Possible title: When Does the Kernel of a Homomorphism Reduce to the Identity?

  • Context: Graduate 
  • Thread starter Thread starter TrickyDicky
  • Start date Start date
  • Tags Tags
    Homomorphisms Kernel
Click For Summary

Discussion Overview

The discussion revolves around the conditions under which the kernel of a homomorphism, specifically in the context of Lie group homomorphisms that are also covering maps, can reduce to the identity. Participants explore examples, particularly the homomorphism from SU(2) to SO(3), and consider the implications of group actions on different spaces.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested

Main Points Raised

  • One participant suggests that if a homomorphism's kernel is the identity, it would be an isomorphism, and questions whether there are cases where the kernel reduces to the identity without the homomorphism being an isomorphism.
  • Another participant clarifies that a homomorphism can be one-to-one (kernel is the identity) but not onto, thus not an isomorphism, and discusses the implications of this in terms of mapping onto a subgroup.
  • A participant provides an example of the homomorphism from SU(2) to SO(3) and questions if there are scenarios where this kernel could reduce to the identity, potentially through group actions on non-vector spaces.
  • Further clarification is sought regarding the term "situations," with participants discussing whether the inquiry pertains specifically to SU(2) and SO(3) or more generally to other groups.
  • One participant introduces the concept of a non-canonical isomorphism, distinguishing it from a canonical one, which does not require prior choices.

Areas of Agreement / Disagreement

Participants express differing interpretations of the initial question and the implications of the kernel's properties, indicating that the discussion remains unresolved with multiple competing views on the conditions for the kernel to reduce to the identity.

Contextual Notes

Participants note that the discussion involves specific examples and definitions that may not be universally applicable, and the exploration of group actions on various manifolds introduces additional complexity to the inquiry.

TrickyDicky
Messages
3,507
Reaction score
28
I'm centering on lie group homomorphisms that are also covering maps from the universal covering group. So that if their kernel was just the identity
they would be isomorphisms.
Are there situations in which the kernel of such a homomorphism would reduce to the identity? I'm thinking of situations where the groups act on different sp
 
Physics news on Phys.org
I'm not clear what your question is. You state that you know that if a homomorphism is an isomorphism then its kernel is the identity. You then ask "Are there situations in which the kernel of such a homomorphism would reduce to the identity?" Do you mean "are the situations in which the kernel is the identity but the homomorphism is NOT an isomorphism?

If that is your question, then, strictly speaking, yes. In order to be an isomorphism, a homomorphism must be both "one to one" and "onto". A homomorphism is "one to one" if and only if its kernel is the identity. But that leaves "onto". It is possible for a homomorphism to be "one to one" yet not be "onto".

However, in such a case, the homomorphism is "onto" a subgroup of the original range group and we normally then focus on the homomorphism being an isomorphism to that subgroup.
 
HallsofIvy said:
I'm not clear what your question is. You state that you know that if a homomorphism is an isomorphism then its kernel is the identity. You then ask "Are there situations in which the kernel of such a homomorphism would reduce to the identity?" Do you mean "are the situations in which the kernel is the identity but the homomorphism is NOT an isomorphism?

If that is your question, then, strictly speaking, yes. In order to be an isomorphism, a homomorphism must be both "one to one" and "onto". A homomorphism is "one to one" if and only if its kernel is the identity. But that leaves "onto". It is possible for a homomorphism to be "one to one" yet not be "onto".

However, in such a case, the homomorphism is "onto" a subgroup of the original range group and we normally then focus on the homomorphism being an isomorphism to that subgroup.
Hi, that was not what I meant, I had some problem with the editor and the question appeared cut.
I was going to give some example: consider the well known homomorphism SU(2)->SO(3), Since SU(2) is the Universal cover of SO(3) the only reason it is not an isomorphism is that its kernel has one more element besides the identity, -I, and my question is if there are situations where the kernel
of this homomorphism is reduced to the identity, amd therefore turned into an isomorphism. Like maybe wnen acting on spaces that are not vector spaces that habe isometries that reduce the kernel.
 
I'm not sure what you mean by "situations". Are you asking about group homomorphisms ##\phi:G\to H## where G and H are not necessarily SU(2) and SO(3), or are you asking specifically about SU(2) and SO(3)?

I suspect that whatever the exact question is, the best answer you will get is that there's a theorem that says that if ##\phi:G\to H## is a homomorphism, then the quotient group ##G/\ker\phi## is isomorphic to ##\phi(G)##.
 
Fredrik said:
I'm not sure what you mean by "situations". Are you asking about group homomorphisms ##\phi:G\to H## where G and H are not necessarily SU(2) and SO(3), or are you asking specifically about SU(2) and SO(3)?

I suspect that whatever the exact question is, the best answer you will get is that there's a theorem that says that if ##\phi:G\to H## is a homomorphism, then the quotient group ##G/\ker\phi## is isomorphic to ##\phi(G)##.
/
That was just an example that fulfills the conditions I stablished. By situations I mean there are group actions on manifolds other than the usual Rn that modify the kernel so that the quotient trivially becomes G/Identity.
 
Basically I guess I was trying to define a non-canonical or unnatural isomorphism, that is, an isomorphism that requires a previous choice, whereas canonical means: "distinguished representative of a class", particularly one that does not require making any choice; this is also known as "natural", as in natural transformation."
 

Similar threads

  • · Replies 3 ·
Replies
3
Views
1K
  • · Replies 12 ·
Replies
12
Views
4K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 21 ·
Replies
21
Views
2K
  • · Replies 10 ·
Replies
10
Views
2K
  • · Replies 26 ·
Replies
26
Views
1K
  • · Replies 2 ·
Replies
2
Views
2K