Rings and Fields

CM10196
Topic 5: Groups, Rings and Fields

A brief introduction to algebra
Guy McCusker
1W2.1
Data plus operations
In programming, and in mathematics, we are often interested in

manipulating certain kinds of data in certain kinds of ways.
That means we are working with

I a collection of data values: that is to say, some kind of set
I a collection of operations on that set: ways of manipulating the data.
Usually, the operations we are using have particular properties that we

exploit; for instance, if we put an element onto the front of a list, and then
look at the front element of the list, we expect to get back the element we
just put there. So we can add to our situation
I a collection of equations which describe how the operations work.
G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 2 / 56
Example: lists
A data structure for “lists of numbers” might look as follows:

I the data we work with are the numbers (e.g. integers) and the lists
I the operations we have are:
I create an empty list
I put an element onto a list
I retrieve the first element of a list (the head)
I retrieve the rest of the elements of the list (another list, called the tail)
I the equations say things like
I if we put n onto a list, the head of the resulting list is equal to n
I if we put n onto a list l, the tail of the resulting list is equal to l
and so on.

Algebra
The field of mathematics known as algebra studies exactly this kind of

situation.
The basic plan of action in algebra is

I observe that certain collections of operations and equations are of
interest in more than one setting
I define an abstract structure, called an algebra, which makes the
operations and equations precise
I study the properties of these algebras at a general level
I apply the general properties to particular instances of the structure to
find out useful facts.
We’ve already done this once, with boolean algebras in Topic 1.
Operations
What do we mean by operations on a set?
In this course, we will only study three kinds of operation:

I binary operations
I unary operations
I nullary operations
Binary operations
A binary operation takes two elements of the set and operates on them to
give a third element.
Example
I addition is a binary operation on the integers
I conjunction and disjunction are binary operations on the truth values
I union and intersection are binary operations on P(A) for any set A
I subtraction of one number from another is a binary operation on the
integers, but not on the natural numbers because for example 3 − 5
does not give a natural number.

Unary operations
A unary operation takes a single element of a set and operates on it to

give a new element.
Example
I The successor (add one) function is a unary operation on the natural
numbers and on the integers
I Multiplication by −1 is a unary operation on the integers.
I Negation is a unary operation on the truth values.
Nullary operations
A nullary operation takes no elements of a sets as input, and gives an
element as output.
That is, a nullary operation is just an element of a set, sometimes called a

constant. When defining an algebra, the constants pick out particular
elements that are of interest.
Example
I The constant values true and false are nullary operations on the
set of truth values.
I Any natural number could be a constant in the set of natural
numbers, but one of particular interest if we’re doing addition is zero,
because it’s a unit for addition.
You shouldn’t have too much trouble imagining that there are other kinds
of operation: ternary ones (three inputs), infinitary ones (infinite number
of inputs) and so on.
Operations are functions
When we describe a set with a binary operation on it, we’ll have to give
two pieces of data:
I the set itself; this will be some set A
I the operation; this will be a function f : A × A −→ A.
For instance, the set N of natural numbers together with the addition
function + : N × N −→ N is a “set with a binary operation”.
If we wanted to specify a unary operation, we’d have to give a function

g : N −→ N.
To specify a nullary operation (a constant), we just give an element n ∈ N.

Monoids
Let’s define an algebra!

Definition
A monoid is given by
I a set A
I a binary operation on A, usually written as a dot (e.g. a · b denotes
the result of applying the operation to elements a and b) or as
concatenation ab
I a constant i.e. an element of A, usually called e, and referred to as
the unit or identity of the monoid.
such that the following equations hold for all a, b, c ∈ A:
associativity (a · b) · c = a · (b · c)
unit a · e = a and e · a = a
Examples of monoids
I The natural numbers with the operation of addition and the constant
0 form a monoid; this works on the integers too.
I The natural numbers with the operation of multiplication and the
constant 1 form a monoid; this works on the integers too.
I The truth values, with the operation of conjunction and the constant
value true form a monoid.
I The truth values, with the operation of disjunction and the constant
value false form a monoid.
I The collection of finite lists of numbers, with the operation of
concatenation and the constant value the empty list, forms a monoid.
What about implication?
Can we make a monoid out of the implication operation?
The answer is no, because it is not associative:
(false → false) → false = false

false → (false → false) = true
On the other hand, the operation of logical equivalence, ≡, is associative,

and true is a good unit for it, so we can make a monoid out of that.

