Why it matters

Every statement about a contested issue arrives wearing a frame — a particular way of carving up the situation that quietly decides what counts as relevant, who the actors are, and what the normal, unmarked state of affairs is. The frame does this work before any explicit claim is made, and it usually does more persuasive work than the claims themselves, because the reader argues inside it without noticing they’ve accepted it. A frame audit is the discipline of pulling the operative frame up into the light: naming the way of seeing the artifact treats as natural, saying what that framing foregrounds and what it hides, and trying an alternative to show the first frame was a choice.

For example: a tax cut described as “tax relief” has already won half the argument. “Relief” presupposes an affliction, an afflicted party who deserves rescue, and a reliever who is the hero — so anyone opposing the cut is, in the logic of the frame, defending the affliction. None of that was argued; it rode in on a single word. Call the same policy “a revenue cut” or “a transfer to high earners” and the actors, the normal state, and the villain all change — same policy, different frame, different conclusion that feels obvious. The audit’s job is to make that swap visible so the reader can see the frame as one option rather than as the shape of reality.

  • What it reveals. The operative frame an artifact runs on — the way of dividing up the issue that fixes what’s relevant, who acts, and what’s treated as normal — together with the metaphors and word choices that install it and the questions it quietly rules out of bounds.
  • How it changes the read. You stop asking “is this claim true?” and start asking “what does this framing make me take for granted before I even get to the claims, and what would the same facts look like framed another way?”
  • When to foreground it. A single persuasive artifact — an op-ed, a press release, a headline, a policy memo — where the wording feels natural and the conclusion feels obvious, and you suspect the framing is doing more work than the stated argument.
  • What you’d miss without it. That the artifact’s force often lives in its vocabulary and its silences, not its logic; check only the explicit reasoning and a perfectly valid argument can still smuggle its conclusion in through the frame you never questioned.
  • Where it misleads. Pushed too hard it slides from surfacing a frame into attacking it — and an audit that argues against the frame loses the trust of readers who hold it; the discipline is to name what the frame does and what it costs without ruling on whether it’s right.

How it works

The cleanest way to see what a frame is doing is to watch one word change everything around it. The linguist George Lakoff built much of his career on a single observation: abstract political ideas are almost always understood through concrete metaphors, and the metaphor we reach for silently fixes how we reason about the idea. Take “tax relief.” Relief is something you give to a sufferer; it implies an affliction, a victim who deserves rescue, and a hero who provides it — and, by quiet implication, a villain who would block the rescue or impose the affliction in the first place. Once “tax relief” is the phrase in the air, anyone arguing to keep the tax has been cast, without a word of argument, as the villain prolonging an affliction. The frame did that. The same policy under “a revenue cut” carries none of it; there’s no sufferer, no hero, just an accounting change with trade-offs. Lakoff’s point, worked out across Metaphors We Live By (with Mark Johnson, 1980) and Don’t Think of an Elephant! (2004), is that you cannot win an argument by negating the other side’s frame — “I’m against tax relief” still activates the relief frame and the victim it presupposes. You have to surface the frame and offer a different one.

That is the first thing a frame audit does: it names the metaphor that’s carrying the load. “Death tax” frames an estate tax as a penalty inflicted on the grieving — death as the trigger, the bereaved as the victim. “Estate tax” frames the very same levy as a tax on accumulated wealth. “Collateral damage” frames killed civilians as an incidental side effect of a properly aimed action; “civilian deaths” frames them as the thing that happened. None of these pairs differ in the facts. They differ in what the words foreground and what they push into the background — and the audit’s job is to quote the operative metaphor and spell out what reasoning it makes easy and what reasoning it makes hard.

But framing is more than metaphor, and here a second tradition comes in. The sociologist Erving Goffman, in Frame Analysis (1974), asked a more basic question: when something happens, what kind of thing do we take it to be? A raised fist is a threat, a salute, a stage gesture, or a joke depending on the frame you read it through — and the frame, not the fist, decides. Goffman showed that every situation is understood through a primary framework that answers “what is going on here?”, and that the same events can be re-keyed — played as a contest, a ceremony, a rehearsal, a technical exercise — each keying changing what the events mean and what response they call for. A frame audit borrows this to ask what the artifact treats as the baseline normal: is a labor dispute being framed as a negotiation (two parties bargaining over terms) or as an attack (one side assaulting the other)? The keying decides who is acting and who is merely reacting, and the artifact almost never states which keying it has chosen — it just proceeds as if its choice were the only sane reading.

The third tradition makes the audit systematic. The communications scholar Robert Entman, in a 1993 paper that gave the field its working definition, argued that to frame is “to select some aspects of a perceived reality and make them more salient” so as to promote a particular problem definition, causal story, moral judgment, and recommended treatment. That four-part anatomy is the audit’s checklist: for the operative frame, what is named as the problem, what is blamed as the cause, who is judged in the moral accounting, and what is prescribed as the fix? Run an artifact through those four and its persuasive engine usually stands exposed — and you often find the whole force concentrated in one slot, a moral verdict riding in a single loaded adjective, with the other three barely argued.

