The BGG Complex on Projective Space
We give a complete construction of the Bernstein-Gelfand-Gelfand complex on real or complex projective space using minimal ingredients.
Gespeichert in:
| Veröffentlicht in: | Symmetry, Integrability and Geometry: Methods and Applications |
|---|---|
| Datum: | 2011 |
| Hauptverfasser: | , |
| Format: | Artikel |
| Sprache: | Englisch |
| Veröffentlicht: |
Інститут математики НАН України
2011
|
| Online Zugang: | https://nasplib.isofts.kiev.ua/handle/123456789/147183 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Zitieren: | The BGG Complex on Projective Space / M.G. Eastwood, Gover A.R.// Symmetry, Integrability and Geometry: Methods and Applications. — 2011. — Т. 7. — Бібліогр.: 25 назв. — англ. |
Institution
Digital Library of Periodicals of National Academy of Sciences of Ukraine| _version_ | 1860194308690280448 |
|---|---|
| author | Eastwood, M.G. Gover, A.R. |
| author_facet | Eastwood, M.G. Gover, A.R. |
| citation_txt | The BGG Complex on Projective Space / M.G. Eastwood, Gover A.R.// Symmetry, Integrability and Geometry: Methods and Applications. — 2011. — Т. 7. — Бібліогр.: 25 назв. — англ. |
| collection | DSpace DC |
| container_title | Symmetry, Integrability and Geometry: Methods and Applications |
| description | We give a complete construction of the Bernstein-Gelfand-Gelfand complex on real or complex projective space using minimal ingredients.
|
| first_indexed | 2025-12-07T18:07:37Z |
| format | Article |
| fulltext |
Symmetry, Integrability and Geometry: Methods and Applications SIGMA 7 (2011), 060, 18 pages
The BGG Complex on Projective Space?
Michael G. EASTWOOD † and A. Rod GOVER †‡
† Mathematical Sciences Institute, Australian National University, ACT 0200, Australia
E-mail: meastwoo@member.ams.org
‡ Department of Mathematics, University of Auckland,
Private Bag 92019, Auckland 1142, New Zealand
E-mail: r.gover@auckland.ac.nz
Received January 30, 2011, in final form June 18, 2011; Published online June 23, 2011
doi:10.3842/SIGMA.2011.060
Abstract. We give a complete construction of the Bernstein–Gelfand–Gelfand complex on
real or complex projective space using minimal ingredients.
Key words: differential complex; BGG complex; projective space; Lie algebra cohomology;
parabolic geometry
2010 Mathematics Subject Classification: 53A20; 17B56; 58D19; 58J70
Dedicated to Willard Miller on his retirement
from the University of Minnesota.
1 Introduction
This article is concerned with differential calculus on real or complex projective space, in-
variant under projective transformations. These transformations constitute a semisimple Lie
group and projective space is hence a homogeneous space of the form G/P for G semisimple.
The subgroup P is parabolic and, more generally, differential geometries modelled on homoge-
neous spaces of this form are known as ‘parabolic’ [5]. Projective space gives rise to projective
differential geometry in this sense. Conformal and CR geometry are included amongst other
examples of parabolic differential geometry. The interplay between the symmetries of projective
space and its invariant differential operators is mediated by representation theory. In this article
we present the ‘BGG complex’ on projective space as perhaps the simplest of these construc-
tions in one of the simplest of settings. We anticipate that our approach will extend to G/P
in general and perhaps to other homogeneous spaces. In [19], Olver introduced complexes of
differential operators on Euclidean space acting between ‘hyperforms’. These are BGG com-
plexes constructed directly, employing only affine invariance and the associated Schur functors.
He constructs some ‘easy’ examples and observes that “Other examples, of greater complexity,
can of course be constructed at will, but the expressions rapidly get out of hand, even in low
dimensional spaces”. We maintain that the extra symmetry that the operators exhibit under
projective transformations allows one to control these expressions more effectively. The fasci-
nating combination of symmetries and differential operators is a defining feature in the work of
Willard Miller to whom we dedicate this article.
The Bernstein–Gelfand–Gelfand (BGG) complexes on RPn or CPn are, by now, well-known
complexes of vector bundles and differential operators between them generalising the de Rham
complex on RPn and the holomorphic de Rham complex on CPn, respectively. An introduction
to such complexes is given in [11] and the particular case of projective space is discussed in [12].
?This paper is a contribution to the Special Issue “Symmetry, Separation, Super-integrability and Special
Functions (S4)”. The full collection is available at http://www.emis.de/journals/SIGMA/S4.html
mailto:meastwoo@member.ams.org
mailto:r.gover@auckland.ac.nz
http://dx.doi.org/10.3842/SIGMA.2011.060
http://www.emis.de/journals/SIGMA/S4.html
2 M.G. Eastwood and A.R. Gover
Usually, their construction involves choosing so-called ‘splitting operators’ [3, 6] constructed
from Kostant’s Laplacian [17] or ‘quabla operator’ [3] or from the Jantzen–Zuckerman translation
functor [24]. Here, we avoid the direct use of splitting operators, instead relying only on diagram
chasing, as is already done in [11, 12] in simple cases. In fact, the BGG complex of holomorphic
differential operators on CP2 was already constructed in this manner [10, p. 351] before it was
realised by John Rice [21] that complexes like this were dual to Lepowsky’s construction [18] on
the level of Verma modules. (On CP1 there is just one family of BGG operators, already singled
out for their invariance in [13, Proposition 2.1].) In this article, for simplicity and cleanliness
we employ a spectral sequence to effect the diagram chasing. We employ projective invariance
to derive explicit formulæ for the operators in the projective BGG complex on RPn in terms of
the usual round metric on the sphere.
As is often done in differential geometry, when it is necessary to write out tensors and
their natural operations and we shall adorn them with upper or lower indices corresponding
to the tangent or cotangent bundle respectively. For example, a vector field can be written
as Xa, a one-form as ωa, and the natural pairing between them as Xaωa in accordance with the
‘Einstein summation convention’. For any tensor φabc we shall write its symmetric part as φ(abc)
and its skew part as φ[abc]. For example, to say that ωab is a two-form is to say that
ωab = −ωba or, equivalently, ωab = ω[ab] or, equivalently, ω(ab) = 0
and then
∇[aωbc] and Xa∇aωbc − 2(∇[bX
a)ωc]a,
for any torsion-free connection ∇a, are the exterior derivative of ωab and the Lie derivative of ωab
in the direction of the vector field Xa, respectively. Such formulæ are not meant to imply any
choice of local coördinates. More precisely, this is Penrose’s ‘abstract index notation’ [20] and
it formalises the conventions used by many classical authors – see, for example, the discussion
of projective differential geometry by Schouten [22].
We shall view RPn as a homogeneous space
RPn = SL(n+ 1,R)/P = G/P, where P =
∗ ∗ · · · ∗
0
... ∗
0
or as a quotient of the ‘projective sphere’
Sn = SL(n+ 1,R)/P = G/P, where P =
λ ∗ · · · ∗
0
... ∗
0
s.t. λ > 0
under the antipodal map. In either case, it is convenient to write the associated Lie algebra as
sl(n+ 1,R) = g = g− ⊕ g0 ⊕ g+, (1)
where
g− =
0 0 · · · 0
∗
... 0
∗
, g0 =
∗ 0 · · · 0
0
... ∗
0
, g+ =
0 ∗ · · · ∗
0
... 0
0
and then p = g0 ⊕ g+.
The BGG Complex on Projective Space 3
From now on we shall discuss only real projective space RPn or its double cover, the sphere Sn.
The complex case is completely parallel with real numbers being replaced by complex numbers
everywhere and by working in the holomorphic category rather than the smooth.
2 An outline of the construction
If V is a finite-dimensional representation of P , we shall denote by V the induced homogeneous
vector bundle on G/P constructed as
V = G×P V = G× V/ ∼, where (g, v) ∼
(
gp, p−1v
)
, ∀ p ∈ P.
