Prove relation between the group of integers and a subgroup

PhysicsRock
Messages
121
Reaction score
19
Homework Statement
Let ##X \subseteq \mathbb{Z}## be a subgroup of ##(\mathbb{Z},0,+)##. Show that an ##l \in \mathbb{N}_0## exists such that ##X = l \cdot \mathbb{Z}##.
Relevant Equations
There are no other equations, so I'm just gonna state the hint that comes with the problem here. It is: In case ##\{ 0 \} \subsetneq X## choose ##l = \min(X \cap \mathbb{N})##.
So, a friend of mine has attempted a solution. Unfortunately, he's having numbers spawn out of nowhere and a lot of stuff is going on there which I can't make sense of. I'm going to write down the entire attempt.

$$
0 \in X \; \text{otherwise no subgroup since neutral element isn't included} \notag
$$

For each additional element ##x##, ##-x## has to be an element as well, to satisfy the condition of an inverse. Now let ##H_+## be the set of all elements of ##X## that are greater or equal to ##0## and likewise for ##H_-## containing all negative integers. Then ##X = H_+ \cup H_-##. Following the hint, let ##l## be the smallest number contained in ##X##. For ##H_+,H_-##, ##l \cdot h = \underbrace{h + h + h + ... + h}_{l-times}## has to be an element of the respective group too, because of closedness.

Now let ##h \in H_+##. Then ##h = l \cdot z + r## with ##r < l## (don't know where the ##r## is coming from and why it has to be less than ##l##). Since ##l## is the smallest positive number and ##r < l##, it follows that ##r = 0## and this ##x = l \cdot z \, \forall \, x \in X##. The process is analogous for ##H_-##.

Does this make sense and if not, what went wrong? Thanks everyone!
 
Physics news on Phys.org
PhysicsRock said:
Homework Statement:: Let ##X \subseteq \mathbb{Z}## be a subgroup of ##(\mathbb{Z},0,+)##. Show that an ##l \in \mathbb{N}_0## exists such that ##X = l \cdot \mathbb{Z}##.
Relevant Equations:: There are no other equations, so I'm just going to state the hint that comes with the problem here. It is: In case ##\{ 0 \} \subsetneq X## choose ##l = \min(X \cap \mathbb{N})##.

So, a friend of mine has attempted a solution. Unfortunately, he's having numbers spawn out of nowhere and a lot of stuff is going on there which I can't make sense of. I'm going to write down the entire attempt.

$$
0 \in X \; \text{otherwise no subgroup since neutral element isn't included} \notag
$$

For each additional element ##x##, ##-x## has to be an element as well, to satisfy the condition of an inverse. Now let ##H_+## be the set of all elements of ##X## that are greater or equal to ##0## and likewise for ##H_-## containing all negative integers. Then ##X = H_+ \cup H_-##. Following the hint, let ##l## be the smallest number contained in ##X##. For ##H_+,H_-##, ##l \cdot h = \underbrace{h + h + h + ... + h}_{l-times}## has to be an element of the respective group too, because of closedness.

Now let ##h \in H_+##. Then ##h = l \cdot z + r## with ##r < l## (don't know where the ##r## is coming from and why it has to be less than ##l##). Since ##l## is the smallest positive number and ##r < l##, it follows that ##r = 0## and this ##x = l \cdot z \, \forall \, x \in X##. The process is analogous for ##H_-##.

Does this make sense and if not, what went wrong? Thanks everyone!
It is correct. Written in a bit confusing way that suggests you didn't understand all steps, but all in all correct.

##X\subseteq \mathbb{Z}## is a subgroup. So ##0\in X## and ##X\neq \emptyset.##

If ##X=\{0\}## then ##X=0\cdot \mathbb{Z}## and we are done.

Next, we assume that ##\{0\}\subsetneq X## is a proper subgroup, i.e. ##X## contains at least one element ##x\neq 0.## With ##x\in X## we also have ##-x\in X## and either ##x## or ##-x## is positive. This means that ##X## contains a number ##x>0.## Since the natural numbers (positive integers) are bounded from below, we can choose the smallest positive ##X\ni l>0.##

We now apply the Euclidean algorithm, i.e. the ordinary division, and divide any number ##x\in X## by ##l.## Unfortunately, we are not allowed to divide ##x## by ##l## since ##\mathbb{Z}## doesn't contain all quotients. What we can do is to subtract ##l## from ##x## as long as the result is greater or equal ##l.## This is basically the long division: ##x=q\cdot l +r.## Subtract ##l## as long from ##x## as ##r\geq l,## here ##q## times, until ##r<l.## We can always perform this step such that ##0\leq r,## because we can select ##q<0## if necessary.

