We want to extend the notion of size of finite sets that is provided by the natural numbers so that we can compare the "size" of any two sets. It is worth noting that the following definition requires some reassurance that we are on mathematically firm ground since we are certainly on the edge of Russell territory. We will discuss this further when we consider alternatives to ZF set theory.
Let be the equivalence relation on the "class" of all set defined by setting if there exists a 1 to 1 onto map . This is the books notation.
We define a second equivalence relation on the class of all set by setting if there exists a 1 to 1 map and a 1 to 1 map .
Finally, we write if the exist a 1 to 1 map .
Exercise: Check that 1. and 2. are equivalence relations. For reference, you need to verify that
and
and
Observation:
For and its subsets, we have essentially shown, that iff
We used this to also show that Note that if we only wanted to show that
, it would have sufficed to present two one to one maps,
defined by m,n23
and
defined by mm,1
Goals:
We show that, in general, for any pair of sets iff
We also will show that for any pair of sets, or
First, the following useful theorem.
Let be a partially ordered set in which every non-empty sub-set has a least upper bound and a greatest lower bound. If is order-preserving, that is
implies , then has a fixed point ( ).
Proof.
Before proceeding with the proof. We look at two examples:
Examples:
1. - The map n n+1 has no fixed point. On the other hand, ,itself a subset, does not have a least upper bound.
2. - where,again, , and ,the usual ordering, on and n for all n
First, note that satifies the hypothesis of Tarski. To see this let . If then it is the least upper bound (and the top element). If and is finite then it, as a finite subset of , has a least upper bound( and again a top element). Finally, if and is not finite, is the least upper bound, in fact the only upper bound( but not the top element)
Next, to see that the conclusion of Tarski holds, let be order preserving. If we have our fixed point. If n for some n then ,since is assumed to be order preserving.
Finally, an induction argument shows that, in general, any order preserving map has a fixed point.
Let
.
Then by definition,
.
Let
Since
,
is
non-empty. Hence exists. For any we have
and , so is an upper bound for . Thus and
But, in general, if then because
implies . Thus and
Finally, we conclude that .
For any pair of sets iff
Proof.
By hypothesis, there exist one-to-one functions
and . We apply the Fixed Point Lemma to the poset and the function where . Here the exponent signifies Set complement.
Note that is order-preserving
Hence, by the Tarski Fixed Point Theorem, there is a set such that
Specifically ,
We can now define a 1 to 1 onto map by , and for .
9.3.1 Lemma: In the setting of the Proof of 9.3 , the union being disjoint.
Proof.
That the union is disjoint simply follows from the fact that . In addition, since is one to one we also can write
, with . Intersecting both sides with gives
But
and thus
or, taking the complements of both sides
9.3.2 Theorem:
In the setting of the Proof of 9.3, then ,
where Moreover the union is a disjoint union of non empty sets.
Proof.
From 9.3.1
That the sets are disjoint follows from the fact that and , hence ,are one to one
We are now about ready to move to a more general discussion of set theory and cardinals. We begin by noting that a more general discussion is appropriate by showing that there are other cardinalities to be considered other than that of and its finite subsets .
For any set ,there is no onto map .
Proof A:( yet another diagonal argument. This is just Russell's argument in a non-paradoxical setting)
Let . Since is onto there is some with .
As usual, by definition if then and if then .
Proof B:( yet another diagonal argument. This is just a disguised version of A.)
Let 0,1. Let be the set of all maps from to . There is an obvious 1 to 1 onto map Specifically,
1 . Hence to prove 9.4 we can show that for any set ,there is no onto map .
Suppose there was such a map. Define by the formula
The rest of the proof is standard. Suppose there was some such that Compute
If 0 then by definition 1 and so on.....
Proof C:( yet another fixed point theorem. )
We first prove the following lemma, which is really a generalization of B.
9.4.1 Lemma : For sets and , suppose there exists an onto map , then every self-map has a fixed point.
Proof:
Consider the "evaluation" map defined by the formula
Like in B, above define by the formula
Again, since is onto there is some such that
We have
on the other hand, by definition,
Thus is a fixed point for
Now, to complete the proof of 9.4, Let 0,1 and note that has a fixed point free self-map. In particular, let be define by the formula
01 and 10 .
Notation:
The standard notation for is .
For , if is a set with we use multiplicative notation for Check that this is well defined.
If is a set with , the standard notation for is , as suggested by the second proof of 9.2 .
The standard notation for is
We will be very interested in We end this Page with a Theorem that is in fact a special case of a general result about infinite cardinals.
Proof:
Let and be the odd and even natural numbers respectively. In full detail,
.
Again, letting 0,1 We invoke Lemma a. on the Background Page to observe that there is a one-one onto map.
Hence,
Let be a finite set with , then
Hint:
Use Lemma 0.3 b. on the Background Page and 9.4 above to prove the theorem by induction for 2 .
Next use 9.3 to prove the result when 22
Solution
One verifies that
2
and
2
By induction, suppose we have shown that
The map is one to one and onto.
Similarly
is one to one and onto.
Thus
is one to one and onto.
Now assume
22 and choose one to one maps and These induce one to one maps.
and
Finally, apply Schroder-Bernstein to and .