Notice that if V is actually a G-module restricted to P , then V is canonically trivialised
V = G×P V ∼= G/P × V by (g, v) 7→ (gP, gv) (2)
as a vector bundle (but not as a homogeneous vector bundle). Hence, in this case V is naturally
equipped with a G-equivariant flat connection ∇ obtained by transporting to V the exterior
derivative d : Λ0⊗V→ Λ1⊗V with values in V. More generally, the coupled de Rham sequence
V
∇−→ Λ1 ⊗ V ∇−→ Λ2 ⊗ V ∇−→ Λ3 ⊗ V ∇−→ · · · ∇−→ Λn−1 ⊗ V ∇−→ Λn ⊗ V → 0 (3)
is exact on the level of germs and provides a resolution of V as a locally constant sheaf on G/P .
This general reasoning holds on any homogeneous space G/P but the following discussion is
specific to RPn or Sn. Suppose V is irreducible as a G-module. In this case we shall see that as
a P -module V is filtered
V = V0 + V1 + V2 + · · ·+ VN−1 + VN (4)
meaning that these are the subquotients listed in a natural order, starting on the left with the
smallest quotient of V. (In other words VN is the smallest P -submodule in the filtration, the
quotient V/VN has a filtration V0 + V1 + V2 + · · · + VN−1, and the meaning is now clear by
induction.) It follows that the bundle V is correspondingly filtered
V = V0 + V1 + V2 + · · ·+ VN−1 + VN (5)
and now we claim that the connection ∇ : V → Λ1⊗V , and consequently the whole complex (3),
is compatible with this filtration (as detailed in Theorem 1 below). The spectral sequence of
a filtered complex [9] now comes into play, having as its E0-level the following.
-
6
p
q
V0 Λ1 ⊗ V0 Λ2 ⊗ V0 Λ3 ⊗ V0 Λ4 ⊗ V0 Λ5 ⊗ V0 · · ·
· · ·
· · ·
· · ·
· · ·
↑∂ ↑∂ ↑∂ ↑∂ ↑∂
V1 Λ1 ⊗ V1 Λ2 ⊗ V1 Λ3 ⊗ V1 Λ4 ⊗ V1
↑∂ ↑∂ ↑∂ ↑∂
V2 Λ1 ⊗ V2 Λ2 ⊗ V2 Λ3 ⊗ V2
↑∂ ↑∂ ↑∂
V3 Λ1 ⊗ V3 Λ2 ⊗ V3
↑∂ ↑∂
V4 Λ1 ⊗ V4
(6)
Here, the precise positioning of the coördinate axes is a matter of convention. The important
property of the E0-level is that the differentials ∂ are simply homomorphisms of vector bundles
and we shall show that they are induced by a complex of G0-modules
V ∂−→ g∗− ⊗ V ∂−→ Λ2g∗− ⊗ V ∂−→ Λ3g∗− ⊗ V ∂−→ · · · ∂−→ Λn−1g∗− ⊗ V ∂−→ Λng∗− ⊗ V,
4 M.G. Eastwood and A.R. Gover
where
G0 =
∗ 0 · · · 0
0
... ∗
0
∈ SL(n+ 1,R)
.
Furthermore, we shall show that this complex defines the Lie algebra cohomology Hr(g−,V),
which in turn has been computed by Kostant [17]. It follows that the E1-level of the spectral
sequence is rather sparse, typically
-
6
p
q
H0 0 0 0 0 0 · · ·
· · ·
· · ·
· · ·
· · ·
0 H1 H2 0 0
0 0 0 H3
0 0 0
0 0
(7)
where H0 = V0 and, in particular, there is precisely one irreducible bundle in each diago-
nal Ep,d−p1 for d fixed. Since (3) resolves V, we know that this spectral sequence is also converging
to V and the only way that this can happen is if the differentials fit together as a resolution
0→ V→ H0 → H1 → H2 → H3 → · · · → Hn−1 → Hn → 0.
This is the required BGG resolution. The rest of the article is devoted to filling in the details
of this argument.
3 The filtering of V as a P -module
Recall the decomposition (1) of g = sl(n+ 1,R) and consider the element
H =
1
n+ 1
n 0 · · · 0
0 −1 0
...
. . .
0 0 −1
∈ g0.
uniquely characterised as lying in the centre of g0 with [H,X] = X, for X ∈ g+. It is called the
grading element [5] of the |1|-graded Lie algebra (1). If the G-module V is restricted to G0, then
it splits into eigenspaces under H. For the standard representation by matrix multiplication on
column vectors for example,
H
0
v1
...
vn
= − 1
n+ 1
0
v1
...
vn
and H
x
0
...
0
=
n
n+ 1
x
0
...
0
. (8)
The BGG Complex on Projective Space 5
Rather than use the actual eigenvalues, which are rational in general, let us subtract the lowest
eigenvalue and write
V = V0 ⊕ V1 ⊕ V2 ⊕ · · · ⊕ VN−1 ⊕ VN (9)
for the eigenspace decomposition, noting that g+ acts by Vj → Vj+1 for all j. It follows that
Vj ≡ Vj ⊕ Vj+1 ⊕ · · · ⊕ VN (10)
are P -submodules of V for all j and we have our filtration (4).
4 The filtered complex Λ• ⊗ V and its spectral sequence
The filtration (10) of V as a P -module certainly induces a filtration
V = V 0 ⊇ V 1 ⊇ V 2 ⊇ · · · ⊇ V N−1 ⊃ V N
of V by G-homogeneous vector bundles on G/P .
Theorem 1. The connection ∇ : V → Λ1⊗V is compatible with this filtration in the sense that
there is a commutative diagram
V
∇−→ Λ1 ⊗ V
∪ ∪
V k ∇−→ Λ1 ⊗ V k−1
for all k = 1, 2, . . . , N .
Proof. Because the assertion is local, without loss of generality it suffices to prove it on the
standard affine coördinate patch, namely
Rn = Q/G0 ⊂ G/P = RPn, where Q =
∗ 0 · · · 0
∗
... ∗
∗
∈ SL(n+ 1,R)
. (11)
Recall (9) that V splits as a G0-module. Hence the same is true of V restricted to this patch:
V |Rn = V0 ⊕ V1 ⊕ V2 ⊕ · · · ⊕ VN−1 ⊕ VN . (12)
To proceed we need a formula for ∇. The appendix discusses various natural constructions on
a general Lie group G, which we now specialise to be the Abelian Lie group
Rn = G− =
1 0 · · · 0
∗
... Id
∗
∈ SL(n+ 1,R)
.
and according to (23) we find that ∇ = d+ θ, where
• V |Rn is trivialised by
G− × V '→ Q×G0 V
∩ ↗ ‖
Q× V V |Rn
to define d : V |Rn → Λ1 ⊗ V |Rn ,
6 M.G. Eastwood and A.R. Gover
• θ : V |Rn → Λ1 ⊗ V |Rn is defined by the same trivialisation together with the Maurer–
Cartan form θ on G−; more specifically,
V |Rn ∼= G− × V θ⊗Id−−−→ Λ1 ⊗ g− ⊗ V Id⊗ρ−−−→ Λ1 ⊗ V ∼= Λ1 ⊗ V |Rn ,
where ρ : g− ⊗ V→ V is the representation of g on V restricted to g−.
Evidently, d preserves the splitting (12) whilst θ sends Vk to Vk−1 for all k = 1, 2, . . . , N . In
particular, ∇ sends V k = Vk ⊕ · · · to Λ1 ⊗ V k−1 = Λ1 ⊗ Vk−1 ⊕ · · · , as required. �
Corollary 1. The complex (3) is compatible with the filtration (5), i.e. ∇ sends Λp ⊗ V k to
Λp+1 ⊗ V k−1.