We now rearrange the equation: ##r=x-q\cdot l=\underbrace{x}_{\in X} - \underbrace{(\underbrace{l+l+\ldots +l}_{q\text{ times}}}_{\in X} \in X.##

However, ##X\ni r\geq 0## can only be the case if ##r=0## by the choice of ##l## as the minimal element among those in ##X##. Thus any arbitrary element ##x\in X## is of the form ##x=q\cdot l+0,## i.e. ##X=\mathbb{Z}\cdot l.##

The distinction between ##H_+## and ##H_-## isn't necessary. The Euclidean algorithm works for all integers in a way such that the remainder of consecutive subtractions are always positive. If necessary, i.e. if ##x<0,## then we perform consecutive additions instead.
 
fresh_42 said:
It is correct. Written in a bit confusing way that suggests you didn't understand all steps, but all in all correct.

##X\subseteq \mathbb{Z}## is a subgroup. So ##0\in X## and ##X\neq \emptyset.##

If ##X=\{0\}## then ##X=0\cdot \mathbb{Z}## and we are done.

Next, we assume that ##\{0\}\subsetneq X## is a proper subgroup, i.e. ##X## contains at least one element ##x\neq 0.## With ##x\in X## we also have ##-x\in X## and either ##x## or ##-x## is positive. This means that ##X## contains a number ##x>0.## Since the natural numbers (positive integers) are bounded from below, we can choose the smallest positive ##X\ni l>0.##

We now apply the Euclidean algorithm, i.e. the ordinary division, and divide any number ##x\in X## by ##l.## Unfortunately, we are not allowed to divide ##x## by ##l## since ##\mathbb{Z}## doesn't contain all quotients. What we can do is to subtract ##l## from ##x## as long as the result is greater or equal ##l.## This is basically the long division: ##x=q\cdot l +r.## Subtract ##l## as long from ##x## as ##r\geq l,## here ##q## times, until ##r<l.## We can always perform this step such that ##0\leq r,## because we can select ##q<0## if necessary.

We now rearrange the equation: ##r=x-q\cdot l=\underbrace{x}_{\in X} - \underbrace{(\underbrace{l+l+\ldots +l}_{q\text{ times}}}_{\in X} \in X.##

However, ##X\ni r\geq 0## can only be the case if ##r=0## by the choice of ##l## as the minimal element among those in ##X##. Thus any arbitrary element ##x\in X## is of the form ##x=q\cdot l+0,## i.e. ##X=\mathbb{Z}\cdot l.##

The distinction between ##H_+## and ##H_-## isn't necessary. The Euclidean algorithm works for all integers in a way such that the remainder of consecutive subtractions are always positive. If necessary, i.e. if ##x<0,## then we perform consecutive additions instead.
Thank you so much. Actually getting it explained is always super helpful. All I got were the equations. My friend didn't explain his steps and his handwriting was hard to read anyway, so I had a very hard time figuring out what each step meant. But this really does explain everything nicely. Thank you again :)
 
Why not simply:

Let ##l## be the least positive element in ##X##. By induction ##l\mathbb Z \subseteq X##.

If there exists positive ##k \in X## but ##k \notin l\mathbb Z## then ##nl < k < (n+1)l## for some ##n##. Hence ##k -nl \in X## with ##0< k -nl < l##. Contradiction.
 
PeroK said:
Why not simply:

Let ##l## be the least positive element in ##X##. By induction ##l\mathbb Z \subseteq X##.

If there exists positive ##k \in X## but ##k \notin l\mathbb Z## then ##nl < k < (n+1)l## for some ##n##. Hence ##k -nl \in X## with ##0< k -nl < l##. Contradiction.
This is the identical proof just without all the technical details and references to the group axioms and the fact that Euclidean rings are principal ideal domains. Whether you write ##k=nl+r## or ##x=ql+r## isn't really a difference.
 
fresh_42 said:
This is the identical proof just without all the technical details and references to the group axioms and the fact that Euclidean rings are principal ideal domains. Whether you write ##k=nl+r## or ##x=ql+r## isn't really a difference.
It seems to me that the OP may be too focused on the details and not yet able to see the outline of a proof for himself. The details rarely sort themselves out without my knowing what I'm trying to do.

I think it's important for the OP to recognise the common proof ideas here, as well as follow a complete step by step solution.
 
  • Like
Likes   Reactions: martinbn
Choosing a minimal element (##l##) from a subset of ##\mathbb{N}## and using the Euclidean division to find a smaller element (##r##) which leads to either a contradiction or to ##r=0## is a standard technique.

Locking ##k## up between ##nl## and ##(n+1)l## not so much.
 

Similar threads

  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 13 ·
Replies
13
Views
3K
  • · Replies 26 ·
Replies
26
Views
2K
  • · Replies 17 ·
Replies
17
Views
5K
  • · Replies 13 ·
Replies
13
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 20 ·
Replies
20
Views
3K