UPSC 2026 Maths Optional Paper 2 Q2a — Step-by-Step Solution
15 marks · Section A
Groups: definition, axioms, examples · Algebra · asked 3× in 14 yrs · Read the full method →
Question
Let G be a finite group generated by two elements u and v of order 2 each. Show that G is isomorphic to the dihedral group of order 2n, where O(uv)=n.
Technique
The recognisable trigger is two involutions: the moment a group is generated by two elements of order 2, put r=uv and s=u — then v=sr, so {r,s} still generates, and the pair (r,s) satisfies the defining dihedral relations rn=s2=e, srs−1=r−1. From those relations alone every element collapses to the normal form rk or rks, giving ∣G∣≤2n; the whole marking weight then sits on proving those 2n elements are distinct, which reduces to the single claim s∈/⟨r⟩. Do not merely assert the isomorphism — exhibit the bijection and check it multiplies correctly.
Solution
Step 0 — Fix the model of the dihedral group.
Write Zn={0,1,…,n−1} with addition mod n. Define the dihedral group of order 2n as the semidirect product Dn=Zn⋊Z2: as a set
Dn={(k,ε):k∈Zn,ε∈{0,1}},∣Dn∣=2n,
with product
(k,ε)(l,δ)=(k+(−1)εl(modn),ε+δ(mod2)).
Put a=(1,0) and b=(0,1). A direct computation gives
so Dn=⟨a,b∣an=b2=e,bab−1=a−1⟩ and, for n≥3, Dn is exactly the symmetry group of the regular n-gon (n rotations ak and n reflections akb).
Why state this first: “the dihedral group of order 2n” must be a group we already know has order 2n, otherwise the counting argument below is circular.
Step 1 — Choose the right generators.
We are given u2=v2=e with u=e, v=e, and G=⟨u,v⟩ finite. Set
r=uv,s=u.
Since G is finite, r has a finite order; by hypothesis
n=O(r)=O(uv).
Step 2 — {r,s} generates G.
Clearly r,s∈G, so ⟨r,s⟩⊆G. Conversely u=s and, since u2=e,
sr=u(uv)=u2v=v.
So both u and v lie in ⟨r,s⟩, whence G=⟨u,v⟩⊆⟨r,s⟩. Therefore
G=⟨r,s⟩.
Step 3 — Verify the three dihedral relations.
rn=e by the definition of n=O(r), and no smaller positive power of r is e.
s2=u2=e, and s=u=e, so O(s)=2.
Using u−1=u and v−1=v,
srs−1=u(uv)u−1=(u2)vu=vu,r−1=(uv)−1=v−1u−1=vu.
Hence
srs−1=r−1.(∗)
Iterating (∗): srks−1=(srs−1)k=r−k, and since s−1=s,
srk=r−ksfor every k∈Z.(∗∗)
Relation (∗∗) is the working form: it lets every s be pushed to the right past any power of r.
Step 4 — Every element of G has the normal form rk or rks.
Let
H={rk:0≤k≤n−1}∪{rks:0≤k≤n−1}.
We show H is a subgroup. All indices are read mod n, legitimate because rn=e. Using (∗∗) and s2=e:
So H is closed under multiplication. Also e=r0∈H; and H is closed under inverses, since (rk)−1=rn−k∈H and
(rks)2=rk−k=e⇒(rks)−1=rks∈H
(so incidentally every element rks is an involution — the “reflections”).
Thus H≤G. Moreover u=s=r0s∈H, and by Step 2 together with (∗∗), v=sr=r−1s∈H. Since G=⟨u,v⟩ is the smallest subgroup containing u and v, we get G⊆H; and H⊆G is clear. Hence
This is where the marks are. Suppose, for contradiction, that s∈⟨r⟩. Then G=⟨r,s⟩=⟨r⟩ is cyclic, hence abelian. Being abelian, u and v commute, so
r2=(uv)(uv)=u(vu)v=u(uv)v=u2v2=e,
and therefore n=O(r)∈{1,2}.
If n=1: ⟨r⟩={e}, so s=e, i.e. u=e — contradicting O(u)=2.
If n=2: ⟨r⟩={e,r} and s=e, so s=r, i.e. u=uv, giving v=e — contradicting O(v)=2.
Both cases are impossible. Hence
s∈/⟨r⟩.(5.1)
Step 6 — The 2n listed elements are distinct, so ∣G∣=2n.
Three checks, and all three are needed:
Rotations are distinct. If ri=rj with 0≤i<j≤n−1 then rj−i=e with 0<j−i<n, contradicting O(r)=n.
Reflections are distinct. If ris=rjs then right-multiplying by s−1 gives ri=rj, so i=j by (1).
No rotation equals a reflection. If ri=rjs for some i,j, then s=ri−j∈⟨r⟩, contradicting (5.1).
So the list in (4.2) has exactly 2n distinct entries and
∣G∣=2n.(6.1)
(Structural aside, worth one line: ⟨r⟩ has index 2 in G by (6.1), hence ⟨r⟩⊴G, and (∗) says s acts on it by inversion — i.e. G=⟨r⟩⋊⟨s⟩, which is precisely the shape of Dn in Step 0.)
Step 7 — Construct the isomorphism explicitly.
By Step 0 the 2n elements ak,akb(0≤k≤n−1) of Dn are distinct and exhaust Dn; by Step 6 the 2n elements rk,rks are distinct and exhaust G. So the assignment
φ:G⟶Dn,φ(rk)=ak,φ(rks)=akb(0≤k≤n−1)
is well defined (each element of G has exactly one normal form) and is a bijection (it matches the two lists term by term).
φ is a homomorphism: in Dn the relations an=b2=e, baj=a−jb give, exactly as in (4.1),
These four cases cover every product in G. Hence φ is a bijective homomorphism, i.e.
G≅Dn,n=O(uv),∣G∣=2n.■
(Alternative one-line finish, if von Dyck’s theorem is available: r,s∈G satisfy the defining relations of Dn=⟨a,b∣an=b2=e,bab−1=a−1⟩, so there is a homomorphism Dn→G with a↦r, b↦s; it is onto because ⟨r,s⟩=G (Step 2), and ∣Dn∣=2n=∣G∣ by Step 6, so it is injective as well. Steps 5–6 are still needed — they supply ∣G∣=2n — so nothing is saved except the four-case check.)
Step 8 — The small cases (honesty at the boundary).
The proof above is uniform in n≥1, but it is worth recording what the conclusion means at the two degenerate values, since “dihedral” is usually pictured as a polygon group:
n=1.uv=e⇒v=u−1=u. The two generators coincide, G=⟨u⟩≅Z2, of order 2n=2. Consistently, D1=Z1⋊Z2≅Z2.
n=2.(uv)2=e gives uvuv=e, so uv=(uv)−1=vu: u and vcommute, and u=v. Then G={e,u,v,uv} is the Klein four-group Z2×Z2, of order 2n=4 — and indeed D2≅Z2×Z2. Note this is abelian: “dihedral of order 4” is the Klein group, not a polygon symmetry group.
n≥3.G is non-abelian (from (∗), sr=r−1s=rs since r2=e) and is the genuine symmetry group of the regular n-gon.
In particular G is abelian ⟺n≤2.
Step 9 — Where finiteness was used.
Finiteness of G was used exactly once — to guarantee n=O(uv)<∞ in Step 1. It is not an extra assumption beyond that: by (6.1), ∣G∣=2O(uv), so G is finite iffuv has finite order. If O(uv)=∞ the same Steps 2–4 give G≅D∞=⟨a,b∣b2=e,bab−1=a−1⟩, the infinite dihedral group.