Integers mod k
Arithmetic modulo k, for some nonzero natural number k, gives us a large
and important class of monoids.
The underlying set we will work with is called Z/kZ (for slightly complex
reasons); all we need to know is that
Z/kZ = {0, 1, 2, . . . , k − 1}.
We can then define addition mod k:
a +k b = the unique c ∈ Z/kZ such that a + b = nk + c for some integer n
and similarly multiplication mod k:
a ×k b = the unique c ∈ Z/kZ such that a × b = nk + c for some integer n

Two monoids on the same set
This gives us two monoids on the same set:

I Z/kZ with the operation of addition mod k, and the constant 0, is a
monoid
I Z/kZ with the operation of multiplication mod k and the constant 1
is a monoid.
Exercise
Suppose we consider the same set but without the element 0, and the
operation of multiplication mod k. Is the operation well-defined on this
smaller set? What if k is a prime?
The question is asking if, given a and b in Z/kZ \ {0}, the number a ×k b
is still in the set.
Functions on a set
Let A be any set, and consider the set AA , that is, the set of all functions
from A to A.
Given f and g : A −→ A, we can form a new function g ◦ f : A −→ A.
We know that
h ◦ (g ◦ f ) = (h ◦ g ) ◦ f
and also that, if we write id for the identity function on A,
id ◦ f = f ◦ id = f .
All this says that

the set of functions from A to itself, with function composition as the binary
operation and id as the constant, form a monoid.

Identities are unique
Let’s prove a fact about monoids in the abstract, i.e. without talking
about a particular monoid.
Theorem
In any monoid, there is exactly one identity element, i.e. exactly one
element e such that e · a = a = a · e for all a.
Proof We know from the definition of a monoid that there is at least
one identity element, so we have to show that there cannot be more.
Suppose e1 and e2 both have the required property. Then
e1 = e1 · e2 = e2
using the unit equation for e2 to get the first equality, and the unit
equation for e1 to get the second.
The monoid of functions
We can find submonoids of the monoid AA :

I since the identity is 1-1, and composing two 1-1 functions gives
another 1-1 function, the set of 1-1 functions forms a monoid with
the same operation (composition) and the same unit (the identity
function).
I similarly, all the onto functions form a monoid
I combining the above observations we can see that the bijective
functions form a monoid too.
This last monoid, which we will call SA , plays a very important role in
mathematics. It is called the symmetric monoid on the set A.
Symmetric monoid on {1, 2, 3}
Here are all six of the bijections from the set {1, 2, 3} to itself.
3 3 3 3 3 3
2 2 2 2 2 2
1 1 1 1 1 1
3 3 3 3 3 3
2 2 2 2 2 2
1 1 1 1 1 1

Symmetries of a triangle
These six maps correspond to the six ways of drawing a triangle with
corners labelled A, B and C:
A A C
C B B C A B
B C B
C A B A A C
Notice that each layout of the triangle can be obtained by rotating and
reflecting the original one.
Inverses
The symmetries of a triangle (or any set) have a special property: each
such map has an inverse, that is to say, another map you can compose it
with to end up with the identity.
Monoids where every element has an inverse are the subject of a large and
important area of algebra called group theory.
To make the idea of “monoid with inverses” precise, we’ll define a new
kind of algebraic structure, called a group.
Groups
Definition
A group is given by
I a set A
I a binary operation on A (written with a dot or with concatenation, as
before)
I a nullary operation (constant) called the unit or identity and written e
I a unary operation called inverse; the inverse of an element a is
written a−1
such that the following equations hold for all a, b, c ∈ A:
associativity (a · b) · c = a · (b · c)
unit a · e = a and e · a = a
inverse a · a−1 = e = a−1 · a.

