Category O
Category O Learning Seminar
Fall 2021
Notes by Patrick Lei
Lectures by Various
Columbia University
Disclaimer
These notes were taken during the seminar using the vimtex package of the editor neovim. Any errors are mine and not the speakers’. In addition, my notes are picture-free (but will include commutative diagrams) and are a mix of my mathematical style and that of the lecturers. If you find any errors, please contact me at plei@math.columbia.edu.
Also, we should note that Fan used different notation than the rest of us because he likes to be special.
Seminar Website:
https://math.columbia.edu/~plei/f21-CO.html
1 Kevin (Sep 29): Review of semisimple Lie algebras and introduction to category
1.1 Review of semisimple Lie algebras
Throughout this lecture, we will work over .
Definition 1.1.1.
A Lie algebra is semisimple if any of the following equivalent conditions hold:
- 1.
is a direct sum of simple Lie algebras (those with no nonzero proper ideals).
- 2.
The Killing form is nondegenerate.
- 3.
The radical (maximal solvable ideal) of is zero.
Some examples of semisimple Lie algebras include , and in some sense (the classification of simple Lie algebras), these are essentially all semisimple Lie algebras.
Now given a semisimple Lie algebra , we will fix a Cartan subalgebra , which is just a maximal abelian subalgebra of semisimple elements. This gives us a root decomposition
where is the subspace of where acts with weight . Some important facts about these root systems are the following:
- •
For all , we have .
- •
For all roots , we have .
- •
If is a root, so is .
In addition, the are required to form a (reduced) root system (denoted ), the precise definition of which is deliberately omitted. Given a choice of Borel subalgebra containing , we obtain a set of positive roots and a set of simple roots. In addition, given a root system , there is a dual root system , whose roots are
Now suppose that is a semisimple Lie algebra with root system . For every , we may choose and , and these determine some . This choice can be made such that .
Recall that the Lie algebra is spanned by the matrices
Then the choice of gives an embedding . These maps, ranging over all , cover all of . Now a basis of is given by and for the simple roots . Therefore, to specify , we only need to give commutation relations for the basis elements.
Now suppose that is some root system. We would like to construct a semisimple Lie algebra with root system . We want to build a semisimple Lie algebra. To do this, choose a set of simple roots , and consider the Lie algebra
where the relations are as follows:
- •
.
- •
We have if and this commutator vanishes otherwise.
- •
.
- •
.
- •
if .
- •
if .
The first four relations are called the Weyl relations and the last two are called the Serre relations. Given this data, we end up with a semisimple Lie algebra with root system . In addition, if is any other semisimple Lie algebra with root system , there is an isomorphism . Moreover, we have a bijection between semisimple Lie algebras and reduced root systems, which restricts to a bijection between simple Lie algebras and irreducible root systems.
| Irreducible root systems | simple Lie algebras |
|---|---|
| exceptional Lie algebras |
We will now discuss the finite-dimensional representation theory of semisimple Lie algebras .
Theorem 1.1.2 (Weyl’s complete reducibility theorem).
Any finite-dimensional representation of is decomposes as a direct sum of simple representations.
Now suppose that is a finite-dimensional -representation. Then has a weight decomposition
These are integral weights, which simply means that for all roots . For any root , and . We would like to think that the raise the weights and lower the weights, so we introduce a partial order. We say that if .
By Weyl’s complete reducibility theorem, it remains to classify the irreducible representations of . These are in bijection with the dominant integral weights, which in particular means that for all . For any dominant weight , there is a unique highest-weight representation . Here, is generated by a single maximal vector of weight . This means that for all positive roots , .
1.2 Introduction to category
We would now like to study infinite dimensional representations of . Of course, this is impossibly complicated in general, so we will impose some finiteness conditions on our representations.
Definition 1.2.1.
The category is the full subcategory of -modules satisfying:
- 1.
is finitely generated as a -module.
- 2.
is -semisimple and has a weight decomposition .
- 3.
is locally -finite, where . Precisely, this means that the generated by any is finite-dimensional.
Here are some facts about category , which are stated without proof.
- •
For all in our category and weights , the weight space is finite-dimensional.
- •
is a Noetherian (everything satisfies the descending chain condition) abelian category.
We will now describe some infinite-dimensional objects in category .
Definition 1.2.2.
For any weight , the Verma module associated to is the module
where is the Borel subalgebra associated to our choice of positive roots and is the -module associated to the -dimensional representation of with weight and the identification .
2 Fan (Oct 06): Beginnings in category : Vermas, central characters, and blocks
Recall the following fact ahout semisimple Lie algebras. If we have a decomposition
where
- 1.
;
- 2.
is a lattice of maximal rank;
- 3.
implies ;
- 4.
,
then is a semisimple Lie algebra.
Now recall that the Weyl group is the group generated by the reflections in the roots. For any , we define the length
Next, there is the Bruhat order, where if and , we say .
Finally, throught this lecture, we will denote the weight lattice by and the root lattice by . In addition, we will denote the -weight space of a module by , and will be the Verma associated to . Also, we will need the notion of the universal enveloping algebra, which we will not write down here.
