MATH20201: Algebraic Structures 1

Dr Marianne Johnson
A 10 credit, level 2, blended learning course, Semester 1, 2021/22
Required knowledge (Foundations of Pure Mathematics & Linear Algebra):

• Binary operations (definition, properties of associativity and commutativity),

• Functions (composition, permutations, linear transformations),
• Number theory (division with remainder, modular arithmetic),
• Matrices (matrix addition, matrix multiplication, invertible matrices, determinant),
• Vector spaces (definition),
• Equivalence relations (definition, equivalence classes, partitions).

Asynchronous resources
Lecture notes and exercises are contained in this document:

Wk 1:First examples of groups Wk 8:Cosets, Lagrange’s Theorem

Wk 2:Examples and notation Wk 9:Homomorphisms
Wk 3:Subgroups Wk 10:Conjugacy
Wk 4:Cyclic subgroups and order Wk 11:Factor groups
Wk 5:Centralisers and Centres Wk 12:First Isomorphism Theorem
Wk 7:Cyclic groups Solutions to Exercises

Lecture videos and quizzes can be accessed via: Blackboard -> AS1 (log-in required)

Synchronous sessions
Each week you should attend one Review session (Mondays 9am) and one Tutorial (check
your personal timetable for details).
Blackboard -> AS1: Live sessions (log-in required)

Communication
Discussion boards: Blackboard -> AS1: Communication (log-in required)
Office hours: Zoom: Mondays 10:00-12:00, Thursdays 16:00-18:00
Email: [email protected]
If you experience problems with this course, please let me know as soon as possible.
There will be course unit feedback surveys in week 2/3, and at the end of the course.
Assessment and feedback
In-semester assessment: 20% timed online test in week 7.
End-of-semester assessment: 80% exam.
Feedback opportunities: quizzes, discussion boards, tutorials, office hours, coursework.
c University of Manchester Algebraic Structures 1 Study Guide, Page 1

Before we begin

Welcome to MATH20201!
This is a blended learning course (see below) taking place in semester 1 of 2021/22.
This document contains all key information about the course, including the lecture
notes, exercises, solutions, and an explanation of how to access the remaining resources
(lecture videos, lecture quizzes, review sessions, tutorials, discussion boards, office hours).
I strongly recommend that you download and save a copy of this file onto
your own device and that you view the document with a pdf reader, rather
than accessing it through a web browser. There are two very good reasons for this:
• If you experience problems with your local internet connection, you will still have
access to your own personal copy of the notes, exercises and solutions for the course,
and will have all the relevant information about course organisation and assessment;
• If you have a PC, laptop or tablet [or a printer!], then you can make notes directly
onto your personal copy of the notes. [However, note that if using a printer, you will
miss out on the navigational links...]

Navigation: Quick links to each section of the notes and other course materials (e.g.
videos, quizzes, etc.) can be found on the front page, and are also embedded throughout
the pdf notes. You will find these links coloured in blue, like the one above. If you are
viewing the notes with a pdf reader such as Adobe Acrobat Reader, then the links may
also be underlined and you may also have access to the ’index’ of sections in a side panel.
If you cannot see the links, or any links seem to be broken, then please let me know.

Blended learning: Activities of the course can be broadly categorised as asynchronous

(directed study that can be performed individually, at your own pace, at a time of your
choosing) or synchronous (activities that we do together at a particular time), as I shall
now explain.

Asynchronous: Lecture Notes, Videos, Quizzes, and Exercises

The notes contained in this document are based on lecture notes originally produced by
Ralph Stöhr, with additions from Peter Rowley, Christopher Frei, and myself.
Each chapter of the notes corresponds to material to be studied in a particular week of
the course, and typically contains:
• Outline of the topics for study and checklist of tasks to be completed in week X,
with links to relevant materials
(e.g. attend review session and tutorial; read lecture notes OR watch lecture videos;
complete quizzes; complete exercises and take time to reflect).
c University of Manchester Algebraic Structures 1 Study Guide, Page 2

• Notes for two Lectures on week X topics;

• Exercises relating to week X topics.
(Solutions to all exercises are provided in the final chapter.)
The notes and exercises accompany the videos and quizzes available at Blackboard ->
AS1 (log-in required). Each Lecture is divided into two subsections and these are
numbered to align with the video content (e.g. Lecture Videos 1.1 and 1.2 correspond to
the two subsections of Lecture 1). Quizzes are aligned to the lectures (e.g. Quiz 1 relates
to Lecture 1, covering topics from videos 1.1 and 1.2). Note: the quizzes do not form part
of the assessment for this course, but they can be helpful in checking your understanding,
and provide an opportunity for feedback. You may also find the quizzes helpful as
practice for the Week 7 test.
Worked solutions to all exercises are provided in one place at the end of this document.
At the end of each set of exercises, there is a link that will take you to the solutions.

Synchronous: Review Sessions and Tutorials

It is very helpful for your learning to be able to discuss and explain key points with your
peers, and to get further insight and feedback from the lecturer in real-time. For this
reason, each week there will be two synchronous (i.e. ‘live’) sessions each week:
• Review Session: Each week there will be a 50-minute Review Session
(Blackboard -> AS1: Live sessions), to review the key points relating to last
week’s topic. The exact format of theses sessions may vary from week to week1 , but
will most likely include:
– Summary of the main ideas you studied last week
– Additional examples
– Discussion of items raised on the discussion board
To get the most out of the Review sessions, you MUST do some work in advance
(i.e. read the lecture notes or watch the lecture videos, and try the relevant
exercises). I will send out an email prompt each week to remind you of what you
should do in advance of the next review session. If you have queries about the
material, then please post them to the discussion boards. I plan to use the
discussion boards (see below) to help to identify areas you would particularly like
me to focus on in the next session; this only works if you participate, so please do!
• Tutorials: Each week there will also be a 50-minute Tutorial Session (Blackboard
-> AS1: Live sessions), where you can practice and discuss the ideas you have
met. The exact format of the tutorials may vary from week to week2 , but will most
likely include:
1
In particular, I may change the format if you tell me that something is not working well.
2
Again, I may change the format of these if you tell me that something is not working well.
c University of Manchester Algebraic Structures 1 Study Guide, Page 3

– Structured exercises, to be worked on in small groups or individually

– Presentation of some solutions and/or recap of key points.

To get the most out of the Tutorial sessions, you MUST do some work in advance
(i.e. quizzes and exercises).

Communication: Discussion Boards, Office hours, Email

Opportunities for further support and discussion are as follows:

• Discussion Boards: I have set up a discussion board ( Blackboard -> AS1:

Communication) for the course on Blackboard. If you have a question about ANY
aspect of the course, please consider asking it via the appropriate thread of the
discussion board: if you are unsure about something, others may be wondering
exactly the same thing.
I will aim to respond to all (reasonable!) questions posted there within 3 working
days, either via the board or during the review sessions. The boards will also
provide a way for you to communicate with each other outside of tutorials. Putting
your own ideas (or even partial ideas) about a mathematical problem into words is
a really important skill to develop. Moreover, you might find that by putting your
heads together, you are able to make greater progress on a problem.
Collaboration and discussion of non-assessed work is encouraged; work
submitted for assessment must be your own.
There are three common sense ground rules for the discussion boards:

(i) Check if your question has been asked/answered elsewhere before posting;
(ii) Be respectful of others in your posts;
(iii) Do not post off-topic (i.e. not relating to the content of this course) material.

• Office Hours: I will hold virtual office hours via Zoom: Mondays 10:00-12:00
and Thursdays 16:00-18:00 each week3 . The password for the session can be
found here: Blackboard -> AS1: Communication. Please do not circulate the
password. This time can also be used to ask questions about the course.

• Email: You can contact me via: [email protected].

When sending e-mail messages to staff, please:

(i) Use your university e-mail account;

(ii) Include a meaningful subject line (e.g. “MATH20201: A question about
conjugacy”);
3
Just in case it becomes necessary to change the time of my office hour for some reason, I will post an
update to Blackboard -> AS1: Communication.
 c University of Manchester Algebraic Structures 1 Study Guide, Page 4

(iii) Use an appropriate form of address, and remember to sign-off in the way that
you prefer to be addressed – for reference my pronouns are she/her/hers, my
title is Dr and you are all welcome to call me Marianne, or Dr Johnson if you
prefer to be formal. (But please don’t use any of
Miss/Ms/Mrs/Madam/Maam/Professor Johnson – thanks!);
(iv) Check whether your question has already been asked/answered on
the discussion boards, and if not consider asking it there.

The reason for points (i) and (ii) is that messages with empty subject lines, or sent
from external e-mail addresses are sometimes blocked by spam filters. How you
prefer to be addressed may not be obvious from your email address, so don’t be shy
to tell me or to correct me.

Assessment and Feedback

The course will be assessed against the Intended Learning Outcomes (ILOs), which state
that after successful completion of this course you will be able to:

ILO1: state the group axioms and identify frequently met examples of groups,

ILO2: define basic concepts in group theory, such as subgroups, conjugacy classes, cyclic
groups, cosets, and factor groups,

ILO3: employ the subgroup criterion to determine whether certain subsets of a group are
subgroups,

ILO4: describe fundamental properties of cosets and factor groups,

ILO5: identify the generators and subgroups of cyclic groups,

ILO6: determine conjugacy classes, cosets and factor groups in certain groups,

ILO7: state, prove and apply Lagrange’s theorem,

ILO8: state and apply the First Isomorphism Theorem for groups.

