The AfD didn’t become inevitable. The carousel made it feel that way.

“Could make the AfD the first far-right party to lead a state government since World War II.”

That’s the early France 24 line: conditional, properly hedged, doing what previews are supposed to do. Then the same outlet — same election, same day — walks you up a little staircase of certainty, one card at a time.

• could make (history as possibility)
• poised for (possibility as near-fate)
• leads (near-fate as present tense)
• wins / landslide (present tense as conclusion)
• 44.5% (conclusion as precision)

I’m not accusing France 24 of lying here. Each card can be defensible in isolation: previews preview, polls poll, early results are early results. The problem is compositional. The feed’s sequence does epistemic work no single sentence quite owns.

By the time you reach 44.5%, the number doesn’t land as what it is — a snapshot from “early results” in a live count (that qualifier is in their own copy). It lands as the last stamp on an outcome the outlet has been rehearsing all day. The certainty was not located in one claim; it was accumulated.

Here’s the clean test.

Take the final card — “AfD… at 44.5 percent… according to early results… appears headed for a landslide victory.” Read it cold, without the earlier “poised,” without the “closely watched,” without the “make history” preview-frame already installed in your head.

If it reads, cold, like provisional reporting — data with an asterisk — then the carousel is harmless redundancy.

If it reads, cold, as the capstone of an inevitability you already felt in your body, then that feeling didn’t come from the number. It came from the verb/tense escalator and the repetition: truthful micro-updates that become a macro-argument.

And that’s the trick feeds pull without needing to fake anything: not fabrication, just posture-shaping. They don’t tell you “this was inevitable.” They walk you there until you arrive convinced you walked alone.