2.1 Definitions
Recall that is the full subcategory of of modules such that
- 1.
is finitely generated over .
- 2.
is -semisimple.
- 3.
is locally -finite.
- 4.
.
- 5.
The set of weights of is contained in some finite union of cones .
Theorem 2.1.1.
The following properties hold for :
- 1.
is Noetherian.
- 2.
is closed under submodules, quotients, finite direct sums, and is abelian.
- 3.
is closed under tensoring with finite-dimensional representations (in fact, if tensoring with is exact and lands in , then must be finite-dimensional).
- 4.
is locally -finite.
- 5.
All are finitely generated over .
2.2 Highest weight modules
Definition 2.2.1.
A vector is a maximal vector if .
Definition 2.2.2.
A module is a highest weight module if there exists a maximal generateing .
Definition 2.2.3.
Let and consider the -module . Then the Verma module for is the module
Note that we have the standard adjunction .
Theorem 2.2.4.
For any highest weight module with highest weight ,
- 1.
, and in particular is -semisimple.
- 2.
All weights of are at most .
- 3.
For any , , and . In addition, .
- 4.
Any quotient of is also a highest weight module with highest weight .
- 5.
Any submodule of a highest weight module with weight is a proper submodule. if is simple, all maximal vectors have weight .
- 6.
There exists a unique maximal submodule, and thus has a unique simple quotient and thus is indecomposable.
- 7.
All simple highest weight modules with highest weight are isomorphic, so .
Corollary 2.2.5.
Let . Then admits a filtration whose successive quotients are highest weight modules.
2.3 Verma modules
Let be a verma module and be the unique simple quotient, and be the unique maximal submodule.
Theorem 2.3.1.
Any simple is isomorphic to for some .
Proposition 2.3.2.
Let be the set of simple roots and . Let such that . Choose a maximal vector. Then
In particular, there exists a nonzero morphism .
Lemma 2.3.3.
We have the commutation relations
2.4 Examples
We will discuss the example of . Let be the operator that outputs the -th diagonal of a matrix. Then let be the root. Let be the matrix such that . In particular, we have .
Then note that if we choose units so that , then . If , then the Verma module for has weights , and the simple module has weights . For non-integral weights, we just have an infinite-dimensional representation. To see this, note that hitting any non-integral weight with will not reach another maximal vector.
2.5 Finite-dimensional modules
Theorem 2.5.1.
For any weight , if and only if is a dominant integral weight. This is equivalent to for all .
This result tells us that weights of are actually symmetric under the Weyl group.
Proof.
First, if the span of is finite-dimensional for , then all of stabilizes . This is because if , then
and thus .
Next, if , then after restriction to , we have , and thus .
Now suppose that . Then after restricting to , the span of is isomorphic to , and in particular it is finite-dimensional. Next, we show that is a sum of finitely many -summands. To see this, consider the sum of all -submodules of . But then if we denote a summand by , we note that is a finite-dimensional representation of , so the natural morphism
lands inside . But then because is a nonzero submodule.
Next, recall that for , then the sets of weights are invariant under reflection across the origin, so we have isomorphisms
Because is a sum of finite-dimensional representations of , for any , consider the finite-dimensional -module containing it. If we add them up, we know that is some finite-dimensional -module . But then the generate , and thus for all , .
Finally, for any orbit of , there exists exactly one representative in the dominant weight lattice, and because there are only finitely many dominant integral weights less than , there must only be finitely many orbits, so is finite-dimensional. □
2.6 Central actions
Here, we will consider the action of on a module . Suppose that is a highest weight module for weight . Then we note that
and therefore . Therefore acts by on any highest weight module of weight , and we call the function a central character. In general, all algebra morphisms arise in this way. Then we have the decomposition
and write for the morphism killing . Then , so is an algebra homomorphism, and we will call this , the Harish-Chandra morphism. In particular, we obtain a morphism
Also, we will consider , where , where
If we can identify the two morphisms on a Zariski-dense subset, they must agree in general. First, note that , and now it suffices to show that for .
To prove this, if there exists such that , then , and thus . In addition, if , we have , and if , we can reverse the roles of because
2.7 More on Harish-Chandra
Consider the twisted Harish-Chandra morphism
This gives us a morphism . In particular, we have
Theorem 2.7.1.
The image of is contained in .
To see this, note that
Theorem 2.7.2.
- 1.
is an isomorphism ;
- 2.
If are linked, then ;
- 3.
Every element of arises in this way.
There is a simple way to see the last two parts of the theorem if we assume some algebraic geometry.
For a central character , consider the module
where depends on . We have a decomposition
which gives us
Now we may define subcategories of given by
Some examples are that all highest weight modules of weight are contained in .
Proposition 2.7.3.
We have a decomposition
We will now consider blocks of category . We say that simple modules are in the same block if there is a nontrivial extension of by . For general , we know that has a finite Jordan-Hölder decomposition because is Artinian, so is in some block if all of its Jordan-Hölder quotients are.
