BN-pairs

Arun Ram
Department of Mathematics and Statistics
University of Melbourne
Parkville, VIC 3010 Australia
aram@unimelb.edu.au

Last updated: 27 November 2014

BN-pairs

A group with BN-pair is a quadruple (G,B,N,S) where G is a group,
B is a subgroup of G,
N is a subgroup of G,
S is a subset of the coset space N/(BN),
and the following axioms are satisfied:

(T1) BN generates G,
BN is a normal subgroup of N,
(T2) S generates the group W=B/(BN) and
each sS has order 2,
(T3) For each sS and wW, (BsB)(BwB)BwBBswB,
(T4) For each sS, (BsB)(BsB)=BBsB.

Every wW is of the form w=n(BN), nN. By BwB is meant the double coset BnB. Since for nN, the double coset BnB depends only on n modulo BN, this is well defined. Similarily we can define wB, Bw, w-1Bw by, respectively, nB, Bn, n-1Bn.

Taking inverses shows axiom (T3) is equivalent to:

For each sS and wW, (BwB)(BsB)BwBBwsB.

The elements sS are called simple reflections. Every element in wW can be written as a product, w=s1sq, of simple reflections.

(Length) For wW, its length, denoted l(w), is the minimal q such that w=s1sq, siS.

(Bruhat Decomposition) The group G is a disjoint union over wW of the double cosets BwB. G=wWBwB, (disjoint union) is called the Bruhat decomposition.

Proof.

G=wWBwB will be proved below.

We show here that Bw1B=Bw2B implies w1=w2.

We prove this by induction on q, assuming l(w1)l(w2)=q.

First note BwB=B implies w=1. This settles the case q=0.

Now suppose Bw1B=Bw2B with l(w1)l(w2)=q1. Then we can find an sS such that l(sw2)=q-1. (If we write w2=s1sq as a product of q simple reflections, s=s1 will do.)

So l(w1)>l(sw2)=q-1. Hence by induction Bw1B and Bsw2B are different, so disjoint.

But (Bsw2B)(BsB)(Bw2B)=(BsB)(Bw1B)(Bw1B)(Bsw1B).

Hence Bsw2B=Bs1wB, l(sw1)l(w1)-1l(w2)=q-1.

By induction sw1=sw2, and we deduce w1=w2.

Let s1,s2,,sq and wW. Then (Bs1s2sqB)(BwB) 1i1<<ipq (Bsi1sipwB) Note: The union is over all substrings of 1<2<<q, including the empty string. Thus the union includes the double coset BwB.

Proof.

The proof is by induction on q using the third BN pair axiom.

The case q=0 is trivial.

Suppose q1. Then using BxyB(BxB)(ByB), (Bs1sqB) (BwB)(Bs1B) (Bs2sqB) (BwB). So by induction, (Bs1sqB) (BwB) (Bs1B) ( 2j1<<jpq (Bsj1sjpwB) ) = 2j1<<jpq (Bs1B) (Bsj1sjpwB).

The third BN-pair axiom (T3) says for sS, (BsB)(BwB) (BwB)(BswB) and thus we get (Bs1sqB) (BwB) 2j1<<jpq ( (Bs1sj1sjpwB) (Bsj1sjpwB) ) = 1i1<i2<<ipq (Bsi1sipB) which proves the lemma.

For IS set WI=s|sI.

PI=BWIB=wWIBwB is a subgroup of G. In particular G=BWB.

Proof.

It is clear that if xPI then x-1PI, since B and WI are both groups.

Suppose x and y are elements of PI. Then for some s1,,sqS and wWI, xBs1sqB andyBwB. Then xy(Bs1sqB)(BwB). So by the lemma xy1i1<<ipq (Bsi1sipwB) PI. Hence PI is a group.

If I=S then PI=BWB is a group containing both B and N. Therefore PS=G.

Let I,JS and let PI=BWIB and PJ=BWJB be as above. Then, for wW, PIwPJ=BWI wWJB.

Proof.

BWIwWJBPIwPJ because BWIPI and WJBPJ.

We now derive the opposite inclusion. Let t1,,tpI and s1,,sqJ. Then by the lemma (and its analogous right hand version), (Bt1tpB)(BwB)(Bs1sq) is contained in a union of double cosets, Bw1wW2 with w1PI and w2PJ.

Hence PIwPJBWIwWJB.

If w has length q, and w=s1,s2,,sq, then BwB=(Bs1B)(Bs2B)(BsqB).

Proof.

By induction on q. The case q=1 is trivial.

For q>1, s2sq and s1sq-1 each have length q-1. By induction Bs2sqB = (Bs2B) (BsqB) Bs1sq-1B = (Bs1B) (Bsq-1B) Hence by BN-pair axiom (T3), (Bs1B) (Bs2B) (BsqB) = (Bs1B) (Bs2sqB) Bs1sqB Bs2sqB, (Bs1B) (Bsq-1B) (BsqB) = (Bs1sq-1B) (BsqB) Bs1sqB Bs1sq-1B. If we can show Bs2sqBBs1sq-1=, then we deduce (Bs1B)(BsqB)Bs1sqB, and since the opposite inclusion is always true, that (Bs1B)(BsqB)=Bs1sqB.

