Fixed-point lemma for normal functions explained

The fixed-point lemma for normal functions is a basic result in axiomatic set theory stating that any normal function has arbitrarily large fixed points (Levy 1979: p. 117). It was first proved by Oswald Veblen in 1908.

Background and formal statement

A normal function is a class function

f

from the class Ord of ordinal numbers to itself such that:

f

is strictly increasing:

f(\alpha)<f(\beta)

whenever

\alpha<\beta

.

f

is continuous: for every limit ordinal

λ

(i.e.

λ

is neither zero nor a successor),

f(λ)=\sup\{f(\alpha):\alpha<λ\}

.It can be shown that if

f

is normal then

f

commutes with suprema; for any nonempty set

A

of ordinals,

f(\supA)=\supf(A)=\sup\{f(\alpha):\alpha\inA\}

.Indeed, if

\supA

is a successor ordinal then

\supA

is an element of

A

and the equality follows from the increasing property of

f

. If

\supA

is a limit ordinal then the equality follows from the continuous property of

f

.

A fixed point of a normal function is an ordinal

\beta

such that

f(\beta)=\beta

.

The fixed point lemma states that the class of fixed points of any normal function is nonempty and in fact is unbounded: given any ordinal

\alpha

, there exists an ordinal

\beta

such that

\beta\geq\alpha

and

f(\beta)=\beta

.

The continuity of the normal function implies the class of fixed points is closed (the supremum of any subset of the class of fixed points is again a fixed point). Thus the fixed point lemma is equivalent to the statement that the fixed points of a normal function form a closed and unbounded class.

Proof

The first step of the proof is to verify that

f(\gamma)\ge\gamma

for all ordinals

\gamma

and that

f

commutes with suprema. Given these results, inductively define an increasing sequence

\langle\alphan\ranglen<\omega

by setting

\alpha0=\alpha

, and

\alphan+1=f(\alphan)

for

n\in\omega

. Let

\beta=\supn<\omega\alphan

, so

\beta\ge\alpha

. Moreover, because

f

commutes with suprema,

f(\beta)=f(\supn<\omega\alphan)

   =\supn<\omegaf(\alphan)

   =\supn<\omega\alphan+1

   =\beta

The last equality follows from the fact that the sequence

\langle\alphan\ranglen

increases.

\square

As an aside, it can be demonstrated that the

\beta

found in this way is the smallest fixed point greater than or equal to

\alpha

.

Example application

The function f : Ord → Ord, f(α) = ωα is normal (see initial ordinal). Thus, there exists an ordinal θ such that θ = ωθ. In fact, the lemma shows that there is a closed, unbounded class of such θ.

References