There are two main components of assessment.

In-semester assessment, 20%: A timed online test in Week 7 , testing ILO1-ILO3.

More information about this will be made available at Blackboard-> AS1: Assessment
& Feedback.
Solutions and feedback to the test will also be posted on Blackboard, after the
assessment has taken place.
c University of Manchester Algebraic Structures 1 Study Guide, Page 5

End-of-semester assessment, 80%: The remaining assessment (testing ILO1-ILO7)

will take place in January (date to be announced by the exams office later in the
semester) More details of the assessment will be posted at: Blackboard-> AS1:
Assessment & Feedback in due course. Feedback on the exam will also be posted to
Blackboard after the assessment has taken place.
You can also get feedback on your progress in the following ways (there is no grade
associated to any of the following activities, but you may find them useful in preparing
for your assessments):

• Completing the weekly multiple choice quizzes

• Participating in tutorials

• Posting items for discussion on the discussion boards

• Asking questions during office hours

Giving feedback on the course

I would be very grateful if you can tell me if something is not working. Please get in
touch via email (at any time) if:

• you have technical difficulties accessing materials;

• you have suggestions on how the synchronous activities (review sessions and
tutorials) could be improved;

• you find any typos in the written materials or verbal slips in the videos.

I will look into any issues reported and respond to these.

There will also be three feedback surveys. The first of these (in week 2/3 provides an
early opportunity to let me know which aspects of the course you think are working well,
and to highlight any problems you have encountered with the course (of course, if there
are problems, please don’t wait until week 3 to tell me!). The second of these (in week
10) provides an informal opportunity for you to give feedback on the resources provided,
and to let me know if there are areas where you feel you may need more support before
the end of the course. The third of these provides a formal opportunity for you to
evaluate the course as a whole; this will take place at the end of the course.

Workload: Planning your study time

You can find an overview of the content contained in each week at Blackboard -> AS1
(log-in required), or by viewing the weekly learning plans at the start of each chapter of
the notes. The breakdown of synchronous and asynchrnous material in a typical week is
c University of Manchester Algebraic Structures 1 Study Guide, Page 6

as follows:
Synchronous Asynchronous
Review session 4 videos to watch [∼ 4 × 18 mins, on average] OR 6-7 pages of notes to read
Tutorial session 2 quizzes AND 1 set of exercises to complete

I suggest that you aim to spend around 6 hours in total per week on the activities for this
course, with around:
• 1 hour reviewing material from the previous week in the Review session;
• 2 hours absorbing new material i.e. reading lecture notes OR watching the lecture
videos;
• 3 hours of active study spent working on exercises, quizzes and participating in
tutorials.
The above is intended as a suggestion of how to spend your time, but how much time it
takes to master a particular topic will vary from person to person.

• If you find that struggle to read/watch and understand the quantity of new material
presented each week, then please talk to me.

• If you find that you can easily absorb the material of the lecture notes/videos in
under 2 hours, then direct any remaining time towards the other activities.

• If you find that the number of exercises in a particular week is overwhelming; do as

many as you can, and follow the suggestions in the guidance to help you prioritise.
(Any remaining exercises can be used later for revision.)

• If you run out of things to do, and would like some suggestions for some non-
compulsory activities, then feel free to ask me.

In order to keep on top of your work and to keep focused, you may find it helpful to:
• allocate specific times during the week to work through the asynchronous
materials, in advance of the related synchronous sessions;
• turn off notifications for email, social media etc. during your study time;
• have ‘pen and paper’ ready whilst reading the notes/watching the videos and
during the synchronous sessions, to take notes and jot down any questions you may
have;
• keep your notes organised and all in one place (e.g. in a notebook or folder);
• copy down exercises, to allow you to take screen breaks while working on these;
• talk to each other in the Tutorials and Discussion Boards;
c University of Manchester Algebraic Structures 1 Study Guide, Page 7

• cross off tasks as you complete them to gauge your progress;

The schedule below gives details of when various activities will take place each week.
Note: There are 9 different tutorial groups for this course; you have been allocated to a
specific group, and should attend only the session marked in your personal timetable.
Details of all sessions are given below, along with times I expect to be in my office and
available.
09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00
Monday Review Office Office
Hour Hour
Tuesday TUT TUT TUT
TUT
Wednesday

Thursday TUT TUT TUT Office Office

TUT TUT Hour Hour
TUT
Friday

The table below gives an overview of which topics you should focus on each week.
Week Beginning Synchronous sessions Asynchronous materials
0 20th September - Welcome + Revision of Y1 material
1 27th September Welcome + Revision of Y1 Lectures 1-2: First examples of groups
2 4th October First examples of groups Lectures 3-4: Examples and notation
3 11th October Examples and notation Lectures 5-6: Subgroups
4 18th October Subgroups Lectures 7-8: Cyclic subgroups and order
5 25th October Cyclic subgroups Lectures 9-10: Centralisers and centres
6 1st November Reading week - no classes! No new material
7 8th November Centralisers and centres Lecture 11: Cyclic groups
Test on material in Lectures 1-10
8 15th November Cyclic groups Lectures 12-13: Cosets, Lagrange’s theorem
9 22nd November Cosets, Lagrange’s theorem Lectures 14-15: Homomorphisms
10 29th November Homomorphisms Lectures 16-17: Conjugacy
11 6th December Conjugacy Lectures 18-19: Factor groups
12 13th December Factor groups + Lectures 20: First isomorphism theorem
First isomorphism theorem
17th January January exam period
24th January
Studying remotely and blended learning: Over the past year we have all (students
and staff alike) had to learn how to work in different ways. Please be kind to yourself and
others as you adjust to new ways of working. Remember that others may find certain
aspects easier/more difficult than you, so please be respectful and supportive of each
c University of Manchester Algebraic Structures 1 Study Guide, Page 8

other. If you are feeling overwhelmed, break down what you need to do into smaller
manageable tasks, and try to do a bit each day. Celebrate when you have achieved what
you set out to do, but don’t be too hard on yourself if some days you don’t achieve
everything you had planned; you are learning and it takes time. Remember to take
breaks [including screen breaks!], and talk to others. Seek help if you need it; don’t leave
this until the end of the course.

Week 0: Warm-up exercises

(These questions only require material from Foundations of Pure Mathematics and
Linear Algebra. The solutions can be found here. If you have questions about this
material, please post them here.)
1. For the set S and ∗ as defined below, determine if ∗ is a binary operation on S.
Give reasons if your answer is NO.

i S = N, a ∗ b = ab
ii S = {..., −3, −1, 1, 3, 5, ...} (the odd integers), a ∗ b = ab

a + b if a is even
iii S = Z, a ∗ b =
ab if b is odd
iv S = P (X) (the power set of a non-empty set X), A∗B =A∩B (the
intersection of the subsets A, B ⊆ X)
v S = P (X), A∗B =A∪B
vi S = R+ = {r ∈ R | r > 0} , a ∗ b = ab
     
1 2 3 1 2 3 1 2 3
vii S = , , , ∗ = ◦ (composition of
2 3 1 3 2 1 1 2 3
permutations)
viii S is the set of all permutations of Ω = {1, 2, 3, 4} fixing at least one element of
Ω, ∗ = ◦
ix S = M2 (Z) (the set of all 2 × 2 matrices with entries in Z), and
A ∗ B = At (At denotes the transpose of A)
x S = M2 (Q) (the set of all 2 × 2 matrices with entries in Q), and
A ∗ B = A−1 B −1
xi S = M2 (Z), A∗B =A+B (addition of matrices)

2. Let S be a finite set with |S| = n. How many binary operations are there on S?

3. For the following binary operations ∗ defined on a set S determine whether or not ∗
is associative, commutative, and whether (S, ∗) has identity element.
Give reasons for each of your answers.
c University of Manchester Algebraic Structures 1 Study Guide, Page 9

(i) S = Z, a ∗ b = a − b,
(ii) S = Q, a ∗ b = 12 (a + b),
(iii) S = N, a ∗ b = b,
(iv) S = {2k | k ∈ Z} , the set of even integers, a ∗ b = ab,
(v) S = {a, b}, and ∗ given by the multiplication table ∗ a b ,
a b b
b b a
(vi) S = {u, v}, and ∗ given by the multiplication table ∗ u v
u u v
v v u
(vii) S is the set of all constant functions from N to N, ∗ = ◦ (composition of
functions). (A function f : N → N is called constant if there is c ∈ N such that
f (x) = c for all x ∈ N.)