Proposition 2.7.4.
If , then is a block of .
To prove this, if are linked, then we have the diagram
giving us an exact sequence
3 Che (Oct 13): Formal characters and applications to finite-dimensional modules
3.1 Weyl character and dimension formulas
Today we will see what Category tells us about finite-dimensional modules. We will fix a semisimple Lie algebra .
Definition 3.1.1.
Given , define the function , where .
Also let be the characteristic function of . Now given , define the convolution product
For example, . Here, we assume that are supported on a finite union of things of the form , where are the non-negative weights. We will call the set of such functions .
Proposition 3.1.2.
- 1.
If is an exact sequence, then .
- 2.
If and is finite-dimensional, then .
Proof.
- 1.
Note that for any such exact sequence.
- 2.
Note that . □
Last time we considered central characters for a highest weight module of weight and highest weight vector
If is a subquotient of the Verma module , then, then . Equivalently, and are linked by some element of the Weyl group. Because is Artinian, for all we have a finite filtration
where all is simple, so they must be isomorphic to some . But then we have
Our goal is now to compute the .
Let be weights linked to lambda arranged such that if , then . Then we should have some identity of the form
This implies that . Note that .
Definition 3.1.3.
Define the Kostant function
Proposition 3.1.4.
. More generally, .
Proof.
By the PBW theorem, we know that is spanned by . This is apparently equivalent to the definition of . □
Definition 3.1.5.
Define the function
Also define
Note that . Also note that
Next, and if is a simple root, then . The reason for this is that but fixes the other positive roots. This implies that .
Theorem 3.1.6 (Weyl character formula).
If , then
Proof.
If we apply to the formula
we obtain
Because , we know that is finite-dimensional and all weight spaces are symmetric. If is a simple root, then we apply to both sides, and we obtain
because . Therefore we see that , so . □
We would now like to compute for dominant integral weights . We want something like
except that both sides here vanish, so this is too naïve. For example, if we consider , we have
If we divide, we actually obtain .
In the general case, let and . Define by extending linearly. Applying to the Weyl charaacter formula, we obtain
Note here that . In the limit, we have and . Therefore we obtain
This is called the Weyl dimension formula.
3.2 Maximal submodules of Verma modules
Theorem 3.2.1.
Let and be simple roots of . Then
is the maximal submodule of .
Remark 3.2.2.
Last time we saw that .
Lemma 3.2.3.
Let . Then
This is proved by induction, so like a certain Fields medalist, we omit the proof.
If correspond to roots , then note that eventually we will have . In fact, four times is enough.
Lemma 3.2.4.
Let be a simple root. Then for any , there exists such that in .
Proof.
We proceed by induction. Suppose that . When , then . For , we have
The first term on the right hand side vanishes by the inductive hypothesis, and the second term becomes by the inductive hypothesis. □
Proof of Theorem.
By the discussion last time and the second lemma, we know that is finite-dimensional. This implies that , but we know that the left hand side is a highest weight module, so . □
Remark 3.2.5.
4 Kevin (Oct 20): Duality and projectives in category
4.1 Duality
Recall from the finite-dimensional story that -representation have duals with -action
for . This is not well-behaved for infinite-dimensional representations (for example, ), so in this case we would like to construct a better-behaved duality functor.
Note that every semisimple Lie algebra has a transpose (if is a matrix Lie algebra, this is literally the transpose) which is an anti-automorphism. Here, we have
This allows us to define1
Definition 4.1.1.
Let . Then the dual of is defined by
Proposition 4.1.2.
.
Proof.
To prove finite generation, note that has finite length (here, because duality preserves formal characters and exchanges quotients and submodules). Clearly the weight spaces are finite-dimensional by assumption, and the weights lie in some union because formal characters are preserved, so we have local -finiteness. □
Here are some more facts about duality.
- •
Duality is a contravariant functor. This is obvious because everything is defined on the level of weight spaces.
- •
There is a natural isomorphism . This is clear because we are taking double duals of finite-dimensional things and adding them up, so in particular duality is an anti-equivalence of categories.
- •
We have . On the other hand, duality for is complicated.
- •
fixes by an exercise in Humphreys.2 In particular, this means that .
4.2 Projectives
Recall that is projective if is right exact.3 Our goal is to prove that has enough projectives (which will mean that we can do homological algebra). The first thing we will do is introduce dominance and antidominance.
Recall that for , contains one dominant weight and one antidominant weight. This gives us two(!) good choices for representatives of . Unfortunately, we care about nonintegral weights,4 and we cannot choose representatives of for general .
Remark 4.2.1.
From now on we will use the action (because this is all we care about), and therefore our new notion of (anti)dominance will not restrict to the old notion of dominance.5
Definition 4.2.2.
A weight is dominant if for all . A weight is antidominant if for all .
Note that this is not the same as the undotted definition. For example, is dominant. Also, the set can have multiple dominant and/or antidominant weights.
Definition 4.2.3.
We define the subgroup
where is the root lattice. We also define
In fact, is the Weyl group of . We may similarly define .
Proposition 4.2.4.
