Theorem Existence and uniqueness of the canonical KL families
✓ theorem proved

Existence and uniqueness of the canonical KL families

For every Coxeter system, exactly one pair \((R,P)\) satisfies the normalization data. It is the pair consisting of the ordinary R construction and the recursive lower-half P construction.

The selected interfaces agree with these constructions: \[ \mathrm{klFamilies}=(R,P),\qquad \mathrm{rPolynomial}(x,y)=R_{x,y},\qquad \mathrm{klPolynomial}(x,y)=P_{x,y}. \]

Relation to the OpenAI notes

Explicit construction of the normalized family.

OpenAI notes: The conventions specify the ordinary R recurrence, the strict P degree bound and reciprocity, referring to the classical Kazhdan–Lusztig normalization. sec:introduction, eq:reciprocity.

Lean (primary): After constructing R, the development defines P by the negative lower half of the proper-upper defect. The reversed R-kernel proves that defect anti-reciprocal, yielding reciprocity. Separate uniqueness inductions identify the selected family with the construction. CanonicalKL:70, CanonicalKL:162, Bruhat:204, CanonicalKL:260, CanonicalKL:282.

Role in the proof: The KB presents the coefficient solution, defect identity, rank induction and seven normalization fields as a continuous construction. The classical selection interfaces are then identified with the explicit formulas.

Proof

Step 1. Supply the three R conditions

Take \(R=\mathrm{OrdinaryR.polynomial}\). Its construction theorem proves:

  • \(R_{x,x}=1\);
  • \(R_{x,y}=0\) when \(x\not\le y\);
  • the ordinary recurrence for every left descent of \(y\).

These are precisely the R_diagonal, R_zero and R_recursion fields.

Step 2. Supply the P values and degree condition

Take \(P=\mathrm{CanonicalP.polynomial}\). The recursive degree theorem gives the diagonal value one, the value zero when \(x\not\le y\), and \(2\,\operatorname{natDegree}(P_{x,y})<d(x,y)\) for \(x<y\).

These provide P_diagonal, P_zero and P_degree.

Step 3. Match the reciprocity sum to the normalization interface

The recursion theorem proves \[ \operatorname{reflect}_{d(x,y)}P_{x,y} =\sum_{z\le y}R_{x,z}P_{z,y}. \] If \(x\not\le z\), its R-factor vanishes, so this is the sum over \([x,y]\).

The interval type in the normalization predicate consists exactly of those elements, carrying the two endpoint inequalities. Its finite instance identifies its finitely supported sum with the same finite sum of values. Therefore the displayed equality supplies the reciprocity field.

All seven fields have now been constructed, proving existence.

Step 4. Apply uniqueness to any normalized pair

Let \((R',P')\) satisfy the same normalization data. The normalized-pair uniqueness theorem applies to it and the pair built in Steps 1–3. It gives \(R'=R\) and \(P'=P\).

Thus the normalization predicate has exactly one satisfying pair.

Step 5. Identify classical selection with the construction

Classical selection of a witness to an inhabited predicate satisfies that predicate. Existence from Step 3 therefore gives NormalizedKL for klFamilies.

Uniqueness identifies klFamilies with \((R,P)\). Taking first components gives equality for rPolynomial; taking second components gives equality for klPolynomial. Evaluating those function equalities at \((x,y)\) yields the stated interfaces.

References

The local source declarations below give the construction and proof.

Lean formalization

OAI.KLInvariance.normalized_interval_sum: CanonicalKL.lean:239.

OAI.KLInvariance.constructed_normalizedKL: CanonicalKL.lean:260.

OAI.KLInvariance.normalizedKL_existsUnique: CanonicalKL.lean:273.

OAI.KLInvariance.klFamilies_spec: CanonicalKL.lean:279.

OAI.KLInvariance.klFamilies_eq_constructed: CanonicalKL.lean:282.

OAI.KLInvariance.klPolynomial_eq_constructed: CanonicalKL.lean:286.

OAI.KLInvariance.rPolynomial_eq_constructed: CanonicalKL.lean:290.

Used by