Post from Truth Social

A five-week-old poll under a two-year-old sign

Nothing in the artifact is current. The poll's field dates read "5/29-31," the screenshot is stamped "4:12 PM · Jun 8, 2026," and the photograph below it shows supporters holding signs that read "MASS DEPORTATION NOW! 2024" — a campaign-cycle image at least eighteen months old. He republished the stack on the morning of July 13 with nothing of his own attached.

That matters because the week behind him contained no immigration event to react to. The documented run is Iran: fresh threats on July 11 after crowds at Khamenei's funeral chanted for his killing, the Ankara press conference where he called Iranian leaders "scum," the plane swap on the way home. Alongside that, a federal judge released $5.8 million in escrowed funds to E. Jean Carroll on July 9, and the housing bill became law on July 10 without his signature. Into that week he inserted a number from May. The post is not commentary on anything happening; it is a mandate being re-shown.

What a number does that a photograph cannot

The rally photograph alone asserts enthusiasm — a crowd, some signs, hands. The poll line above it converts enthusiasm into arithmetic: "🟢 Support: 56% / 🔴 Oppose: 44%," attributed to "Harvard Harris | 5/29-31 | 1,725 RV." Publishing the two together is a claim that the people in the photograph are the majority, sourced to an institution he does not otherwise cite. The borrowed authority is the point; a poll he did not commission and a verified account he does not run testify on his behalf, which is a cleaner form of validation than saying it himself.

The formulation he chose to amplify is the maximal one — "Deporting Every illegal immigrant," not the criminal-conviction carve-out that usually softens the question. A twelve-point split is rendered as a green dot against a red dot, two colors that do not encode margins. What gets published as consensus is 56.

The 44 percent have no color but red

The binary is the image's, and selecting it adopts it. There is no third category in a green-dot/red-dot layout: a population either backs removing every member of another population or is on the other side of the line. The people to be removed appear nowhere except as the object of the verb on the sign. That absence is what makes the post readable as policy rather than as cruelty — no one is named, mocked, or described, and the sadism score stays low because there is no one in the frame to enjoy the suffering of. The flattening is categorical instead of personal.

Elevated, and why it stops there

"MASS DEPORTATION NOW!" is removal of a population category on a demanded timetable, and pairing it with a majority figure supplies a permission structure: the thing is popular, therefore the thing is legitimate, therefore proceed. That is the recognizable shape of rhetoric that precedes administrative violence. What is missing is the rest of the pattern — no individual or institution is identified as an obstacle, no grievance is articulated against a named person, and nothing is asked of the audience. Followers are shown that they are the majority, not pointed anywhere. The rating rests on the content of the demand he chose to circulate, not on any instruction in it.

Who picked the screenshot

Posted 13:17 UTC — 9:17 a.m. Eastern, wherever in his usual July rotation he was — inside a same-day run of media-only posts. That is a scrolling-and-forwarding session, and forwarding favorable polling screenshots is long-standing personal behavior for him rather than staff work; aides tend to produce captions, attributions and framing. But with no words at all there is no spelling, no punctuation habit, no drift to read, and the timing sits in the ambiguous mid-morning band that fits either hand. The score below is about who selected and pressed post, not who wrote anything, and it is weakly held.

The one language sample in the frame — "Support for Deporting Every illegal immigrant," with its stray capital — was typed by @OpenSourceZone, so his own production is not observable here in any form. Longitudinal comparison would need his captioned posts from the same week.