The following are equivalent:
- 1.
is dominant.
- 2.
for all .
- 3.
for all .
- 4.
for all .
Proof.
Clearly 1 implies 2, and 2 implies 1 because positive roots are sums of simple roots with nonnegative coefficients. To prove that 2 is equivalent to 3, note that
Finally, to see that 3 is equivalent to 4, note that 4 implies 3 automatically. To prove that 3 implies 4, we induct on . If with , we see that
It is clear that while is a nonnegative multiple of . Because of the length condition, we see that is positive. □
Corollary 4.2.5.
The orbit has a unique (anti)dominant weight.
Proof.
This is because 1 is equivalent to 4 in the proposition. □
Theorem 4.2.6.
- 1.
If is dominant, then is projective.
- 2.
If is projective and is finite-dimensional, then is projective.
- 3.
has enough projectives.
Proof.
- 1.
Consider and suppose is a maximal weight vector with weight (coming from a map ). Assume that . Our goal is to lift to a maximal weight vector in , but because is surjective, we can lift to . If is maximal, then we are done, so suppose that is not maximal.
In this case, there exists such that is a maximal vector with weight greater than . However, this weight must be linked to , so by dominance of , it cannot exist.
- 2.
Here, we use the tensor-Hom adjunction
Because is exact and is exact, the functor is exact and thus is projective.6
- 3.
The first thing we want to do is to find projectives mapping onto . For large , is dominant. This implies that is projective, but then is projective.
In fact, there exists a surjection . To see this, if is a -module and is a -module, then
This is known as the tensor identity and is apparently not obvious unless you have the arrogance level of a certain Chinese mathematician. Because , we obtain
The surjection comes from the fact that the lowest weight of is , so we can kill all of the higher weights.
The rest of the proof is simply homological algebra. For a general , because is Artinian, we can induct on the length of . First consider a short exact sequence
By assumption, there exists a surjection , and this morphism lifts to . If does not surject onto , then cannot intersect (otherwise it would contain all of and thus surject onto ). This implies that , which splits the exact sequence. □
By standard homological algebra, because is Artinian and has enough projectives, then has projective covers (i.e. unique minimal projectives surjecting onto ). If we define to be the projective cover of , the are precisely the indecomposable projectives. Therefore every projective is a direct sum of .
Theorem 4.2.7.
- 1.
has a standard filtration, which is a filtration with subquotients that are Verma modules.
- 2.
(BGG reciprocity) The multiplicity of in the composition series for is given by
5 Fan (Oct 27): more on tHe strUcture, simplicity criterioN, ANd embedding of verma modules
We will be discussing the following results during this lecture:
Theorem 5.0.1.
Let . Then if and only if is -antidominant (this is the notion of antidominance that we discussed last time).
Theorem 5.0.2.
For any , if has , then .
5.1 Basic facts
Recall that the socle of a module is defined to be the direct sum of its simple submodules. A fact from ring theory1 is that if is left noetherian and has no right zero-divisors, then any two left ideals intersect nontrivially.
Proposition 5.1.1.
for all , has a unique simple submodule.
Proof.
Recall that , but then simple submodules of are ideals in , which intersect nontrivially. □
Theorem 5.1.2.
For any ,
- 1.
Any nonzero is an injection.
- 2.
.
- 3.
If is the unique simple submodule, then .
Proof.
- 1.
is determined by the image for some of the highest weight vector. But then , and this follows from the fact that has no zero divisors.
- 2.
If are morphisms and is the unique simple submodule of , then are both simple and thus isomorphic, so there exists such that , and so because it is not injective.
- 3.
Suppose is the unique simple submodule of . Then we have the sequence
and so must inject into and thus . □
Proposition 5.1.3.
For , if is a reduced word expression, then there exists a sequence of embeddings
Proof.
We will induct on . Choose some . Recall that if and only if . Note that . Thus if , then . We now compute
Then we know . □
5.2 Proof of first theorem
If is -antidominant, we will prove that . Recall that antidominance is equivalent to for all . But the Jordan-Holder factors of look like for with is linked to . This is the same as . But then the only weight that can appear is , so .
Now we will prove that if , then is antidominant. This is hard, so we will assume for now that . Because is simple, then . Now assume that for some , . Then
and so . This gives a contradiction.
We now finish the proof in the general case. Assume there exists such that . Then . By the second theorem, we have an injection , and thus is simple.
5.3 More basic facts
Proposition 5.3.1.
Let and . If , then:
- •
If , then .
- •
If , then .
Proof.
In the first case, then we know that
and so .
In the second case, let be maximal vectors of . Then we know that
and because , is a maximal vector. Also because , there exists such that .
It is a fact that if is a nilpotent Lie algebra and , for any there exists such that . Applying this fact to , , and , there exists such that
But then
If , then we know . On the other hand, if , we have
but this is equal to , so we are done. □
5.4 Proof of second theorem
We will prove this result in the integral case. Assume that and . If , then there exists such that . Thus there exists such that if we write , then
If we define and analogously and assume that and thus , then we have
where so that
Note that this is the difference of a multiple of and a sum of positive roots.