Proof. In fact, in the trivialisation V |Rn ∼= G− × V employed in the proof of Theorem 1
Λp ⊗ V |Rn ∼= Λp ⊗ V 3 ω ⊗ v ∇7−→ dω ⊗ v + (−1)pω ∧ (θ ρ)v ∈ Λp+1 ⊗ V ∼= Λp+1 ⊗ V |Rn
and the conclusion is manifest. �
According to this corollary, we may now consider the spectral sequence of the filtered complex
∇ : Λ• ⊗ V on G/P , the E0-level of which is
Ep,q0 = Λp+q ⊗ V−q with differential ∂ : Ep,q0 → Ep,q+1
0 , (13)
where we have chosen to normalise (6) by placing V0 at the origin. By construction, the bundle Vk
on G/P is the homogeneous bundle induced from Vk = Vk/Vk+1 as a P -module (cf. (10)). We
already know that, as an eigenspace for the grading element H in the centre of g0, the vector
space Vk is a G0-module. By regarding it as the quotient Vk/Vk+1 we are equivalently making Vk
into a P -module by decreeing that G+ act trivially. By construction, the E0-differential is G-
equivariant. Furthermore, its definition
0 → Λp+1 ⊗ V k−2 → Λp+1 ⊗ V k−1 → Λp+1 ⊗ Vk−1 → 0
∇ ↑ ∇ ↑ ∂ ↑
0 → Λp ⊗ V k−1 → Λp ⊗ V k → Λp ⊗ Vk → 0
and the Leibniz rule ensure that it is linear over the functions. In other words ∂ is a G-equivariant
homomorphism of homogeneous bundles and, as such, must be induced by a homomorphism of
P -modules Λp(g/p)∗ ⊗Vk → Λp+1(g/p)∗ ⊗Vk−1, which we shall also denote by ∂. In fact, from
the formula for ∇ displayed in the proof of Corollary 1, we see that ∂ is induced by
Λp(g/p)∗ ⊗ V = Λpg∗− ⊗ V 3 vβγ···δ
∂7−→ ρ[αvβγ···δ] ∈ Λp+1g∗− ⊗ V = Λp+1(g/p)∗ ⊗ V, (14)
where V is regarded as a P -module by restricting the G action to G0 and decreeing that G+ act
trivially. Since g− is Abelian, this formula agrees with (26) for the Koszul differential (used in [8]
to define Lie algebra cohomology). As observed in the appendix, this is a complex ofG0-modules.
Alternatively, we could come to the same conclusion by restricting attention to a standard affine
coördinate chart G− ∼= Rn ↪→ RPn and noticing that the kernel of d : Λp⊗V→ Λp+1⊗V consists
precisely of the left-invariant V-valued p-forms under the action of G− on itself. Since ∇ = d+∂,
we may now invoke the second realisation of Hr(g−,V) from Theorem 5 in the Appendix.
To proceed we need to be explicit concerning the irreducible representation V of SL(n+1,R).
In fact, one usually deals with complex representations of SL(n+1,C) (starting with sl(n+1,C))
but here there is no real difference and we shall adopt the notation from [2] in denoting such
representations by attaching non-negative integers to its Dynkin diagram
V = • • • • • • •· · ·
a1 a2 a3 a4 a5 · · · an−1 an
The BGG Complex on Projective Space 7
(meaning that −[a1, a2, . . . , an] is the lowest weight of this representation with respect to the
standard basis of fundamental weights). With this notation, here is the conclusion of Kostant’s
computation [17] of Lie algebra cohomology. We describe the result as an SL(n,R)-module (by
attaching non-negative integers to an A-series Dynkin diagram with one fewer nodes) and fix
the action of G0 by specifying how the grading element H ∈ g0 acts.
H0(g−,V) = • • • • • •· · ·
a2 a3 a4 a5 · · · an−1 an H ; −c
H1(g−,V) = • • • • • •· · ·
a1 + a2 + 1 a3 a4 a5 · · · an−1 an H ; −c+ a1 + 1
H2(g−,V) = • • • • • •· · ·
a1 a2 + a3 + 1 a4 a5 · · · an−1 an H ; −c+ a1 + a2 + 2
H3(g−,V) = • • • • • •· · ·
a1 a2 a3 + a4 + 1 a5 · · · an−1 an H ; −c+ a1 + a2 + a3 + 3
...
...
...
...
...
Hn−1(g−,V) = • • • • • •· · ·
a1 a2 a3 a4 · · · an−2 an−1 + an + 1 H ; −c+ a1 + · · ·+ an−1 + n− 1
Hn(g−,V) = • • • • • •· · ·
a1 a2 a3 a4 · · · an−2 an−1 H ; −c+ a1 + · · ·+ an−1 + an + n
(15)
where
c =
na1 + (n− 1)a2 + (n− 2)a3 + · · ·+ 2an−1 + an
n+ 1
(obtained by acting on [a1, a2, . . . , an] with the first column of the inverse Cartan matrix for
sl(n+ 1)). Notice that each cohomology is an irreducible representation of G0.
Theorem 2. The E1-level of the spectral sequence of the filtered complex ∇ : Λ• ⊗ V consists
of irreducible homogeneous vector bundles on RPn under the action of G = SL(n+ 1,R). Only
the following terms are non-zero
E0,0
1 ! H0(g−,V)
Ea1+1,−a1
1 ! H1(g−,V)
Ea1+a2+2,−a1−a2
1 ! H2(g−,V)
Ea1+a2+a3+3,−a1−a2−a3
1 ! H3(g−,V)
...
...
...
...
...
EN+n−1−an,−N+an
1 = E
a1+a2+a3+···+an−1+n−1,−a1−a2−a3−···−an−1
1 ! Hn−1(g−,V)
EN+n,−N
1 = E
a1+a2+a3+···+an−1+an+n,−a1−a2−a3−···−an−1−an
1 ! Hn(g−,V)
meaning that the bundle in question is induced by the G0-module as listed and extended trivially
as a P -module.
Proof. According to (13), we already know that Λr ⊗ V is spread along the rth diagonal
Λr ⊗ V = Λr ⊗ V0 + Λr ⊗ V1 + Λr ⊗ V2 + · · · = Er,00 + Er+1,−1
0 + Er+2,−2
0 + · · ·
of the E0-level of the spectral sequence and that the E0-differential is induced by the Koszul
differential (14) defining the Lie algebra cohomology Hr(g−,V). We see from (15) that each of
these Hr(g−,V) is irreducible and so it follows that the rth diagonal of the E1-level consists of
a single irreducible homogeneous vector bundle. To complete the proof it suffices to locate the
position of this bundle along this particular diagonal and, to do this, the action of the grading
element H turns out to be sufficient. More precisely,
H acts by −c on V0 ∴ H acts by −c+ k on Vk,
H acts by −1 on g− ∴ H acts by p on Λpg∗−,
∴ H acts by −c+ k + p on Λpg∗− ⊗ Vk,
∴ H acts by −c+ p on Λp+qg∗− ⊗ V−q! Ep,q0 .
8 M.G. Eastwood and A.R. Gover
Therefore H acts by −c + p on the G0-module inducing Ep,q1 and so, from the action of H in
table (15), the bundle induced by Hr(g−,V) is located at Ea1+a2+···+ar+r,−a1−a2−···−ar
1 .
Finally, we are claiming that N = a1 + a2 + · · ·+ an. To see this we note that, since H acts
on V0 by −c, it acts on VN by N − c. On the other hand, in accordance with the action of the
longest element of the Weyl group,
V∗ = • • • • • • •· · ·
an an−1 · · · a5 a4 a3 a2 a1 = V∗N + V∗N−1 + · · ·+ V∗2 + V∗1 + V∗0
and so H acts on V∗N by −c′, where
c′ =
nan + (n− 1)an−1 + · · ·+ 3a3 + 2a2 + a1
n+ 1
.
