8.04 Deck group

Below the video you will find accompanying notes and some pre-class questions.

Notes

Deck group

(0.00) Let p:YX be a path-connected covering space. Let G=π1(X,x), H=pπ1(Y,y) (where yp1(x)) and let NHG be the normaliser of H, that is the largest subgroup of G containing H for which HNH is normal. Then Deck(Y,p)NH/H.
In particular, if Y is a normal cover then H is normal in G, so NH=G and Deck(Y,p)G/H=π1(X,x)/pπ1(Y,y) as we suspected.
(2.48) We will write down a surjective homomorphism ϕ:NHDeck(Y,p) whose kernel is H. The first isomorphism theorem will then imply that Deck(Y,p)NH/H.

(3.57) We define ϕ as follows. Given an element βNH we have βpπ1(Y,y)β1=pπ1(Y,y),

and we also have βpπ1(Y,y)β1=pπ1(Y,σβ(y)), so by the existence and uniqueness results for deck transformations, there exists a unique deck transformation Fβ:YY such that Fβ(y)=σβ(y). We define ϕ(β)=Fβ.

(6.00) We need to check that ϕ is a homomorphism, that ϕ is surjective, and that kerϕ=H.

To show that ϕ is a homomorphism: Fβ1(Fβ2(y))=Fβ1(σβ2(y))=σβ1σβ2(y)=σβ1β2(y)=Fβ1β2(y),

where we used that the monodromy βσβ is a homomorphism.

(7.46) To see that ϕ is surjective, given a deck transformation F we want to find βNH such that F=Fβ. Let α be a path in Y from y to F(y). Let β=pα. This β is a loop in X because p(y)=p(F(y))=x and F(y)=σβ(y) by definition of monodromy. Therefore F=Fβ, because these covering transformations agree at y.

(10.38) To see that kerϕ=H, suppose that βkerϕ (so that Fβ=idY). Therefore y=Fβ(y)=σβ(y), so y is fixed by the monodromy around β. This means that the unique lift of β starting at y is a loop ˜β, so [β]=p[˜β]pπ1(Y,y). This shows that kerϕpπ1(Y,y). The inclusion pπ1(Y,y)kerϕ is an exercise.

Deck group of the universal cover

(13.23) If p:YX is a simply-connected covering space then Deck(Y,p)π1(X,x).

For example:

Pre-class questions

  1. Show that the deck group of a simply-connected covering space of X is isomorphic to π1(X,x).
  2. Show that pπ1(Y,y)kerϕ.
  3. There is a gap in the proof of surjectivity of ϕ. Can you find it? Can you fix it?

Navigation

Comments, corrections and contributions are very welcome; please drop me an email at j.d.evans at lancaster.ac.uk if you have something to share.

CC-BY-SA 4.0 Jonny Evans.