79
ISSN 1364-0380
Geometry & Topology
G
T
G G TT TG T T
G
G T
T
G
G T GG TT
G T G
GG G T TT
Volume 2 (1998) 79{101
Published: 3 June 1998
Flag Manifolds and the Landweber{Novikov Algebra
Victor M Buchstaber
Nigel Ray
Department of Mathematics and Mechanics, Moscow State University
119899 Moscow, Russia
and
Department of Mathematics, University of Manchester
Manchester M13 9PL, England
Email: buchstab@mech.math.msu.su and nige@ma.man.ac.uk
Abstract
We investigate geometrical interpretations of various structure maps associated
with the Landweber{Novikov algebra S and its integral dual S . In particular, we study the coproduct and antipode in S , together with the left and
right actions of S on S which underly the construction of the quantum (or
Drinfeld) double D(S ). We set our realizations in the context of double complex cobordism, utilizing certain manifolds of bounded flags which generalize
complex projective space and may be canonically expressed as toric varieties.
We discuss their cell structure by analogy with the classical Schubert decomposition, and detail the implications for Poincare duality with respect to double
cobordism theory; these lead directly to our main results for the Landweber{
Novikov algebra.
AMS Classication numbers
Primary: 57R77
Secondary: 14M15, 14M25, 55S25
Keywords: Complex cobordism, double cobordism, flag manifold, Schubert
calculus, toric variety, Landweber{Novikov algebra.
Proposed: Haynes Miller
Seconded: Gunnar Carlsson, Ralph Cohen
Copyright Geometry and Topology
Received: 23 October 1997
Revised: 6 January 1998
80
1
Victor M Buchstaber and Nigel Ray
Introduction
The Landweber{Novikov algebra S was introduced in the 1960s as an algebra
of cohomology operations in complex cobordism theory, and was subsequently
described by Buchstaber and Shokurov [6] in terms of dierential operators on
a certain algebraic group. More recently, both S and its integral dual S have
been studied from alternative viewpoints [15], [18], [22], reflecting the growth
in popularity of Hopf algebras throughout mathematics. Nevertheless, the interpretations have remained predominately algebraic, although the underlying
motivations have ranged from theoretical physics to combinatorics.
Our purpose here is to provide a purely geometric description of S , incorporating its structure maps and certain left and right actions by S ; the importance
of the latter is their contribution to the adjoint action, which gures prominently in the construction of the quantum (or Drinfeld) double D(S ). We
work in the context of double complex cobordism, whose properties we have
developed in a preliminary article [8]. So far as we are aware, double cobordism
theories rst appeared in [20], and in the associated work [23]. To emphasize
our geometric intent we return to the notation of the 60s, and write bordism
and cobordism functors as Ω ( ) and Ω ( ) throughout.
The realizations we seek are provided by a family of bounded flag manifolds
with various double U {structures. These manifolds were originally constructed
by Bott and Samelson [4] without reference to flags or U {structures, and were
introduced into complex cobordism theory in [21]. We consider their algebraic
topology in detail, describing computations in bordism and cobordism theory
which provide the essential link with the Landweber{Novikov algebra, and are
related to the generalized Schubert calculus of Bressler and Evens [5]. These
results appear to be of independent interest and extend to the topological study
of other toric manifolds [3], [9], as well as being related to Magyar’s program
[17] for the description of arbitrary Bott{Samelson varieties in combinatorial
terms. We hope to record such extensions in a future work.
For readers who seek background information in algebra, combinatorics, and
geometry, we suggest the classic books by Kassel [14], Aigner [2], and Griths
and Harris [12] respectively.
We begin in section 2 by summarizing prerequisites and notation connected
with double complex cobordism, recalling the coecient ring ΩDU and its
subalgebra G , together with the canonical isomorphism which identies them
with the Hopf algebroid AU
and its sub-Hopf algebra S respectively. In section
3 we study the geometry and topology of the bounded flag manifolds B(Zn+1 ),
Geometry and Topology, Volume 2 (1998)
Flag Manifolds and the Landweber-Novikov Algebra
81
describing their toric structure and introducing the posets of subvarieties XQ
which serve to desingularize their cells. In section 4 we dene the basic U {
and double U {structures on XQ which underlie the geometrical realization of
G , and use them to compute ΩU (XQ ) and ΩU (XQ ); the methods extend to
double cobordism, although several aspects of duality demand extra care. We
apply this material in section 5 to calculate ΩDU {theory characteristic numbers
of the XQ , interpreting the results by means of the calculus of section 3. Under
the canonical isomorphism, realizations of the relevant structure maps for S and S follow immediately.
We use the following notation and conventions without further comment.
Given a complex m{plane bundle over a nite CW complex, we let ? denote
the complementary (n − m){plane bundle in some suitably high-dimensional
trivial bundle Cn .
We write AU for the algebra of complex cobordism operations, and AU
for its
continuous dual HomΩU (AU ; ΩU ), forcing us in turn to write S for the graded
Landweber{Novikov algebra, and S for its dual HomZ (S ; Z); neither of these
notations is entirely standard.
Several of our algebras are polynomial in variables such as bk of grading 2k ,
where b0 is the identity. An additive basis is therefore given by monomials of
the form b!1 1 b!2 2 : : : b!nn , which we denote by b! , where ! is the sequence of
nonnegative, eventually zero integers (!1 ; !2 ; : : : ; !n ; 0 : : : ). The set of all such
!
+! . Given any ! , we
sequences forms
Pan additive semigroup, and b! b = b
write j!j for 2 i!i , which is the grading of b . We distinguish the sequences
(m), which have a single nonzero element 1 and are dened by b(m) = bm
for each integer m 1. It is often convenient to abbreviate the formal sum
P
n
k0 bk to b, and write (b)k for the component of the nth power of b in grading
2k ; negative values of n are permissible.
When dualizing, we choose dual basis elements of the form c! , dened by
hc! ; b i = !; ; this notation is designed to be consistent with our convention on
gradings, and to emphasize that the elements c! are not necessarily monomials
themselves.
The authors are indebted to many colleagues for enjoyable and stimulating
discussions which have contributed to this work. These include Andrew Baker,
Sara Billey, Francis Clarke, Fred Cohen, Sergei Fomin, Sergei Gelfand, Christian
Lenart, Peter Magyar, Haynes Miller, Jack Morava, Sergei Novikov, and Neil
Strickland.
Geometry and Topology, Volume 2 (1998)
82
2
Victor M Buchstaber and Nigel Ray
Double complex cobordism
In this section we summarize the appropriate parts of [8], concerning the notation and conventions of double complex cobordism, and operations and cooperations in the corresponding homology and cohomology theories.
Double complex cobordism is based on manifolds M whose stable normal bundles are equipped with a splitting of the form = ‘ r . We refer to an
equivalence class of such splittings as a double U {structure (‘ ; r ) on M ,
writing (M ; ‘ ; r ) if the manifold requires emphasis. It is helpful to think of ‘
and r as the left and right normal bundles of the structure, respectively. We
may follow Stong [24] and Quillen [19] in setting up the corresponding bordism
and cobordism functors geometrically, taking necessary care with the double
(X) is
indexing inherent in the splitting. Cartesian product ensures that ΩDU
a graded ring for any space or spectrum X . Both functors admit an involution
induced by interchanging the order of ‘ and r , and we nd it convenient
to write (M ) for (M ; r ; ‘ ). The coecient ring ΩDU is the double complex
cobordism ring.
