Yatir Halevi
Posted on Monday, 10 February, 2003 - 09:45 am:

I have in front of me a proof to the theorem: ''If each of two sets M and N can be mapped injectively into the other, then there is a bijection from M to N.''

I'm having trouble following the proof. Could somebody help me? :)

Here is the proof:

Let f:MN and g:NM be injections. For each AM define the set F(A)M by:

F(A):=M\g(N\f(A))

that is, any subset AM is mapped to f(A)N, then we take its complement N\f(A) in N, then its complement is mapped back to obtain g(N\f(A))M, and of this we finally take to complement in M.

Suppose that there is some A0 with F( A0 )= A0 , that is, such that g(N\f( A0 ))=M\A0 . Then, clearly, (Not very clear to me... :) ),

φ(x):=

f(x) if x A0

g-1 (x) if x A0

provides a bijection from M to N. Hence it remains to find such a ''fixed set'' A0 . (I also have a proof for that, and that was easy to understand)

Thanks in advance,

Yatir

Chris Purcell
Posted on Monday, 10 February, 2003 - 11:30 am:

Firstly, g is an injection. Since M\A0 =g(N\f( A0 )), that means every x A0 must equal g(y) for some yN, and since g is an injection that element is unique. Hence defining φ(x):= g-1 (x) on M\A0 makes sense, giving us a bijection between M\A0 and N\f( A0 ).

Secondly, f is an injection. That means φ also gives us a bijection between A0 and φ( A0 )f( A0 ) simply by virtue of being injective on it.

Now A0 and M\A0 are disjoint and together make up all of M. Equally, φ( A0 )f( A0 ) and φ(M\A0 )N\f( A0 ) are disjoint and together make up all of N. φ is thus a bijection from M to N.

Sorry if this is still impenetrable - clearly is always hard to explain!

Chris

Yatir Halevi
Posted on Monday, 10 February, 2003 - 12:08 pm:

Got it! Thanks.
I just needed someone to spell it out for me

Yatir