Theorem Prime-denominator membership in a projective target
✓ theorem proved

Prime-denominator membership in a projective target

Let \(R\) be a unique factorization domain, \(E\) a torsion-free \(R\)-module, and \(M\subseteq E\) a projective submodule. Fix \(x\in E\). Assume \[ \exists c\ne0,\quad cx\in M, \] and, for every prime element \(p\), \[ \exists a,\quad p\nmid a,\quad ax\in M. \] Then \(x\in M\). The free-target theorem is the special case where \(M\) is free.

Relation to the OpenAI notes

General module interface for the local membership test.

OpenAI notes: The free-target lemma uses a polynomial coefficient ring and a finite free ambient module. Basis coordinates in the fraction field lie in every height-one local ring, so their denominators are units. edge:lattice.

Lean (primary): The free-target interface allows a torsion-free ambient module over a UFD and states generic span through nonzero scalar multipliers. Its proof removes prime factors from those multipliers. A separate projective-target version obtains finite-support coordinates through a free-module section and retraction. ModuleAlgebra:45, ModuleAlgebra:106, IntegralPairings:1904.

Role in the proof: The proofs below expose the exact generic-span and prime-denominator witnesses used by local-to-global membership.

Proof

Step 1. Obtain coordinates with a retraction

There is a canonical free module on the underlying elements of \(M\), whose vectors are finitely supported coefficient families. Its evaluation map \[ \varepsilon:R^{(M)}\twoheadrightarrow M,\qquad (c_m)\longmapsto\sum_m c_m m \] is surjective: the single basis vector indexed by \(m\) evaluates to \(m\).

Projectivity of \(M\) gives a linear section \(s:M\to R^{(M)}\), with \(\varepsilon s=1_M\). These maps provide free coordinates for the divisibility calculation while retaining a map back into \(M\).

Step 2. Divide a projective vector using a prime relation

Suppose \(p\) is prime, \(p\nmid a\), and \(m,n\in M\) satisfy \[ am=pc\,n. \] Apply the linear map \(s\) and take a coordinate: \[ a(s(m))_i=pc(s(n))_i. \] Primality and \(p\nmid a\) imply \(p\mid(s(m))_i\) for every \(i\).

The vector \(s(m)\) has finite support. Divide those finitely many coefficients by \(p\), putting zero outside that support, to obtain a finitely supported vector \(z\) with \(pz=s(m)\).

Apply \(\varepsilon\): \[ p\,\varepsilon(z)=\varepsilon(s(m))=m. \] Thus \(m\) is divisible by \(p\) inside \(M\). If \(M\) is free, its basis coordinates and inverse basis map give this same calculation directly.

Step 3. Remove a prime factor from a generic denominator

Start with \(cx\in M\), \(c\ne0\), and factor \(c\). If \(c\) is a unit, then \(x=c^{-1}(cx)\in M\).

Otherwise write \(c=pc_1\), with \(p\) prime and \(c_1\ne0\). Choose the local witness \(a\) with \(p\nmid a\) and \(ax\in M\). In \(M\), set \[ m=pc_1x,\qquad n=ax. \] Commutativity of scalar multiplication gives \(am=pc_1n\). Step 2 supplies \(m'\in M\) with \(pm'=m\).

Step 4. Cancel in the ambient module

Inside \(E\), the last equality is \(pm'=pc_1x\). Since \(p\ne0\) and \(E\) is torsion-free, \[ m'=c_1x. \] Thus the smaller denominator \(c_1\) also sends \(x\) into \(M\).

Step 5. Finish the finite factorization induction

A nonzero element of a UFD is a unit times finitely many prime factors, counted with multiplicity. Repeating Steps 3–4 removes one factor at a time. The final unit case gives \(x\in M\).

The hypotheses for every prime concern the fixed vector \(x\), so they remain available at each factor-removal step.

References

The local source declarations below give the construction and proof.

Lean formalization

OAI.KLInvariance.ProjectiveLattice.finsupp_divide: IntegralPairings.lean:1870.

OAI.KLInvariance.ProjectiveLattice.prime_divide_of_relation: IntegralPairings.lean:1886.

OAI.KLInvariance.ProjectiveLattice.mem_of_prime_local_membership: IntegralPairings.lean:1904.

OAI.KLInvariance.EdgeLattice.mem_of_prime_local_membership: ModuleAlgebra.lean:45.

Used by