We may recombine the left and right normal bundles to obtain a forgetful homomorphism : ΩDU (X) ! ΩU (X); conversely, we may interpret any standard
U {structure as either of the double U {structures (; 0) or (0; ), thereby inducing multiplicative natural transformations ‘ and r : ΩU (X) ! ΩDU (X),
which are interchanged by . All these transformations have cohomological
counterparts, and the compositions ‘ and r reduce to the identity.
Given an element of ΩU (X) or ΩU (X), we write ‘ () and r () as ‘ and
r respectively.
From the homotopy theoretic viewpoint, it is convenient to work in any of
the currently fashionable categories which admit well-behaved smash products; a coordinate-free approach suces, as described in [10], for example. The
Pontryagin{Thom construction then ensures that the double complex bordism
and cobordism functors are represented by the Thom spectrum MU ^ MU ,
which we label as DU , and the cobordism ring ΩDU is identied with the
homotopy ring (DU ). The transformation is induced by the product map
on MU , whilst ‘ and r are induced by the left and right inclusion of MU in
DU respectively, using the unit S 0 ! MU on the opposite factor.
We may also identify the homotopy
ΩU (MU ), adopting the convention
taking the argument as the second
Hopf algebroid AU
of cooperations
Geometry and Topology, Volume 2 (1998)
ring of MU ^ MU with the ΩU {algebra
of [1] (and most subsequent authors) in
factor. Of course, ΩU (MU ) is also the
in complex bordism theory. The Thom
83
Flag Manifolds and the Landweber-Novikov Algebra
isomorphism ΩU (MU ) = ΩU (BU + ) provides a further description, whose ring
structure is induced by the Whitney sum map on the Grassmannian BU ; it is
U (BU )
commonly used to transfer the standard polynomial generator n in Ω2n
U
to the polynomial generator bn in Ω2n (MU ), for each n 0. Monomials ! in
the n are dual to the universal Chern classes c! in ΩU (BU ), and monomials
b! in the bn are dual to the Landweber{Novikov operations s! in the algebra of
complex cobordism operations AU . The Landweber{Novikov algebra S is the
sub-Hopf algebra generated additively by the s! , with coproduct induced by the
Cartan formulae; its integral dual S is the polynomial subalgebra Z[b1 ; b2 ; : : : ]
U
of AU
, with coproduct induced from that of A by restriction. We combine
our isomorphisms as
ΩDU = ΩU (MU ) = ΩU (BU + );
(2.1)
referring to the rst as the canonical isomorphism, and to the composition as
h. An analysis of the Pontryagin{Thom construction conrms that h maps the
double cobordism class of any (M ; ‘ ; r ) to the cobordism class of the singular
U {manifold r : M ! BU .
2 (CP 1 ), arising
There are two complex orientation classes x‘ and xr in ΩDU
from the rst Chern class x in ΩU2 (CP 1 ); indeed, DU is the universal example
of a doubly complex oriented spectrum. More generally, there are left and right
(BU ), dual to monomials !
Chern classes c ;‘ and c!;r in ΩDU
‘ and r in
ΩDU (BU ). We obtain mutually inverse expansions
X
X
xr =
gn xn+1
and
x
=
gn xn+1
(2.2)
‘
r
‘
n0
n0
2 (CP 1 ), where g and g
DU for all n and are interchanged
in ΩDU
n lie in Ω2n
n
by the involution . For n > 0 they are annihilated by the transformation
, whilst g0 = g0 = 1. As documented in [8], the image of gn under the
canonical isomorphism is bn , and the isomorphism h of (2.1) therefore satises
U (BU ), for each n 0.
h(gn ) = n in Ω2n
+
These observations arise from minor manipulations with the denitions, and
suggest that we introduce the polynomial subalgebra G of ΩDU , generated
by the elements gn (or, equivalently, by the elements gn ) for n 0. We may
then incorporate our previous remarks and formulate the geometric viewpoint;
we also appeal to [21], recalling the construction of singular manifolds : B n !
U (CP 1 ), where B n is an iterated 2{sphere bundle
CP 1 to represent n in Ω2n
which admits a bounding U {structure for each n 0.
Geometry and Topology, Volume 2 (1998)
84
Victor M Buchstaber and Nigel Ray
Proposition 2.3 The canonical isomorphism identies G with the dual of
the Landweber{Novikov algebra S in AU
; a representative for the generator
n
?
gn is given by (B ; ; ), for each n 0.
We shall apply Proposition 2.3 to realize the coproduct and antipode of S ,
given by
X
(bn ) =
(b)n+1
and
(bn ) = (b)−(n+1)
;
(2.4)
n
n−k ⊗ bk
k0
and the left and right actions of S on S , given by
hy; s‘ ai = h(s)y; ai
and
hy; sr ai = hys; ai;
here s and y lie in S , and the actions on a in S extend naturally to AU
.
Equivalently, we may write
X
X
s‘ a =
h(s); a1 ia2
and
sr a =
hs; a2 ia1
(2.5)
P
where (a) =
a1 ⊗ a2 , conrming that either of the left or right actions
determines (and is determined by) the coproduct .
We consider the algebra A = Ω (DU ) of operations in double complex
DU
DU
cobordism theory, whose continuous ΩDU {dual is the corresponding Hopf alge
broid of cooperations ADU
= ΩDU (DU ). An element s of S yields operations
s‘ ⊗ 1 and 1 ⊗ sr by action on the rst or second factor MU of DU , leading to
the left and right Landweber{Novikov operations s ;‘ ⊗ 1 and 1 ⊗ s!;r , which
commute in ADU by construction. It follows that ADU contains the subalgebra
S‘ ⊗ Sr , and that ADU
contains the subalgebra S;‘ ⊗ S;r = Z[bj;‘ ⊗ 1; 1 ⊗ bk;r :
j; k 0]; these are integrally dual Hopf algebras. Of course S‘ ⊗ Sr acts on
the coecient ring ΩDU , and we need only unravel the denitions in order to
express the result in terms of the canonical isomorphism.
Proposition 2.6 The canonical isomorphism identies the actions of the algebras S‘ ⊗ 1 and 1 ⊗ Sr on ΩDU with the left and right actions of S on AU
respectively; in particular G is closed under the action of S‘ ⊗ Sr .
Since S is cocommutative, the image of the coproduct : S ! S‘ ⊗ Sr
denes a third subalgebra Sd of ADU . The canonical isomorphism identies
the resulting diagonal action of Sd on G with the adjoint action of S on S ;
this is fundamental to the formation of the quantum double D(S ) [14], and
underlies the description of D(S ) as a subalgebra of ADU [7], [8].