We conclude that N = c+ c′ = a1 + a2 + a3 + · · ·+ an−1 + an, as required. �
As outlined in Section 2, the E1-level of this spectral sequence is rather sparse (7) and
Theorem 2 says exactly how sparse. In particular, since there is only one non-zero bundle on
each diagonal of the E1-level, the general theory of spectral sequences provides a complex of
differential operators
H0 → H1 → H2 → H3 → · · · → Hn−1 → Hn → 0
whose cohomology on the level of sheaves coincides with that of (3), namely V for ker : H0 → H1
and otherwise zero. The bundle Hr is induced on RPn = G/P from the G0-module Hr(g−,V)
extended trivially as as P -module. This is the BGG resolution (constructed on a general G/P
by Lepowsky [18] on the level of generalised Verma modules). More explicitly, in the discussion
following Corollary 1, in any affine coördinate patch we identified ∇ : Λp ⊗ V → Λp+1 ⊗ V as
d+ ∂ where V |Rn is trivialised as Rn × V and
Ep,q0 = Λp+q ⊗ V−q ∼= Λp+q ⊗ V−q
d−→ Λp+q+1 ⊗ V−q ∼= Λp+q+1 ⊗ V−q = Ep+1,q
0
is the exterior derivative with values in V−q. Adding these operators to the diagram (6) gives
a double complex as employed by Baston [1, p. 120]. We conclude that in any affine coördinate
patch, our spectral sequence (of a filtered complex) coincides with Baston’s spectral sequence
(of a double complex). In particular, the operators Hk → Hk+1 are obtained as zigzag compo-
sitions of the first order differential operators d together with choices of algebraic splittings of
the operators ∂ (Baston and others use the algebraic adjoint ∂∗ introduced by Kostant [17]).
This confirms that the resulting operators Hk → Hk+1 are differential. By using the spectral
sequence of a filtered complex as we have done, it is manifest that the differential operators
Hk → Hk+1 are independent of any choice of splittings and that the whole construction and
resulting BGG complex is G-equivariant. It is also clear that the first steps in this approach
can be taken on an arbitrary homogeneous space G/P .
5 Formulæ for the BGG operators
Already, the affine invariance of the operators occurring in the BGG resolution on RPn fixes
their formulæ with respect to the flat connection on Rn ⊂ RPn as follows (affine invariance
being ensured by noting that Q as in (11) is acting on Rn = Q/G0 by affine transformations).
Firstly, consider the symbol of the BGG differential operator Hr → Hr+1. As a homomorphism
of homogeneous bundles
⊙sΛ1 ⊗Hr → Hr+1 for some s, it is induced by a homomorphism of
G0-modules⊙sg∗− ⊗Hr(g−,V)→ Hr+1(g−,V)
The BGG Complex on Projective Space 9
for some s, which we can determine just from the action of the grading element. Specifically,
we know that
H acts by −c+ a1 + a2 + · · ·+ ar + r on Hr(g−,V)
H acts by −1 on g−
∴ H acts by s on
⊙sg∗−.
It is immediate that s = ar+1+1. Furthermore, from the Littlewood–Richardson rules (e.g., [14]),
as an sl(n,R)-module Hr+1(g−,V) occurs with multiplicity one in the following decomposition⊙ar+1+1g∗− ⊗Hr(g−,V) = · · · ⊕Hr+1(g−,V)⊕ · · ·
‖ ‖
• • •· · ·
ar+1 + 1 0 · · · 0 ⊗ • • • •· · · · · ·
a1 · · · ar + ar+1 + 1 ar+2 · · · an = ⊕ • • • •· · · · · ·
a1 · · · ar ar+1 + ar+2 + 1 · · · an⊕
The symbol of the BGG operator Hr → Hr+1 is thus determined uniquely (up to scale) as
induced by the projection onto this summand. Similarly, there are no lower order terms since
no appropriate invariant homomorphisms are available. We record this conclusion as follows.
Theorem 3. Each bundle Hr in the BGG complex on RPn is an irreducible tensor bundle and
the differential operator ∇(s) : Hr → Hr+1 is given by φ 7→ π(∇sφ) where ∇ is the flat affine
connection and π :
⊙sΛ1⊗Hr → Hr+1 is the unique projection onto this irreducible summand.
Equivalently, the bundles Hr and the operators between them are exhibited as Young tableau
in [12]. It is often useful, however, to be able to write the BGG complex globally on RPn
without recourse to affine coördinates and projective invariance. For this, we shall use the
round metric on the sphere and the corresponding Levi Civita connection on Sn or RPn. By
way of normalisation, if gab denotes the round metric and δa
b the identity endomorphism on the
tangent bundle, let us choose the radius of the sphere so that
Rab
c
d = δa
cgbd − δbcgad where (∇a∇b −∇b∇a)Xd = Rab
c
dX
d (16)
defines the Riemann curvature tensor. For any irreducible covariant tensor bundle E on RPn,
let us write ∇2 for the composition⊙rΛ1 ⊗ E ∇◦∇−−−→ Λ1 ⊗ Λ1 ⊗
⊙rΛ1 ⊗ E →
⊙r+2Λ1 ⊗ E
and g for the composition⊙rΛ1 ⊗ E g⊗Id−−−→
⊙2Λ1 ⊗
⊙rΛ1 ⊗ E →
⊙r+2Λ1 ⊗ E.
Theorem 4. If s is odd, the operator ∇(s) : Hr → Hr+1 is given by
π
(
(∇2 + (s− 1)2g) · · · (∇2 + 16g)(∇2 + 4g)∇
)
and, if s is even,
π
(
(∇2 + (s− 1)2g) · · · (∇2 + 9g)(∇2 + g)
)
.
Proof. It is convenient to use some notation from [12] where the projective BGG operators are
written as acting between covariant tensor bundles specified by weighted Young tableau:
· · ·· · ·
· · ·· · ·
· · ·
· · ·· · ·
· · ·· · · · · ·· · ·
· · · · · ·· · · · · ·
� -b
(w)
∇(s)
−−−→
· · ·· · ·
· · ·· · ·
· · ·
· · ·· · ·
· · ·· · · · · ·· · ·
· · · · · ·· · · · · ·
� -b+s
(w) ....
10 M.G. Eastwood and A.R. Gover
Here, if it is the rth row to which the boxes on the right hand side are being added and there are
a total of v boxes on the left hand side (so that v is the valence of the corresponding tensor),
then w + r = s+ v + b (see [12]). In particular, if we consider only first order operators
· · ·· · ·
· · ·· · ·
· · ·
· · ·· · ·
· · ·· · · · · ·· · ·
· · · · · ·· · · · · ·
� -b
(w)
∇−−−→
· · ·· · ·
· · ·· · ·
· · ·
· · ·· · ·
· · ·· · · · · ·· · ·
· · · · · ·· · · · · ·
� -b+1
(w) ,
(17)
then w = v + b + 1 − r. More generally, taking conventions from [12], if ∇ and ∇̂ are two
torsion-free connections in the same projective class
∇̂aφb = ∇aφb −Υaφb −Υbφa
for some 1-form Υa, then for φbc···d having symmetries and projective weight specified by the
left hand side of (17),
π(∇̂aφbc···d) = π
(
∇aφbc···d + (w − (v + b+ 1− r))Υaφbc···d
)
,
where π is the Young projector corresponding to the right hand side of (17). We may iterate
this formula, adding more boxes to the rth row (assuming that there is room to do so) and, each
time, both b and v increase by one. Suppressing indices, after s iterations the result is that
π(∇̂sφ) = π
(
(∇+ (k − 2s+ 2)Υφ) · · · (∇+ (k − 4)Υφ)(∇+ (k − 2)Υφ)(∇+ kΥ)φ
)
,
where we are writing k = w − (v + b+ 1− r). In particular, if s = w + r − v − b as it is in the
case of a BGG operator, then this iteration reads
π(∇̂sφ) = π
(
(∇− kΥφ) · · · (∇+ (k − 4)Υφ)(∇+ (k − 2)Υφ)(∇+ kΥ)φ
)
,
where k = s−1. So far, this conclusion holds under any projective change of connection but now
we specialise to the case of the round connection ∇ on the sphere, being projectively equivalent
to the flat connection ∇̂ (under gnomonic projection). The general formula [12, (3.4)] for the
change in the Ricci tensor specialises to
0 = gab −∇aΥb + ΥaΥb or, suppressing indices, ∇Υ = g + Υ2.
