# PR95: historical priority unresolved

The note gives a verified exact counterexample to the broad nonzero-magnitude statement printed as Ohtsuki Conjecture 7.5, p.474. It does not claim the first counterexample, exhaustive novelty clearance, a newly resolved historical problem, continuous openness, or originality of the underlying RT formulas.

The collection is Geometry & Topology Monographs 4, nominally 2002, actually published 1 June 2004. The conjecture is attributed to Guadagnini and Pilo, Three-manifold invariants and their relation with the fundamental group, CMP 192 (1998), 47-65, DOI [10.1007/s002200050290](https://doi.org/10.1007/s002200050290). The available [1996 preprint v1](https://arxiv.org/abs/hep-th/9612090v1) gives the broad setting and its SU(2) lens-space theorem; the final journal full text was not compared.

## Direct missing source

Takahito Kuriya, The LMO invariant and the Guadagnini-Pilo conjecture for lens spaces, is listed as a preprint by his [Kyushu institutional record](https://www2.math.kyushu-u.ac.jp/coe/staff/staff_sub/kuriya.html), cited in the [2004 Topology Symposium proceedings](https://www.mathsoc.jp/~topology/topsymp/kouenshu/ts2004all.pdf), printed p.54 reference [12], and associated with an exact-title talk on 21 February 2003 in a [Kyushu report](https://www2.math.kyushu-u.ac.jp/coe/report/pdf3/report-6.pdf), printed p.121. No verified DOI or journal publication is asserted for it.

Its full text, complete conclusions and hypotheses could not be obtained. We cannot establish that it does not already contain or circumscribe this result. Its title, public author record, an unverified aggregator abstract and later perturbative results cannot supply the missing finite-level comparison. No inference about its original language is made. This is the decisive reason priority remains unresolved.

## Bounded comparisons and boundaries

The proof input is [Hansen-Takata arXiv:math/0209403v2](https://arxiv.org/abs/math/0209403v2), Eq.(12), Sections 4-5 and Theorem 5.1. The journal publication is JKTR 13(5) (2004), 617-668, DOI [10.1142/S0218216504003342](https://doi.org/10.1142/S0218216504003342); its final full text and possible later corrections were not obtained. The note identifies the actual preprint formula source.

The history audit examined selected primary passages of Kuriya's 2008 On the LMO conjecture, Kuriya-Le-Ohtsuki 2012, Guadagnini-Mancarella 2010, Hikami 2005/2006, Ionicioiu-Williams 1998, Garoufalidis-Le-Marino 2008 and Oriti's 2003 thesis. It authenticated no earlier qualifying pair in those passages. Perturbative recovery, equal-magnitude complex-value separation and rank-one tests do not alone answer this finite-level full SU(5) comparison. These are bounded read outcomes, not absence theorems.

Later selected reads included [Gang arXiv:0912.4664v2](https://arxiv.org/abs/0912.4664v2), [Kubo-Yokoyama arXiv:2108.09300v2](https://arxiv.org/abs/2108.09300v2) and the [final JHEP article](https://doi.org/10.1007/JHEP04(2022)074). Gang's final pagination is 1119-1128, JKPS 74 (2019), DOI [10.3938/jkps.74.1119](https://doi.org/10.3938/jkps.74.1119); its final full text remains unread. The selected general-q localization formula is U(N) and retains a sector-sign assumption, so no full ordinary SU(N) conclusion was inferred from it.

Kubo-Yokoyama's later ordinary SU(N) WZNW comparison is explicitly spin-independent; proposed background fields are distinguished from spin dependence. A relative phase discrepancy can vanish for even shifted coupling. The selected lens models and tables have q=1 and do not authenticate an earlier same-theory pair L(5,1), L(5,2). No blanket dismissal of all SU(N) expressions as spin invariants is warranted.

A bounded check of Hansen-Takata's older sl4 lens-space remark found equal nonzero magnitudes in four exact reconstructions, leaving a discrepancy with its printed complex-separation statement. It remains an unresolved source discrepancy, neither a proved historical erratum nor evidence of firstness; it is unnecessary for the present proof.

Unread or access-limited sources include Sokolov 1997, Takata 1996 projective-category full text, a reliably readable complete Zhang-Carey 1996 source, Yamada 1995, the final HT article, Gang final text and unexamined portions of the later works. Bibliographies and searches are not exhaustive.

No individual was contacted and no outreach was prepared. Extensive AI assistance and absence of conventional human peer review are disclosed. The publication qualification permits reporting the checkable result with unresolved priority; it does not resolve the missing source.
