14 Dual isogenies and the Weil pairing

Let K be a perfect field and E∕K an elliptic curve.

Proposition 14.1. Assuming that:

  • Φ⊂E(K¯) be a finite Gal⁡(K¯∕K)-stable subgroup

Then there exists an elliptic curve E′∕K and a separable isogeny ϕ:E→E′ defined over K, with kernel Φ, such that every isogeny ψ:E→E″ with Φ⊂ψ factors uniquely via ϕ:
 E                E′′


ψϕ∃ unique  E ′

Proof. Omitted (see Silverman, Chapter III, Proposition 4.12). □

Proposition 14.2. Assuming that:

  • ϕ:E→E′ an isogeny of degree n

Then there exists a unique isogeny ϕ^:E′→E such that ϕ^ϕ=[n].

Proof. Case ϕ is separable: We have |ker⁡ϕ|=n, hence ker⁡ϕ⊂E[n]. Apply Proposition 14.1 with ψ=[n].

Case ϕ is inseparable: omitted.

Uniqueness: Suppose ψ1ϕ=ψ2ϕ=[n]. Then (ψ1−ψ2)ϕ=0, so deg⁡(ψ1−ψ2) deg ⁡ϕ=0. Then deg⁡(ψ1−ψ2)=0, hence ψ1=ψ2. □

Remark.

  • (i) Write E1∼E2 to mean “E1 and E2 are isogenous”. Then ∼ is an equivalence relation.
  • (ii) deg⁡[n]=n2 gives that deg⁡ϕ^= deg ⁡ϕ and [n]^=[n].
  • (iii) Note:
    ϕϕ^ϕ=ϕ∘[n]E=[n]E′∘ϕ

    Hence ϕϕ^=[n]E′.

    In particular, ϕ^^=ϕ.

  • (iv) If E→ψE′→ϕE″ then ϕψ^=ψ^ϕ^.
  • (v) If ϕ∈End⁡(E) then
    ϕ2−(Tr⁡ϕ)ϕ+ deg ⁡⁡ϕ=0.

    So

    ([Tr⁡ϕ]−ϕ)ϕ⏟=ϕ^=[ deg ⁡⁡ϕ].

    Hence [Tr⁡ϕ]=ϕ+ϕ^.

Lemma 14.3. Assuming that:

  • ϕ,ψ∈Hom⁡(E,E′)

Then ϕ+ψ^=ϕ^+ψ^.

Proof.

  • (i) If E=E′, then this follows from Tr⁡(ϕ+ψ)=Tr⁡(ϕ)+Tr⁡(ψ).
  • (ii) In general, let α:E′→E be any isogeny (e.g. ϕ^). Then: (αϕ+αψ)^=αϕ^+αψ^(ϕ+ψ)^α^=(ϕ^+ψ^)α^

    Hence the result follows. □

Remark. In Silverman’s book he proves Lemma 14.3first, and uses this to show deg⁡:Hom⁡(E,E′)→ℤ is a quadratic form.

Notation.

sum⁡:Div⁡(E)→E∑np(P)⏟formal sum↦∑nP(P)⏟addusinggrouplaw

Recall

E →≅Pic⁡0(E)P↦[(P)−(0)]

Therefore sum⁡D↦[D] for all D∈Div⁡0(E).

We deduce:

Lemma 14.4. Assuming that:

  • D∈Div⁡(E)

Then D∼0 if and only if deg⁡D=0 and sum⁡D=0E.

We will now discuss Weil pairing.

Let ϕ:E→E′ be an isogeny of degree n, with Dual isogeny ϕ^:E′→E. Assume char⁡K∤n (hence ϕ, ϕ¯ separable).

We define the Weil pairing

eϕ:E[ϕ]×E′[ϕ^]→μn.

Let T∈E′[ϕ]. Then nT=0, so there exists f∈K¯(E′)∗ such that

div⁡(f)=n(T)−n(0).

Pick T0∈E(K¯) with ϕ(T0)=T. Then

ϕ∗(T)−ϕ∗(0)=∑P∈E[ϕ](P+T0)−∑P∈E[ϕ](P)

has sum

nT0=ϕ^ϕT0=ϕ^T=0.