4. Take all examples from Exercise 1 for which the answer was YES, and determine
for those binary operations whether or not they are associative, commutative, and
whether (S, ∗) has an identity element.
c University of Manchester Algebraic Structures 1 Week 1 Checklist

Resources for Week 1

First examples of groups

(Click here to return to the front page navigation)

This course is all about certain algebraic structures known as groups. The asynchronous
materials for this week focus on the definition of a group, and some basic properties
(uniqueness of the identity element; uniqueness of inverses; cancellation). Concepts will
be demonstrated using examples from settings familiar to you from previous study (we
will see groups of numbers, residues, permutations, matrices). By the end of this week,
you should be able to state the group axioms and identify frequently met examples of
groups (ILO1).
Synchronous Sessions:[2 hours of timetabled study]
 REVIEW: Attend Review Session and Tutorial (Topic: Welcome + Revision of Y1)
Asynchronous Tasks: [4 hours directed private study]

 READ: Lecture 1 [∼ 3 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 1.1: What is a group? [20 mins 37s]
Lecture Video 1.2: Basic properties of groups [11 mins 14s]

 CHECK UNDERSTANDING: Lecture 1 Quiz [Suggested study time = 10 minutes]

 READ: Lecture 2 [∼ 3.5 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 2.1: Groups of numbers [33 mins 16s]
Lecture Video 2.2: Groups of matrices [13 mins 8s]

 CHECK UNDERSTANDING: Lecture 2 Quiz [Suggested study time = 10 minutes]

 PRACTICE AND REFLECT: Do Exercises 5-10.

Take some time to consider which activities you found straightforward, and which
you found more challenging. Post any queries to the discussion board before the
next review session. [Suggested study time = 2 hours]
c University of Manchester Algebraic Structures 1 Week 1, Lecture 1, Page 1

Lecture 1
Groups

1.1 What is a group?

A group is a fundamental algebraic structure that occurs frequently throughout
mathematics. Before giving the formal definition, we start with two quite distinct
examples.
Example 1.1 (The integers with addition). We can add two integers a, b ∈ Z together to
produce a new integer a + b ∈ Z. You know from school that addition of integers has some
nice properties:
1. ∀a, b, c ∈ Z, a + (b + c) = (a + b) + c (addition is associative),

2. ∀a ∈ Z, a + 0 = 0 + a = a (0 is a identity element for addition),

3. ∀a ∈ Z, a + (−a) = (−a) + a = 0 (−a is an inverse to a for addition) .

Moreover, addition of integers has the property that
4. ∀a, b ∈ Z, a + b = b + a (addition is commutative).
Example 1.2 (Permutations of a finite set with function composition). Recall from
Foundations of Pure Mathematics that a permutation of Nn = {1, 2, . . . , n} is a bijective
function f : Nn → Nn . We write Sn to denote the set of all permutations of Nn . A
permutation f ∈ Sn can be written in standard notation as a table
 
1 2 ··· n
f= .
f (1) f (2) · · · f (n)

For example,  
1 2 3
f=
2 3 1
denotes the permutation f ∈ S3 with f (1) = 2, f (2) = 3 and f (3) = 1. The identity
 
1 2 ··· n
idn =
1 2 ··· n

is the permutation that leaves Nn unchanged, i.e. idn (x) = x for all x ∈ Nn . Moreover,
every permutation f ∈ Sn has an inverse permutation f −1 ∈ Sn that satisfies

∀x, y ∈ Nn , f (x) = y ⇐⇒ f −1 (y) = x. (1.1)

c University of Manchester Algebraic Structures 1 Week 1, Lecture 1, Page 2

You can find f −1 by reading the table of f bottom-to-top. Any two permutations f, g ∈ Sn
can be composed to produce another permutation

g ◦ f ∈ Sn , defined by g ◦ f (x) = g(f (x)) ∀x ∈ Nn .

This composition has some nice properties, similar to the addition of integers:

1. ∀f, g, h ∈ Sn , f ◦ (g ◦ h) = (f ◦ g) ◦ h (composition is associative),

2. ∀f ∈ Sn , f ◦ idn = idn ◦f = f (idn is a identity element for composition),

3. ∀f ∈ Sn , f ◦ f −1 = f −1 ◦ f = idn (f −1 is an inverse to f for composition).

All these properties are easy to prove, and you have done so in Foundations of Pure
Mathematics. For example, for the third property, let x ∈ Nn and write y := f (x). Then,
using (1.1),
f −1 ◦ f (x) = f −1 (f (x)) = f −1 (y) = x = idn (x).
Since this holds for all x ∈ Nn , we have shown that f −1 ◦ f = idn . Note that it is not true
in general that f ◦ g = g ◦ f for f, g ∈ Sn .

Recall that a binary operation on a set G is a function ∗ : G × G → G, so ∗ assigns to

every pair (g, h) ∈ G × G an element g ∗ h ∈ G. Note that we have written g ∗ h instead
of ∗(g, h). In both of the above examples, we considered a set G, namely G = Z or
G = Sn , and a binary operation ∗ on G, namely ∗ = + or ∗ = ◦, that has some nice
properties. This leads us to the following definition.

Definition 1.3 (Group axioms). A group is a pair (G, ∗), where G is a non-empty set
and ∗ : G × G → G is a binary operation on G with the following properties:

(G1) ∗ is an associative binary operation, that is:

∀g, h, k ∈ G, g ∗ (h ∗ k) = (g ∗ h) ∗ k,

(G2) there exists a identity element e ∈ G, that is:

∃e ∈ G, ∀g ∈ G, e ∗ g = g ∗ e = g,

(G3) each g ∈ G has an inverse g 0 ∈ G, that is:

∀g ∈ G, ∃g 0 ∈ G, g ∗ g 0 = g 0 ∗ g = e.

Example 1.4 (Examples 1.1 and 1.2 revisited). By Example 1.1, (Z, +) is a group with
identity element e = 0, and n0 = −n is an inverse of n ∈ Z. By Example 1.2, (Sn , ◦) is a
group with identity element e = idn , and f 0 = f −1 is an inverse of f ∈ Sn .
c University of Manchester Algebraic Structures 1 Week 1, Lecture 1, Page 3

1.2 Basic properties of groups

In Example 1.1 it is easy to see that 0 is the only identity element of (Z, +), and that −n
is the only inverse of n ∈ Z. Likewise, in Example 1.2, idn is the only identity element of
(Sn , ◦)) and f −1 is the only inverse of f ∈ Sn ). These facts are not just true for Z or Sn ;
they follow directly from the group axioms (G1), (G2), (G3) and thus hold for all groups.
Lemma 1.5 (Uniqueness of identity element; uniqueness of inverse).
Let (G, ∗) be a group with identity element e.
1. The identity element e is unique. That is, if ẽ is another identity element, then
e = ẽ,
2. Each g ∈ G has a unique inverse g 0 . That is, if g 00 is another inverse of g, then
g 0 = g 00 .
Proof. 1. As e is an identity element and ẽ ∈ G, we have e ∗ ẽ = ẽ by (G2). On the other
hand, ẽ is also an identity element and e ∈ G, so we have e ∗ ẽ = e by (G2) for ẽ. This
shows that e = ẽ.
2. Since g 0 and g 00 are inverses of g, we have g 00 ∗ g = e and g ∗ g 0 = e. Therefore, and also
using (G1) and (G2),

g 0 = e ∗ g 0 = (g 00 ∗ g) ∗ g 0 = g 00 ∗ (g ∗ g 0 ) = g 00 ∗ e = g 00 .

In this course, we study results like Lemma 1.5 that hold for all groups. We will see many
examples of groups, and we study the groups (Sn , ◦) in detail.
Remark 1.6 (The inverse of the identity; the inverse of an inverse). Let (G, ∗) be a
group with identity element e, and denote the (unique) inverse of g ∈ G by g 0 . Here are
some simple observations:
1. The identity element is its own inverse, that is, e0 = e (since e ∗ e = e by (G2)).
2. For all g ∈ G, we have (g 0 )0 = g (since g ∗ g 0 = g 0 ∗ g = e0 is symmetric in g and g 0 ).
Here is another simple fact that follows immediately from the group axioms.
Lemma 1.7 (Cancellation). Let (G, ∗) be a group, and a, b, c ∈ G. Then
1. a∗b=a∗c implies that b = c,
2. b∗a=c∗a implies that b = c.
Proof. 1. Multiplying the equation a ∗ b = a ∗ c on the left by the inverse a0 of a gives

a0 ∗ (a ∗ b) = a0 ∗ (a ∗ c).

By (G1), we can re-bracket on both sides to obtain

(a0 ∗ a) ∗ b = (a0 ∗ a) ∗ c.
c University of Manchester Algebraic Structures 1 Week 1, Lecture 1, Page 4

By (G3) we know that a0 ∗ a = e, giving

e ∗ b = e ∗ c.

By (G2), e is the identity element of G, and hence b = c.

2. The proof of the second part is very similar: try it as an exercise!
c University of Manchester Algebraic Structures 1 Week 1, Lecture 2, Page 1

Lecture 2
Examples and non-examples of groups

In this lecture we look at several examples and non-examples of groups. These examples
will come up frequently throughout the course.