By analogy with standard cobordism theory the action of S‘ ⊗ Sr on ΩDU
may be expressed in terms of characteristic numbers, since the operation s ;‘ ⊗
Geometry and Topology, Volume 2 (1998)
Flag Manifolds and the Landweber-Novikov Algebra
85
s!;r corresponds to the Chern class c ;‘ ⊗ c!;r under the appropriate Thom
(BU BU ). So the action of s
isomorphism ADU = ΩDU
+
;‘ ⊗ s!;r on the
cobordism class of (M ; ‘ ; r ) is given by the Kronecker product
hc
;‘ (‘ )c!;r (r ); i
(2.7)
in ΩDU , where in ΩDU (M ) is the canonical orientation class represented by
the identity map on M . The left and right actions of S are therefore given by
restriction, yielding hc ;‘ (‘ ); i and hc!;r (r ); i respectively. Our procedure
for computing the actions of S‘ and Sr on G in Theorem 5.4 is now revealed;
we take the double U {cobordism class of (M ; ‘ ; r ), form the Poincare duals of
c ;‘ (‘ ) and c!;r (r ) respectively, and record the double U {cobordism classes
of the resulting source manifolds.
3
Bounded flag manifolds
In this section we introduce our family of bounded flag manifolds, and discuss
their topology in terms of a cellular calculus which is intimately related to the
Schubert calculus for classic flag manifolds. Our description is couched in terms
of nonsingular subvarieties, anticipating applications to cobordism in the next
section. We also invest the bounded flag manifolds with certain canonical U {
and double U {structures, and so relate them to our earlier constructions in
ΩDU . Much of our notation diers considerably from that introduced in [21].
We shall follow combinatorial convention by writing [n] for the set of natural
numbers f1; 2; : : : ; ng, equipped with the standard linear ordering < . Every
interval in the poset [n] has the form [a; b] for some 1 a b n, and
consists of all m satisfying a m b; our convention therefore dictates
that we abbreviate [1; b] to [b]. It is occasionally convenient to interpret [0]
as the empty set, and [1] as the natural numbers. We work in the context
of the Boolean algebra B(n) of nite subsets of [n], ordered by inclusion. We
decompose each such subset Q [n] into maximal subintervals I(1)[ [I(s),
where I(j) = [a(j); b(j)] for 1 j s, and assign to Q the monomial b! , where
!i records the number of intervals I(j) of cardinality i for each 1 i n;
we refer to ! as the type of Q, noting that it is independent of the choice of
n. We display the elements of Q in increasing order as fqi : 1 i dg,
and abbreviate the complement [n] n Q to Q0 . We also write I(j)+ for the
subinterval [a(j); b(j) + 1] of [n + 1], and Q^ for Q [ fn + 1g. It is occasionally
convenient to set b(0) to 0 and a(s + 1) to n + 1.
We begin by recalling standard constructions of complex flag manifolds and
some of their simple properties, for which a helpful reference is [13]. We work in
Geometry and Topology, Volume 2 (1998)
86
Victor M Buchstaber and Nigel Ray
an ambient complex inner product space Zn+1 , which we assume to be invested
with a preferred orthonormal basis z1 , : : : , zn+1 , and we write ZE for the
subspace spanned by the vectors fze : e 2 Eg, where E [n+1]. We abbreviate
Z[a;b] to Za;b (and Z[b] to Zb ) for each 1 a < b n + 1, and write CP (ZE )
for the projective space of lines in ZE . We let V − U denote the orthogonal
complement of U in V for any subspaces U < V of Zn+1 , and we regularly
abuse notation by writing 0 for the subspace which consists only of the zero
vector. A complete flag V in Zn+1 is a sequence of proper subspaces
0 = V0 < V1 < < Vi < < Vn < Vn+1 = Zn+1 ;
of which the standard flag Z0 < < Zi < < Zn+1 is a specic example.
The flag manifold F (Zn+1 ) is the set of all flags in Zn+1 , topologized as the
quotient U (n + 1)=T of the unitary group U (n + 1) by its maximal torus.
The flag manifold
is a nonsingular complex projective algebraic variety of di
mension n+1
,
whose
cells e are even dimensional, indexed by elements of the
2
symmetric group Sn+1 , and partially ordered by the decomposition of into
a product of transpositions. The closure of every e is an algebraic subvariety,
generally singular, known as the Schubert variety X . Whether considered as
cells or subvarieties, the e dene a basis for the integral homology and cohomology groups H (F (Zn+1 )) and H (F (Zn+1 )), which are integrally dual.
The manipulation of cup and cap products and Poincare duality in these terms
is known as the Schubert calculus for F (Zn+1 ).
An alternative description of H (F (Zn+1 )) is provided by Borel’s computations
with the characteristic homomorphism H (BT ) ! H (U (n + 1)=T ), induced
by the canonical torus bundle U (n + 1)=T ! BT . Noting that H (BT ) is
a polynomial algebra on two dimensional generators xi for 1 i n, Borel
identies H (F (Zn+1 )) with the ring of coinvariants under the action of Sn+1 .
In this context, xi is the rst Chern class of the line bundle over F (Zn+1 )
obtained by associating Vi − Vi−1 to each flag V .
The interaction between the Schubert and Borel descriptions of the cohomology of F (Zn+1 ) is a fascinating area of combinatorial algebra and has led to a
burgeoning literature on the subject of Schubert polynomials, beautifully introduced in MacDonald’s book [16].
We call a flag U in Zn+1 bounded if each i{dimensional component Ui contains the rst i − 1 basis vectors z1 , : : : , zi−1 , or equivalently, if Zi−1 < Ui
for every 1 i n + 1. We dene the bounded flag manifold B(Zn+1 ) to be
the set of all bounded flags in Zn+1 , topologized as a subvariety of F (Zn+1 );
it is straightforward to check that B(Zn+1 ) is nonsingular, and has dimension
Geometry and Topology, Volume 2 (1998)
Flag Manifolds and the Landweber-Novikov Algebra
87
n. Clearly B(Z2 ) is isomorphic to the projective line CP (Z2 ) with the standard complex structure, whilst B(Z1 ) consists solely of the trivial flag. We
occasionally abbreviate B(Zn+1 ) to Bn , in recognition of its dimension.
The algebraic torus (C )n is contained in Zn , and each of its points t determines
a line Lt < Zn+1 with basis vector t + zn+1 . We may therefore embed (C )n
in B(Zn+1 ) as an open dense subset, by assigning the bounded flag
0 < Lt < Lt Z1 < < Lt Zi < < Lt Zn−1 < Zn+1
to each t. The standard action of (C )n on this torus extends to the whole of
B(Zn+1 ) by coordinatewise multiplication on Zn (xing zn+1 ), and therefore
imposes a canonical toric variety structure [11].
There is a map ph : B(Zn+1 ) ! B(Zh+1;n+1 ) for each 1 h n, dened by
factoring out Zh . Thus ph (U ) is given by
0 < Uh+1 − Zh < < Ui − Zh < < Un − Zh < Zh+1;n+1
for each bounded flag U . Since Zi−1 < Ui for all 1 i n + 1, we deduce that
Zh+1;i−1 < Ui − Zh for all i > h + 1, ensuring that ph (U ) is indeed bounded.
