with a named alternative to show that frames are choices, not neutral descriptions, and to surface divergent policy implications

The analysis concludes with cross-cutting critical observations (frame consistency, agency distribution, moral economy) and a six-paragraph rhetorical synthesis.

About the outputs

Each analysis is presented in two forms:

The written analysis is a structured document working through all four tasks, with exemplar quotes, semantic frame element tables, bridging language tables, and a full rhetorical conclusion.

The audit dashboard visualizes key patterns across the analysis: frame family distribution, Entman function coverage, the contrastive pairs viewer, and an interactive panel for browsing the critical observations and conclusion paragraphs.

Provides a forensic breakdown of how those metaphors work, what they make salient, what they conceal, and how they activate pre-existing moral frameworks in the audience.

Extended processing summaries​

Some outputs include an "Extended Processing Summary" section representing the model's intermediate reasoning before producing the final analysis. These are included selectively when diagnostically interesting. Basically, they are computational artifacts useful for assessing prompt effectiveness, and don’t suggest evidence of cognition. See the How this works page for why the first-person framing of these summaries is itself a framing choice worth examining.

The experiment

Can a model be taught to perform the kind of frame analysis that political communication scholars do, identifying what cognitive work a text performs, what worldviews it activates, and whose interests it serves?

Let me know what you think,Troy

Discourse Depot © 2026 by TD is licensed under CC BY-NC-SA 4.0