Examples of groups
I The integers with addition as the operation, 0 as the unit, and −n as

the inverse of n, form a group.
I The non-zero rational numbers with multiplication as the operation, 1
as the unit, and 1/x as the inverse of x, form a group.
I The bijective functions from a set A to itself, with composition as the
operation, the identity function as unit, and the inverse of a function
as inverse, form a group SA .
Non-examples of groups!
Some non-examples:
I The natural numbers with addition as the operation do not form a
group because there’s no inverse for any n > 0.
I The integers with multiplication do not form a group because no
number other than 1 has an inverse.
I The rationals with multiplication do not form a group because 0 has
no inverse.
I The 1-1 functions on a set do not form a group because a function
which is not bijective does not have an inverse.
Inverses are unique

We already know (from the fact about monoids) that a group has exactly
one identity element.
We can also show that every element has exactly one inverse:
Theorem
In any group, each element a has exactly one inverse, i.e. there is exactly
one element b such that a · b = e = b · a.
Proof We know that a has at least one inverse because a−1 must exist
by definition of a group. So we just show that there cannot be two
inverses.
Suppose b1 and b2 are both inverses of a. Then
b1 = b1 · e = b1 · (a · b2 ) = (b1 · a) · b2 = e · b2 = b2 .

Note that we used the property of associativity, as well as the fact that b1
and b2 are inverses of a and that e is a unit. All the group axioms are at
work here!
Facts about inverses
Knowing that inverses are unique lets us demonstrate a few simple facts
which hold in every group.
Theorem
For any element a of a group. (a−1 )−1 = a.
Proof Since inverses are unique, we just have to show that a is an

inverse of a−1 . But this just means that
aa−1 = e = a−1 a
which is true because a−1 is the inverse of a.
Exercise
Show that, for any elements a and b of a group,
(ab)−1 = b −1 a−1 .
Cancellation laws
Theorem
In any group, for any a, b and c,
I if a · b = a · c then b = c
I if b · a = c · a then b = c.
Exercise
Prove this. It’s easy. Remember inverses exist.
The group (Z/nZ)+
For any natural number n, the set
{0, 1, 2, . . . , n − 1}
can be given the structure of a group:

I the operation is addition modulo n
I the unit is 0
I the inverse of a is n − a.
The notation (Z/nZ)+ is used to denote this group.

The group (Z/pZ)∗
Let p be any prime. We can turn the set
{1, 2, . . . , p − 1}
into a group under the operation of multiplication mod p:

I 1 is the identity element
I it turns out that for any a in the set there is some b such that
a ×p b = 1, so inverses exist.
The notation (Z/pZ)∗ is used for this group.
Let’s check that inverses exist for the case p = 5:

I the inverse of 1 is 1: 1 × 1 = 1
I the inverse of 2 is 3: 2 × 3 = 6 = 1 mod 5
Multiplication tables
Sometimes, for small groups, it can help to write out a multiplication

table, showing how the binary operation of the group works.
For (Z/5Z)∗ , the multiplication table looks like this.
· 1 2 3 4
1 1 2 3 4
2 2 4 1 3
3 3 1 4 2
4 4 3 2 1
Cyclic groups
In this group, the element 3 is called a generator: if we look at the

sequence
3, 3 · 3, 3 · 3 · 3, . . .
we reach every element of the group: the sequence is the same as
3, 4, 2, 1, 3, 4, 2, 1, . . .
Because multiplying by 3 takes us round and round this loop, hitting all
the elements as we go, the group is called cyclic.
Exercise
Check that 2 is also a generator for the group. What about 1 and 4?

