If X is infinite, then X and X×X are equinumerous.
Does this proof look fine or contain gaps? Do you have suggestions? Many thanks for your dedicated help!
My attempt:
We denote A and B are equinumerous by A∼B.
Let Y={(A,f)∣A∈P(X) and f:A→A×A is bijective }. We define a partial order < on Y by (A1,f1)<(A2,f2)⟺A1⊆A2 and f1⊆f2
Since X is infinite, there exists A⊆X such that A∼N (Here we assume Axiom of Countable Choice). Thus A∼N∼N×N∼A×A.
Hence there is a bijection f:A→A×A, and consequently f∈Y and Y≠∅.
For any chain Z in Y, let F={f∣∃A∈P(X),(A,f)∈Z} and f∗=⋃f∈Ff.
- f∗ is a mapping
For (a,x1),(a,x2)∈f∗, there are (A1,f1),(A2,f2)∈Z such that (a,x1)∈f1 and (a,x2)∈f2. Since Z is a chain, we can safely assume (A1,f1)<(A2,f2). It follows that f1⊆f2. Thus (a,x1),(a,x2)∈f2 and consequently x1=x2 by the fact that f2 is a mapping.
Let A∗=domf∗, then A∗=⋃f∈Fdomf.
- f∗ is surjective
For all f∈F, f is a bijection from domf to domf×domf, and thus ranf=domf×domf.
ranf∗=⋃f∈Franf=⋃f∈F(domf×domf)⊆(⋃f∈Fdomf)×(⋃f∈Fdomf)= domf∗×domf∗ =A∗×A∗. Thus ranf∗⊆A∗×A∗.
For any (a1,a2)∈A∗×A∗, since A∗=⋃f∈Fdomf, there exist (A1,f1),(A2,f2)∈Z such that a1∈A1 and a2∈A2. Since Z is a chain, we can safely assume that (A1,f1)<(A2,f2), then A1⊆A2, then a1,a2∈A2, then (a1,a2)∈A2×A2. Since f2 is bijective, there exists a∈A2⊆A∗ such that f(a)=f∗(a)=(a1,a2)∈A2×A2⊆A∗×A∗.
Hence there exists a∈A∗ such that f∗(a)=(a1,a2) for any (a1,a2)∈A∗×A∗. Thus f∗ is surjective.
- f∗ is injective
Assume (a1,x),(a2,x)∈f∗. Then there exists (A1,f1),(A2,f2)∈Z such that (a1,x)∈f1 and (a2,x)∈f2. Since Z is a chain, we can safely assume (A1,f1)<(A2,f2). It follows that f1⊆f2. Thus f1(a1)=f2(a1) and consequently f2(a2)=x=f1(a1)=f2(a1). Hence f2(a2)=f2(a1) and consequently a2=a1 by the fact that f2 is injective. It follows that f∗ is injective.
To sum up, f∗:A∗→A∗×A∗ is bijective and hence A∗∼A∗×A∗ and f∗∈Y. Furthermore, f∗ is an upper bound of Z by definition. Thus Y satisfies the requirement of Zorn's Lemma and hence has a maximal element (ˉA,ˉf).
- X∼X×X
First, we prove X∼ˉA. If not, then X∖ˉA is infinite and consequently there exists C⊆X∖ˉA such that C∼N∼N×N∼C×C. Thus there is a bijection f_:C→C×C. Hence g=ˉf∪f_ is a bijection from ˉA∪C to (ˉA×ˉA)∪(C×C). Hence ˉA∪C∼(ˉA×ˉA)∪(C×C).
Since C∼N, (ˉA×ˉA)∪(C×C)∼ˉA×ˉA∼(ˉA∪C)×(ˉA∪C) by Lemma. It follows that ˉA∪C∼(ˉA∪C)×(ˉA∪C) and consequently there exists a bijection G:ˉA∪C→(ˉA∪C)×(ˉA∪C). Thus (ˉA∪C,G)∈Y and ˉf⊊G. This contradicts the maximality of (ˉA,ˉf). Hence X∼ˉA∼ˉA×ˉA∼X×X. This completes the proof.
Update: I fixed Part 4 as follows:
We denote:
X∼Y⟺ X and Y are equinumerous.
|X|+|Y|:=|A∪B| for X∼A, Y∼B, and A∩B=∅.
|X|.|Y|:=|X×Y|.
Lemma 1: If ℵ0≤|Z|, |X|+|Y|=|Z|, and |X|<|Z|. Then |Y|=|Z|. (I presented a proof here)
Lemma 2: If ℵ0≤|X|=|Y|. Then |X|+|Y|=|X|. (I presented a proof here)
- X∼X×X
We first prove ˉA∼X. Assume the contrary that |ˉA|≠|X|, then |ˉA|<|X|.
We have ˉA∪(X∖ˉA)=X, ˉA∩(X∖ˉA)=∅, and |ˉA|<|X|. Then |X∖ˉA|=|X| by Lemma 1. It follows that there exists C⊆X∖ˉA such that |C|=|ˉA|<|X∖ˉA|=|X|. Consider D=(C×ˉA)∪(ˉA×C)∪(C×C)
a. (ˉA×ˉA)∪D=(ˉA×ˉA)∪(C×ˉA)∪(ˉA×C)∪(C×C)=(ˉA∪C)×(ˉA∪C) by the definition of Cartesian product.
b. Since |C|=|ˉA|, |C×ˉA|=|ˉA×C|=|C×C|=|ˉA×ˉA|=|ˉA| [since ˉf:ˉA→ˉA×ˉA is bijective].
c. Since |C×ˉA|=|ˉA×C|=|C×C|=|ˉA×ˉA|=|ˉA|≥ℵ0 and they are pairwise-disjoint [Since ˉA∩C=∅], |D|=|ˉA| by Lemma 2.
d. Since |C|=|ˉA|=|D|, |C|=|D|. Thus there exists a bijection f_:C→D.
Let G=ˉf∪f_, then G:ˉA∪C→(ˉA×ˉA)∪D is bijective since ˉf,f_ are bijective, or equivalently G:ˉA∪C→(ˉA∪C)×(ˉA∪C) is bijective. It follows that ˉf⊊G and (ˉA∪C,G)∈Y. This contradicts the maximality of (ˉf,ˉA). Hence X∼ˉA∼ˉA×ˉA∼A×X.
No comments:
Post a Comment