We may readily check that ph is the projection of a ber bundle, with ber
B(Zh+1 ). In particular, p1 has ber CP (Z2 ), and after n − 1 applications we
may exhibit B(Zn+1 ) as an iterated bundle
B(Zn+1 ) ! ! B(Zh;n+1 ) ! ! B(Zn;n+1 );
(3.1)
where the ber of each map is isomorphic to CP 1 . This construction was
introduced in [21].
We dene maps qh and rh : B(Zn+1 ) ! CP (Zh;n+1 ) by letting qh (U ) and rh (U )
be the respective lines Uh −Zh−1 and Uh+1 −Uh , for each 1 h n. We remark
that qh = q1 ph−1 and rh = r1 ph−1 for all h, and that the appropriate qh and
rh may be assembled into maps qQ and rQ : B(Zn+1 ) ! Q CP (Zh;n+1 ), where
h varies over an arbitrary subset Q of [n]. In particular, q[n] is an embedding
which associates to each flag U the n{tuple (U1 ; : : : ; Uh −Zh−1 ; : : : ; Un −Zn−1 ),
and describes B(Zn+1 ) as a projective algebraic variety.
We proceed by analogy with the Schubert calculus for F (Zn+1 ). To every flag
U in B(Zn+1 ) we assign the support S(U ), given by fj 2 [n] : Uj 6= Zj g, and
consider the subspace
eQ = fU 2 B(Zn+1 ) : S(U ) = Qg
for each Q in the Boolean algebra B(n). For example, e; is the singleton
consisting of the standard flag.
Geometry and Topology, Volume 2 (1998)
88
Victor M Buchstaber and Nigel Ray
Lemma 3.2 The subspace eQ B(Zn+1 ) is an open cell of dimension 2jQj,
whose closure XQ is the union of all eR for which R Q in B(n).
Proof If Q = [j I(j), then eQ is homeomorphic to the cartesian product
j eI(j) , so it suces to assume that Q is an interval [a; b]. If U lies in e[a;b]
then Ua−1 = Za−1 and Ub+1 = Zb+1 certainly both hold; thus e[a;b] consists of
those flags U for which qj (U ) is a xed line L in CP (Za;b+1 ) n CP (Za;b ) for
all a j b. Therefore e[a;b] is a 2(b − a + 1){cell, as sought. Obviously
eR X[a;b] for each R Q, so it remains only to observe that the limit of a
sequence of flags in e[a;b] cannot have fewer components satisfying Uj = Zj ,
and must therefore lie in eR for some R [a; b].
Clearly X[n] is B(Zn+1 ), so that Lemma 3.2 provides a CW decomposition for
Bn with 2n cells.
We now prove that all the subvarieties XQ are nonsingular, in contrast to the
situation for F (Zn+1 ).
Proposition 3.3 For any Q [n], the subvariety XQ is dieomorphic to the
cartesian product j B(ZI(j)+ ).
Proof We may dene a smooth embedding iQ : j B(ZI(j)+ ) ! B(Zn+1 ) by
choosing the components of iQ (U (1); : : : ; U (s)) to be
(
Za(j)−1 U (j)i if k = a(j) + i − 1 in I(j)
Tk =
(3.4)
Zk
if k 2 [n + 1] n Q;
where U (j)i < ZI(j)+ for each 1 i b(j) − a(j) + 1; the resulting flag is
indeed bounded, since Za(j);a(j)+i−1 < U (j)i holds for all such i and 1 j s.
Any flag T in B(Zn+1 ) for which S(T ) Q must be of the form (3.4), so that
iQ has image XQ , as required.
We may therefore interpret the set
X (n) = fXQ : Q 2 B(n)g
as a Boolean algebra of nonsingular subvarieties of B(Zn+1 ), ordered by inclusion, on which the support function S : X (n) ! B(n) induces an isomorphism
of Boolean algebras. Moreover, whenever Q has type ! then XQ is isomorphic
to the cartesian product B1!1 B2!2 : : : Bn!n , and so may be abbreviated to B ! . In
this important sense, S preserves types. We note that the complex dimension
jQj of XQ may be written as j!j.
The following quartet of lemmas is central to our computations in section 4.
Geometry and Topology, Volume 2 (1998)
Flag Manifolds and the Landweber-Novikov Algebra
89
Lemma 3.5 The map rQ0 : B(Zn+1 ) ! Q0 CP (Zh;n+1 ) is transverse to the
subvariety Q0 CP (Zh+1;n+1 ), whose inverse image is XQ .
Proof Let T be a flag in B(Zn+1 ). Then rh (T ) lies CP (Zh+1;n+1 ) if and
only if Th+1 = Th Lh for some line Lh in Zh+1;n+1 . Since Zh < Th+1 , this
condition is equivalent to requiring that Th = Zh , and the proof is completed
by allowing h to range over Q0 .
Lemma 3.6 The map qQ0 : B(Zn+1 ) ! Q0 CP (Zh;n+1 ) is transverse to the
subvariety Q0 CP (Zh+1;n+1 ), whose inverse image is dieomorphic to B(ZQ^ ).
Proof Let T be a flag in B(Zn+1 ) such that qh (T ) lies CP (Zh+1;n+1 ), which
occurs if and only if Th = Zh−1 Lh for some line Lh in Zh+1;n+1 . Whenever
this equation holds for all h in some interval [a; b], we deduce that Lh actually
lies in Zb+1;n+1 . Thus we may describe T globally by
Tk = Z[k−1]nQ Ui ;
where Ui lies in ZQ^ , and i is k − j[k − 1] n Qj. Clearly Ui−1 < Ui and
Zfq1 ;:::;qi−1 g < Ui for all appropriate i, so that U lies in B(ZQ^ ). We may
now identify the required inverse image with the image of the natural smooth
embedding jQ : B(ZQ^ ) ! B(Zn+1 ), as sought.
We therefore dene YQ to consist of all flags T for which the line Th − Zh−1
lies in ZQ^ for every h in Q0 . It follows that YQ is isomorphic to Bk whenever
Q has cardinality k ; for example, Y[n] is B(Zn+1 ) itself and Y; consists of the
single flag determined by T1 = Zn+1 . The set
Y(n) = fYQ : Q 2 B(n)g
is also a Boolean algebra of nonsingular subvarieties.
Lemma 3.7 For any 1 m n − h, the map qh : B(Zn+1 ) ! CP (Zh;n+1 ) is
transverse to the subvariety CP (Zh+m;n+1 ), whose inverse image is dieomorphic to Y[h;h+m−1]0 .
Proof Let T be a flag in B(Zn+1 ) such that qh (T ) lies CP (Zh+m;n+1 ), which
occurs if and only if Th = Zh−1 Lh for some line Lh in Zh+m;n+1 . Following
the proof of Lemma 3.6 we immediately identify the required inverse image with
Y[h−1][[h+m;n] , as sought.
Geometry and Topology, Volume 2 (1998)
90
Victor M Buchstaber and Nigel Ray
Lemma 3.8 The following intersections in B(Zn+1 ) are transverse:
XQ \ XR = XQ\R
and YQ \ YR = YQ\R whenever Q [ R = [n];
(
XQ;R if Q [ R = [n]
and XQ \ YR =
;
otherwise;
where XQ;R denotes the submanifold XQ\R B(ZR^ ). Moreover, m copies of
Yfhg0 may be made self-transverse so that
Yfhg0 \ \ Yfhg0 = Y[h;h+m−1]0
for each 1 h n and 1 m n − h.