Order of a group and an element
The number of elements in a finite group is called its order. So the group
we just saw is called the cyclic group of order four.
Given any element a of a group, we can look at the elements
a, a · a, a · a · a, ...
These are referred to as
a, a2 , a3 , ...
We also define a0 = e, the identity element.
The order of element a is the smallest n > 1 such that an = e.
Order of an element
Theorem
In a finite group, every element has a finite order.
Proof Write down the sequence of elements

a, a2 , a3 , a4 , . . .
This is an infinite sequence of elements of the group, but the group is
finite so at some point we must find two elements the same. That is, there
must be some m and n (with m < n, say) such that
am = an .
But this means am = am · an−m . If we now multiply both sides by (am )−1 ,
we get
e = (am )−1 · am = (am )−1 · am · an−m = e · an−m = an−m
so the order of a is at most n − m.
Cyclic groups again
A finite group of order n is a cyclic group if it has a generator.
A generator is just an element of order n, i.e. an element a such that
a, a2 , a3 , . . . , an−1
are all different, but

an = e,
which of course means that any ak with k ≥ n is equal to one of
e, a, a2 , a3 , . . . , an−1 .

The group (Z/nZ)+ again
For any n, the group (Z/nZ)+ is a cyclic group of order n.
The element 1 is always a generator: because the operation is addition

(mod n), we have
11 = 1
12 = 2
..
.
1k = k
..
.
1n−1 = n − 1
1n = 0
Logarithms
In the real numbers, any positive number x has a logarithm base 2, which
is to say a number y such that
2y = x.
In any cyclic group with generator g , every element a can be written as g n

for some n. By analogy with the above, we refer to this n (actually, we use
the smallest such n) as the logarithm base g of a.
The difficulty of computing the logarithm of a number in (Z/pZ)∗ is the

key to the cryptographic methods that make secure internet transactions
possible.
If you’re interested, research the Diffie-Hellman algorithm (easy exercise in

using Google.)
Groups of symmetries
Another important collection of examples of groups are the symmetries on

regular polygons. That is, for a given regular polygon, like an equilateral
triangle or a square, pentagon etc., consider all the ways of rotating and
reflecting it.
We’ve already seen this group (as a monoid) in the case of a triangle. In
that case, every permutation of the set {1, 2, 3} gave us an element of the
group.
In general, for an n-sided polygon, the group of symmetries has 2n

elements, while the group of all permutations of {1, 2, 3, . . . , n} has n!
elements, so they’re not the same.

Symmetries of the square
Below are the eight elements of the group of symmetries of a square.
A B D A C D B C
D C C B B A A D
D C C B B A A D
A B D A C D B C
Symmetries and permutations

The symmetries of the square correspond to bijections on {A, B, C , D} in
an obvious way.
For each picture, construct a map which takes A to the element in the
top-left, B to the top-right, C to the bottom-right and D to the
bottom-left.
So for instance,
D A
C B
corresponds to the map
A 7→ D
B 7→ A
C 7→ B
D 7→ C
Subgroups
If we have a group, sometimes we can find a subset of its elements which
also forms a group, with the same operation, unit and inverse as the
original group.
For instance, consider the set of permutations (bijective functions) on the

three-element set {A, B, C }.
There are 6 of them, corresponding to the symmetries of the triangle we

drew before.
Three of these correspond to rotations, without any reflection. Since

rotating and then rotating again gives another rotation, these three
elements are closed under composition.
Thus these three themselves form a group.
(It’s a cyclic group, again: rotating once is a generator, so we have the

cyclic group of order three.)
Rotations of the triangle
Here are all the symmetries of the triangle, with the ones that are not
rotations drawn in grey.
A A C
C B B C A B
B C B
C A B A A C
Rotational permutations
Here are all six of the bijections from the set {1, 2, 3} to itself, with the
ones that are not rotations greyed out. Notice the way these bijections
correspond to the symmetries of the triangle on the previous slide.
3 3 3 3 3 3
2 2 2 2 2 2
1 1 1 1 1 1
3 3 3 3 3 3
2 2 2 2 2 2
1 1 1 1 1 1
Abelian groups
A group is called Abelian if its binary operation is commutative: that is, if
ab = ba
for all a and b in the group.
All the groups we’ve seen that are based on addition or multiplication of
numbers are Abelian, because addition and multiplication are themselves
commutative.

