Skip to content
7Analytica

Method 2 min read

Reading framing without reading minds

In plain words: the same event can be told in very different ways. This shows how to see the difference without guessing what anyone meant.

7Analytica editorialLanguage / Method

Word choice shapes what a report appears to claim. It can be examined closely without guessing at anyone’s intent.

The idea in one picture

“Clashes” and “attack” can describe the same event. So can “militants” and “fighters”, or “annexed” and “integrated”. The facts underneath may be identical. The impression left with the reader is not. Framing is not a flaw in a report so much as a property of all language, and any serious reading of coverage has to account for it.

Describe the choice, not the motive

It is easy to slide from noticing a word to accusing the writer of an agenda. That step requires evidence about the writer that the text alone cannot supply. A more careful practice stays with what is visible: this report uses a verb that assigns responsibility to one party, while that report uses a passive construction that leaves the actor unnamed. The reader can draw their own inference.

What can be compared

  • Who is the grammatical subject of each action, and who is left out of the sentence.
  • Whether numbers are given as counts, as percentages or as vague quantities, and against what baseline.
  • Which side’s statements are quoted directly and which are paraphrased.
  • Intensity of vocabulary: neutral, hedged or charged, measured against how other outlets describe the same event.

Why comparison needs a baseline

A word is charged or neutral only relative to how else the event could have been described. Reading one article tells you little. Reading ten reports of the same event, side by side, shows where the language diverges and by how much. The divergence is the finding; the individual words are only its evidence.

Framing is measured by comparison. A single report has no baseline.

Keeping the reading honest

Linguistic analysis can mislead if it is presented as certainty. Counts of loaded terms depend on a word list someone chose. Translation can introduce or remove charge. The method should show its word lists, its sample and its limits, and it should be used to prompt closer reading rather than to settle it.

The linguistic audit in 7Analytica follows that rule: it marks the passages that differ, shows how, and stops short of saying why.

Related reading

See the record these notes describe.

Request access