2.1 Groups of numbers

Groups from addition
Example 2.1. We have already seen in the last lecture that (Z, +) is a group. From basic
arithmetic, it is clear that (Q, +), (R, +) and (C, +) are also groups. Indeed, in each case,
addition is associative, 0 is the identity element, and −x is the inverse of any number x.
Example 2.2. The set 2Z of even numbers is a group under addition: the sum of two
even numbers is even, so + is a binary operation on 2Z. Addition is associative, and
0 ∈ 2Z is the identity element. The inverse of x ∈ 2Z is −x ∈ 2Z.
Example 2.3. We can also take G to be Rn , Qn , Cn or, more generally, any vector
space. Then (G, +) is a group. (Exercise: Look back at the vector space axioms from
Linear Algebra. Notice that three of the axioms are the group axioms, written additively.)
Example 2.4. The set 2Z + 1 = {. . . , −3, −1, 1, 3, . . .} of odd is not a group under
addition: the sum of two odd numbers is even, so + is not even a binary operation (let
alone, an associative binary operation) on 2Z + 1.
Example 2.5. The set of natural numbers N = {1, 2, . . .} is not a group under addition.
Although + is an associative binary operation on N, notice that for all n, m ∈ N, we have
n + m > n, so there is no identity element. (Since there is no identity element, there are
also no inverses.)

Groups from multiplication

Example 2.6. Let C∗ = C r {0}, the set of all non-zero complex numbers. Then (C∗ , ×)
is a group:
• × is a binary operation on C∗ (the product of two non-zero complex numbers is
again a non-zero complex number).
• multiplication of numbers is associative, so (G1) holds.
• 1 ∈ C∗ and 1 × z = z × 1 = z for all z ∈ C∗ , so 1 is the identity element and (G2)
holds.
c University of Manchester Algebraic Structures 1 Week 1, Lecture 2, Page 2

• for any z ∈ C∗ , we have 1/z ∈ C∗ and z(1/z) = (1/z)z = 1. Hence, 1/z is the
inverse of z and (G3) holds.
Example 2.7. With Q∗ = Q r {0}, R∗ = R r {0}, and R+ = {x ∈ R : x > 0}, we have
the following further examples of groups of numbers under multiplication: (Q∗ , ×),
(R∗ , ×), but also (R+ , ×) and ({1, −1}, ×).
Example 2.8. The full set of complex numbers with multiplication, (C, ×), does not
form a group, because 0 ∈ C does not have an inverse: there is no z ∈ C such that
z × 0 = 1.
Example 2.9. The natural numbers with multiplication, (N, ×), are not a group: there
is an identity element 1 ∈ N, but 2, 3, 4, . . . do not have inverses.

Groups from modular arithmetic

Write Zn = {0, 1, . . . , n − 1}. Recall that addition modulo n is the binary operation ⊕
defined as follows: for a, b ∈ Zn , we have a ⊕ b = r, where r ∈ Zn is the remainder of a + b
divided by n, that is, a + b = q × n + r, where q ∈ Z and r ∈ Zn . Notice that
a ⊕ b = b ⊕ a for all a, b ∈ Zn .
Lemma 2.10 ((Zn , ⊕) is a group). Let n ∈ N. Then (Zn , ⊕) is a group. The identity
element is 0 ∈ Zn , and the inverse of a ∈ Zn is given by
(
n−a if a 6= 0,
a0 =
0 if a = 0.

Proof. Since a ⊕ b = r ∈ Zn for all a, b ∈ Zn , we see that ⊕ is a binary operation on Zn .

(G2): For all a ∈ Zn , we have a + 0 = 0 + a = a = 0 × n + a. Hence, 0 ⊕ a = a ⊕ 0 = a
and 0 is indeed an identity element.
(G3): If a 6= 0, then n − a ∈ Zn . We have a + (n − a) = n = 1 × n + 0, so
a ⊕ (n − a) = (n − a) ⊕ a = 0 and a0 = n − a is indeed inverse to a when a 6= 0.
Since 0 + 0 = 0 × n + 0, we have 0 ⊕ 0 = 0, and hence 00 = 0.
(G1): For associativity, we show that both a ⊕ (b ⊕ c) and (a ⊕ b) ⊕ c are the remainder
of a + b + c divided by n. Indeed, using long division write b + c = q1 n + r1 and
a + r1 = q2 n + r2 with q1 , q2 ∈ Z and r1 , r2 ∈ Zn . Then
a ⊕ (b ⊕ c) = a ⊕ r1 = r2 .
On the other hand, we have a + b + c = a + q1 n + r1 = q1 n + (a + r1 ) = (q1 + q2 )n + r2 , so
r2 is indeed the remainder of a + b + c divided by n. Analogously, write a + b = q3 n + r3
and r3 + c = q4 n + r4 with q3 , q4 ∈ Z and r3 , r4 ∈ Zn . Then
(a ⊕ b) ⊕ c = r3 ⊕ c = r4 ,
and a + b + c = q3 n + r3 + c = (q3 + q4 )n + r4 , so r4 is also the remainder of a + b + c
divided by n. Hence, r2 = r4 and ⊕ is indeed associative.
c University of Manchester Algebraic Structures 1 Week 1, Lecture 2, Page 3

Remark 2.11 ((Zn , ⊗) is not a group). Similarly, one can define multiplication modulo
n by a ⊗ b = r, where r ∈ Zn is the remainder of ab divided by n, that is, ab = qn + r for
q ∈ Z and r ∈ Zn . This is also an associative binary operation on Zn , and Zn contains an
identity element (namely, 0 if n = 1 and 1 if n ≥ 2). However, (Zn , ⊗) is not a group for
n ≥ 2, as 0 does not have an inverse: 0 ⊗ a = 1 would mean that 0 = 0 × a = q × n + 1
for some q ∈ Z, which is impossible for n ≥ 2.
Example 2.12 ((Zn \ {0}, ⊗) is a group if and only if n is prime). As in previous
examples, one can ask whether (Zn r {0}, ⊗) is a group. The answer here is slightly more
intricate: (Zn r {0}, ⊗) is a group if and only if n is prime.
(Exercise: If n = ab where a, b are positive integers with a 6= 1 6= b, show that ⊗ is not a
binary operation on Zn r {0}.)

2.2 Groups of matrices

You know from Linear Algebra that multiplication of matrices is associative, i.e.
A × (B × C) = (A × B) × C for any matrices A, B, C of the correct dimensions so that
we can multiply them.
For n ∈ N, write GL(n, R) for the set of all invertible (n × n)-matrices with entries in R:
GL(n, R) = {A ∈ Mn (R) | det(A) 6= 0}.
Example 2.13 (The group of all real invertible matrices). (GL(n, R), ×) is a group
under matrix multiplication. Indeed, the product of two invertible (n × n)-matrices is
again invertible, so we have a binary operation on GL(n, R). As recalled above, matrix
multiplication is associative. The (n × n)-identity matrix
 
1 0 0 ··· 0
0 1 0 · · · 0
 
0 0 1 · · · 0
In =  
 .. .. .. . . .. 
. . . . .
0 0 0 ··· 1
has the property that In × A = A × In = A for all (n × n)-matrices A, so it is the identity
element. Finally, each A ∈ GL(n, R) has an inverse by definition.
Definition 2.14 (The general linear group). The group (GL(n, R), ×) is called the
General Linear Group of degree n over R.
Example 2.15 (Invertible matrices over other fields). Similarly, we have the general
linear groups
(GL(n, Q), ×) and (GL(n, C), ×).
More generally, for every field K, we have the general linear group (GL(n, K), ×). We
shall often write simply GL(n, K), where the operation of matrix multiplication is
understood, but not forgotten. For example, GL(n, Zp ) is the group of invertible matrices
with entries in the field Zp with p elements, under the operation of matrix multiplication.
c University of Manchester Algebraic Structures 1 Week 1, Lecture 2, Page 4

Remark 2.16. The set Mn (R) of all n × n-matrices over R is not a group under matrix
multiplication. Notice that matrix multiplication is an associative binary operation on this
set, and In is an identity element contained in this set. However, Mn (R) contains
matrices which are not invertible. For example: the matrix whose entries are all 0.
However, Mn (R) is a group under matrix addition. Indeed, you may remember from
Linear Algebra that this set is a vector space over R. Alternatively, check that the group
axioms hold in this case.

‘Additive’ and ‘multiplicative’ notation

Remark 2.17. There are two commonly used notations for inverses.

1. Some symbols for binary operations, such as ∗, ×, ◦, ·, suggest that the group is
written “multiplicatively”. For such multiplicative groups, we write g −1 instead of g 0
for the inverse of g (as we did in Example 1.2).
When a group (G, ∗) is written in multiplicative notation, we follow the same
convention as for usual multiplication and omit the operation: we write gh instead
of g ∗ h.

2. Other symbols, such as +, ⊕, suggest “additive notation”. For a group in additive

notation, we write −g instead of g 0 for the inverse of g (as we did in Example 1.1).

We often write simply G instead of (G, ∗) to denote a group. This is just for ease of
notation; the operation ∗ is an important aspect of a group and should not be forgotten
even if it is not written explicitly.
c University of Manchester Algebraic Structures 1 Week 1, Exercises, Page 1

Week 1: Exercises

Guidance: This week I would like you to practice determining whether a given set G
with a binary operation ∗ is a group. Questions 5-8 are all of this kind; do as many of
these exercises as you need to become comfortable with this procedure. Question 9 tests
your understanding of one of our favourite examples - the general linear group. Question
10 tests your understanding of inverses.