A non-Abelian group
The group of permutations on {1, 2, 3} is not Abelian: consider composing

the two maps
3 3 3 3
2 2 2 2
1 1 1 1
Composing one way round, 1 is sent to 1 and then to 3; composing the

other way around, 1 is sent to 3 and then to 2. So the two compositions
do not give the same map.
Rotations are Abelian
On the other hand, if we restrict to the rotations we get another Abelian

group.
In fact, any cyclic group is Abelian: since every element can be written as
g n where g is the generator, we always have
g n · g m = g n+m = g m+n = g m · g n .
Rings
Notice that lots of our examples of groups seemed to be generalizations of

ideas from arithmetic: multiplication or addition played a key role, for
example.
In arithmetic, though, we are interested in both multiplication and addition

at the same time (and more besides). That is to say, we care about more
than one binary operation.
To talk about these things in general, we introduce the structure of a ring.

Rings
Definition
A ring is given by
I a set A
I a binary operation called multiplication on A, written ×
I a binary operation called addition on A, written +
such that
I there is an element 0, and a unary operation taking each element a to
−a, which make the set A with operation + into an Abelian group.
I there is an element 1 which makes the set A with operation × into a
monoid
I multiplication distributes over addition:
a × (b + c) = (a × b) + (a × c)
(a + b) × c = (a × c) + (b × c)
Warning
What we’ve defined is sometimes called a ring with 1 or unitary ring:

sometimes rings are not required to have the unit for the multiplication
operation.
Examples
I The integers with the usual operations of addition and multiplication

of course form a ring.
I The rationals and the reals are rings with the usual operations.
I The set Z/nZ is a ring, with the operations of addition mod n and
multiplication mod n. Note that in this case −a is the number n − a.

Some simple facts
Theorem
In any ring, for any a and b:
I 0×a=0=a×0
I (−a) × b = a × (−b) = −(a × b)
Proof For the first one, we calculate

0 + (0 × a) = 0 × a = (0 + 0) × a = (0 × a) + (0 × a)
and then use the cancellation property of the group to deduce that
0 = 0 × a.
For the second one, we show that (−a) × b is an inverse of a × b with
respect to +:
(a × b) + (−a × b) = (a + −a) × b = 0 × b = 0
and then use the fact that inverses are unique to deduce the equation we
need.
Fields
Rings talk about addition and multiplication, but not division. We can
request further structure to arrive at a field.
Definition
A field is a ring such that
I 0 6= 1
I multiplication is commutative
I for every a 6= 1, there exists a−1 such that a × a−1 = a−1 × a = 1.
To put it another way, the non-zero elements of the ring form an Abelian
group.
Examples
I The integers do not form a field because division is not possible

I The rationals and the reals do form fields
I The rings Z/pZ where p is prime are fields.
I If n is not prime, Z/nZ is not a field, because multiplicative inverses
do not exist.

No zero divisors
Theorem
In any field, if a × b = 0 then at least one of a and b is zero.
Proof Suppose a × b = 0. If a = 0, we’re done. If a 6= 0, then a−1

exists, and we have
b = 1 × b = (a−1 × a) × b = a−1 × (a × b) = a−1 × 0 = 0.