Proof The rst three formulae follow directly from the denitions, and dimensional considerations ensure that the intersections are transverse. The manifold XQ;R is dieomorphic to j YR(j) as a submanifold of B(Zn+1 ), where
Q = [j I(j) and R(j) = I(j) \ R for each 1 j s.
Since Yfhg0 is dened by the single constraint Uh = Zh−1 Lh , where Lh is a line
in Zh+1;n+1 , we may deform the embedding jfhg0 (through smooth embeddings,
in fact) to m − 1 further embeddings in which the Lh is constrained to lie in
Z[h;n+1]nfh+i−1g , for each 2 i m. The intersection of the m resulting
images is determined by the single constraint Lh < Zh+m;n+1 , and the result
follows by applying Lemma 3.7.
It is illuminating to consider the toric structure of B(Zn+1 ) in these terms.
Proposition 3.9
(1) For each 1 h n, the projection qh : B(Zn+1 ) ! CP (Zh;n+1 ) is equivariant with respect to an action of the torus (C )n−h+1 , and the equivariant ltration
CP (Zh ) CP (Zh;i ) CP (Zh;n+1 )
lifts to an equivariant ltration of the irregular values of qh .
(2) The quotient of B(Zn+1 ) by the action of the compact torus T n is homeomorphic to the n{cube I n .
Proof For (1), we choose the subtorus of (C )n in which the rst h − 1 coordinates are 1; in particular, when h = 1 the result refers to toric structures
on B(Zn+1 ) and CP n . For (2), we proceed inductively from the observation
that the invariant submanifolds of the action of T n are the subvarieties XQnR;Q
Geometry and Topology, Volume 2 (1998)
Flag Manifolds and the Landweber-Novikov Algebra
91
for all pairs R Q [n]; in particular, the xed points are standard flags in
the subvarieties B(ZQ^ ), and so display the vertices of the quotient in bijective
correspondence with the subsets Q.
The second part of Proposition 3.9 refers to the structure of B(Zn+1 ) as a toric
manifold [9], and may be extended by algebraic geometers to a more detailed
description of the associated fan [11].
4
Normal structures and duality
In this section we describe the basic U { and double U {structures on the varieties XQ , and compute their cobordism rings. We pay special attention to
Poincare duality, which makes delicate use of the normal structures and is of
central importance to our subsequent applications.
We consider complex line bundles γi and i over B(Zn+1 ), classied respectively by the maps qi and ri for each 1 i n. We set γ0 to 0 and 0 to γ1 ,
which are compatible with the choices above and enable us to write
γi i i+1 n = C n−i+2
(4.1)
for every 0 i n. We may follow [21] in using (3.1) to obtain an expression of
the form R = (n+1
i=2 γi )R for the tangent bundle of B(Zn+1 ), as prophesied
by the toric structure; so (4.1) leads to an isomorphism = ni=2 (i − 1)i . We
refer to the resulting U {structure as the basic U {structure on B(Zn+1 ). We
emphasize that these isomorphisms are of real bundles only, and that the basic
U {structure is not compatible with any complex structure on the underlying
variety. On B(Z2 ), for example, the basic U {structure is that of a 2{sphere
S 2 , rather than CP 1 . Indeed, the basic U {structure on B(Zn+1 ) extends over
the 3{disc bundle associated to γ1 R for all values of n, so that B(Zn+1 )
U .
represents zero in Ω2n
L
By virtue of (4.1) we may introduce the double U {structure ( ni=1 ii ; γ1 ),
which we again label basic; equivalently, we rewrite ‘ as γ = −(γ1 γn ).
The basic double U {structure does not bound, however, as we shall see in
Proposition 4.2. Given any cartesian product of manifolds B(Zn+1 ), we also
refer to the product of basic structures as basic.
Proposition 4.2 With the basic double U {structure, B(Zn+1 ) represents gn
in ΩDU ; if ‘ and r are interchanged, it represents gn .
Geometry and Topology, Volume 2 (1998)
92
Victor M Buchstaber and Nigel Ray
Proof It suces to apply Proposition 2.3 for gn , because the bundle γ1 over
B(Zn+1 ) coincides with the bundle of [21] over B n . The result for gn follows
by applying the involution .
Corollary 4.3 The cobordism classes of the basic double U {manifolds XQ
give an additive basis for G as Q ranges over nite subsets of [1].
Proof It suce to combine Propositions 3.3 and 4.2, remarking that XQ represents g! whenever Q has type ! .
Henceforth we shall insist that Bn denotes B(Zn+1 ) (or any isomorph) equipped
exclusively with the basic double U {structure.
Proposition 4.4 Both X (n) and Y(n) are Boolean algebras of basic U {
submanifolds, in which the intersection formulae of Lemma 3.8 respect the
basic U {structures.
Proof It suces to prove that the pullbacks in Lemmas 3.5, 3.6 and 3.7 are
compatible with the basic U {structures. Beginning with Lemma 3.5, we note
that whenever h over B(Zn+1 ) is restricted by iQ to a factor B(ZI(j)+ ), we
obtain k+1 if h = a(j)+k lies in I(j) and γ1 if h = a(j)−1; for all other values
of h, the restriction is trivial. Since the construction of Lemma 3.5 identies
(iQ ) with the restriction of h h as h ranges over Q0 , we infer an isomorphism
(iQ ) = (j γ1 ) C n−j−jQj over XQ (unless 1 2 Q, in which case the rst γ1
is trivial). Appealing to (4.1), we then verify that this is compatible with the
basic structures in the isomorphism XQ = iQ ( B(Zn+1 ) ) (iQ ), as claimed.
The proofs for Lemmas 3.6 and 3.7 are similar, noting that the restriction of h
to YQ is k if h = qk lies in Q, and is trivial otherwise, and that the restriction
of γh is γk if h = qk lies in Q, and is γk+1 if qk is the greatest element of Q
for which h > qk (meaning γ1 if h < q1 , and the trivial bundle if h > qk for all
k ). Since the construction of Lemma 3.6 identies (jQ ) with the restriction
of h γh as h ranges over Q0 , we infer an isomorphism
(jQ ) =
s+1 M
a(j) − b(j − 1) − 1 γc(j)
(4.5)
j=1
P
over YQ , where c(j) = j + j−1
i=0 (b(i) − a(i)). This isomorphism is also com ( B(Zn+1 ) ) (j ), once more by
patible with the basic structures in YQ = jQ
Q
appeal to (4.1).
Geometry and Topology, Volume 2 (1998)
Flag Manifolds and the Landweber-Novikov Algebra
93
The corresponding results for double U {structures are more subtle, since we are
free to choose our splitting of (iQ ) and (jQ ) into left and right components.
Corollary 4.6 The same results hold for double U {structures with respect
to the splittings (iQ )‘ = 0 and (iQ )r = (iQ ), and (jQ )‘ = (jQ ) and
(jQ )r = 0.
