Can an Onto Function Exist for Every Subset of a Given Set?

  • Thread starter Thread starter cragar
  • Start date Start date
  • Tags Tags
    Functions Proof
Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
3 replies · 2K views
cragar
Messages
2,546
Reaction score
3

Homework Statement


Given any set A, there does not exist a function f:A→P(A)
It wants us to do a proof by contradiction and assume there is an onto function.
For each element [itex]a \in A[/itex] f(a) is a particular subset
of A. The assumption that f is onto means that every subset of A appears as f(a)
for some [itex]a \in A[/itex] To arrive at a contradiction, we will produce
a subset [itex]B \subseteq A[/itex] that is not equal to f(a) for
any [itex]a \in A[/itex]
[itex]B=\{a\in A:a\notin f(a)\}[/itex]
If f is onto then it must be the case that B=f(a') for some a' in A.
the contradiction arises when we consider if a' is an element of B.
First show that the case that a' is in B leads to a contradiction.
Now finish the argument by showing that the case a' is not in B
is equally unacceptable.

The Attempt at a Solution


If a' is in B then a' is in A
and if f is onto then f(a') is in B
if f(a') is in B then f(a') is some a
in A but this is a contradiction
because we assumed that a was not in f(a).
If a' is not in B then it is not in A and therefore
it is not in anything.
Am I headed the right direction?
 
Last edited:
Physics news on Phys.org
cragar said:

The Attempt at a Solution


If a' is in B then a' is in A
and if f is onto then f(a') is in B
if f(a') is in B then f(a') is some a
in A but this is a contradiction
because we assumed that a was not in f(a).
If a' is not in B then it is not in A and therefore
it is not in anything.

This is somewhat confused.

Given [itex]f[/itex], [itex]B[/itex] consists of exactly those elements [itex]a \in A[/itex] which are not members of their image, [itex]f(a) \subset A[/itex]. However there could be other elements of A which are members of their image.

The point is that if [itex]f[/itex] is onto, then by definition [itex]B \subset A[/itex] must be the image of some [itex]a' \in A[/itex], so that [itex]f(a') = B[/itex]. Note that, by assumption, [itex]a' \in A[/itex]. Whether [itex]a' \in B[/itex] is another question.

So what happens if [itex]a' \in B[/itex]? That would mean that [itex]a'[/itex] is not a member of its image. But its image is B. This is a contradiction.

Now you need to show that a similar contradiction arises if you assume [itex]a' \notin B[/itex].
 
cragar said:
If a' is in B then a' is in A
a' was chosen to be in A.
and if f is onto then f(a') is in B
a' was chosen such that f(a')=B. It cannot be "in B".
is image f(a)=a
a is an element of A; f(a) is a subset of A. They cannot be equal.
Start again at "If a' is in B then " and use the definition of B to deduce something about a'.