Groups
(requires material up to Lecture 1)

5. Which of the following sets are groups with respect to the binary operation given?
In each case, give detailed reasons for your answer.
 
a a
(i) The set of all 2 × 2 matrices of the form where a ∈ R, a 6= 0; under
0 0
matrix multiplication.
 
a b
(ii) The set of all 2 × 2 matrices of the form where a, b ∈ R, a, b 6= 0;
0 0
under matrix multiplication.
(iii) The set Z × Z = {(a, b); a, b ∈ Z} with multiplication

(a, b)(n, m) = (a + n, b + m) for all (a, b), (n, m) ∈ Z × Z.

(iv) The set Z × Z with multiplication

(a, b)(n, m) = (a + n, (−1)n b + m) for all (a, b), (n, m) ∈ Z × Z.

(v) The power set P (X) of a non empty set X under the binary operation of
intersection of sets.
(vi) P (X) as in (v) with the associative binary operation

AB = (A ∪ B) \ (A ∩ B), for allA, B ⊆ X.

(You may take for granted that this operation is associative!)

6. For the set G and the binary operation ∗ on G as given below, determine if (G, ∗)
is a group.
If your answer is NO, explain which axioms fail to hold.
c University of Manchester Algebraic Structures 1 Week 1, Exercises, Page 2

(i) G = Q\ {0} , a ∗ b = ab (multiplication of numbers),

(ii) G = {1, −1} , a ∗ b = ab (multiplication of numbers),
(iii) G = Z, a ∗ b = ab (multiplication of numbers),
(iv) G = {f, g} , where f and g are functions from Z to Z defined by
f (n) = n, g(n) = −n ∀n ∈ Z, with ∗ = ◦ (composition of functions),
   
1 0 0 1
(v) G = , , ∗ = matrix multiplication,
0 1 1 0
   
1 2 3 1 2 3
(vi) G = , , ∗ = ◦ (composition of permutations),
1 2 3 3 2 1
   
1 2 3 1 2 3
(vii) G = , , ∗ = ◦ (composition of permutations),
1 2 3 2 3 1
(viii) G = {0, 1} , a ∗ b = ab (multiplication of numbers),
(ix) G is the set of odd integers, a ∗ b = ab (multiplication of numbers).

7. Prove that R × R \ {(0, 0)} is a group under the binary operation defined by

(a, b)(c, d) = (ac − bd, bc + ad), for all (a, b), (c, d) ∈ R × R \ {(0, 0)}.

8. Let G = R\ {1} , and a ∗ b = a + b − ab. Prove that (G, ∗) is a group.

9. Make a list of all elements of the group GL(2, Z2 ).

10. Let G be a group and g, h ∈ G. Show that (gh)−1 = h−1 g −1 .

Solutions to the exercises can be found here.

c University of Manchester Algebraic Structures 1 Week 2 Checklist

Resources for Week 2

Examples and notation

(Click here to return to the front page navigation)

This week we will look at a particularly interesting family of examples; the symmetric
groups. We will also consider some useful notation for taking ‘powers’ of a fixed element
(more on this idea later), and look at the idea of a multiplication table of a finite group.
By the end of this week you should be able to perform routine computations in the
symmetric groups (ILO1) and define some basic concepts in group theory, such as
abelian groups, multiplication tables etc. (ILO2).
Synchronous Sessions:[2 hours of timetabled study]

 REVIEW: Attend Review Session and Tutorial (Topic: Week 1 - Groups)

Asynchronous Tasks: [4 hours of directed private study]

 READ: Lecture 3 [4 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 3.1: The symmetric group (Sn , ◦) [8 mins 45s]
Lecture Video 3.2: Notations for permutations [20 mins 41s]

 CHECK UNDERSTANDING: Lecture 3 Quiz [Suggested study time = 10 minutes]

 READ: Lecture 4 [∼ 3.5 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 4.1: Abelian groups, Multiplication tables [13 mins 43s]
Lecture Video 4.2: Powers of group elements [14 mins 19s]

 CHECK UNDERSTANDING: Lecture 4 Quiz [Suggested study time = 10 minutes]

 PRACTICE AND REFLECT: Do Exercises 11-21.

Take some time to consider which activities you found straightforward, and which
you found more challenging. Post any queries to the discussion board before the
next review session. [Suggested study time = 1 21 hours]
c University of Manchester Algebraic Structures 1 Week 2, Lecture 3, Page 1

Lecture 3
The symmetric groups

3.1 The symmetric group (Sn , ◦)

Let n ∈ N and recall that by Sn we denote the set of all permutations of
Nn = {1, 2, . . . , n}, that is, the set of all bijective (i.e. 1-1 and onto) functions from Nn to
itself. We have already seen that (Sn , ◦) is a group, where ◦ is the composition of
functions defined by g ◦ f (x) = g(f (x)).
Indeed, the composite of two bijective functions is again bijective, so ◦ is a binary
operation on Sn . Composition is associative, as

f ◦ (g ◦ h)(x) = f (g(h(x))) = (f ◦ g) ◦ h(x).

The identity map  

1 2 3 ··· n
idn =
1 2 3 ··· n
is the identity element, and the inverse of an arbitrary permutation
 
1 2 ··· n
σ=
f (1) f (2) · · · f (n)

can be found by reading the table bottom-to-top.

Definition 3.1 (The symmetric groups). The group (Sn , ◦) is called the symmetric group
of degree n.

We know from Foundations of Pure Mathematics that Sn has n! elements.

Example 3.2 (The symmetric groups of degree 1, 2, 3). For n = 1, we have

 
1
S1 = .
1

For n = 2, we have    
1 2 1 2
S2 = , .
1 2 2 1
For n = 3, we have
           
1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
S3 = , , , , , .
1 2 3 1 3 2 3 2 1 2 1 3 2 3 1 3 1 2
c University of Manchester Algebraic Structures 1 Week 2, Lecture 3, Page 2

Example 3.3 (Computations in S3 ). In (S3 , ◦), let

   
1 2 3 1 2 3
f= , g= .
2 1 3 2 3 1
Then we have    
1 2 3 −1 1 2 3
g◦f = and f = .
3 2 1 2 1 3
We have found g ◦ f by computing
g ◦ f (1) = g(f (1)) = g(2) = 3,
g ◦ f (2) = g(f (2)) = g(1) = 2,
g ◦ f (3) = g(f (3)) = g(3) = 1.
Similarly, we have found f −1 by computing
f −1 (1) = (the element that f maps to 1) = 2,
f −1 (2) = (the element that f maps to 2) = 1,
f −1 (3) = (the element that f maps to 3) = 3.
A warning about composition of permutations:
Remark 3.4. Remember that elements of Sn are bijective functions and ◦ is composition
of functions. If g, f ∈ Sn , then g ◦ f denotes the function Nn → Nn defined by
g ◦ f (x) = g(f (x)). Notice that f is performed before g here, not the other way around.
When composing functions, one should read from right to left. This notation for function
composition is widely used in the other areas of mathematics you are studying (e.g.
calculus, analysis).
[Aside: In some textbooks on algebra functions are written with their arguments on the
left (that is, writing (x)f , rather than f (x)). This has the neat advantage that
composition of functions can be read from left to right (because what we usually call g ◦ f
would instead be called f ◦ g)! Make sure that you check which convention is used in any
textbooks that you consult, as f ◦ g is not equal to g ◦ f in general.]

3.2 Notations for permutations

The standard two-line notation for permutations can be quite cumbersome and is not
very informative. Cycle notation is more concise and offers more information at a glance.
Definition 3.5 (Cycles and transpositions). Let α1 , . . . , αr ∈ Nn be r distinct elements.
Then the permutation f ∈ Sn defined by
f (α1 ) = α2 , f (α2 ) = α3 , . . . , f (αr−1 ) = αr , f (αr ) = α1 , and
f (α) = α for α ∈ Nn r {α1 , . . . , αr }
is called a cycle of length r, or an r-cycle. We denote it by f = (α1 · · · αr ).
A 2-cycle (α1 α2 ) is called a transposition.
c University of Manchester Algebraic Structures 1 Week 2, Lecture 3, Page 3

According to the definition, the r-cycle (α1 · · · αr ) maps each αi to αi+1 , except for αr ,
which is mapped to α1 . All other elements of Nn are left fixed.
Example 3.6.
1. Every 1-cycle (α) is the identity idn .
(Indeed, from the definition, α is mapped to α, and all other elements are fixed.
Thus all elements are fixed.)
2. In (S3 , ◦), we have
   
1 2 3 1 2 3
(1 3) = and (1 3 2) = .
3 2 1 3 1 2

Definition 3.7 (Disjoint cycles). Cycles (α1 · · · αr ) and (β1 · · · βs ) are said to be disjoint,
if
{α1 , . . . , αr } ∩ {β1 , . . . , βs } = ∅.
Example 3.8. The cycles (1 2 4) and (3 5) in S5 are disjoint, the cycles (1 2 3) and
(3 4 5) are not.
Lemma 3.9 (Disjoint cycles commute). Let (α1 · · · αr ) and (β1 · · · βs ) be two disjoint
cycles in Sn . Then

(α1 · · · αr ) ◦ (β1 · · · βs ) = (β1 · · · βs ) ◦ (α1 · · · αr ).

In other words, disjoint cycles commute.

