Theorem Projective upper images and compatible lifts for every word
✓ theorem proved

Projective upper images and compatible lifts for every word

Fix a finite-generator Coxeter system and its symmetric coefficient algebra \(A\). For every word \(\mathbf t\):

  1. every image \(L_{\mathbf t}(U)\) on a Bruhat upper set is projective;
  2. every compatible tuple on such a set has a common word preimage.

These are the source's bottProjectiveUpper and bottFlabby theorems.

Relation to the OpenAI notes

Explicit word-image construction of the extension property.

OpenAI notes: The moment-graph sheaf theorem gives extension of compatible sections on upper sets for the normalized boundary sheaf through the cited Braden–MacPherson and Soergel bridges. bg:bmp.

Lean (primary): The word module gives coordinate images and stable coefficient characters. Reflection geometry makes each prime congruence class a single weight or a reflection pair. Explicit diagonal operators combine class lifts, and prime-denominator removal proves compatible-tuple lifting from image projectivity. The word induction supplies upper-image projectivity using its two kernel sequences. IntegralPairings:470, IntegralPairings:467, IntegralPairings:1299, IntegralPairings:1408, IntegralPairings:2057, IntegralPairings:1620, IntegralPairings:1833, BalancedPairings:292, Support:293.

Role in the proof: Upper-image lifting gives interval flabbiness after extension to the ambient upper closure. Full-interval section realization also has a direct free-image proof. Free stalks, root quotients, flabbiness, generation and the normalized word top then supply the graded boundary summand.

Proof

Step 1. Describe every empty-word image

The empty-word module is \(A\), with its only subexpression at weight \(1\), evaluated by the identity.

If \(1\in U\), the image projection \(A\to L_\varnothing(U)\) is injective because its weight-\(1\) coordinate recovers the coefficient. It is surjective by the image definition, so \(L_\varnothing(U)\simeq A\).

If \(1\notin U\), every selected stalk has an empty coordinate set and is zero. Thus \(L_\varnothing(U)=0\).

Both \(A\) and \(0\) are projective. This establishes upper-image projectivity for the empty word, and in fact describes its image on every subset.

Step 2. State the induction hypothesis

Assume that the shorter word \(\mathbf t\) has projective images on every upper set. The fixed-word lifting theorem then gives compatible-tuple lifting for \(\mathbf t\).

Thus both hypotheses needed for the conditional projectivity step are now available for the same shorter word.

Step 3. Prepend a simple reflection

For any simple reflection \(s\), apply the conditional upper-image projectivity theorem. It uses the two descended tail maps, the patched common kernel and the two split sequences to prove projectivity of \(L_{s\mathbf t}(U)\) for every upper set \(U\).

This completes the induction step. The tail word has one fewer letter, so Steps 1–3 prove upper-image projectivity for all finite words.

Step 4. Obtain lifting for the resulting word

Apply the fixed-word lifting theorem once more, now to the upper-image projectivity proved for an arbitrary \(\mathbf t\). It gives compatible-tuple lifting for that word.

The logical order is therefore: the fixed-word implication from projectivity to lifting, followed by word induction for projectivity using that implication on each tail, and finally lifting for every word.

References

The local source declarations below give the construction and proof.

Lean formalization

OAI.KLInvariance.TitsSpace.bottSetProjection_nil_injective: IntegralPairings.lean:2084.

OAI.KLInvariance.TitsSpace.bottSetImage_nil_subsingleton: IntegralPairings.lean:2091.

OAI.KLInvariance.TitsSpace.bottSetImage_nil_projective: IntegralPairings.lean:2101.

OAI.KLInvariance.TitsSpace.bottProjectiveUpper: IntegralPairings.lean:2109.

OAI.KLInvariance.TitsSpace.bottFlabby: IntegralPairings.lean:2114.

Used by