In this equation Υ is viewed as a tensor but if it is viewed as an operator φ
Υ7−→ Υφ, then we
should write ∇Υ = Υ∇ + g + Υ2. This equation allows us to deal with the iterated formula
above. For example, when k = 2, also bearing in mind that as operators ∇ and g commute,
π
(
∇̂3φ
)
= π
(
(∇− 2Υ)∇(∇+ 2Υ)
)
= π
(
∇3 − 2Υ∇2 + 2∇(∇Υ)− 4Υ(∇Υ)
)
= π
(
∇3 − 2Υ∇2 + 2∇
(
Υ∇+ g + Υ2
)
− 4Υ
(
Υ∇+ g + Υ2
))
= π
(
∇3 − 2Υ∇2 + 2(∇Υ)∇+ 2g∇+ 2(∇Υ)Υ− 4Υ2∇− 4gΥ− 4Υ3
)
= π
(
∇3 − 2Υ∇2 + 2
(
Υ∇+ g + Υ2
)
∇+ 2g∇+ 2(∇Υ)Υ− 4Υ2∇− 4gΥ− 4Υ3
)
= π
(
∇3 + 4g∇+ 2
(
Υ∇+ g + Υ2
)
Υ− 2Υ2∇− 4gΥ− 4Υ3
)
= π
(
∇3 + 4g∇+ 2Υ(∇Υ)− 2Υ2∇− 2gΥ− 2Υ3
)
= π
(
∇3 + 4g∇+ 2Υ
(
Υ∇+ g + Υ2
)
− 2Υ2∇− 2gΥ− 2Υ3
)
= π(∇3 + 4g∇) = π
((
∇2 + 4g
)
∇
)
,
as advertised in the statement of the theorem. For higher k, direct calculations rapidly get out
of hand. Instead, it suffices to prove the following lemma in which we have isolated the required
algebra (and then we prove the lemma by indirect means). �
The BGG Complex on Projective Space 11
Remark 1. Our curvature normalisation (16) implies that the Ricci tensor on our round sphere
is given by
Rab ≡ Rcacb = (n− 1)gab.
Thus, the metric gab coincides with 1
n−1Rab, a tensor generally known in projective differential
geometry [12] as the Schouten tensor or Rho-tensor Pab. Replacing g by P in the formulæ of
Theorem 4 gives expressions that are valid on any space of constant curvature. More generally,
there is a Rho-tensor that arises in similar contexts [4] within parabolic differential geometry [5].
Lemma 1. Define an associative algebra R = R〈∇,Υ〉 with generators subject to the ‘Riccati
relation’ ∇Υ = Υ∇+ 1 + Υ2. Then the following identities hold in R. If k is even, then
(∇− kΥ)(∇− (k − 2)Υ) · · · (∇+ (k − 2)Υ)(∇+ kΥ) = (∇2 + k2) · · · (∇2 + 16)(∇2 + 4)∇
and, if k is odd, then
(∇− kΥ)(∇− (k − 2)Υ) · · · (∇+ (k − 2)Υ)(∇+ kΥ) = (∇2 + k2) · · · (∇2 + 9)(∇2 + 1).
Proof. The algebra R may be realised by the following differential operators on the circle
f(θ)
∇7−→ df(θ)/dθ, f(θ)
Υ7−→ (tan θ)f(θ).
To see this, note that these operators certainly satisfy the Riccati relation and we are required,
therefore, to show that they satisfy no further relations. WithinR we may normalise any element
as follows. By induction, the Riccati relation extends to
∇Υ` = Υ`∇+ `
(
Υ`−1 + Υ`+1
)
, ∀ ` ≥ 1
whence
∇kΥ` = ∇k−1
(
Υ`∇+ `
(
Υ`−1 + Υ`+1
))
=
(
∇k−1Υ`
)
∇+ `
(
∇k−1Υ`−1
)
+ `
(
∇k−1Υ`+1
)
and it follows by induction on k that
∇kΥ` = Υ`∇k + k`
(
Υ`−1 + Υ`+1
)
∇k−1
+ 1
2k(k − 1)`
(
(`− 1)Υ`−2 + 2`Υ` + (`+ 1)Υ`+2
)
∇k−2 + · · · ,
where the ellipsis · · · denotes terms of lower order in ∇ with coefficients that are real polynomial
in Υ. It follows that every element of R can be written uniquely in the form
k∑
p=0
Ap(Υ)∇p
for suitable real polynomials Ap(Υ). In our claimed realisation, such an element is represented
by the differential operator
k∑
p=0
Ap(tan θ) dp/dθp
and now it suffices to observe (by acting on 1, θ, θ2, . . . , θk near 0) that such a differential
operator vanishes if and only if all the polynomials Ap(Υ) are zero. (More precisely, we should
restrict the action of such operators to smooth functions on (−π/2, π/2) or some other suitable
function space.)
Having realised R by differential operators, we are reduced to proving identities amongst
these operators. This is accomplished in the following lemma. �
12 M.G. Eastwood and A.R. Gover
Lemma 2. Let D denote the differential operator f(θ) 7→ (cos2 θ)(df(θ)/dθ) on the circle. Then
the following identities hold. If k is even, then
1
cosk θ
d
dθ
(
Dk
(
f(θ)
cosk θ
))
=
(
d2
dθ2
+ k2
)
· · ·
(
d2
dθ2
+ 16
)(
d2
dθ2
+ 4
)
d
dθ
f(θ)
and, if k is odd, then
1
cosk θ
d
dθ
(
Dk
(
f(θ)
cosk θ
))
=
(
d2
dθ2
+ k2
)
· · ·
(
d2
dθ2
+ 9
)(
d2
dθ2
+ 1
)
f(θ).
Proof. Writing Dk+1 for the operators on the left-hand-sides of the displays in this lemma, the
following identity is easily verified
Dk+3 =
(
d
dθ
− (k + 2) tan θ
)
Dk+1
(
d
dθ
+ (k + 2) tan θ
)
.
It follows that, in our realisation of the algebra R, we obtain the expressions on the left-hand-
sides of the displays in the Lemma 1.
On the other hand, the operators on the right-hand-sides of the claimed identities, in the
current lemma, are characterised up to scale as annihilating the functions
cos(kθ), sin(kθ), cos((k − 2)θ), sin((k − 2)θ), cos((k − 4)θ), . . . .
Since all operators have dk+1/dθk+1 as leading term, it is therefore sufficient to show that the
left hand sides of these purported identities have the same property. Notice that there is an
invertible relationship
cos(mθ) = 2m−1 cosm θ + · · · = Tm(cos θ),
where Tm is the mth Chebyshev polynomial of the first kind and a similar invertible relationship
sin(mθ) = (sin θ)
(
2m−2 cosm−1 θ + · · ·
)
= (sin θ)Um−1(cos θ)
where Um−1 is the (m−1)st Chebyshev polynomial of the second kind. Only the degree of these
Chebyshev polynomials concerns us and it now suffices to show that Dk+1 annihilates the first
k + 1 of
1,
sin θ
cos θ
,
1
cos2 θ
,
sin θ
cos3 θ
,
1
cos4 θ
,
sin θ
cos5 θ
,
1
cos6 θ
,
sin θ
cos7 θ
, . . . .
