The Overton Window is the range of ideas that can be said out loud in public without the speaker being treated as strange. Not the ideas that are correct. Not even the ideas that are popular. The ideas that get a straight face. Everything inside the window can be argued about normally; everything outside it gets argued about as a symptom. * The model came from Joseph P. Overton, a senior vice president at the Mackinac Center for Public Policy in Midland, Michigan, who sketched it in the mid-1990s under the clunkier name "window of political possibilities." His picture was a yardstick. The full spectrum of options on any issue runs the whole length of it, but at any given moment only a short section is politically feasible, and that section is what politicians can safely stand on. Overton died in 2003, when the ultralight plane he was piloting crashed near Caro, Michigan. His colleague Joseph Lehman carried the idea forward and named it after him. * The version that spread later sorts the yardstick into six degrees of acceptability: unthinkable, radical, acceptable, sensible, popular, and policy. An idea does not jump from the first to the last. It walks, one notch at a time, and the walking is mostly done in public conversation rather than in legislatures. * That last part is the bit almost everybody gets backwards. The Mackinac Center's own explanation is emphatic that the window moves when the opinions of the general public move, and politicians then adjust to stay inside it — not the other way round. Politicians are followers in this model, not drivers. The strategic corollary is the one Milton Friedman gave to people who wanted change: keep the unfashionable alternative alive and researched and ready, because sooner or later "the politically impossible becomes politically inevitable." * So where does the news come in? Nobody measures the window. It is not published anywhere. It is performed, every day, in the way stories are worded — and that makes a newsroom one of the main places the window is maintained, whether or not anyone in the building thinks of it that way. * The maintenance happens in small parts of speech. Which position gets "critics say" bolted onto the end of it, and which one gets reported as the baseline. Who is introduced as an expert, who as an activist, and who as a hardliner. Whether a proposal is called a plan or a demand. Whether the money question — how would you pay for this? — is asked of both sides or only of the side that wants something new. Whether a position appears in the story as a quote from a named human being, or as the narrator's own aside. None of those choices is a lie. All of them tell you where the edges of the window are that morning. * For a case you can check, look at how a wedding page moved. Gallup first asked Americans about legal recognition of same-sex marriage in 1996, and 27% said yes. On August 18, 2002, The New York Times announced that its Styles pages would carry same-sex commitment ceremonies alongside weddings, with then-managing editor Howell Raines acknowledging the religious and legal controversy and saying the ceremonies merited coverage in their own right. The first such announcement — Daniel Gross and Steven Goldstein — ran on September 1, 2002. That was twelve years after a much smaller paper, the Everett Herald in Washington state, had already started doing it. * The wedding page did not cause the change and it did not lead it. What it did was record the exact moment a newsroom decided an idea had crossed from "controversial" into "coverable in the ordinary way" — from radical to acceptable, in Overton's vocabulary. Gallup's number reached 70% in 2021 and peaked at 71% in 2022 and 2023. Twelve years earlier, the small paper in Everett had made the same call while the national consensus still said it was strange. * And then the yardstick slid back. Gallup's Values and Beliefs survey, conducted May 1-17, 2026, put support at 65% — six points off that peak. Windows are not ratchets. They move in both directions, and anyone who tells you history only bends one way is trying to sell you a mood, not a model. * The most useful thing this concept gives an ordinary reader is a way to catch one specific trick: the phrase "politically unrealistic," and its cousins "not serious," "won't fly," "has no path." Every one of those is a claim about the room. Almost every one is printed as though it were a claim about the policy. And the reasoning underneath it runs in a circle — the idea is unrealistic because serious people do not back it, and serious people do not back it because it is unrealistic. A policy can poll at sixty percent and still be described in national coverage as a fringe position, because the window is not drawn by the sixty percent. It is drawn by the smaller group who own the microphones, and it is drawn with adjectives. * Working out where a story has put the edges is not hard once you look for it. Count the modifiers: if one side's proposal is "sweeping" and "far-left" while the other's is just what the bill does, you have found the frame. Notice who gets asked how they will pay for it, and who gets to leave the price offstage. Notice whether the objection to an idea comes out of a named mouth or out of the narrator's own voice, because the second is the newsroom putting its thumb on the yardstick. Notice whether an idea is placed on the spectrum with any evidence at all — a poll number, a vote count — or purely by vibe. And notice which ideas simply never appear, since the edge of a window is quieter than anything inside it. * None of this requires a conspiracy, and the model was not built by anybody's enemy — it came out of a free-market think tank that used it to plan its own long game. The window is just a running, unsigned consensus about which ideas get discussed with a straight face today. It is maintained by people who never announce that they are maintaining it. Which is the point: once you know it exists, "everyone agrees" stops being a description of reality and turns back into what it always was, a description of who counts as everyone.