A motivating example
Consider the following equation, where x \in \mathbb{Z}:

    \[ 5+x=11 \]

You have most likely solved many equations such as these and would be able to conclude what x equals in a second or so. However, if we take our time solving the problem and consider all the steps necessary, we stumble upon the underlying rules that allow us to solve such equations. Writing out each step fully, we’d get something like this:

    \[ x+5=11 \]

    \[ \implies (x+5)+(-5)=11+(-5) \]

    \[ \implies x+(5+(-5))=6 \]

    \[ \implies x+0=6 \]

    \[ \implies x=6 \]

Writing all of this may seem a little long-winded but it reveals the rules- known as axioms- that allow us to do all of this. Firstly, we are able to write that (x+5)+(-5) is the same as x+(5+(-5)) due to the more general rule that:

    \[ (a+b)+c=a+(b+c) \quad a,b,c \in \mathbb{Z} \]

This is known as the property of associativity. Essentially, we have freedom over the order in which we add things. Next, the fact that 5+(-5)=0 is a specific case of the more general rule that:

    \[ a+(-a)=0 \quad a \in \mathbb{Z} \]

This arises from the fact that (-a) is the additive inverse of a. Two integers are additive inverses of one another when they sum to 0 and you will probably be aware of the fact that every integer has a distinct additive inverse. Lastly, we know that x+0=x since we know that generally:

    \[ 0+a=a+0=a \quad a \in \mathbb{Z} \]

Summarised in words, when we add 0 to an integer, we just get that integer. We refer to 0 as the identity element. We also know one final thing: when we add two integers, we get another integer. Written succinctly, we have that:

    \[ \forall a,b \in \mathbb{Z},  a+b \in \mathbb{Z} \]

This is referred to as closure.

The Four Axioms
When we have some operation * and it is acting over some set, S, we call the set with the operation a group when it satisfies the four axioms mentioned before:

    \[ \text{Associativity: } \forall a,b,c \in S,  a\( * \)(b\( * \)c)=(a\( * \)b)\( * \)c \]

    \[ \text{Identity: } \exists e \in S \text{ s.t. } \forall a \in S,e\( * \)a=a\( * \)e=a \]

    \[ \text{Inverses: } \forall a \in S,\exists a^{-1} \in S \text{ s.t. } a\( * \)a^{-1}=a^{-1}\( * \)a=e \]

    \[ \text{Closure: } \forall a,b \in S,  a\( * \)b \in S \]

We denote a group with set S and operation * by (S, *). There are many groups you likely have dealt with many times such as (\mathbb{Z},+) and (\mathbb{Q}\setminus \{0\}, \times). The set of all matrices with entries in \mathbb{R} and matrix multiplication as the operation is not a group since matrices that have a determinant equal to zero don’t have inverses. However, if we consider matrix multiplication with entries in \mathbb{R} over the set of invertible matrices, this does in fact form a group.

Examples
While the groups previously mentioned certainly are fascinating in their own right, we will now shift our attention to a group that we tend to not deal with much in pre-university mathematics, but is of enormous significance in group theory. I am speaking of the symmetric group on n letters, which revolves around rearrangements of the set \{1,2,3…n\}. For the sake of simplicity, we’ll begin with S_3, the symmetric group on 3 letters. Here are some elements of the group:

    \[ \begin{pmatrix} 1 & 2 & 3 \\ 2 & 3 & 1 \end{pmatrix},\  \begin{pmatrix} 1 & 2 & 3 \\ 3 & 1 & 2 \end{pmatrix},\  \begin{pmatrix} 1 & 2 & 3 \\ 2 & 1 & 3 \end{pmatrix} \]

Our operation is composition; here is an example of composing two elements:

    \[ \begin{pmatrix} 1 & 2 & 3 \\ 2 & 1 & 3 \end{pmatrix} \circ \begin{pmatrix} 1 & 2 & 3 \\ 3 & 2 & 1 \end{pmatrix} = \begin{pmatrix} 1 & 2 & 3 \\ 3 & 1 & 2 \end{pmatrix} \]

Focusing on a single number, we see that the path of 3 is 3 \rightarrow 1 \rightarrow 2. Notice that we read compositions of symmetries right to left like how we read the composition of functions. Generally, some element \begin{pmatrix} 1 & 2 & 3 \\ a & b & c \end{pmatrix} corresponds to rearranging the numbers 1, 2, 3 to a, b, c where a, b, c are distinct integers from 1 to 3. Notice that similarly to matrices, when composing two elements, we consider the element on the right first and then the element on the left. It is true that S_3 satisfies the four necessary axioms to be classified as a group, and generally, we have that S_n is a group (the proof of this is left as an exercise for the reader).