This says that 0 has no interesting divisors; it is therefore said that a field
has no zero divisors.
That’s why Z/nZ is not a field if n is not prime: if we have a × b = n

where a and b are not zero or one, then a ×n b = 0, so the structure
cannot be a field.
And on into algebra. . .
The topic of algebra concerns the study of these kinds of structure in great
depth. There are many amazing things that can be constructed and
proved using the kinds of ideas we’ve seen here.
We have barely scratched the surface.
The most important ideas are here, though: the idea that we can
abstractly capture the kinds of operations and equations we make use of,
and reason about these structures at a general level.
Appendix 1: proving that (Z/pZ)∗ is a group

We need to show that for every element a there is some b such that
a ×p b = 1
i.e. that ab = np + 1 for some integer n.
To do this, we use a very interesting lemma:

Lemma
Let a and b be integers, and let c be the least positive integer of the form
ax + by , where x and y are integers. Then c is the greatest common
divisor of a and b.
Once we’ve got this lemma, the fact we need follows immediately: since
the gcd of a and p is 1, we get
1 = ax + py
for some x and y , and then a ×p x = 1.
Proving the lemma
We first show that the number c divides both a and b.
We know there is some m with 0 ≤ m < c such that a = nc + m. But then
m = a − nc
= a − nax − nby
= a(1 − nx) + b(−ny )
Since c is the smallest integer of this form, and since m < c, we must
have m = 0 i.e. c divides a. Similarly we show c divides b.
Proving the lemma
Finally we show that any number that divides both a and b divides c. This
is easy.
Suppose n divides a and b. Then n divides ax and also by , so it divides

ax + by , for any x and y . Hence n divides c.
So c is a divisor of both a and b, and any other common divisor divides c.

Thus c is the greatest common divisor of a and b.

Rings and Fields

Uploaded by

Copyright:

Available Formats

Rings and Fields

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Rings and Fields

Uploaded by

Copyright:

Available Formats

CM10196

Topic 5: Groups, Rings and Fields

Data plus operations

In programming, and in mathematics, we are often interested in

That means we are working with

Usually, the operations we are using have particular properties that we

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 2 / 56

A data structure for “lists of numbers” might look as follows:

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 3 / 56

The field of mathematics known as algebra studies exactly this kind of

The basic plan of action in algebra is

We’ve already done this once, with boolean algebras in Topic 1.

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 4 / 56

What do we mean by operations on a set?

In this course, we will only study three kinds of operation:

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 5 / 56

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 6 / 56

A unary operation takes a single element of a set and operates on it to

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 7 / 56

That is, a nullary operation is just an element of a set, sometimes called a

Operations are functions

If we wanted to specify a unary operation, we’d have to give a function

To specify a nullary operation (a constant), we just give an element n ∈ N.

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 9 / 56

Let’s define an algebra!

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 10 / 56

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 11 / 56

What about implication?

Can we make a monoid out of the implication operation?

The answer is no, because it is not associative:

(false → false) → false = false

On the other hand, the operation of logical equivalence, ≡, is associative,

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 12 / 56

Z/kZ = {0, 1, 2, . . . , k − 1}.

We can then define addition mod k:

a +k b = the unique c ∈ Z/kZ such that a + b = nk + c for some integer n

and similarly multiplication mod k:

a ×k b = the unique c ∈ Z/kZ such that a × b = nk + c for some integer n

Two monoids on the same set

This gives us two monoids on the same set:

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 14 / 56

Given f and g : A −→ A, we can form a new function g ◦ f : A −→ A.

All this says that

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 15 / 56

Suppose e1 and e2 both have the required property. Then

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 16 / 56

The monoid of functions

We can find submonoids of the monoid AA :

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 17 / 56

Symmetric monoid on {1, 2, 3}

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 18 / 56

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 19 / 56

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 20 / 56

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 21 / 56

I The integers with addition as the operation, 0 as the unit, and −n as

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 22 / 56

G.A.McCusker@bath.ac.uk (1W2.1) CM10196 Topic 5: Groups, Rings, Fields 23 / 56

Inverses are unique

Proof Since inverses are unique, we just have to show that a is an