Continuing, note that are linked to . These satisfy , , and . Thus there exists such that and . But then
and in fact we obtain a multiple of . Now there exists and , so . To construct this, note that
This implies that flips and thus . Also, . Therefore,
But then the coefficient of in is positive, and thus .
We then have , and therefore either or , and so we are done.
We sketch the rest of the proof. Fix and is an integer. Then define
Then define . We know that . But now is Zariski-dense in , so we need to prove that is a closed subscheme of . But now exactly when there exists such that . But now , and this is equivalent to
and this is equivalent to
Finally we have . If we write , we have a map
by an argument about passing from to . But now our chain of equivalences continues to for all . But now is given by the rank of the map being lower than usual, so it is a closed subscheme.
6 Patrick (Nov 03): CAts, bOndage, and why you can’T do representAtioN theory without Geometry
Note: these are the speaker’s notes.
6.1 Introduction
Our goal is to prove the following theorem, due to Bernstein-Gelfand-Gelfand.
Theorem 6.1.1 (BGG).
Let .
- 1.
If is strongly linked to , then . This implies .
- 2.
If , then is strongly linked to .
We first need to define what it means for weights to be strongly linked. Then we will state a result of Jantzen that we will use to prove this theorem. Next, we will do some linear algebra before proving the result of Jantzen. Finally, we will use Jantzen to prove BGG. Note that this is a different order than in which Humphreys does things.
Example 6.1.2 (Cursed example).
Consider . Number the simple roots and fundamental weights as usual. This means that the simple roots and fundamental weights are
Write as . This means that
Consider . Then we note that but if , then . But this implies that
and therefore . However, apparently there is no embedding of inside .
The reason for this is actually that if you inspect the Bruhat order for . The weight in linked to is . Here, we note that while . However, and are incomparable.
6.2 Strong linkage
The first thing we need to do is to define what it means for two weights to be strongly linked.
Definition 6.2.1.
Let . Write if of there exists a root such that . Note that this is equivalent to . More generally, if or there exists such that
we say that is strongly linked to and write .
Note that if we repeatedly apply Theorem 4.6 (the second big theorem in Fan’s lecture), we immediately obtain the first part of the theorem. The second part of the proof is more involved.
Before we proceed, we need to discuss the Bruhat order on the Weyl group (in fact, this can be defined for any Coxeter group, but we will not need this). Let be a regular integral weight which is antidominant. Then suppose that is a positive root and that for some . By the second main theorem of the previous lecture, we know that
But we know that this is equivalent to
Because is antidominant, we know . This tells us that if , then
All of this is reversible, so by definition we see that if and only if . This implies, after we prove the main theorem, that
Corollary 6.2.2.
Let be a regular antidominant weight and let . Then if and only if .
6.3 Jantzen filtration
Proposition 6.3.1 (Jantzen).
Let . Then has a filtration by submodules
such that for large enough and the following conditions hold:
- 1.
Each nonzero quotient has a nondegenerate contravariant form in the sense of 3.14.
- 2.
is the unique maximal submodule of .
- 3.
At the level of formal characters, we have
This filtration is called the Jantzen filtration and the formal character formula is called the Jantzen sum formula. We will write
Also note that the summation on the right side of the sum formula is over a set of positive roots which we will call . In fact, if but , the sum formula tells us that .
Example 6.3.2.
Let be regular, antidominant, and integral. Then we know that is the set of all satisfying . But this is the same as
but of course this is equivalent to by defintion of being integral and antidominant. Therefore in fact here .
Here are some reasonable questions about these things:
- 1.
Is the Jantzen filtration unique relative to properties such as the sum formula and existence of nondegenerate contravariant forms on the quotients?
- 2.
What are the composition factor multiplicities in each filtration layer ?
- 3.
Are the filtration layers semisimple? If so, does the filtration coincide with one of the standard module filtrations with semisimple quotients?
- 4.
How does the Jantzen filtration behave with respect to ?
A more precise formulation of the last question is called the Jantzen conjecture, and this can only be proved (as of whenever the book was published) using geometric tools like Beilinson-Bernstein localization. For completeness, it is stated here:
Conjecture 6.3.3 (Jantzen).
Suppose that and set . Then if while if .
Example 6.3.4.
Let . Suppose that is regular, antidominant, and integral. We will write down the Jantzen filtration for and then deduce that all composition factors have multiplicity . Let be the simple roots and write . The sum formula tells us that
But this implies that the Jantzen filtration is
This implies that there are four composition factors each with multiplicity .
6.4 Proofs of results
We will now prove Jantzen’s theorem and apply it to prove the main theorem. I assume that everyone is familiar with the basic theory of finitely generated modules over a principal ideal domain.1
Let be a principal ideal domain and be a prime element. Suppose that is a free -module of rank with a nondegenerate symmetric bilinear form . Let be the determinant of the bilinear form – this is well-defined up to a unit. Let , which is a vector space over . For , define
For notation, we will write and for the image of under the bilinear form. We will write for a basis of and for the dual basis.