Proof. Write f = (α1 · · · αr ) and g = (β1 · · · βs ). We must show that f ◦ g = g ◦ f , that is,
for any x ∈ Nn , f (g(x)) = g(f (x)). There are three cases:
Case 1: If x ∈ {α1 , . . . , αr }. Then also f (x) ∈ {α1 , . . . , αr }. Since f and g are disjoint,
x, f (x) ∈
/ {β1 , . . . , βs }, and thus g(x) = x, g(f (x)) = f (x). Putting these facts
together, we get f (g(x)) = f (x) = g(f (x)), as desired.
Case 2: If x ∈ {β1 , . . . , βs }. This is similar to the first case. [Exercise: write out the
details.]
Case 3: Otherwise, x ∈ Nn r ({α1 , . . . , αr } ∪ {β1 , . . . , βs }). In this case,
f (x) = g(x) = x, and thus f (g(x)) = x = g(f (x)).

We will frequently use the following fact, which you know from Foundations of Pure
Mathematics.
Theorem 3.10 (Every permutation is a product of pairwise disjoint cycles).
If f is a permutation in Sn , then there exist t ≥ 1 and g1 , . . . , gt ∈ Sn such that
f = g1 ◦ · · · ◦ gt and g1 , . . . , gt are pairwise disjoint cycles.
c University of Manchester Algebraic Structures 1 Week 2, Lecture 3, Page 4

Remark 3.11.
1. The slogan of the previous result is “Every permutation can be written as a product
of pairwise disjoint cycles”. Here “product” means function composition ◦, and we
allow for products of length 1 (since f could be a cycle itself ).
2. All elements of Nn that are fixed by f can be included in this product as 1-cycles,
e.g. (1 3)(2) and (1 3) to denote the same element of S3 .
We will often omit the 1-cycles in practice.
3. The decomposition of f as a product of disjoint cycles is unique up to the inclusion
(or not) of 1-cycles and the order in which the cycles are written.
4. When all the 1-cycles are included, then clearly the lengths of the cycles involved in
the decomposition of f ∈ Sn will sum up to n, as every j ∈ Nn appears in exactly
one cycle.
Example 3.12 (Writing a permutation  as a product of disjoint cycles).
1 2 3 4 5 6
Let us write the permutation f = as a product of disjoint cycles.
3 6 2 5 4 1
First, let us try to find the cycle containing 1. Starting at 1, by applying f repeatedly, we
find: f (1) = 3, f (3) = 2, f (2) = 6, and f (6) = 1. Thus one of the cycles in our
decomposition will be (1 3 2 6).
The smallest element of N6 not yet considered is 4. By applying f repeatedly we find
f (4) = 5 and f (5) = 4, giving the cycle (4 5). Composing these two cycles, we obtain
f = (1 3 2 6)(4 5) = (4 5)(1 3 2 6).
The procedure of the above example can be turned into a proof of Theorem 3.10.
Proof of Theorem 3.10. Pick any element α1 ∈ {1, . . . , n}. Consider the sequence
αi+1 = f (αi ) for i = 1, 2, . . .. Because Nn is finite, it is clear that this sequence must
contain repetitions. Suppose then that the first r terms are distinct, and that
αr+1 ∈ {α1 , . . . , αr }.
We claim that αr+1 = α1 . Suppose for contradiction that αr+1 = αj for some
j ∈ {2, . . . , r}. Then f (αr ) = αr+1 = αj = f (αj−1 ). Since f is injective, this implies
αr = αj−1 , contradicting that the first r terms are distinct. So we must have αr+1 = α1 .
Considering the r-cycle g1 = (α1 · · · αr ), we have shown that f (x) = g1 (x) for all
x ∈ {α1 , . . . , αr }.
If {α1 , . . . , αr } =
6 Nn , then choose β1 ∈ Nn r {α1 , . . . , αr } and consider the sequence
βi+1 = f (βi ) for i = 1, 2, . . .. Note that βi ∈ / {α1 , . . . , αr } for all i, as f is injective. By the
same argument as before, we can compute a cycle g2 = (β1 · · · βs ) with f (x) = g2 (x) for
x ∈ {β1 , . . . , βs }.
We can iterate this procedure, and since Nn is finite, the process must terminate, yielding
disjoint cycles g1 , g2 , g3 , . . . , gt , such that f (x) = gi (x) for all x that appear in gi .
Therefore, f = g1 ◦ · · · ◦ gt .
c University of Manchester Algebraic Structures 1 Week 2, Lecture 4, Page 1

Lecture 4
Abelian groups, multiplication tables, powers

4.1 Abelian groups, Multiplication tables

We give a name to groups satisfying the fourth property of (Z, +) from Example 1.1.

Definition 4.1 (Abelian group). A group (G, ∗) is called abelian1 (or commutative), if
∗ is commutative, i.e.
∀g, h ∈ G, g ∗ h = h ∗ g.

Example 4.2.

• All of our groups of numbers (Z, +), (Q, +), (R, +), (C, +), (Q∗ , ×), (R∗ , ×),
(C∗ , ×), (R>0 , ×), ({−1, 1}, ×) are abelian.

• The residue groups (Zn , ⊕) are abelian.

• The symmetric group (Sn , ◦) is not abelian for n ≥ 3, since

(1 2) ◦ (2 3) 6= (2 3) ◦ (1 2).

Indeed, (1 2) ◦ (2 3) = (1 2 3), but (2 3) ◦ (1 2) = (1 3 2).

(How did we compute this? Element by element! For example,
(1 2) ◦ (2 3)(1) = (1 2)(1) = 2, (1 2) ◦ (2 3)(2) = (1 2)(3) = 3, and
(1 2) ◦ (2 3)(3) = (1 2)(2) = 1, and thus (1 2) ◦ (2 3) = (1 2 3).)

• The general linear group GL(n, R) is not abelian for n ≥ 2. For example,
     
0 1 1 1 1 1 0 1
6= .
1 0 0 1 0 1 1 0

Indeed,    
0 1 1 1
LHS = RHS = .
1 1 1 0
1
after Niels Henrik Abel (1802–1829)
c University of Manchester Algebraic Structures 1 Week 2, Lecture 4, Page 2

Multiplication tables
Let G = {g1 , g2 , . . . , gn } be a finite group with n distinct elements g1 , . . . , gn . We can
write down its multiplication table, which completely determines the group operation ∗:
∗ g1 g2 ··· gn
g1 g1 ∗ g1 g1 ∗ g2 ··· g1 ∗ gn
g2 g2 ∗ g1 g2 ∗ g2 ··· g2 ∗ gn
.. .. .. ... ..
. . . .
gn gn ∗ g1 gn ∗ g2 ··· gn ∗ gn

Example 4.3. The multiplication table of the group (Z3 , ⊕) is

⊕ 0 1 2
0 0 1 2
1 1 2 0
2 2 0 1

Multiplication tables of groups have an important property.

Lemma 4.4 (Group multiplication table has no repeats in a row (or a column)).
Let G be a finite group. Then in any row (or column) of the multiplication table of G,
each element of G appears exactly once.

Proof. Write G = {g1 , g2 , . . . , gn } with distinct elements g1 , . . . , gn . The i-th row of the
multiplication table is
gi g1 , gi g2 , . . . , gi gn . (∗)
All these elements are distinct: indeed, if gi gj = gi gk for j 6= k, then Lemma 1.7
(cancellation) implies that gj = gk , contradicting the assumption that the n elements
g1 , . . . , gn are distinct. Hence the n elements in (∗) are distinct. This can only happen if
each of the n elements of G appears exactly once. The proof for columns is analogous.

4.2 Powers of group elements

Remark 4.5. Recall that when a group (G, ∗) is written in multiplicative notation, we
write gh instead of g ∗ h. By (G1), we can write ghk to denote both of the products g(hk)
and (gh)k. Using this and induction, one can also see that products of multiple elements
g1 g2 · · · gn are independent of how they are bracketed and thus well defined.

Definition 4.6 (Positive powers of a group element). For g ∈ G and n ∈ N, we define

g n = gg · · · g .
| {z }
n times

Lemma 4.7 (Properties of powers). Let G be a group, g ∈ G, and m, n ∈ N. Then

c University of Manchester Algebraic Structures 1 Week 2, Lecture 4, Page 3

1. g m g n = g m+n
2. (g m )n = g mn .
Proof. 1. By definition: g m g n = (g · · · g )(g · · · g ) = g · · · g = g m+n .
| {z } | {z } | {z }
m times n times m+n times
2. By definition: (g m )n = (g · · · g )n = (g · · · g )(g · · · g ) · · · (g · · · g ) = g · · · g = g mn .
| {z } | {z } | {z } | {z } | {z }
m times m times m times m times mn times
| {z }
n times

Definition 4.8 (Integer powers of a group element). We extend this notation to integer
powers: for g ∈ G and n ∈ Z, we write
 n
 g , for n > 0,
n
g = e, for n = 0,
 −1 |n|
(g ) , for n < 0.
The rules of Lemma 4.7 remain valid for integer powers. That is, for any g ∈ G and any
m, n ∈ Z, we have
g m g n = g m+n and (g m )n = g mn .
Example 4.9. For any group G and g ∈ G, we have
g 4 g −2 = ggggg −1 g −1
= ggg gg −1 g −1
|{z}
=e
−1
= gg gg
|{z}
=e
= gg
= g 2 = g 4−2
Example 4.10. Take G = GL(2, R) and
   
1 1 1 1
g= , h= .
0 1 −1 1
Then

       
3 1 1
1 1 1 1 1 1 1 2 1 3
g = = = ,
0 1
0 1 0 1 0 1 0 1 0 1
    
−2 −1 −1 1 −1 1 −1 1 −2
g =g g = = .
0 1 0 1 0 1
Moreover,
    
2 1 1 1 1 0 2
h = =
−1 1 −1 1 −2 0
    
−2 1/2 −1/2 1/2 −1/2 0 −1/2
h = = .
1/2 1/2 1/2 1/2 1/2 0
c University of Manchester Algebraic Structures 1 Week 2, Lecture 4, Page 4

Remark 4.11 (Powers in additive notation). For groups (G, +) in additive notation,
such as (Z, +) or (Zn , ⊕), using the notation g n to denote g + · · · + g seems unnatural!
| {z }
n times
Thus for additively written groups (G, +), if g ∈ G and n ∈ N, we write

ng = g + · · · + g ,
| {z }
n times

which is extended to integer multiples by


 ng, for n > 0,
ng = e, for n = 0,
|n|(−g), for n < 0.