So there exists g∈K¯(E)∗ such that

div⁡(g)=ϕ∗(T)−ϕ∗(0).

Now

div⁡(ϕ∗f)=ϕ∗(div⁡f)=n(ϕ∗(T)−ϕ∗(0))=div⁡(gn)

Therefore ϕ∗f=cgn for some c∈K¯∗.

Rescaling f, we can say without loss of generality c=1, i.e. ϕ∗f=gn.

For S∈E[ϕ] we get τS∗(div⁡g)=(div⁡g) so τS∗g=ζg for some ζ∈K¯∗, i.e. ζ=g(X+S)g(X) is independent of choice of X∈E(K¯).

Now

ζn=g(X+S)ng(X)n=f(ϕ(X+S))f(ϕ(X))=1

since S∈E[ϕ]. Hence ζ∈μn.

We define

eϕ(S,T)=g(X+S)g(X).

Proposition 14.5. e is bilinear and non-degenerate.

Proof.

  • (i) Linearity in first argument: e(S1+S2,T)=g(X+S1+S2)g(X+S2)g(X+S2)g(X)=e(S,T)e(S2,T)
  • (ii) Linearity in second argument: Let T1,T2∈E′[ϕ^]. div⁡(f1)=n(T1)−n(0)ϕ∗f=g1ndiv⁡(f2)=n(T2)−n(0)ϕ∗f2=g2n

    There exists h∈K¯(E′) such that

    div⁡(h)=(T1)+(T2)−(T1+T2)−(0).

    Then put f=f1f2hn and g=g1g2ϕ∗h.

    Check: div⁡(f)=n(T1+T2)−n(0). Yes.

    ϕ∗f=ϕ∗f1ϕ∗f2(ϕ∗h)n=(g1g2ϕ∗(h))n=gn.

    Therefore

    e(S,T1+T2)=g(X+S)g(X)=g1(X+S)g1(X)g2(X+S)g2(X)h(ϕ(X))h(ϕ(X+S))⏟=1since S∈E[ϕ]=e(S,T1)e(S,T2)
  • (iii) e is non-degenerate. Fix T∈E′[ϕ^]. Suppose e(S,T)=1 for all S∈E[ϕ]. Then τS∗g=g for all S∈E[ϕ].

    Have K¯(E)∕ϕ∗K¯(E′) is a Galois extension with Galois group E[ϕ] (S∈E[ϕ] acts on K¯(E) via τS∗).

    Therefore g=ϕ∗h for some h∈K¯(E′). So ϕ∗f=gn=ϕ∗(hn). So f=hn, and hence div⁡(h)=(T)−(0). So T=0. □

We’ve shown E′[ϕ]↪Hom⁡(E[ϕ],μn). It is an isomorphism since #E[ϕ]=#E′[ϕ^]=n.

Remark.

  • (i) If E,E′,ϕ are defined over K then e is Galois equivariant, i.e. ∀⁡σ∈Gal⁡(K¯∕K), ∀⁡S∈E[ϕ], ∀⁡T∈E′[ϕ],
    e(σS,σT)=σ(e(S,T)).
  • (ii) Taking ϕ=[n]:E→E (so ϕ^=[n]) gives
    en:E[n]×E[n]→μn2μn.

Corollary 14.6. Assuming that:

  • E[n]⊂E(K)

Then μn⊂K.

Proof. Let T∈E[n] have order n. Non degeneracy of en implies that there exists S∈E[n] such that en(S,T) is a primitive n-th root of unity, say ζn.

Then

σ(ζn)=σ(en(S,T))=en(σS,σT)=ζn

for all σ∈Gal⁡(K¯∕K). So ζn∈K. □

Example. There does not exist E∕ℚ with E(ℚ)tors≅(ℤ∕3ℤ)2.

Remark. In fact, en is alternating, i.e. en(T,T)=1 for all T∈E[n].

(This implies en(S,T)=en(T,S)−1).