Lemma 6.4.1.
- 1.
We have the identity
- 2.
For all , the modified bilinear form induces a nondegenerate form on .
Proof.
Write for some . Then writing , we know that . On the other hand, we know that if and only if
for all . But if we write , then we know that is spanned by for all where together with for the such that . But this implies that
and in particular for .2 This means the sum is defined, and therefore we have
For the second part, it ic clear that this form takes values in , but we need to check that there is an induced form on . But here, if , we see that
But if , the same argument gives us . This gives us a form on . Finally we know that if , then , and thus we conclude that the form is nondegenerate. □
Before we continue, we will state some facts about contravariant forms that were not covered by Kevin.
Definition 6.4.2.
Let be a symmetric bilinear form on a module . Then is contravariant if for all and .
It is not even clear that these exist, but here are some properties:
- 1.
If has a contravariant form, then for .
- 2.
If is a highest weight module, then it has a unique (up to scalars) nonzero contravariant form.
- 3.
If is a submodule, then is also a submodule.
- 4.
Any contravariant form on a highest weight module must induce the zero form on its maximal submodule. Moreover, the maximal submodule is the radical of the form and the form is nondegenerate if and only if is simple.
Proof of Jantzen.
The reason we did all of that linear algebra above is because we will extend to the field . Also write . Then apparently we can adapt the entire theory of Lie algebras to and . I will not check this, and you should not ask me to do so. We will use the lemma to construct filtrations here and then kill to obtain the Jantzen filtration.
Let be a weight and write . Clearly is antidominant, and thus is simple. But now we know that the contravariant form on is nondegenerate by Theorem 3.15 in Humphreys. Here, recall that the contravariant form satisfies for all and the identity
Now recall that is an -form. This induces an -form . Each weight space is a free -module of finite rank and the contravariant form is nondegenerate after restricting to each weight space. If is the space of positive linear combinations of simple roots, then write for the -form of . Set
These form a decreasing filtration of .
Now we can quotient by the ideal and we obtain a decreasing filtration on . By the lemma, the quotients have induced nondegenerate contravariant forms. Also, we know that the filtration of each individual weight space eventually ends at , and thus for large enough we have because only finitely many weights are linked to .
Next, we note that is a highest weight module with a nondegenerate contravariant form and therefore is simple. Because has a unique simple quotient, we see that . First, we will express the sum
in terms of the determinants of the contravariant forms on the steps of the filtration. Here, by some magic involving Shapovalov elements that was discovered by Jantzen and Shapovalov independently while working on their PhD theses, the determinant of the contravariant form on the weight space of is given by
where is the Kostant partition function. Note that the Kostant partition function gives the number of ways to write a weight as a sum of positive roots with non-negative coefficients. This is actually unique only up to a unit, but this means that is well-defined. But now we know that
This is not a multiple of unless . But this means that and for this term is . But this implies that for fixed and , the contribution is given by
Finally, we compute formally and obtain the result
This comes from a variable change in . But now we precisely have
and therefore if we fix the sum becomes
as desired. □
Remark 6.4.3.
Remember that disappears when we kill . Remember we only needed the fact that . We may consider what ahppens if we replace by another weight, but a definitive answer to this is only provided by Belinson-Bernstein using a geometric approach.
Proof of BGG.
We will finally prove the BGG theorem. We will induct on the number of linked weights . If is minimal in its linkage class, then we know , so we are done. Now suppose that and . But this implies that
But now the sum formula for the Jantzen filtration gives us
for some . By the inductive hypothesis, there exists such that
Because , we are done. □
7 Fan (Nov 10): why You shoUld care about the bgg resolutioN, aNd more on homologicAl consideratioNs in category o
7.1 The BGG resolution
Let be a dominant (in the old sense) weight. A BGG resolution is something of the form
Later, we will see that there is something called the BGG resolution, but we will probably not prove it.
Theorem 7.1.1 (Weak BGG).
There exists a resolution
such that the standard filtration of consists of the weights .
A fact is that For all subalgebras there exists a resolution of the trivial resolution
where the differentials are given by
Lemma 7.1.2.
For a finite dimensional representation of which is -semisimple, the standard filtration of contains precisely the weights of , respecting multiplicity.
Recall the following theorem from your first course on Lie algebras:
Theorem 7.1.3 (Lie).
For a finite dimensional representation of a solvable , there exists
such that , and there exists a nonzero and such that for all .
Proof of lemma.
Using the previous theorem, we have a filtration
and the quotients are , where is a weight of . □
Lemma 7.1.4.
For a central character and having a standard filtration, we have
Proof.
Choose a standard filtration
such that . Then we have
a standard filtration of , and also we note that . □
Proof of weak BGG for .
Consider the resolution
Taking central characters with , we have a new resolution
But we also know that
Now recall that . Also we will use the fact that if has
then . This tells us that
To prove the general case, we will use the translation functor . This is given by projecting onto , tensoring by , and then including into .
Proof in general.
