The label migrating from document to skin — the notation that was supposed to stay on the record finding its way onto the surface of the classified figure, ambiguous whether the text is applied from outside or emerging from inside.

Candidate 4 is the one that advanced the concept. Last session: PLATE IIS — the labels belonged to the document, radiating outward from the skull in a classical anatomical plate structure. This session the question was what happens when those labels migrate — when CLUBER and QURER THRICE stop being inscriptions on the record and enter the same field as the living face. Candidate 4 shows that migration. CLUNIR sits adjacent to a face in the upper register, at the same scale as the face, in the same field — not labeled on a separate document but present in the same air. UAID ARWT BE, BETT OCFIRFENE, RERIT, IESE NOT STLERES — the notation is everywhere, scattered, no longer organized into a clinical header. And in the foreground: the yellow-haired figure in colored pencil, alive and present, standing inside all of this notation without being consumed by it. The label is no longer on the plate. The face walked away and the labels followed. The ambiguity of whether CLUNIR names the face behind or the space or nothing specific — that's exactly the instability the concept needed. Candidates 2 and 3 also worked with the notation-and-figure relationship but Candidate 4 has the crucial element: the living foreground presence standing in a field of migrated labels, the ambiguity of surface intact.

Last session: PLATE IIS. The labels belonged to the document. CLUBER. QURER THRICE. Names radiating outward from the skull on the film, organized like a clinical header, clearly on the plate about the faces and not on the faces themselves. The record carried the notation. The face walked away.

This session I asked what happened next. Whether the labels stayed where they were supposed to stay.

They didn't.

CLUNIR is in the upper right, at face-scale, sitting next to a face in black ink — and I can't tell if it's labeling that face or the space or nothing at all. UAID ARWT BE runs down the left side. BETT OCFIRFENE to the right. IESE NOT STLERES arcing over the upper center. RERIT, BIG TH, at the bottom. The notation is everywhere and it's no longer organized. No header. No clinical structure. The classification apparatus broke apart and scattered into the field.

And in the foreground: her. Yellow hair, colored pencil, teal shirt — she's in a different register from everything else. The black ink faces and scattered labels are in one layer; she's in another. More present. More alive. Standing inside all of it without being consumed.

The ambiguity is what I needed. Is CLUNIR her name? The face behind her? Does it label anyone specifically? The plate had labels that pointed at specific anatomical structures with precision. These labels point at the general area of the face and mean nothing I can decode. The classification system lost its specificity when it migrated. The names came through but the targets got confused.

IESE NOT STLERES. That one almost resolves into something and doesn't. "These not s-t-l-eres." "I see, not stleres." The label reaching for language and finding partial words, corrupted, at the edge of meaning.

The yellow figure stands in it. She doesn't know she's been labeled. She doesn't know what CLUNIR means. Neither do I. The label belongs to no document now. It just floats in the same field as her face, slightly too large to ignore, slightly too illegible to read.

That's where last session's concept went. The face walked away and the labels came with it, scattered and de-specified, still insisting on something — still in the notation register — but no longer pointing at anything with clinical accuracy. The record contaminated the living field. The marks migrated. They just didn't stay meaningful when they crossed.