The remaining move is the one readers find most clarifying, and it follows straight from Entman’s word select: to make some things salient is to leave others dark. So the audit keeps two columns — what the artifact includes and emphasizes, and what it excludes or downplays — because frames work as much by silence as by speech. The piece that frames a factory closure entirely as a story about quarterly costs has, by that framing, made the workers’ livelihoods, the town’s tax base, and the executives’ compensation simply not part of the discussion — not refuted, just unmentioned, which is more effective. Then comes the test that proves the whole thing: the audit sketches at least one counterframe — the same facts organized under a rival frame, with Entman’s four questions answered again — to demonstrate that the operative frame was a choice and not the shape of the world. Crucially, the audit stops there. It surfaces the frame and names its costs; it does not declare the frame wrong. That restraint is the point: a frame audit that stayed neutral can be trusted by a reader who likes the frame and a reader who hates it alike, and that shared trust is exactly what makes the surfacing useful.

Framework & implementation

Output contract

The deliverable is a fixed set of sections, so the audit is auditable rather than an impression: Operative Frames Named (each in alternative-comparable vocabulary, with where it surfaces in the artifact), Lakoff Metaphor Inventory (each metaphor quoted, with the reasoning it licenses and the reasoning it makes harder), Goffman Primary Framework and Keyings (the presupposed baseline situation, any keying applied, any fabrication flag), Entman Four Functions per Frame (problem definition, causal interpretation, moral evaluation, treatment recommendation — each with quoted evidence), Selection and Salience Inventory (two columns: included-and-emphasized, excluded-or-downplayed), an optional Presupposition and Nominalisation Audit (word- and grammar-level mechanisms, each quoted with the framing work it does), a Counterframe (the same issue under a rival frame, with the four functions briefly re-populated), and Confidence per Finding (how firmly each rests on quoted text versus structural inference). When the real question turns out to be frame-rejection rather than frame-surfacing, the contract flags the sideways route rather than quietly switching stance.

Origin and evidence

The audit braids three research traditions. George Lakoff’s conceptual-metaphor theory — Metaphors We Live By, with Mark Johnson (1980), and the applied political treatment in Don’t Think of an Elephant! (2004) — supplies the core insight that abstract reasoning runs on concrete metaphor and that the metaphor silently fixes the inference. Erving Goffman’s Frame Analysis (1974) supplies the sociological layer: the primary framework that answers “what is going on here?” and the keyings that re-cast the same events as contest, ceremony, or rehearsal. Robert Entman’s “Framing: Toward Clarification of a Fractured Paradigm” (Journal of Communication, 1993) supplies the operational anatomy used throughout media studies — framing as selection and salience in the service of a problem definition, a causal story, a moral evaluation, and a treatment recommendation. The optional word-and-grammar layer draws on critical discourse analysis (Fairclough, van Dijk, Wodak) for presupposition, nominalization, and agentless-passive mechanisms.

Applications and common uses

  • Op-ed and editorial analysis. The native use: surface the frame an opinion piece runs on, the metaphors carrying it, and the questions it rules out of bounds — separate from whether its claims are true.
  • Headline and wire-copy framing. A headline frames a story in a handful of words (“halts after safety scare” versus “after latest incident”); the audit names what the wording foregrounds and backgrounds before the reader reaches the article.
  • Policy and advocacy language. “Tax relief,” “death tax,” “border crisis,” “pro-life” — terms engineered to win the argument by framing it; the audit unpacks the presupposition each installs.
  • Think-tank and institutional dispatches. Pieces whose persuasion lives in a naturalized worldview rather than explicit claims — the audit names the worldview in vocabulary that lets a reader see it as one option.
  • Media-literacy and comparative reading. Setting how two outlets frame the same event side by side — the same facts, different problem-definitions and treatments — to make the framing itself the object of study.

Failure modes and when not to use it

  • Frame-naturalization. Naming the operative frame in the artifact’s own terms (“the way the economy works”) instead of alternative-comparable vocabulary leaves it invisible — the reader nods along as if it were reality. The mode forces vocabulary that travels across rival frames.
  • Silence-blindness. Cataloguing only what the artifact says and not what it omits misses half the framing, since frames work by selection. The two-column inventory makes the exclusion column mandatory.
  • Stance-slippage into attack. The audit’s whole value rests on staying suspending; the moment it starts arguing the frame is wrong, it forfeits the trust of readers who hold the frame. The mode names what the frame does and costs, and routes rejection elsewhere rather than performing it.

When not to reach for it. When you want the argument’s skeleton — its premises, inferences, and whether the logic holds — that is structure, not framing, and the argument-audit mode fits. When the question is whether the piece hangs together with itself — internal contradictions, equivocation, claims that undercut each other — that is the coherence-audit mode. When you want a verdict that the piece is propaganda and a full diagnosis of its techniques, that is a rejection task for the propaganda-audit mode, not a suspending surface. And when two whole paradigms are in play and you want them set against each other, that is the frame-comparison mode — this one audits a single frame; it does not compare or rank.

  • Argument Audit — the molecular sibling in the same territory: when you want the full teardown — frame and logical structure and propaganda function — composed in one pass and synthesized across all three.
  • Coherence Audit — the neutral-stance sibling that examines whether the artifact’s reasoning hangs together with itself, rather than what way of seeing it installs.
  • Propaganda Audit — the adversarial-stance sibling for when the operative question is rejection: not just what frame is in play but whether the piece functions as propaganda, and how.
  • Frame Comparison — the mode for when two or more whole frames are in play across artifacts and the task is to set them side by side, rather than surface the single frame this mode audits.