Let . Then recall we have a resolution
Tensoring by , which is finite-dimensional, we obtain
Finally, projecting to , we have
Now if we consider , we have the filtration
Also we know that . We will use the fact that if , then . Therefore
Projecting onto the central character , we have
Now we know that for some and this means that
and thus . Because and , we know that
Using the previous equality for , we see that , which means that and . □
Theorem 7.1.5.
Let .
- 1.
If , then and .
- 2.
If and , then .
Proof of Strong BGG.
We induct on . The base case is length . Here, we have
Because have the same length, the vanishes, and so the exact sequence must split. In the inductive step, we can pull the direct sum out of the Ext functor, and each individual Ext vanishes, so we are done. □
This argument in fact proves that weak BGG is strong BGG. Note that this was not available to BGG in their original paper.
Proof of theorem.
By proposition 3.1 of Humphreys, we know . Thus there exists a non-split extension
We know that there exists a lift . If , then it must be and so , and so it splits. This means that . Now take the standard filtration
We know that contains weights at least with appearing exactly once. But then by BGG reciprocity, we know that
and this is equivalent to .
Because . We can consider the images . Choose the smallest such that . Thus has a submodule isomorphic to the image of If we quotient out more stuff, we get a copy of inside . But now , and thus . By definition, .
Now we will state another fact. Let be integral, dot-regular, and -antidominant. Then if and only if . □
7.2 Why you should care
We will now explain why you should care about the BGG resolution.
Theorem 7.2.1 (Bott, Kostant).
Let . Then
as -modules, where is the longest element of .
Proof.
Recall that . But we know that . We now consider the BGG resolution of , and so we have
Note that . Therefore
□
Apparently there are two facts that if differ by length , then either there are two between them or there is nothing. Also, the BGG resolution is apparently unique.
Next, for a -antidominant weight, we can define .
Theorem 7.2.2.
The projective dimension of is , the projective dimension of is , and the homological dimension of is .
We will now state some more facts. First, if have standard filtrations, then .
Theorem 7.2.3.
The following are equivalent:
- 1.
has a standard filtration;
- 2.
for all ;
- 3.
for all ;
- 4.
();
- 5.
().
Proposition 7.2.4.
We have the identity
8 Fan (Nov 17): VERmas and sImpleS undEr the translation functor, facets, chambers, anD walls
The reason we should care about translation functors is the following:1 for any antidominant , is self dual. Also,
Before we continue, we will do some more homological stuff. If we consider , this is actually the same as . Of course, this does not remain true in the derived category, but we also have . If we consider
we need to be careful about how we derive things. Recall that Lie algebra cohomology is computed by the Koszul complex
Note that is most definitely not -finite, but somehow can be computed in the category of -modules that are -semisimple and locally -finite. Now we have
8.1 The actual content of this lecture
Definition 8.1.1.
Let such that . Then . Denote by the representative of the orbit of under the undotted action. Then the translation functor is given by
Proposition 8.1.2.
is exact, preserves projectives, commutes with taking duals, and has a standard filtration.
Proof.
Exactness is obvious. Preserving projectives and commuting with duals are both from Kevin’s talk, and having a standard filtration follows from having a standard filtration follows from
and the only that works is . □
Proposition 8.1.3.
For compatible, the functors are both left and right adjoints.
From now on, all chambers will be discussed with respect to .
Definition 8.1.4.
A facet is a subset of determined by . is given by
We can define by making our inequalities weak. We can also define by allowing and by allowing .
We will assume the following facts without proof:
- 1.
is a root system. It has the Weyl group given by the subgroup of fixing pointwise.
- 2.
If there exist such that , then for all .
For any , let be spanned by over . Let be such that for all , .
Definition 8.1.5.
Define .
We will use the following facts without proof:
- 1.
For all , and .
- 2.
We have .
- 3.
If , then .
Lemma 8.1.6.
For compatible with and , for all , .
Theorem 8.1.7.
Let be antidominant and compatible. For and ,
for all .
Proof.
We consider . The standard filtration of this is given by
Also, because . But . Also . By the lemma, for any , . But now only can appear in the standard filtration, and so we are done. □
Corollary 8.1.8.
For the setting of the theorem, if has a standard filtration, then has a standard filtration.2
Proof.
Induct on the size of the standard filtration. Consider
Applying the translation functor, we have
By the inductive hypothesis, we are done. □
Proposition 8.1.9.
In the same setting, we have
Proof.
Consider . Applying the translation functor, we have
If is nonzero, it is a highest weight module. But now we have
and this becomes
This gives a morphism . But now recall that , and therefore, . □
Theorem 8.1.10.
Consider the same setting as before. Then if and only if .
Proof.
We know that is exact, so it does not increase the length of the Jordan-Hölder filtration. Thus it brings a unique to . But , and thus .
Now let be a facet in . Suppose that . For all , . Thus and . But now or , and we know that , so . Thus
for all . Because is the Weyl group of , for all . Thus , so . Therefore brings to .
If , we know that . But now because
except
9 Kevin (Dec 01): Kazhdan-Lusztig theory
So far, we have cared about the Verma modules
Conjecture 9.0.1 (Kazhdan-Lusztig).
