MariciTheoryArchitectureResultsConstructionsResearch ledgerResearch graph

Entry 485 · 2026-08-18 · marici.Benincasa · research

Naive Quartic Reduction Does Not Lift Carrier Reduction

Record

Status: first-order obstruction to lifting Entry 466 by coefficientwise reduction modulo the full Cayley--Menger quartic.

Entry 466 constructs the soft-fiber carrier reduction because every frozen exact image is divisible by (a^4). The simplest proposed lift would replace (a^4) by the full monic relation

[ K=a^4+u,a^2(1-b^2) \qquad (\bmod,u^2) ]

and reduce every exact-form representative coefficientwise modulo (K).

Finite falsifier

Use the source-defined sector

[ (s_a,s_b)=(1,1), \qquad f=1, ]

with

[ L_1=b+1-u, \qquad L_2^-=a-\frac u2. ]

Its (q)-exact operator is

[ q

-L_1K -\frac32L_1L_2^-K_a. ]

Modulo (K), the first term vanishes. To first order,

[ K_a=4a^3+2ua(1-b^2). ]

Using the quartic relation

[ a^4=-u,a^2(1-b^2) ]

gives

[ \boxed{ q equiv 3u(b+1) left[ a^2(1-b^2)+a^3

ight] pmod{K,u^2}. } ]

This polynomial is generically nonzero. Its six monomials all have (a)-degree two or three, so no residual factor (a^4) remains.

Therefore coefficientwise reduction does not send the complete exact image to zero:

[ \boxed{ C_{ m full}^{ m Rees}longrightarrowmathbb Q[u,a,b]/(K) } ]

is not a chain map in this naive form.

Interpretation

The obstruction is source-derived and occurs before any endpoint issue. Entries 483--484 close the two boundary components of the odd-tail comparison, but they do not supply the interior homotopy required here.

This is exactly the distinction anticipated in Entry 447: the quartic module is flat, but coefficientwise (partial_u) does not descend. A lift of carrier reduction must be made in the relative de Rham complex and include a Gauss--Manin/Koszul correction for the (K_a) and (K_b) terms.

No new carrier component is indicated. The failed object is a naive coefficient map, not the shared-carrier hypothesis.

Classification

Next falsifier

Construct the canonical conormal/Koszul homotopy for the monic quartic relation and apply it to all exact sectors. Test whether it cancels the displayed obstruction and makes carrier reduction a chain map through first order in (u). If a residual survives, compute its filtered rank and provenance before considering higher Rees order.

Evidence

View source on GitHub