# Saar's dolls grip the invariant model — until the material itself turns out to already be someone else's emission

**Artwork**:   
**Artist**: Arnaud Quercy  
**Asset**: IDS-RNI0130  
**Serial**: NAN-RDG000073

The double-invariant model grips cleanly for freshly made work, but the boundary this case exposes is what happens when the raw material is an appropriated found object already bearing someone else's encoded intent — a recursive, stacked-codex situation the framework's two-layer picture doesn't explicitly anticipate, though it doesn't refute it either.

## Context

### What the source says [@source]

Betye Saar has collected Black dolls — some handmade, some mass-manufactured, some embodying racist caricature (mammy, golliwog, topsy-turvy figures) — for decades, and since the 1970s has folded them into assemblages, prints, and, since the pandemic, watercolors. She describes the dolls as having 'an energy of their past life, of the little child who loved them,' and the exhibition frames her practice as one of collecting, sensing the object's residual energy, and then 'transforming and releasing the object onto the world' — recontextualizing derogatory origins into images of celestial flight or self-possession, without erasing the dolls' physical history.

### Where the lens grips

P3a's double-layer invariant maps cleanly onto the case: the physical layer (the caricatured doll form — mammy, golliwog — literally manufactured stereotype) is received as constraint, undissolved by any reframing; the intentional layer (Saar's alchemizing, reclaiming logic) is latent and requires the curators' and artist's own dialogue with Hyperallergic to be activated — exactly P19a's condition that explicability completes the ricochet. P8 (Seriality) also grips: the same doll-material recurs and proliferates across a 1972 print, a 1975 assemblage, and 2020s watercolors, each a variation rather than a conclusion.

### Where it breaks

P3a assumes the physical invariant is codex-free material fact — dimensions, colors, forms, received as neutral constraint. Here the physical substrate is not neutral: it is itself already an emission of a prior, different codex (the racist manufacturer's caricature logic, plus the sedimented history of a 'child who loved them'). Saar's intentional invariant is not encoded onto raw material but layered onto someone else's already-encoded wave. The model doesn't state whether this counts as one invariant structure or two stacked ones — this is contestable: a defender could say the manufacturer's caricature logic is just part of the 'physical' layer (pattern, color, form) with no separate intentional status, in which case P3a holds cleanly; the case slips only if one insists the caricature was itself already an intentional encoding (a codex) rather than mere material fact.

### Propositional grid [@p31]

> **Read this grid.** “Not evaluated” does **not** mean the source has nothing to say about that proposition. It means **the reading did not instruct it**. That is not a result, and it counts in no denominator. A silence, by contrast, is a finding: the reading touched the proposition without concluding, or noted its absence itself.

**Evaluated**

<table class="prop-grid"><colgroup><col class="pg-id"><col class="pg-title"><col class="pg-audit"><col class="pg-verdict"></colgroup><thead><tr><th>proposition</th><th></th><th>audit status</th><th>relational verdict</th></tr></thead><tbody>
<tr><td><strong>P3a</strong></td><td>Double Structure of the Invariant</td><td>not audited</td><td>pressure-against</td></tr>
<tr><td><strong>P8</strong></td><td>Seriality</td><td>not audited</td><td>evidence-for</td></tr>
<tr><td><strong>P19a</strong></td><td>Explicability as Condition</td><td>not audited</td><td>evidence-for</td></tr>
</tbody></table>

**Not evaluated (34)** — P1, P2, P3, P4, P5, P6, P7, P9, P10, P11, P12, P13, P14, P14a, P14b, P15, P16, P16a, P17, P18, P18a, P19, P20, P21, P22, P23, P24, P25, P26, P27, P28, P29, P30, P31

---

*Overall verdict: **Bounded** — taxonomy of the August 2026 pass, predating the move to relational verdicts. Kept as metadata: the gap between the two forms is data, not residue. This reading’s judgement is in the grid above.*

Reading of 2026-05-26 · model `claude-sonnet-5` · source 6,834 chars · 1 attempt(s)

---

### First pass — kept as written

*Reading produced by the T50 RSS pass, before the boundary lens. It is reproduced UNCORRECTED. Keeping it is the point: without it, a changed title is just a changed title — with it, what the second pass refused becomes legible.*

> Saar's collection and gift practice instantiate a codex of liberation through material constraint. The doll—a loaded object carrying centuries of racist iconography—becomes the site where she engineers diffraction. By collecting, preserving, and now gifting these objects to an institution, she forces a ricochet: the receiver (museum, public) arrives expecting racist artifact, encounters instead an archive of refusal and reclamation. The physical invariant (the doll's form) remains; the intentional invariant (why these dolls, arranged this way, given now) reframes what the object can mean. The gift itself is a codex—a system of constraints that says: 'these objects belong to collective memory, not private possession.' Generative loss occurs in the transfer: the dolls lose their isolation, gain their historical weight.

<sub>T50 RSS assessment · 2026-05-26 · remplacée par `boundary-lens-2026-08`</sub>