Since
D
1
cos2` θ
= 2`
sin θ
cos2`−1 θ
and D
sin θ
cos2`−1 θ
= (2`− 1)
1
cos2(`−1) θ
− 2(`− 1)
1
cos2(`−2) θ
,
this follows easily by induction. �
Remark 2. More generally, the formula
Dk+1f = u−k−2
(
u2 d
dx
)k+1
u−kf, where
(
d2
dx2
+ Φ
)
u = 0
is used in [15] to derive expressions for BGG operators in conformal geometry and in [4] these
expressions are extended to parabolic geometries in general. Lemma 2 concerns the case Φ = 1.
The BGG Complex on Projective Space 13
6 An example
The following is amongst the simplest of non-trivial examples. Let us take n = 2 so G = SL(3,R)
and let V = R3, regarded as column vectors with SL(3,R) acting by left matrix multiplication.
We saw in (8) how H splits R3 and in Section 3 that this results in the filtering
R3 = R2 + R as a P -module
where SL(2,R) ⊂ P acts on R2 as the standard representation and acts trivially on R. Dropping
projective weights, the corresponding bundle V on RP2 is filtered
V = V0 + V1 = T + Λ0,
where T is the tangent bundle and Λ0 is the trivial bundle. The E0-level (6) of our spectral
sequence becomes
-
6
p
q
T Λ1 ⊗ T Λ2 ⊗ T 0
↑∂ ↑∂
0 Λ0 Λ1 Λ2
and one checks that ∂ : Λ0 → Λ1 ⊗ T and ∂ : Λ1 → Λ2 ⊗ T are given by
µ 7→ δb
cµ and µb 7→ δ[a
cµb],
respectively. In this simple case, we do not need Kostant’s Theorem to see that
∂ : Λ0 → Λ1 ⊗ T is injective with cokernel = (Λ1 ⊗ T )◦,
∂ : Λ1 → Λ2 ⊗ T is an isomorphism,
where (Λ1 ⊗ T )◦ denotes the trace-free part of Λ1 ⊗ T . Therefore, the E1-level (7) is
-
6
p
q
T → (Λ1 ⊗ T )◦ 0 0
0 0 0 Λ2
and we obtain
0→ R3 → T → (Λ1 ⊗ T )◦ → Λ2 → 0
as the resulting BGG complex. Writing out the flat connection ∇ on V in terms of the round
connection gives σc
µ
∇7−→
∇bσc + δb
cµ
∇bµ− σb
and
σb
c
µb
∇7−→
∇[aσb]
c + δ[a
cµb]
∇[aµb] + σ[ab]
for the two operators in V → Λ1 ⊗ V → Λ2 ⊗ V and, noting that
Λ1 ⊗ V 3
σb
c
−∇cσbc
∇7−→ −
0
∇c∇[aσb]
c + gc[aσb]
c
,
we obtain a formula for the projectively invariant operator (Λ1 ⊗ T )◦
∇(2)
−→ Λ2 in agreement with
Theorem 4.
14 M.G. Eastwood and A.R. Gover
Appendix: Lie algebra cohomology as a geometrical construction
Although in this article we shall need only the cohomology of an Abelian Lie algebra, we take the
opportunity here to describe the cohomology of a general Lie algebra g in terms of differential
geometry on G, a Lie group whose Lie algebra is g. We believe that for a general parabolic
geometry, we shall need this geometric interpretation for a nilpotent Lie algebra. Suppose V
is a G-module and use the same notation for the corresponding representation of g. Following
but adapting [8], we are going to present the Lie algebra cohomology Hr(g,V) as a geometrical
construction on G. Beware that G is no longer the Lie group SL(n+ 1,R) as it was until now.
This section is written to be self-contained with the aim of being useful elsewhere. This material
is well-known to experts and implicit in [8] but we believe it worthwhile laying out the details.
We shall view G as a homogeneous space under its own action on the left. Its tangent
bundle TG is then regarded as a homogeneous bundle and can be identified as G× g, where g
is the Lie algebra of G. It is convenient to write this isomorphism as
X 7→ X θ for vector fields X on G, (18)
where θ is a 1-form on G with values in g known as the Maurer–Cartan form [23]. To compute
with θ it is convenient to write functions on G with values in g as Xα and then (18) becomes
Xa 7→ Xα ≡ θαaXa with inverse Xα 7→ Xa ≡ φaαXα,
where φaαθ
β
a = δα
β and θαaφ
b
α = δa
b. A vector field Xa on G is left-invariant if and only if the
corresponding function Xα : G→ g is constant. Choosing any torsion-free affine connection ∇a
on G and expanding the definition ∇a(θαbXb) = 0, the left-invariant vector fields are those that
satisfy
θαb∇aXb −Xb∇aθαb = 0
from which it follows easily that the Lie bracket of two left-invariant vector fields is again left-
invariant. Since the left-invariant vector fields on G are of the form φ(X) for X ∈ g we may
define the Lie bracket on g by transportation:
φ([X,Y ]) = [φ(X), φ(Y )] for X,Y ∈ g. (19)
For computational purposes, let us write [X,Y ]γ = Γαβ
γXαY β for the Lie bracket on g. Then
we can write out (19) explicitly as
φcγΓαβ
γXαY β = [φaαX
α, φbβY
β]c = φaαX
α∇a(φcβY β)− φbαY α∇b(φcβXβ)
= (φaα∇aφcβ − φaβ∇aφcα)XαY β
or, in other words, as
Γαβ
γφcγ = φaα∇aφcβ − φaβ∇aφcα.
Bearing in mind that φbβθ
γ
b = δβ
γ whence θγb∇aφ
b
β + φbβ∇aθ
γ
b = 0, we may rewrite this as
Γαβ
γ = −φaαφbβ∇aθ
γ
b + φaβφ
b
α∇aθ
γ
b = −φaαφbβ(∇aθγb −∇bθ
γ
a)
and finally as
∇aθγb −∇bθ
γ
a + Γαβ
γθαa θ
β
b = 0. (20)
The BGG Complex on Projective Space 15
This formula employs an arbitrary torsion-free connection. Without indices and without this
connection it is more usually written as
dθ + 1
2 [θ, θ] = 0 or dθ + θ ∧ θ = 0. (21)
In other words, the definition (19) of the Lie bracket on g is equivalent to (20) or (21), usually
known as the Maurer–Cartan equation [23].
Now suppose V is a G-module and use the same notation for the corresponding g-module.
There are two canonically defined connections on the vector bundle V = G × V over G. One
of them is the evident flat connection, ignoring the action of G on V. We shall denote it by d
since it is the exterior derivative acting on functions with values in V. The other one takes the
isomorphism from (2)
V = G× V ∼= G× V by (g, v) 7→ (g, gv) (22)
and pulls back the evident flat connection on the right hand side as was done in Section 2 and
we shall denote this one by ∇ as was done there. To relate these two connections more explicitly
suppose f : G→ V is a section of V that is constant after the twisting (22). It means that the
function g 7→ gf(g) is constant. If so, then for fixed g ∈ G and X ∈ g, the function
R 3 t 7→ getXf
(
getX
)
∈ V
is constant. Equivalently, the function t 7→ etXf(getX) is constant and so
0 =
d
dt
(
etXf
(
getX
))∣∣∣
t=0
=
d
dt
f
(
getX
)∣∣
t=0
+Xf(g) = (φ(X)f +Xf)(g),
where this last equality is due to the flow of the left-invariant vector field φ(X) being the one-
parameter subgroup of right-translations g 7→ getX (see, e.g. [25]). For computational purposes,
let us write Xα 7→ Xαρα where ρα ∈ g∗ ⊗ End(V) for the action of g on V. Then g 7→ gf(g) is
constant if and only if
0 = φ(X)f +Xf = Xαφaαdaf +Xαραf = Xa(daf + θαa ραf)
for all left-invariant vector fields Xa. It follows that the connection ∇a on V is given by
f 7→ ∇af = daf + θαa ραf or, without indices, as f 7→ ∇f = df + θf, (23)
where θ ∈ Λ1⊗g is the Maurer–Cartan form. As a check, the differential in the coupled de Rham
sequence (3) is
Λp ⊗ V 3 ω ∇7−→ dω + θ ∧ ω ∈ Λp+1 ⊗ V
and the Maurer–Cartan equation (21) shows that the composition V ∇→ Λ1 ⊗ V ∇→ Λ2 ⊗ V is
given by
∇2f = d(df + θf) + θ ∧ (df + θf) = (dθ + θ ∧ θ)f = 0
and the connection ∇ is flat, as expected. More generally, the whole sequence
0→ V ∇−→ Λ1 ⊗ V ∇−→ Λ2 ⊗ V ∇−→ · · · ∇−→ Λp ⊗ V ∇−→ Λp+1 ⊗ V ∇−→ · · · (24)
is a complex.
