How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
Compatible signs exist on the Bruhat graph
Statement
There is a function from the set of arrows of the Bruhat graph to such that for every square the product of the four signs is : . Consequently the two saturated paths of a rank-two interval always carry opposite total signs. Moreover, for any two compatible signings there are vertex signs , with , such that on every cover.
Facts & Assumptions
Given: The finite Weyl group , its Bruhat covers oriented downwards, and its rank-two diamonds.
Bruhat order has the subword property and the right lifting property: if , and for a simple reflection , then and . If both descend under , then . These are proved in Bruhat intervals of rank two are diamonds from Bruhat order on a finite Weyl group. A nonidentity element has a simple right descent, and the unique longest element reverses all positive roots (Finite Weyl strong exchange and deletion, Finite Weyl positive roots and simple reflections, Finite Weyl closed chambers and stabilizers).
Every interval of length two has exactly two middles (Bruhat intervals of rank two are diamonds).
Proof
Induct on to sign every cover in the principal ideal with product on every diamond. For there are no covers. Choose a simple right descent of and put . By induction sign all edges in . Every descends under : otherwise lifting would give . Moreover , by descent monotonicity. Set for these outside vertices. If is any other edge with outside, then also descends: if , lifting gives , and equality of lengths forces , the excluded vertical edge. Thus , both vertices lying in , and is a diamond. Define . The first factor is already defined, either by induction when , or as when is outside. This assigns each edge once and makes every such side diamond negative.
Let be a diamond in . If , all its vertices lie in , so its product is by induction. Suppose is outside. If neither nor equals , both descend by step 1.1. Then descends too: if , lifting gives ; equality of lengths forces , a contradiction. The four translated vertices are distinct, lie in , and form a diamond by descent monotonicity and their lengths. Each of the four side diamonds for the edges of has product : step 1.1 gives this if is outside, and induction gives it if . Multiplying these four products and the product of the translated diamond leaves exactly the product of , because each vertical edge and each translated edge occurs twice. Hence its product is .
The remaining case, after exchanging , is . Here descends, and ; it has length and differs from unless . If , lifting against gives , so . Thus is precisely the side diamond for , already made negative in step 1.1. If , the vertices give three diamonds: , and . Indeed follows from and the lengths, from descent monotonicity applied to , and , , are given covers. The first two are side diamonds, negative by step 1.1 or induction; the last lies in because . Multiplying their three products cancels all extra edges twice and leaves exactly the product of . It is therefore .
Steps 2.1 and 2.2 exhaust all diamonds, proving the induction. Every lies below the longest element: repeatedly append a simple reflection that increases length, producing Bruhat covers. Length is bounded on the finite group, so this stops at an element with for every simple root by the simple-reflection criterion. Every positive root is a nonnegative combination of simple roots, so reverses all positive roots and is the longest element by [F1]. Thus take to be that element to obtain a signing on all of . In a diamond the two path products satisfy , hence , the required consequence. For the empty signing satisfies the assertion vacuously. The construction uses only recursion on a finite group and selection from finite sets, so no infinite Choice principle is used.
For the last assertion put , whose product on each diamond is . Induct on to find and on . The identity ideal is immediate. With as in step 1.1, take the inductively supplied signs on and set for every outside . This handles vertical edges. Every other edge with outside has the side diamond of step 1.1, with . Also , either by induction when or by the new definition otherwise. Its diamond identity gives ; substituting the known ratios yields . All remaining edges lie in . Taking to be the longest element completes this finite induction and proves the assertion.
Depends on
Used by
- Unsigned Bruhat edge sums need not square to zero Counterexample
- The BGG differential from signed Verma maps Definition
- Sign cancellation in an A2 Bruhat diamond Example
- The A2 BGG resolution with six Verma summands Example
- The BGG differential squares to zero Proposition
Dependency tree · two levels
8 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.
Sources
- A. Rocha-Caridi, Splitting criteria, Trans. AMS 262 (1980), Lemma 10.4 and its proof, pp. 354-355 (standard reference, not scraped)
- J. Bernstein, I. Gelfand and S. Gelfand, Differential operators on the base affine space and a study of g-modules, Lemma 10.4 and Sec. 11 (author-hosted scan) (standard reference, not scraped)
- Fan Zhou, The classical and the functorial BGG resolutions (Columbia thesis 2021), Part I Lemma (10.3,10.4), p. 10 (standard reference, not scraped)
- N. Hemelsoet and R. Voorhaar, A computer algorithm for the BGG resolution, arXiv:1911.00871, Prop. 2.3 and Sec. 4.2 (standard reference, not scraped)