For all m, n ∈ Z and g ∈ G, we then have the identities

ng + mg = (n + m)g and n(mg) = (nm)g.

Note that these are not new results, they are just different, more intuitive ways of writing
our earlier results for powers, when the operation is additive.
c University of Manchester Algebraic Structures 1 Week 2, Exercises, Page 1

Week 2: Exercises

Guidance: This week I would like you to practice (a) working with permutations (see
exercises 11-18) and (b) determining multiplication tables (see exercises 19-21).
Questions 11-17 are very straightforward, testing your knowledge of the various notations
used for permutation. Do as many of these exercises as you need to become comfortable
with the notation. Question 18 is more interesting: make sure to try that one! Questions
19-21 allow you to practice working with some familiar associative binary operations.
The symmetric group
(requires material up to Lecture 3.)
   
1 2 3 4 5 1 2 3 4 5
11. Let f = , g= .
5 4 1 2 3 2 1 5 4 3
(i) Work out g ◦ f, f ◦ g, g ◦ g and f ◦ f (in standard notation).
(ii) Write f, g, g ◦ f, f ◦ g, g ◦ g and f ◦ f as a composite of disjoint cycles.

12. Write the following permutations of N5 = {1, 2, 3, 4, 5} in standard notation.

(i) (1 3 4) ◦ (2 5) ,
(ii) (1 3 4) ◦ (1 3 4) ◦ (1 3 4) ,
(iii) (2 3 1) ◦ (3 2 1) ,
(iv) (1 3 2 4) ◦ (5 3 1) ◦ (2 4) .

13. Which of the following equations are true?

(i) (1 2 3 4 5) = (3 4 5 1 2) ,
(ii) (1 2 3 4 5) = (2 3 4 1 5) ,
(iii) (5 4 3 2 1) = (3 2 1 5 4) ,
(iv) (5 4 3 2 1) = (1 5 4 2 3) .

14. Write  
1 2 3 4 5 6 7 8 9
f=
4 9 8 5 6 1 3 7 2
and  
1 2 3 4 5 6 7 8 9
g=
5 9 1 3 4 8 7 2 6
as composites of disjoint cycles.
c University of Manchester Algebraic Structures 1 Week 2, Exercises, Page 2

15. Write the following permutations as composites of disjoint cycles:

(i) (1 2 3 4) ◦ (1 3) ◦ (2 4) ◦ (1 4 3 2)
(ii) (1 2 3 4 5) ◦ (1 3 4 2) ◦ (1 5 4 3 2)
(iii) (1 5) ◦ (1 4) ◦ (1 3) ◦ (1 2)

16. Write the following permutations in S9 as composites of disjoint cycles, including

1-cycles for fixed elements:

(a) (1 2 3 4) ◦ (3 2 5 1)−1 ◦ (2 7 8),

(b) (1 2 3) ◦ (2 3 4 5 6) ◦ (7 8 9) ◦ (1 2 3)−1 .

17. Using cycle notation, make a list of all elements of the symmetric group S4 .

18. Prove that every non-trivial permutation of Nn = {1, 2, 3, ..., n} can be written as a
composite of less than n transpositions.

Multiplication tables
(Requires material up to Lecture 4.)

19. Let G = {e, a, b, c}, where

       
1 0 −1 0 0 −1 0 1
e= , a= , b= , c= ,
0 1 0 −1 1 0 −1 0

and let ∗ be matrix multiplication. Write out the multiplication table for (G, ∗) and
prove that it is a group.

20. Work out the multiplication tables for (Z4 , ⊕) and (Z4 , ⊗).

21. Let G = {e, a, b, c} be the set of permutations of N4 = {1, 2, 3, 4} where

e = id4 , a = (1 2), b = (3 4), c = (1 2) ◦ (3 4)

and let ∗ = ◦ (composition of permutations). Write out the multiplication table

for (G, ∗) and prove that it is a group.

Solutions to the exercises can be found here.

c University of Manchester Algebraic Structures 1 Week 3 Checklist

Resources for Week 3

Subgroups

(Click here to return to the front page navigation)

This week we will consider the question of which subsets of a group G turn out to be a
group themselves, with respect to the group operation of G. Such subsets are called
subgroups of G and we shall see some concrete examples of these, paying special attention
to particularly interesting and useful subgroups of invertible matrices. We shall show
that the intersection of two subgroups of a given group G is again a subgroup of G,
whilst the union of two subgroups need not be. By the end of this week you should be
able to explain the concept of a subgroup (ILO2) and employ the subgroup criterion to
determine whether certain subsets of a group are subgroups (ILO3).
Synchronous Sessions:[2 hours of timetabled study]
 REVIEW: Attend Review Session and Tutorial (Topic: Week 2 - Symmetric groups)
Asynchronous Tasks: [4 hours of directed private study]

 READ: Lecture 5 [∼ 3 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 5.1: The subgroup criterion [16 mins 42s]
Lecture Video 5.2: Examples of subgroups [22 mins 18s]

 CHECK UNDERSTANDING: Lecture 5 Quiz [Suggested study time = 10 minutes]

 READ: Lecture 6 [∼ 3 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 6.1: Some subgroups of GL(n, R) [43 mins 4s]
Lecture Video 6.2: Intersection of subgroups [7 mins 18s]

 CHECK UNDERSTANDING: Lecture 6 Quiz [Suggested study time = 10 minutes]

 PRACTICE AND REFLECT: Do Exercises 22-24.

Take some time to consider which activities you found straightforward, and which
you found more challenging. Post any queries to the discussion board before the
next review session. [Suggested study time = 1 hour]
c University of Manchester Algebraic Structures 1 Week 3, Lecture 5, Page 1

Lecture 5
Subgroups

5.1 The subgroup criterion

Let G be a group and H ⊆ G a subset. Then for any two elements g, h ∈ H, the binary
operation of G gives a meaning to the product gh. This will be an element of G, which
may or may not be in H.
Definition 5.1 (Closure). We say that H is closed under the binary operation of G, if

∀g, h ∈ H, gh ∈ H.

Example 5.2. Let G = Z, the additive group of integers. We consider the subsets

H1 = {. . . , −4, −2, 0, 2, 4, 6, . . .} ⊆ Z,
H2 = {1, 2, 3, 4, 5, 6, . . .} ⊆ Z,
H3 = {. . . , −3, −1, 1, 3, 5, 7, . . .} ⊆ Z,

i.e. the even integers, the positive integers and the odd integers. The subsets H1 and H2
are closed under the operation +, but H3 is not closed (e.g. 1 + 1 = 2 ∈ / H3 ).
If a subset H is closed under the binary operation of G, then restricting the binary
operation of G to pairs of elements in H gives us a binary operation on the subset H. In
this case, we say that “The group operation of G induces a binary operation of H.” The
subset H, with the binary operation coming from G, then may or may not be a group.
Example 5.3. Consider the subsets H1 , H2 ⊆ Z from Example 5.2. We have already
seen in Example 2.2 that the even integers H1 are a group, and in Example 2.5 that the
positive integers H2 do not form a group.
Definition 5.4 (Subgroup). A non-empty subset H of a group G is called a subgroup, if
H is closed under the binary operation of G and H is a group with the induced operation.
We write H ≤ G to indicate that H is a subgroup of G.
Remark 5.5 (The definition expanded). In detail: A non-empty subset H ⊆ G is a
subgroup if
• ∀g, h ∈ H, gh ∈ H, (closure)

• ∀g, h, k ∈ H, (gh)k = g(hk), (G1)

• ∃ẽ ∈ H, ∀h ∈ H, ẽh = hẽ = h, (G2)

c University of Manchester Algebraic Structures 1 Week 3, Lecture 5, Page 2

• ∀h ∈ H, ∃h0 ∈ H, hh0 = h0 h = ẽ. (G3)

The first condition is the closure condition and remaining conditions are the usual group
axioms, where we have written ẽ for the identity element of H (since e is already reserved
for the identity element of G) and h0 for the inverse of h ∈ H (since we denote the
inverse of g ∈ G by g −1 ).
It turns out that we can minimize the conditions to be checked, as we shall show next.
Remark 5.6 (The definition contracted). Here are some comments on the conditions
above:

• The closure condition (∀g, h ∈ H, gh ∈ H) is essential. Without it the binary

operation of G does not induce a binary operation on H, so we definitely do not
have a subgroup.