Proof One extra fact is required in the calculation for iQ , namely that γ1 on
B(Zn+1 ) restricts trivially to XQ (or to γ1 if 1 2 Q).
At this juncture we may identify the inclusions of XQ in F (Zn+1 ) with certain
of the desingularizations introduced by Bott and Samelson [4]. For example,
X[n] is the desingularization of the Schubert variety X(n+1;1;2;:::;n) , and the
resolution map is actually an isomorphism in this case. Moreover, the corresponding U {cobordism classes form the cornerstone of Bressler and Evens’s
calculus for ΩU (F (Zn+1 )). In both of these applications, however, the underlying complex manifold structures suce. The basic U {structures become vital
when investigating the Landweber{Novikov algebra (and could also have been
used in [5], although an alternative calculus would result). We leave the details
to interested readers.
We now use the basic structures on XQ to investigate Poincare duality in
bordism and cobordism, beginning with the CW decomposition for B(Zn+1 )
which stems from Lemma 3.2. Since the cells eQ occur only in even dimensions,
the corresponding homology classes xH
Q form a basis for the integral homology
groups H (B(Zn+1 )) as Q ranges over B(n). Applying HomZ determines a
dual basis Hd (xH
Q ) for the cohomology H (B(Zn+1 )); we delay clarifying the
cup product structure until after Theorem 4.8 below, although it may also be
deduced directly from the toric properties of B(Zn+1 ).
U (B(Z
We introduce the complex bordism classes xQ and yQ in Ω2jQj
n+1 )), represented respectively by the inclusions iQ and jQ of the subvarieties XQ and
YQ with their basic U {structures. By construction, the fundamental class in
H
H2jQj (XQ ) maps to xH
Q in H2jQj (B(Zn+1 )) under iQ ; thus xQ maps to xQ
under the Thom homomorphism ΩU (B(Zn+1 )) ! H (B(Zn+1 )). The Atiyah{
Hirzebruch spectral sequence for ΩU (B(Zn+1 )) therefore collapses, and the
classes xQ form an ΩU {basis as Q ranges over B(n). The classes x[n] and y[n]
coincide, since they are both represented by the identity map. They constiU (B(Z
tute the basic fundamental class in Ω2n
n+1 )), with respect to which the
Poincare duality isomorphism is given by
Pd (w) = w \ x[n]
Geometry and Topology, Volume 2 (1998)
94
Victor M Buchstaber and Nigel Ray
U
in Ω2(n−d)
(B(Zn+1 )), for any w in ΩU2d (B(Zn+1 )).
An alternative source of elements in ΩU2 (B(Zn+1 )) is provided by the Chern
classes
xi = c1 (γi )
and
yi = c1 (i )
for each 1 i n. It follows from (4.1) that
xi = −yi − yi+1 − − yn
(4.7)
Q
Q
for every i. Given Q [n], we write Q xh as xQ and Q yh as y Q in
2jQj
ΩU
(B(Zn+1 )), where h ranges over Q in both products.
We may now discuss the implications of our intersection results of Lemma 3.8 for
the structure of ΩU (B(Zn+1 )). It is convenient (but by no means necessary) to
use Quillen’s geometrical interpretation of cobordism classes, which provides a
particularly succinct description of cup and cap products and Poincare duality,
and is conveniently summarized in [5].
Theorem 4.8 The complex bordism and cobordism of B(Zn+1 ) satisfy
0
0
(1) Pd(xQ ) = yQ and Pd (y Q ) = xQ ;
(2) the elements fyQ : Q [n]g form an ΩU {basis for ΩU (B(Zn+1 ));
(3) Hd(xQ ) = xQ and Hd (yQ ) = y Q ;
(4) there is an isomorphism of rings
ΩU (B(Zn+1 )) = ΩU [x1 ; : : : ; xn ]=(x2i = xi xi+1 );
where i ranges over [n] and xn+1 is interpreted as 0.
Proof For (1), we apply Lemma 3.6 and Proposition 4.4 to deduce that
0
2jQ0 j
xQ in ΩU (B(Zn+1 )) is the pullback of the Thom class under the collapse
0
map onto M ((jQ )). Hence xQ is represented geometrically by the inclusion
0
jQ : YQ ! B(Zn+1 ), and therefore Pd(xQ ) is represented by the same singular
0
U {manifold in Ω2jQj (B(Zn+1 )). Thus Pd (xQ ) = yQ . An identical method
0
works for Pd (y Q ), by applying Lemma 3.5. For (2), we have already shown
that the xQ form an ΩU {basis for ΩU (B(Zn+1 )). Thus by (1) the y Q form
a basis for ΩU (B(Zn+1 )), and therefore so do the xQ by (4.7); the proof is
concluded by appealing to (1) once more. To establish (3), we remark that the
cap product xQ \ xR is represented geometrically by the ber product of jQ0
and iR , and is therefore computed by the intersection theory of Lemma 3.8.
Geometry and Topology, Volume 2 (1998)
Flag Manifolds and the Landweber-Novikov Algebra
95
Bearing in mind the crucial fact that each basic U {structure bounds (except
in dimension zero!), we obtain
hxQ ; xR i = Q;R
(4.9)
and therefore that Hd (xQ ) = xQ , as sought. The result for Hd (yQ ) follows
similarly. To prove (4) we note that it suces to obtain the product formula
x2i = xi xi+1 , since we have already demonstrated that the monomials xQ form
a basis in (2). Now xi and xi+1 are represented geometrically by Yfig0 and
Yfi+1g0 respectively, and products are represented by intersections; according
to Lemma 3.8 (with m = 2), both x2i and xi xi+1 are therefore represented by
the same subvariety Yfi;i+1g0 , so long as 1 i < n. When i = n we note that
xn pulls back from CP 1 , so that x2n = 0, as required.
For any Q [n], we obtain the corresponding structures for the complex bordism and cobordism of XQ by applying the Künneth formula to Theorem 4.8.
Using the same notation as in B(Zn+1 ) for any cohomology class which restricts
along (or homology class which factors through) the inclusion iQ , we deduce,
for example, a ring isomorphism
ΩU (XQ ) = ΩU [xi : i 2 Q]=(x2i = xi xi+1 );
(4.10)
where xi is interpreted as 0 for all i 2
= Q.
The relationship between the classes xi and yi in ΩU (B(Zn+1 )) is described by
(4.7), but may be established directly by appeal to the third formula of Lemma
3.8, as in the proof of Theorem 4.8; for example, we deduce immediately that
xi yi = 0 for all 1 i n. When applied with arbitrary m, the fourth formula
of Lemma 3.8 simply iterates the quadratic relations, and produces nothing
new.
The results of Theorem 4.8 extend to any complex oriented cohomology theory
as usual; in particular, we may substitute double complex cobordism, so long
as we choose left or right Chern classes consistently throughout. To understand
duality, however, we must also attend to the choice of splittings provided by
Corollary 4.6, and the failure of formulae such as (4.9) because the manifolds
Bn are no longer double U {boundaries. Since, by (2.7), duality lies at the
heart of our applications to the Landweber{Novikov algebra, we treat these
issues with care below.