Consider the linear mapping
Λpg∗ ⊗ V 3 vαβ···γ
θ7−→ vab···c ≡ θαa θ
β
b · · · θ
γ
c vαβ···γ ∈ Γ(G,Λp ⊗ V)
in which V is just a passenger (i.e. plays no rôle). We shall refer to the resulting V-valued p-form
as left-invariant just as we would if V were absent.
16 M.G. Eastwood and A.R. Gover
Lemma 3. The connection ∇ : Λp ⊗ V→ Λp+1 ⊗ V preserves left-invariance.
Proof. We use the Maurer–Cartan equation (20) to compute
∇[avbc···d] = d[a(θ
β
b θ
γ
c · · · θδd]vβγ···δ) + θα[aθ
β
b θ
γ
c · · · θδd]ραvβγ···δ
= p(d[aθ
ε
b)θ
γ
c · · · θδd]vεγ···δ + θα[aθ
β
b θ
γ
c · · · θδd]ραvβγ···δ
= −(p/2)Γαβ
εθα[aθ
β
b θ
γ
c · · · θδd]vεγ···δ + θα[aθ
β
b θ
γ
c · · · θδd]ραvβγ···δ
= θα[aθ
β
b θ
γ
c · · · θδd]ραvβγ···δ + (−1)p(p/2)θα[aθ
β
b θ
γ
c · · · θδd]Γαβ
εvγ···δε
= θαa θ
β
b θ
γ
c · · · θδd
(
ρ[αvβγ···δ] + (−1)p(p/2)Γ[αβ
εvγ···δ]ε
)
, (25)
as required. �
In fact (25) shows that ∇θv = θ∂v, where
∂ : Λpg∗ ⊗ V→ Λp+1g∗ ⊗ V is given by
vβγ···δ 7→ ρ[αvβγ···δ] + (−1)p(p/2)Γ[αβ
εvγ···δ]ε. (26)
It also follows that
0→ V ∂−→ g∗ ⊗ V ∂−→ Λ2g∗ ⊗ V ∂−→ · · · ∂−→ Λpg∗ ⊗ V ∂−→ Λp+1g∗ ⊗ V ∂−→ · · · (27)
is a complex of g-modules. Alternatively, this may be directly verified from (26) using
• ρ[αρβ] = 1
2Γαβ
γργ (i.e. that ρ : g→ End(V) is a representation),
• Γ[αβ
δΓγ]δ
ε = 0 (i.e. the Jacobi identity in g),
• (Xv)βγ···δ = Xαραvβγ···δ + (−1)ppXαΓα[β
εvγ···δ]ε (the action of g on Λpg∗ ⊗ V).
We have shown that there are two ways of defining Lie algebra cohomology as follows.
Theorem 5. The Lie algebra cohomology Hr(g,V) may be defined as either
• the cohomology of the Koszul complex (27), or
• the cohomology of the complex (24) restricted to left-invariant forms.
Remark 3. Although we use this theorem in the main body of this article, it is easily avoided.
In tackling a general parabolic geometry, however, we believe that Theorem 5 will be essential.
Remark 4. As a minor variation on this construction, suppose G is enlarged to Q, a semi-direct
product
Q = G0 nG i.e. Id→ G�Q −→ G0 → Idi
and suppose that V extends to a representation of Q. We identify G with the Q-homogeneous
space Q/G0, noting that when the action of Q on G = Q/G0 is restricted to G it coincides with
its usual action of G on itself by left translation. The Q-homogeneous bundle V ≡ Q ×G0 V
on Q/G0 is equipped with a flat connection ∇ by dint of the canonical trivialisation
V = Q×G0 V 3 (q, v) 7→ (qG0, qv) ∈ Q/G0 ⊗ V = G× V,
which clearly coincides with ∇ defined by (22). This connection is Q-equivariant. Consequently,
not only does the twisted de Rham complex
V
∇−→ Λ1 ⊗ V ∇−→ Λ2 ⊗ V ∇−→ · · · ∇−→ Λp ⊗ V ∇−→ Λp+1 ⊗ V ∇−→ · · ·
coincide with (24) and thereby compute the Lie algebra cohomology Hr(g,V) when restricted
to G-invariant forms, but also the complex (27) is automatically one of Q-modules where the
Q-action on V is as supposed and the Q-action on g∗ is induced by the conjugation action of Q
on G (bearing in mind that G is a normal subgroup of Q).
The BGG Complex on Projective Space 17
Addendum: A canonical connection on G
Again, although it is unnecessary for the current article and already known to experts, we
suspect that the following optional extra will be invaluable in dealing with a general parabolic
geometry. Since we already have established suitable notation in the Appendix above, we take
the opportunity of presenting it here. Our canonical connection Da was introduced in [7] as the
‘(+)-connection’ and Lemma 4 is stated without proof as [16, Proposition 2.12].
The trivialisation T ∗G = G × g∗ provided by the Maurer–Cartan form also equips G with
a canonical flat affine connection Da defined by
Daωb ≡ θβb daωβ,
where ωβ ≡ φcβωc and da on the right hand side of this equation simply takes the gradient of
a function with values in g∗. If we expand using any torsion-free affine connection ∇a
Daωb = θβb∇a(φ
c
βωc) = ∇aωb + (θβb∇aφ
c
β)ωc = ∇aωb − (φcβ∇aθ
β
b )ωc,
then we see that, for f a smooth function,
DaDbf −DbDaf = (−φcβ∇aθ
β
b + φcβ∇bθβa )Dcf = −φcγ(∇aθγb −∇bθ
γ
a)Dcf
and so the canonical connection Da has torsion
Tab
c = −(∇aθγb −∇bθ
γ
a)φcγ .
Alternatively, from (20) we see that
Tab
cθγc = −(∇aθγb −∇bθ
γ
a) = Γαβ
γθαa θ
β
b ,
which we record as the following lemma.
Lemma 4. The torsion of Da coincides with the Lie bracket on g under the Maurer–Cartan
parallelism.
Notice that Daθ
β
b = 0. It is another way to characterise Da and, indeed, is the main point of
this construction as follows.
Lemma 5. Even locally, the kernel of the induced operator
D : Λp → Λ1 ⊗ Λp
is the left-invariant forms on G.
Proof. Recall that the left-invariant forms are obtained as
Λpg∗ 3 vαβ···γ
θ7−→ vab···c ≡ θαa θ
β
b · · · θ
γ
c vαβ···γ ∈ Γ(G,Λp)
and it clear that such forms are annihilated by D. Conversely, since D is flat and all covariant
constant sections are already accounted for, there can be no more, even locally. �
Remark 5. Of course, this lemma also holds for V-valued differential forms where the connec-
tion is trivially coupled with V and it is this that we have in mind in constructing the BGG
complex in general.