What about the set \{a+bi \in \mathbb{C} : a^2+b^2=1\} under multiplication- is this a group? Let’s have a look. It is clear that 1, which is thankfully in our set, will act as our identity, and we have associativity since complex multiplication is generally associative. The inverse of some a+bi is a-bi as (a+bi)(a-bi)=a^2+b^2=1. To prove that we have closure, we consider some elements a+bi and c+di and find their product then show it too is a complex number of the form x+yi that satisfies x^2+y^2=1:

    \[ (a+bi)(c+di)=(ac-bd)+(ad+bc)i \]

    \[ (ac-bd)^2+(ad+bc)^2=a^2 c^2+b^2 d^2+a^2 d^2+b^2 c^2=(a^2+b^2 )(c^2+d^2 )=1 \]

And thus, we have shown that the set \{a+bi \in \mathbb{C} : a^2+b^2=1\} under multiplication is indeed a group. We could have taken a more geometric approach too of course; our set is essentially all of the possible movements around the unit circle in the complex plane. Any two movements around the unit circle can just be considered as a singular combined movement, so we have closure. We can choose to not move around the unit circle at all and this would be our identity. It doesn’t matter in which order we apply movements around the unit circle so we have associativity. Finally, we clearly have an inverse for each movement since we just move in the opposite direction the same distance. My geometric explanation of course lacks rigour and isn’t as precise as my previous justification however it can provide us with some intuition as to whether or not we are dealing with a group.

Lastly, let’s consider a group predicated on rotating and flipping a square. We allow ourselves to rotate the square by 0, 90, 180, or 270 degrees and to flip the square about its 4 lines of symmetries. Our identity is clearly a rotation of 0 degrees as this keeps the square fixed. It’s quite clear that we have associativity and closure (if you don’t find this clear then spend some time convincing yourself of it), and we have inverses as the flips are self-inverses while the inverse of a n degree rotation is simply a 360-n degree rotation. Note that we are implicitly using modulo 360 here. This group is known as D_8, the dihedral group of order 8. Generally, a polygon with n vertices has 2n symmetries and the group of these symmetries is denoted D_{2n}. Note however that some writers denote the group as D_{n} in reference to the number of edges of the polygon. Conventions differ amongst different writers and so elsewhere you may see the group I have spoken of referred to as D_4

Cyclic groups
In group theory, the word order can have two different meanings. When referring to some group G, the order of G, written as \lvert G \rvert, means the number of elements in the set G. If the order of a group is finite then we say that it’s a finite group; otherwise, we refer to it as an infinite group. There is also another definition, but before defining that I’ll first define a few more things.

Let G be a group with a \in G. Then, for n \in \mathbb{Z}^{+}, we define a^{n}=\prod_{k=1}^{n}a. Bare in mind that although you may be accustomed to thinking of aa as meaning a \times a, this is not necessarily the case. For example, if the operation is addition, then aa=a+a and correspondingly a^{2}=a+a. A group G is said to be cyclic if there exists an element a such that \forall g \in G, \exists i \in \mathbb{Z} such that g=a^{i}. That is, every element in G is just some power of a. If this is the case then we say that G is generated by a and write that G= \langle a \rangle.

For example, the additive group \mathbb{Z} is cyclic as \mathbb{Z}= \langle 1 \rangle. Another example is the multiplicative group of complex fourth roots of unity, \{1,-1,i,-i\}. This group is cyclic and is generated by i.