Q
Q
We are particularly interested in the left and right Chern classes xQ
‘ , y ‘ , xr
2jQj
and yrQ in ΩDU (Bn ), and we seek economical geometric descriptions of their
DU (B ) for the homology
Poincare duals. We continue to write xR and yR in Ω2jRj
n
Geometry and Topology, Volume 2 (1998)
96
Victor M Buchstaber and Nigel Ray
classes represented by the respective inclusions of XR and YR with their basic
double U {structures.
DU
Proposition 4.11 In Ω2(n−jQj)
(Bn ), we have that
0
Pd (xQ
‘ ) = yQ
0
and
0
Pd (yrQ ) = xQ ;
0
Q
whilst Pd (xQ
r ) and Pd(y‘ ) are represented by the inclusion of YQ and XQ
with the respective double U {structures
YQ − ((jQ ) γ1 ); (jQ ) γ1
and ( XQ − iQ γ1 ; iQ γ1 );
for all n 0.
Proof The rst two formulae follow at once from Corollary 4.6, by analogy
with (1) of Theorem 4.8. The second require the interchange of the left and
right components of the normal bundles of jQ and iQ respectively, plus the
(γ ) is always γ , whatever Q.
observation that jQ
1
1
Proposition 4.11 extends to XQ by the Künneth formula, which we express
in terms of restriction along iQ in our applications below; it also extends to
general doubly complex oriented cohomology theories in the obvious fashion.
It inspires many interesting cobordism calculations, of which we oer a single
example.
Proposition 4.12 The map qh : Bn ! CP n−h+1 represents either of the expressions
n+1−h
X
gn−m m;‘
or
m=0
n+1−h
X
gn−j (g)m
j−m m;r
jm=0
DU (CP n−h+1 ), for each 1 h n.
in Ω2n
Proof The coecient of m;‘ in the rst expression is given by hxm
h;‘ ; x[n] i; by
Proposition 4.11, this is gn−m when 1 m n − h + 1, and zero otherwise,
as required. To convert the result into the second expression, we dualize the
expansion (2.2).
Geometry and Topology, Volume 2 (1998)
Flag Manifolds and the Landweber-Novikov Algebra
5
97
Applications
In our nal section, we apply the duality calculations to realize the left and
right actions of the Landweber{Novikov algebra on its dual; some preliminary
combinatorics is helpful.
Fixing the subset Q = [sj=1 I(j) of [n], we consider the additive semigroup
H(Q) of nonnegative integer sequences h of the form
P (h1 ; : : : ; hn ), where hi = 0
for all i 2
= Q; for any such h, we set jhj = 2 i hi . Whenever h satises
Pb(j)
i=l hi b(j) − l + 1 for all a(j) l b(j), we dene the subset hQ Q by
m
X
fm :
hi < m − l + 1 for all a(j) l m b(j)g;
i=l
otherwise, we set hQ = Q. It follows that hQ = Q \ h[n] for all h in H(Q),
and we introduce the subset S(h) [s] of indices j for which I(j) \ hQ 6= ;.
We also identify the subsemigroup K(Q) H(Q) of sequences k for which ki
is nonzero only if i = a(j) for some 1 j s.
For each h in H(Q) and k in K(Q), our applications require us to invest the
manifold XQ;(h+k)[n] of Lemma 3.8 with a non-basic double{U structure. In
terms of the decomposition S(h+k) YI(j)\(h+k)[n] , this is given by
(5.1)
(γ − ka(j) γ1 ); (ka(j) + 1)γ1 ;
S(h+k)
S(h+k)
k
and we denote the resulting double{U manifold by XQ;(h+k)[n]
. For example,
when h is 0 and k has a single nonzero element ka(j) = m for some 1 j s
k
and m b(j)−a(j), then XQ;(h+k)[n]
reduces to the manifold XQn[a(j);a(j)+m−1]
with double U {structure
γ j−1 (γ − mγ1 ) γ s−j ; γ1j−1 (m + 1)γ1 γ1s−j :
(5.2)
This case is important enough to motivate the notation XPm:j (omitting the : j
if s = 1) for any XP whose basic double U {structure is similarly amended on
m:j
its j th factor YI(j) ; in particular, (5.2) describes XQn[a(j);a(j)+m−1]
.
We may now apply Proposition 4.11 to compute the eect of the left and right
actions of S on S under the canonical isomorphism. To ease computations
with the left action we consider the monomial basis of tangential Landweber{
Novikov operations s for AU
; under the universal Thom isomorphism, these
correspond to the Chern classes ? c induced by the involution ? of complementation on BU . There are therefore expressions
X
s =
;! s! ;
(5.3)
!
Geometry and Topology, Volume 2 (1998)
98
Victor M Buchstaber and Nigel Ray
where the ;! are integers
and
P
P the summation ranges over sequences ! for
which j!j = j j and
!i i . For each Q [n], it is also helpful to
partition K(Q) and H(Q) into compatible blocks K(Q; ) and H(Q; ) for
every indexing sequence ; each block consists of those sequences k or h which
have i entries i for each i 1, and all other entries zero. Thus, for example,
jhj = j j for all h in H(Q; ). Any such block will be empty whenever
is
incompatible with Q in the appropriate sense.
Theorem 5.4 Up to double U {cobordism, the actions of S‘ and Sr on additive generators of G are induced by
X
X
k
s ;‘ (XQ ) =
XQ;h[n]
and
s!;r (XQ ) =
XQ;k[n]
H(Q; )
K(Q;!)
respectively.
Proof We combine (2.7) with Proposition 4.11, recalling that c is evaluated
on any sum of line bundles ri=1 i by forming the symmetric sum of all monomials c1 (1 )i1 : : : cr (r )ir , where i of the exponents take the value i for each
(B ) allows us to
1 i r. We note that the product structure in ΩDU
n
m
[i;i+m−1]
replace any xi (either left or right) by x
when [i; i + m − 1] Q,
and zero otherwise; indeed, the denitions of H(Q) and K(Q) are tailored
exactly to these relations. For s ;‘ (XQ ) we set k = 0, and observe that
c ;‘ (‘ ) = iQ c ;‘ (Q γi ). For s!;r (XQ ) we set h = 0, and observe in turn
that c!;r (r ) = iQ c!;r (γa(1) γa(s) ). The computations are then straightforward, although the bookkeeping demands caution.
Recalling (5.1), we may combine the left and right actions by
X
k
s ;‘ ⊗ s!;r (XQ ) =
XQ;(h+k)[n]
;
H(Q; ); K(Q;!)
from which the diagonal action follows immediately. If we prefer to express the
action of S‘ in terms of the standard basis s! , we need only incorporate the
integral relations (5.3).
Readers may
Ln observe that our expression in section 3 for ‘ as the sum of line
bundles
i=1 ii appears to circumvent the need to introduce the tangential
operations s . However, it contains n(n + 1)=2 summands rather than n, and
their Chern classes yi are algebraically more complicated than the xi used
above, by virtue of (4.7). These two factors conspire to make the alternative
calculations less palatable, and it is an instructive exercise to reconcile the two
Geometry and Topology, Volume 2 (1998)
99
Flag Manifolds and the Landweber-Novikov Algebra
approaches in simple special cases. The apparent dependence of Theorem 5.4
on n is illusory (and solely for notational convenience), since ki and hi are zero
whenever i lies in Q0 .