18 M.G. Eastwood and A.R. Gover
Acknowledgements
Eastwood is supported by the Australian Research Council. Gover is partly supported by the
Royal Society of New Zealand (Marsden Grant 10-UOA-113). The authors would like to thank
Katharina Neusser for many valuable conversations and the referees for their useful corrections
and suggestions.
References
[1] Baston R.J., Almost Hermitian symmetric manifolds. II. Differential invariants, Duke Math. J. 63 (1991),
113–138.
[2] Baston R.J., Eastwood M.G., The Penrose transform. Its interaction with representation theory, Oxford
University Press, New York, 1989.
[3] Calderbank D.M.J., Diemer T., Differential invariants and curved Bernstein–Gelfand–Gelfand sequences,
J. Reine Angew. Math. 537 (2001), 67–103, math.DG/0001158.
[4] Calderbank D.M.J., Diemer T., Souček V., Ricci-corrected derivatives and invariant differential operators,
Differential Geom. Appl. 23 (2005), 149–175, math.DG/0310311.
[5] Čap A., Slovák J., Parabolic geometries. I. Background and general theory, Mathematical Surveys and
Monographs, Vol. 154, American Mathematical Society, Providence, RI, 2009.
[6] Čap A., Slovák J., Souček V., Bernstein–Gelfand–Gelfand sequences, Ann. of Math. (2) 154 (2001), 97–113,
math.DG/0001164.
[7] Cartan É., Schouten J.A., On the geometry of the group-manifold of simple and semi-simple groups, Nederl.
Akad. Wetensch. Proc. Ser. A 29 (1926), 803–815.
[8] Chevalley C.C., Eilenberg S., Cohomology theory of Lie groups and Lie algebras, Trans. Amer. Math. Soc.
63 (1948), 85–124.
[9] Chow T.Y., You could have invented spectral sequences, Notices Amer. Math. Soc. 53 (2006), 15–19.
[10] Eastwood M.G., A duality for homogeneous bundles on twistor space, J. London Math. Soc. (2) 31 (1985),
349–356.
[11] Eastwood M.G., Variations on the de Rham complex, Notices Amer. Math. Soc. 46 (1999), 1368–1376.
[12] Eastwood M.G., Notes on projective differential geometry, in Symmetries and Overdetermined Systems of
Partial Differential Equations, IMA Vol. Math. Appl., Vol. 144, Springer, New York, 2008, 41–60.
[13] Eastwood, M.G., Tod, K.P., Edth – a differential operator on the sphere, Math. Proc. Cambridge Philos.
Soc. 92 (1982), 317–330.
[14] Fulton W., Harris J., Representation theory, Graduate Texts in Mathematics, Vol. 129, Springer-Verlag,
New York, 1991.
[15] Gover A.R., Conformally invariant operators of standard type, Quart. J. Math. Oxford Ser. (2) 40 (1989),
197–207.
[16] Kobayashi S., Nomizu K., Foundations of differential geometry. II, Wiley Interscience, New York – London –
Sydney, 1969.
[17] Kostant B., Lie algebra cohomology and the generalized Borel–Weil theorem, Ann. of Math. (2) 74 (1961),
329–387.
[18] Lepowsky J., A generalization of the Bernstein–Gelfand–Gelfand resolution, J. Algebra 49 (1977), 496–511.
[19] Olver P., Differential hyperforms I, http://www.math.umn.edu/~olver/a_/hyper.pdf.
[20] Penrose R., Rindler W., Spinors and space-time. Vol. 1. Two-spinor calculus and relativistic fields, Cambridge
Monographs on Mathematical Physics, Cambridge University Press, Cambridge, 1984.
[21] Rice J.W., Private communication, January 1985.
[22] Schouten J.A., Ricci-calculus. An introduction to tensor analysis and its geometrical applications, Springer-
Verlag, Berlin, 1954.
[23] Sternberg S., Lie algebras, http://www.math.harvard.edu/~shlomo/docs/lie_algebras.pdf.
[24] Vogan D.A., Representations of real reductive Lie groups, Progress in Mathematics, Vol. 15, Birkhäuser,
Boston, Mass., 1981.
[25] Woodhouse N.M.J., Geometric quantisation, 2nd ed., Oxford University Press, New York, 1992.
http://dx.doi.org/10.1215/S0012-7094-91-06306-4
http://dx.doi.org/10.1515/crll.2001.059
http://arxiv.org/abs/math.DG/0001158
http://dx.doi.org/10.1016/j.difgeo.2004.07.009
http://arxiv.org/abs/math.DG/0310311
http://dx.doi.org/10.2307/3062111
http://arxiv.org/abs/math.DG/0001164
http://dx.doi.org/10.2307/1990637
http://dx.doi.org/10.1112/jlms/s2-31.2.349
http://dx.doi.org/10.1007/978-0-387-73831-4_3
http://dx.doi.org/10.1017/S0305004100059971
http://dx.doi.org/10.1017/S0305004100059971
http://dx.doi.org/10.1093/qmath/40.2.197
http://dx.doi.org/10.2307/1970237
http://dx.doi.org/10.1016/0021-8693(77)90254-X
http://www.math.umn.edu/~olver/a_/hyper.pdf
http://dx.doi.org/10.1017/CBO9780511564048
http://dx.doi.org/10.1017/CBO9780511564048
http://www.math.harvard.edu/~shlomo/docs/lie_algebras.pdf
1 Introduction
2 An outline of the construction
3 The filtering of V as a P-module
4 The filtered complex V and its spectral sequence
5 Formulæ for the BGG operators
6 An example
Appendix: Lie algebra cohomology as a geometrical construction
Addendum: A canonical connection on G
References
|
| id | nasplib_isofts_kiev_ua-123456789-147183 |
| institution | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| issn | 1815-0659 |
| language | English |
| last_indexed | 2025-12-07T18:07:37Z |
| publishDate | 2011 |
| publisher | Інститут математики НАН України |
| record_format | dspace |
| spelling | Eastwood, M.G. Gover, A.R. 2019-02-13T18:38:35Z 2019-02-13T18:38:35Z 2011 The BGG Complex on Projective Space / M.G. Eastwood, Gover A.R.// Symmetry, Integrability and Geometry: Methods and Applications. — 2011. — Т. 7. — Бібліогр.: 25 назв. — англ. 1815-0659 2010 Mathematics Subject Classification: 53A20; 17B56; 58D19; 58J70 DOI:10.3842/SIGMA.2011.060 https://nasplib.isofts.kiev.ua/handle/123456789/147183 We give a complete construction of the Bernstein-Gelfand-Gelfand complex on real or complex projective space using minimal ingredients. This paper is a contribution to the Special Issue “Symmetry, Separation, Super-integrability and Special Functions (S⁴)”. The full collection is available at http://www.emis.de/journals/SIGMA/S4.html.
 Eastwood is supported by the Australian Research Council. Gover is partly supported by the Royal Society of New Zealand (Marsden Grant 10-UOA-113). The authors would like to thank Katharina Neusser for many valuable conversations and the referees for their useful corrections and suggestions. en Інститут математики НАН України Symmetry, Integrability and Geometry: Methods and Applications The BGG Complex on Projective Space Article published earlier |
| spellingShingle | The BGG Complex on Projective Space Eastwood, M.G. Gover, A.R. |
| title | The BGG Complex on Projective Space |
| title_full | The BGG Complex on Projective Space |
| title_fullStr | The BGG Complex on Projective Space |
| title_full_unstemmed | The BGG Complex on Projective Space |
| title_short | The BGG Complex on Projective Space |
| title_sort | bgg complex on projective space |
| url | https://nasplib.isofts.kiev.ua/handle/123456789/147183 |
| work_keys_str_mv | AT eastwoodmg thebggcomplexonprojectivespace AT goverar thebggcomplexonprojectivespace AT eastwoodmg bggcomplexonprojectivespace AT goverar bggcomplexonprojectivespace |