Now for the other definition of order. Let G be a group and a \in G. The order of a, denoted \lvert a \rvert, is the smallest positive integer n such that a^{n}=e, assuming that such n exists, in which case a has finite order. If no such n exists then we say that a has infinite order.

Proposition: Let G be a group and a \in G. Suppose i,j \in \mathbb{Z}. Then:

1. If a has infinite order, then a^{i}=a^{j} if and only if i=j
2. If \lvert a \rvert=n< \infty, then a^{i}=a^{j} if and only if i \equiv j \pmod{n}

Proof: (1) Suppose that a^{i}=a^{j} but i \neq j. Without loss of generality, suppose that i>j Then a^{i}(a^{j})^{-1}=a^{j}(a^{j})^{-1}=e. That is, a^{i-j}=e. But i-j is a positive integer, and this contradicts the assumption that a has infinite order.

(2) Suppose that a^{i}=a^{j}. Once again, a^{i-j}=e. Using the Euclidean division algorithm, write i-j=nq+r with q,r \in \mathbb{Z} and 0 \leq r<n. Then e=a^{i-j}=a^{nq+r}=(a^{n})^{q}a^{r}. But a^{n}=e and thus a^{r}=e. But since n is the smallest positive integer having this property and r<n, it must be that r=0. And hence n \rvert (i-j) as desired. Conversely, suppose that i \equiv j \pmod{n}. Then let us write i-j=nk for some k \in \mathbb{Z}. But we now have that a^{i-j}=a^{nk}=(a^n)^{k}=e^{k}=e. Thus a^{i-j}a^{j}=ea^{j} and hence a^{i}=a^{j} as desired. \quad \blacksquare

In these two proofs, we have assumed that if a is an element of some group and m and n are integers, then a^{m}a^{n}=a^{m+n}, (a^{m})^{n}=a^{mn} and a^{-n}=(a^{-1})^{n}. Thankfully, these are all true. The proofs are not difficult and so are left as exercises for the reader.

A useful corollary of the second proposition is that for a group G with some element a of order n< \infty and some integer i, a^{i}=e if and only if n \rvert i

Subgroups
For some group G, we define H, a subset of G, to be a subgroup of G if H is also a group under the same operation and write H\le G. For every group, the group itself and \{e\} are both subgroups. We refer to a subgroup as a proper subgroup when it is a group other than G itself.

For example, we can see that \mathbb{Z} is a subgroup of \mathbb{Q} and both are subgroups of \mathbb{R} with addition as the operation. So, what determines whether a group is a subgroup of another group or not?

Theorem: Let G be a group and H a subset of G. Then H\le G if and only if:
1. e \in H
2. \forall a,b \in H, ab \in H
3. \forall a \in H, a^{-1} \in H

Proof: If H is a subgroup of G, it must have an identity element f such that ff=f but we also have that ef=f. Hence ff=ef \implies f=e as required. By the definition of a group, the second and third condition must hold. Conversely, let’s suppose that the conditions hold. Clearly, the conditions guarantee the existence of an identity element, closure, and the existence of inverses of each element. Associativity is inherited from G and so H satisfies the four axioms and is indeed a subgroup of G. \quad \blacksquare

Commutativity
A convenient feature of addition over the integers is that when adding two integers, it doesn’t matter what order we add them; 5+7=7+5 and generally:

    \[ a+b=b+a \quad a,b \in \mathbb{Z} \]

This property is known as commutativity and is present in practically all of the groups we tend to encounter in pre-university mathematics. However, there are many groups that don’t have this property. Matrix multiplication, for example, isn’t commutative, and nor is the composition of symmetries. Groups in which all elements commute are known as abelian groups. It can be useful to know which elements of a group commute regardless of whether a group is abelian or not, and so we refer to elements of a group, G, that commute with all other elements in G as the centre of the group, Z(G). For example, the centre of the group with operation matrix multiplication acting on the set of invertible 2 \times 2 matrices with entries in \mathbb{R} is the set of matrices:

    \[ \left\{ \left( \begin{array}{c} a \\ 0 \end{array} \begin{array}{c} 0 \\ a \end{array} \right) : a \in \mathbb{R} \setminus \{0\} \right\} \]

Trending