We may specialize Theorem 5.4 to the cases when
and ! are of the form
(m) for some integer 0 m jQj, or when Q = [n] (so that we are dealing
with polynomial generators of G ), or both. We obtain
s(m);‘ (XQ ) =
X
X b(j)−m+1
j
XQnI(j) YI(j)n[i;i+m−1]
i=a(j)
and
s(m);r (XQ ) =
X
m:j
XQn[a(j);a(j)+m−1]
;
(5.5)
j
where the summations range over all j with b(j) − a(j) m − 1, and
X
s ;‘ (X[n] ) =
Yh[n]
H([n]; )
and
s!;r (X[n] ) =
(
m
X[m+1;n]
0
when ! = (m)
otherwise:
(5.6)
These follow from (5.1), and the facts that K(Q; (m)) consists solely of sequences containing a single nonzero entry m in some position a(j), and K([n]; !)
is empty unless ! = (m) for some 0 m n.
We might expect Theorem 5.4 to provide geometrical conrmation that G
is closed under the action of S‘ ⊗ Sr on ΩDU , as noted in Proposition 2.6;
k+1
however, it remains to show that XkQ
lies in G ! Currently, we have no direct
geometrical proof of this fact.
We now turn to the structure maps of S , continuing to utilize the canonical
isomorphism to identify G and G ⊗ G with S and S ⊗ S respectively.
We express monomial generators of G ⊗ G as double U {cobordism classes
of pairs of basic double U {manifolds (XQ ; XR ), where Q and R range over
independently chosen subsets of [n].
Proposition 5.7 Up to double U {cobordism, the coproduct and the antipode of the dual of the Landweber{Novikov algebra are induced by
X
k
XQ 7!
(XQ;k[n]
; XQnkQ )
and
XQ 7! (XQ )
K(Q)
respectively.
Geometry and Topology, Volume 2 (1998)
100
Victor M Buchstaber and Nigel Ray
Proof For , we combine the right action of Theorem 5.4 with (2.5), and the
observation that XQnkQ is isomorphic to B ! for each k in K(Q; !). For , we
refer to Proposition 4.2.
Corollary 5.8 When equipped with the double U {structure
s
s
(γ − m(j)γ1 ); (m(j) + 1)γ1 ;
j=1
j=1
Q
m(j)+1
DU for any sequence of natthe manifold XQ represents j (g)b(j)−a(j)+1 in Ω2jQj
ural numbers m(1), m(2), : : : , m(s).
Proof If we consider the coproduct for Q = [n] in Proposition 5.7, we deduce
m
that X[m+1;n]
represents (g)m+1
n−m by appeal to (2.4). The result for general XQ
follows by applying this case to each factor YI(j) .
Corollary 5.8 is particularly fascinating because it describes how to represent
an intricate (but important) polynomial in the cobordism classes of the basic
Bn by perturbing the double U {structure on a single manifold XQ .
For a nal comment on Proposition 5.7, we note that the elements of ΩDU ⊗ΩU
ΩDU may be represented by threefold U {manifolds. Under the canonical isomorphism, the coproduct on the Hopf algebroid ADU
is then induced by map
ping the double U {cobordism class of each (M ; ‘ ; r ) to the threefold cobordism class of (M ; ‘ ; 0; r ), and the diagonal on G follows by restriction. Theories of multi U {cobordism are remarkably rich, and have applications to the
study of iterated doubles and Adams{Novikov resolutions; we reserve our development of these ideas for the future.
References
[1] J Frank Adams, Stable Homotopy and Generalized Homology, Chicago Lectures in Mathematics, University of Chicago Press (1974)
[2] Martin Aigner, Combinatorial Theory, Springer{Verlag (1979)
[3] Anthony Bahri, Martin Bendersky, The KO{theory of toric manifolds,
preprint, Rider University (1997)
[4] Raoul Bott, Hans Samelson, Application of the theory of Morse to symmetric
spaces, American J. Math. 80 (1958) 964{1029
[5] Paul Bressler, Sam Evens, Schubert calculus in complex cobordism, Transactions of the AMS 331 (1992) 799{813
Geometry and Topology, Volume 2 (1998)
Flag Manifolds and the Landweber-Novikov Algebra
101
[6] V M Buchstaber, A B Shokurov, The Landweber{Novikov algebra and formal vector elds on the line, Funktsional Analiz i Prilozhen 12 (1978) 1{11
[7] Victor M Buchstaber, Semigroups of maps into groups, operator doubles, and
complex cobordisms, from: \Topics in Topology and Mathematical Physics", S P
Novikov, editor, AMS Translations (2) 170 (1995) 9{31
[8] Victor M Buchstaber, Nigel Ray, Double cobordism and quantum doubles,
preprint, University of Manchester (1997)
[9] Michael W Davis, Tadeusz Januszkiewicz, Convex polytopes, Coxeter orbifolds and torus actions, Duke Math. J. 62 (1991) 417{451
[10] A D Elmendorf, The Grassmannian geometry of spectra, J. of Pure and Applied Algebra 54 (1988) 37{94
[11] William Fulton, Introduction to Toric Varieties, Annals of Math. Studies, No.
131, Princeton UP (1993)
[12] Phillip Griths, Joseph Harris, Principles of Algebraic Geometry, Wiley
Classics Library, John Wiley and Sons (1994)
[13] Howard H Hiller, Geometry of Coxeter Groups, Research Notes in Mathematics, No. 54, Pitman (1982)
[14] Christian Kassel, Quantum Groups, Graduate Texts in Mathematics, volume
155, Springer{Verlag (1995)
[15] Toshiyuki Katsura, Yuji Shimizu, Kenji Ueno, Complex cobordism ring
and conformal eld theory over Z, Mathematische Annalen 291 (1991) 551{571
[16] Ian G Macdonald, Notes on Schubert Polynomials, Publications du Laboratoire de Combinatoire et d’Informatique Mathematique, volume 6, Universite
du Quebec a Montreal (1991)
[17] Peter Magyar, Bott{Samelson varieties and conguration spaces, preprint,
Northeastern University (1996)
[18] Sergei P Novikov, Various doubles of Hopf algebras: Operator algebras on
quantum groups, and complex cobordism, Uspekhi Akademii Nauk SSSR 47
(1992) 189{190
[19] Daniel G Quillen, Elementary proofs of some results of cobordism theory using
Steenrod operations, Advances in Math. 7 (1971) 29{56
[20] Nigel Ray, SU and Sp Bordism. PhD thesis, Manchester University (1969)
[21] Nigel Ray, On a construction in bordism theory, Proceedings of the Edinburgh
Math. Soc. 29 (1986) 413{422
[22] Nigel Ray, William Schmitt, Combinatorial models for coalgebraic structures, preprint, University of Memphis (1997)
[23] Nigel Ray, Robert Switzer, On SUSU bordism, Oxford Quarterly J. Math.
21 (1970) 137{150
[24] Robert E Stong, Notes on Cobordism Theory, Princeton UP (1968)
Geometry and Topology, Volume 2 (1998)