For all
where
where
This result was proved by Beilinson-Bernstein and by Brylinski-Kashiwara using the theory of
9.1 Combinatorics
Definition 9.1.1.
Let
- •
ifT w T w ′ = T w w ′ ;ℓ ( w w ′ ) = ℓ ( w ) + ℓ ( w ′ ) - •
if( T s + 1 ) ( T s − q ) = 0 is a simple reflection.s
The Hecke algebra is defined to be
and
Apparently this comes from number theory,1 and you are supposed to specialize
- •
If we set
,q = 1 specializes toℋ .ℤ [ W ] - •
If we set
to be a prime power, thenq is the algebra of intertwining operators on the space of functions on the flag variety ofℋ ⊗ ℤ [ q ± 1 / 2 ] ℂ .G ( 𝔽 q )
The Kazhdan-Lusztig polynomials
Theorem 9.1.2 (Kazhdan-Lusztig).
For any
The
Proof.
First define
We can check that
The relations and the fact that
Now we will prove that if the
Comparing coefficients of
After cancelling, we obtain
Note that the first term on the left has only positive powers of
We will now prove the existence of the
Clearly,
Here,
9.2 Geometry
Let
We can check that
Suppose we have a variety
- •
Each
is closed;X i - •
is smooth of pure dimensionX i − X i − 1 ;i - •
The stratification is a Whitney stratification.
Given such a stratification, let
Denote the inclusions
This is independent of the choice of Whitney stratification. Of course, we may now define intersection cohomology
We can relate intersection cohomology to the Kazhdan-Lusztig stuff as follows:
Theorem 9.2.1 (Kazhdan-Lusztig).
- 1.
We have
P y , w ( q ) = ∑ i dim IH y 2 i ( ℬ ¯ w ) q i , where
means relative toIH y .ℬ y ⊆ ℬ ¯ w - 2.
We have the relation
∑ i = 0 ℓ ( w ) IH 2 i ( ℬ ¯ w ) q i = ∑ x ≤ w q ℓ ( x ) P x , w ( q ) .
Corollary 9.2.2.
10 Patrick (Dec 08): Koszul duality for people who aren’t Peter May
Note: these are the speaker’s notes.
Before we begin, we will establish some notation.
Notations 10.0.1.
Let
10.1 Koszul duality
Definition 10.1.1.
A graded ring
such that
Example 10.1.2.
The Koszul resolution
shows that
There is an alternative characterization of Koszul rings.
Definition 10.1.3.
Let
In the cases we will consider, any simple
Proposition 10.1.4.
Let
- 1.
is Koszul.A - 2.
For any pure
-modulesA of weightsM , N we havem , n wheneverext A i ( M , N ) = 0 .i ≠ m − n - 3.
We have
wheneverext A i ( A 0 , A 0 ⟨ n ⟩ ) = 0 .i ≠ n
We will list some more properties of Koszul rings:
Proposition 10.1.5.
If
Proof.
Define
such that
is an injective resolution of
whenever
Definition 10.1.6.
A graded ring
is surjective and the kernel
Proposition 10.1.7.
Any Koszul ring is quadratic.
We will now assume that all
Definition 10.1.8.
Let
Remark 10.1.9.
Clearly
Proposition 10.1.10.
Let
Proof.
We will use the fact (without proof) that the Koszul complex of
with differential given by
Therefore, we have a bigraded space
where the differential has bidegree
where the differential has degree
Theorem 10.1.11.
Let
We will write
We will now give a numerical criterion for a ring to be Koszul. Suppose that the
Lemma 10.1.12.
Suppose that
Theorem 10.1.13.
Let
Now we are finally able to define a derived category version of Koszul duality, which is what we really wanted. Write
Theorem 10.1.14.
Let
We will not prove this result, but we will construct a functor. Let
with the differentials coming from the Koszul complex and from
where
In fact,
The inverse functor is given as follows. Let
This will have anticommuting differentials given by
Then we consider the total differential
One can check that
We will write
Theorem 10.1.15.
There is a canonical isomorphism
Theorem 10.1.16.
Let
10.2 Parabolic-singular duality
Let
Let
Then the simple objects in
Let
The main theorem of this part of the talk is
Theorem 10.2.1.
Suppose that
where the summations are over
Note that
Corollary 10.2.2 (Soergel ’90).
Let
Moreover,
The proof of the theorem is geometric and developing the machinery would take to long, so we will prove the equality of dimensions for the first isomorphism. Using BGG reciprocity and the Kazhdan-Lusztig conjectures, we have
By an argument involving localization (translating the problem to geometry), this is precisely the dimennsion of
Now write
Corollary 10.2.3.
There is a ring isomorphism
Proof.
By the theorem, there is an equivalence
Proposition 10.2.4.
The rings
Proof.
First, note that
This tells us that
To prove this, consider that
To prove that
In fact, we have
and we have
by some magic of Kazhdan-Lusztig, which says that
Remark 10.2.5.
You could also prove that