But by Bruhat decomposition, Bs2sqBBs1sq-1 if and only if s2sq=s1s2sq-1 which on left multiplication by s1 gives s1s2s1=s2sq-1, contradicting l(w)=q.

For wW and sS there are a priori three possibilities for the length of sw or ws: either l(w), l(w)+1 or l(w)-1. In fact the first possibility never occurs, as the following proposition, which strengthens the previous, one shows.

Suppose wW and sS.
Then l(sw)l(w) if and only if (BsB)(BswB) =BswB, l(sw)l(w) if and only if (BsB)(BswB) =BswBBwB,
Similarly for multiplication of w by s on the right.

Proof.

We proceed by induction on q=l(w), the case q=0 being trivial.

Suppose l(sw)l(w)=q.

Write w=ws, l(w)=q-1, sS,

Then by induction (BsB)(BwB)=BswB and (BwB)(BsB)=BwsB=BwB.

Thus we have (BsB) (BwB) = (BsB) (BwB) (BsB) = (BswB) (BsB) (BswsB) (BswB)(axiom) = BswBBswB We cannot have (BsB)(BwB)=BswBBwB unless BwB=BswB. But then, by Bruhat decomposition we deduce, sw=w, giving l(sw)=q-1, contradicting l(sw)l(w)=q.

Hence (BsB)(BswB)=BswB.

Now consider the case l(sw)l(w).

This is equivalent to l(ssw)=l(w)l(sw). Then by the result above, BwB=BsswB= (BsB)(BswB). Hence we deduce (BsB)(BwB) = (BsB) (BsB) (BwB) = (BBsB) (BwB)(axiom) = BwB(BsB) (BwB) BwBBswB = BwBBswB. This together with (T3) shows (BsB)(BwB)=BwBBsw in this case.

(Cancellation Properties)

(a) Suppose w=s1s2sq, l(w)=q, and l(sw)<l(w).
Then for some i, sw=s1sˆisq.
Similarity if l(ws)<l(w), ws=s1sˆisq for some i.
(b) Suppose we express wW as a word in S: w=s1s2sq.
Then if l(w)<q, w=s1sˆisˆjsq, for some i,j.

Proof.

(a) Set wi=s1si. Then l(wi)=i.
Let i be minimal such that l(swi)<l(wi). Such i exist as this is the case for i=q. Hence l(swi-1)=i>i-1=l(wi-1) and l(swi)<l(wi). By the proposition above, (BsB)(Bwi-1B) =Bswi-1Band (BsB)(Bwi)B= BswiBBwiB. We calculate (BsB)(Bwi-1B) (BsiB)= { (Bswi-1B) (BsiB) BswiB Bswi-1B, (BsB)(BwiB)= BswiBBwiB. Thus we must have equality in the first case and Bswi-1B=BwiB. By Bruhat decomposition we deduce, swi-1=wi. This gives sw=ss1si-1 sis1=s1 si-1sisis1 =s1sˆisq.
(b) Put wj=s1s2sj. Let j1 be minimal such that l(wj)<j. (There is at least one such j, j=q, as l(w)=l(wq)<q). Then l(sjwj-1)=l(wj)j-1=l(wj-1). By the cancellation lemma there is an i such that s1sj-1sj=s1sˆisj-1. Postmultiplication by sj+1sq, gives w=s1sˆi sˆjsq.

We deduce immediately from the cancellation results:

If w=s1sp and l(w)=q, then w=si1siq for some sequence, 1i1<<iqp.

Since the elements of any WI are products of siI we deduce:

If wWI, IS, has length l(w)=q then w can be expressed w=s1sq, all siI.

This means that the length of any wWI, defined relative to WI, i.e. with respect to the generators sI of WI, is the same as its length defined with respect to the set of generators S of W.

In particular as l(w)=1 only if w=sS, we deduce:

We have sWI if and only if sI.

This generalizes.

Suppose l(w)=q and w=s1sq. Then wWI, IS, if and only if each siI.

Proof.

The if part follows from the definition of WI.

For the converse we work by induction on q=l(w). The case q=1 is the corollary above.

For q>1, s2sq has length q-1.

By the previous corollaries we can write w=t1t1, all tiI.

Then using the cancellation results: for some i, s2sq=s1t1tq=t1tˆjtqWI.

Hence by induction s2,,sqI. From this we deduce s1WI, which we know implies s1I, also.

From the above, given w=s1sq of length q, we have wWIWJ if and only if each siIJ. This together with the definition PK=BWKB, KS proves:

(a) WIWJ=WIJ.
(b) PIPJ=PIJ.

page history