Ah — this is a very precise algebraic point, and yes, it is one of those “miracles” people like to point out in the hendecagon story.
Let me state it cleanly and then explain why it matters.
The “miracle”: an unexpected irreducible cubic
Start with the obvious object:
ζ=e2πi/11,ζ11=1
The cyclotomic polynomial is
Φ11(x)=x10+x9+⋯+x+1
which is irreducible of degree 10 over Q.
Now consider the real quantities
uk=ζk+ζ−k=2cos(112πk),k=1,…,5
These five numbers are Galois conjugates, and together they generate the real subfield
Q(cos(2π/11))
which has degree 5 over Q.
So far, nothing miraculous: degree 5 → non-constructible.
Where the “miracle” happens
Now define the following grouped sums:
AB=u1+u3+u4=u2+u5
(or equivalent partitions; the exact grouping depends on a choice of automorphism).
When you compute the minimal polynomial satisfied by A, something striking occurs:
It satisfies a cubic polynomial that is irreducible over Q.
In other words, although the full real subfield has degree 5, this particular symmetric combination collapses to degree 3 — but not further.
This is the “miracle” people refer to.
Why this is surprising
You might expect:
-
Either the expression stays degree 5, or
-
It drops all the way to degree 1 or 2 (which would imply constructibility)
But instead it lands exactly at degree 3.
That means:
No square-root tower can reach a cubic extension.
Galois-theoretic explanation (what’s really going on)
The Galois group of Q(ζ11)/Q is:
(Z/11Z)×≅C10
Complex conjugation cuts this down to the real subfield, with Galois group:
C5
Now here’s the key point:
Those orbits can have size 3, producing elements whose stabilizer has order gcd(5,3)5, giving rise to a degree-3 minimal polynomial.
That group-action accident is the “miracle.”
It’s not generic; it depends delicately on:
-
the prime 11,
-
the structure of (Z/11Z)×,
-
and the particular trigonometric combinations chosen.
Why this matters geometrically
This cubic irreducibility explains:
-
why some neusis / origami constructions reduce the problem to solving a cubic first,
-
why certain “almost Euclidean” constructions for the hendecagon look tantalizingly close,
-
and why algebraic manipulations alone don’t rescue classical constructibility.
You get:
degree 5⟶degree 3⟶still impossible
So the hendecagon fails in the smallest possible way.
Big picture takeaway
The hendecagon isn’t just non-constructible —
it is minimally non-constructible.
That’s why people describe this cubic factorization as a “miracle”:
the algebra almost collapses into Euclidean territory, but not quite.
If you want next, I can:
-
write down the explicit cubic polynomial,
-
show how this connects to origami axiom O6, or
-
compare this phenomenon with n=7 and n=13 where no such miracle occurs.