1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: Group set proof

  1. Dec 6, 2012 #1
    1. The problem statement, all variables and given/known data
    Let (G,*) be a finite group of even order. Prove that there exists some g in G such that g≠e and g*g=e. [where e is the identity for (G,*)]

    2. Relevant equations
    Group properties

    3. The attempt at a solution
    Let S = G - {e}. Then S is of odd order, and let T={g,g^-1: g[itex]\in[/itex]S}.
    Then [itex]\exists[/itex]h[itex]\in[/itex]S such that h[itex]\notin[/itex]T. Since G is a group, h*h must equal e.

    Does that work?
  2. jcsd
  3. Dec 6, 2012 #2
    You have a good idea here, nearly perfect, but the last step seems not obvious at first. Who says that h*h must be equal to e? All we know (since G is a group) is that h*h must lie in G. To show that h*h = e, you also need h∉S. (Then h*h ∈ G - S = {e}.)
  4. Dec 6, 2012 #3


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    How is that possible? You have defined T in such a way that [itex]S \subset T[/itex]. You have the right idea, though. If you partition G such that each element of g is partnered with its inverse, then e will be partnered with itself. If you assume that no other element has this property, then you should be able to obtain a contradiction.
  5. Dec 6, 2012 #4
    If h[itex]\notin[/itex]S, then h=e, which isn't what we want. I was trying to show that if you take all the elements and their inverses out of S, then you're left with a single element whose inverse must be itself. There's probably a better way to show this.
  6. Dec 6, 2012 #5
    How about collecting all elements that aren't their own inverses in one set, i.e. S = { g∈G : g-1≠g }. Then for any element x∈G, either x ∈ G - S, or x and x-1 are both in S. "Both" is important here, since it means that S has an even number of elements.

    So what's left in G - S?
  7. Dec 6, 2012 #6
    That's good, thanks!
  8. Dec 6, 2012 #7
    It might be interesting for the OP that this result has a significant generalization: If G is a group and if p is a prime number such that p divides the order of G, then there exists an element g in G with order p.

    This is called Cauchy's theorem.
  9. Dec 6, 2012 #8
    So |x|[itex]\in[/itex]G = p? Would this relate to a group being cyclic?
  10. Dec 6, 2012 #9
    I don't really get your notation of [itex]|x|\in G=p[/itex] :confused:

    I mean to say that there exists a [itex]x\in G[/itex] such that [itex]|x|=p[/itex] (assuming |x| means order for you).

    If G has prime order, then this result shows indeed that there exists an element in G with prime order. So G must be cyclic. Is this a relationship that you are talking about?
  11. Dec 6, 2012 #10
    Yeah that's exactly it, I was being sloppy. My understanding of a cyclic group is that there exists some element of G, say a, such that, for every x in G, x=a^n for some integer n.

    How does this follow from |x|=p?
  12. Dec 6, 2012 #11
    It doesn't follow in general.

    I'm saying the following: If G has order p, then there exists an element a in G that has order p. So [itex]\{1,a,a^2,...,a^{p-1}\}[/itex] has p elements and is a subset of G. Since G also has p elements, we must have [itex]G=\{1,a,a^2,...,a^{p-1}\}[/itex]. So G is cyclic.
  13. Dec 7, 2012 #12
    Look at all elements of the form xn with n=0,...,(p-1). If you can show that they're all distinct (i.e. xn ≠ xm if n ≠ m), then these p elements must make up all of G.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook