Cantor’s Theorem played a large part in pushing set theory towards axiomatization. Here is a
conundrum motivated by the theorem: If a ‘set’ is just a collection of objects, then we may consider
the ‘set of all sets.’ Call this A. Now consider the power set of A. Since P(A) is a set of sets, it must
be a subset of A, whence
|
P(A)
|
≤
|
A
|
. However, by Cantor’s Theorem, we have
|
A
|
⪇
|
P(A))
|
.
The conclusion is the manifest absurdity
|
A
|
⪇
|
A
|
The remedy is a thorough definition of ‘set’ which prevents the collection of all sets from being
considered a set. This is where axiomatic set theory begins.
A word on the limits of proof
Throughout this course we have learned about some of the basic methods and and concepts used
by the mathematician. In particular, we learned about various types of proof and how to use these
proofs to demonstrate the truth of statements about mathematical objects. As we finish the course, it
makes sense to reflect on the limits of our methods.
In the early 20th century, the discovery of various paradoxes and contradictions led to a foun-
dational crises in mathematics. After all, it is difficult to build a house if you have cracks in your
foundation! The result was an effort to put all of mathematics on a rigorous axiomatic basis by for-
mulating a list of reasonable axioms from which all of mathematics could be derived, using basic
logical reasoning. This axiomatic foundation ideally would satisfy the following conditions:
8.2.1 consistency, i.e. no contradiction would be derivable from the axioms;
8.2.2 completeness, i.e. all true mathematical statements would be derivable from the axioms.
The hope for such a foundation was crushed in 1931, when a young logician by the name of Kurt
G
¨
odel published his famous Incompleteness Theorems which showed that no such axiomatic system
could exist. Essentially, G
¨
odel showed that in any consistent axiomatic system that was strong
enough to produce some basic arithmetic, there must be statements which are neither derivable nor
refutable from the axioms. Perhaps even worse, no such system can prove its own consistency.
While the strongest aims of some of the early 20th century attempts at an axiomatic foundation
cannot be accomplished, the research of that time was able to provide a foundation that most modern
mathematicians deem adequate for current work. Perhaps the most popular approach is to base all
of mathematics on set theory – you will see as your studies progress that many of the objects you
study can be formalized as sets together with functions and relations between sets. We have seen
in Chapter 7 that functions and relations are just themselves sets. Even numbers like 0, 1, 2 or
12
19
or
3.14 . . . can be thought of as sets, if one desires. In turn, set theory is often axiomatized using the ZFC
axioms (short for Zermelo-Fraenkel set theory with the Axiom of Choice).
While the ZFC axioms are subject to the limitations imposed by G
¨
odel’s theorems, they have
proven themselves by being able to formalize most of the mathematics actually used by current
mathematicians, and have so far not produced any inconsistencies. Thus most mathematicians feel
little need to dwell on the foundational issues of the previous century.
Reading Questions
8.2.1 A set A is uncountable if and only if .
235