• Associativity (∀g, h, k ∈ H, (gh)k = g(hk)) holds automatically. Indeed, it holds for

all g, h, k ∈ G, and hence, in particular, for all g, h, k ∈ H.

• If the second group axiom (∃ẽ ∈ H, ∀h ∈ H ẽh = hẽ = h) is satisfied, then it turns
out that the identity element ẽ of H coincides with the identity element e of G:

Proof that ẽ = e. Since e is the identity element of G we have eẽ = ẽ.

Since ẽ is the identity element of H and ẽ ∈ H we also have ẽẽ = ẽ.
Hence eẽ = ẽẽ in G, and so e = ẽ by cancellation (Lemma 1.7), applied in G.

Thus checking whether the second group axiom holds in H is equivalent to checking
whether e ∈ H.

• Now the final group axiom reads as follows: ∀h ∈ H, ∃h0 ∈ H, hh0 = h0 h = e.

Since the inverse of h in G is unique (Lemma 1.5), this is equivalent to the
condition that ∀h ∈ H, h−1 ∈ H.

From the above comments, we see that a non-empty subset H of a group G with identity
element e is a subgroup of G if and only if
(a) ∀g, h ∈ H, gh ∈ H,

(b) e ∈ H

(c) ∀h ∈ H, h−1 .
It is now not hard to prove the following theorem which provides an easy criterion for
deciding whether or not a given subset of a group is a subgroup.
Theorem 5.7 (The subgroup criterion). Let G be a group. A non-empty subset H of G
is a subgroup if and only if the following two conditions both hold:
c University of Manchester Algebraic Structures 1 Week 3, Lecture 5, Page 3

(i) ∀g, h ∈ H, gh ∈ H,

(ii) ∀h ∈ H, h−1 ∈ H.

Proof. Suppose H is a subgroup of G, i.e. conditions (a)-(c) from Remark 5.6 are
satisfied. Then in particular (i) holds as it is the same as the condition (a), and (ii) holds
as it is the same as condition (c).
Conversely, suppose that conditions (i) and (ii) hold for a non-empty subset H of a group
G. We need to check that each of the conditions (a)-(c) from Remark 5.6 are satisfied.
Conditions (a) and (c) are the same as (i) and (ii) and so hold by our assumption.
For condition (b), take an element g ∈ H (this can be done since H 6= ∅).
By (ii), we then have g −1 ∈ H, and so by (i) we also have e = gg −1 ∈ H.

5.2 Examples of subgroups

Example 5.8 (Two very easy examples). In any group G with identity element e, the set
{e} is a subgroup of G. This subgroup is called the trivial subgroup. Similarly, the group
G is itself is a subgroup: G ≤ G. A subgroup H of G is called proper if H 6= G.

All of the following examples use Theorem 5.7.

Example 5.9. Q ≤ R. For a, b ∈ Q, we have a + b ∈ Q, so (i) holds. Moreover, −a ∈ Q,

so (ii) holds (recall that we denote the inverse of a by −a in additive notation).

Example 5.10. N is not a subgroup of Z. Although the sum of two natural numbers is
natural, the inverse −n of n ∈ N is not in N.

Example 5.11. H = {z ∈ C : |z| = 1} ≤ C∗ . The product of two complex numbers of

modulus 1 is again a complex number of modulus 1, and the inverse of a complex number
of modulus 1 is also of modulus 1.

Example 5.12. N is not a subgroup of Q∗ . Although the product of two natural numbers
is natural, N is not closed under taking inverses: for example, 2 ∈ N but 2−1 = 1/2 ∈
/ N.

Example 5.13. {0, 3, 6} ≤ Z9 . From the multiplication table

⊕ 0 3 6
0 0 3 6
3 3 6 0
6 6 0 3
we see that this set is closed under addition modulo 9, and that each element has an
inverse (i.e., the identity element 0 appears in every row and column).

Example 5.14. H = {0, 3, 6} is not a subgroup of Z8 . For example, 3 ⊕ 6 = 1 in Z8 ,

and 1 ∈
/ H, so H is not closed under the operation of addition modulo 8.
c University of Manchester Algebraic Structures 1 Week 3, Lecture 5, Page 4

Example 5.15. {id3 , (1 2 3), (1 3 2)} ≤ S3 . From the multiplication table

◦ id3 (1 2 3) (1 3 2)
id3 id3 (1 2 3) (1 3 2)
(1 2 3) (1 2 3) (1 3 2) id3
(1 3 2) (1 3 2) id3 (1 2 3)
we see that this set is closed under composition of permutations, and that each element
has an inverse.

Example 5.16. H = {id3 , (12), (13)} is not a subgroup of S3 . Note that although the
inverse of each element of H is contained in H, H is not closed: for example,
(12)(13) = (132 is not contained in H.
c University of Manchester Algebraic Structures 1 Week 3, Lecture 6, Page 1

Lecture 6
Examples of subgroups

6.1 Some subgroups of GL(n, R)

The general linear groups have lots of subgroups, which is part of the reason why they
are very interesting objects in mathematics. Again, in all of our examples we apply the
subgroup criterion (Theorem 5.7).
  
1 n
Example 6.1. H = : n ∈ Z ≤ GL(2, R): indeed, we have
0 1
    
1 n 1 m 1 n+m
= ∈ H,
0 1 0 1 0 1
so (i) holds, and
 −1  
1 n 1 −n
= ∈ H,
0 1 0 1
so (ii) holds.
Example 6.2. The Special Linear Group
SL(n, R) = {A ∈ GL(n, R) | det A = 1}
is a subgroup of GL(n, R): the subset SL(n, R) is not empty, as it contains, e.g., the
identity matrix. Moreover, if det A = det B = 1, then det(AB) = (det A)(det B) = 1 and
det(A−1 ) = 1/ det A = 1. Hence, the two conditions of the subgroup criterion are satisfied.
Example 6.3. The subgroup of non-zero scalar matrices:
  

 λ 0 ··· 0 

 0 λ ··· 0 
 

Scal(n, R) =  .. .. . . ..  : λ ∈ R, λ 6= 0 ≤ GL(n, R)
 

  . . . .  

 
 0 0 ··· λ 

Indeed, the product of two scalar matrices is a scalar matrix, and the inverse of a scalar
matrix is also one. Hence, Scal(n, R) ≤ GL(n, R) by the subgroup criterion.
Example 6.4. The subgroup of invertible diagonal matrices:
  

 λ1 0 · · · 0 

 0 λ 2 · · · 0 
 

D(n, R) =  .. .  : λi ∈ R, λ1 · · · λn 6= 0 ≤ GL(n, R).
 
.. . .
 .
 . . ..  

 
 0 0 · · · λn 
c University of Manchester Algebraic Structures 1 Week 3, Lecture 6, Page 2

Again, products and inverses of diagonal matrices are diagonal matrices, so we can apply
the subgroup criterion.

Example 6.5. The subgroup of upper triangular matrices:

  

 λ 1 ∗ · · · ∗ 

 0 λ 2 · · · ∗ 
 

T(n, R) =  ..  : λi ∈ R, λ1 · · · λn 6= 0 ≤ GL(n, R).
 
.. . .

  . . . ∗  

 
 0 0 · · · λn 

Note: The stars above the diagonal are to indicate arbitrary independent real values.
To apply the subgroup criterion here, we must know that the product of two upper
triangular matrices is again upper triangular, and that the inverse of an upper triangular
matrix is again upper triangular. The first fact follows from the definition of matrix
multiplication (exercise: try it!). The second fact then follows from Gauss-Jordan
elimination, noting that to
find the inverse of an upper triangular matrix, one can perform a sequence of elementary
row operations each of which corresponds to left multiplication by an upper triangular
elementary matrix.

Example 6.6. The subgroup of upper uni-triangular matrices:

 

 1 ∗ ··· ∗  
 0 1 · · · ∗  
 
UT(n, R) =  .. .. . .  ≤ GL(n, R)
 

  . . . ∗  

 0 0 ··· 1  

Note: The stars above the diagonal are to indicate arbitrary independent real values.
Similarly, we have the subgroups of lower triangular and lower uni-triangular matrices
Tl (n, R) and UTl (n, R).

Remark 6.7. The subgroups of GL(n, R) from the previous examples are related to each
other as follows:

UT(n, R) ≤ SL(n, R)
UT(n, R) ≤ T(n, R)
D(n, R) ≤ T(n, R)

The groups Scal(n, R) and D(n, R) are abelian for all n ≥ 1. On the other hand, T(n, R)
is not abelian for all n ≥ 2, and UT(n, R) is not abelian for all n ≥ 3.

Remark 6.8. All these subgroups may also be defined for the general linear groups over
other fields, such as Q and C.
c University of Manchester Algebraic Structures 1 Week 3, Lecture 6, Page 3

6.2 Intersection of subgroups

The intersection of two subgroups of a group G is again a subgroup.

Lemma 6.9 (Intersection of subgroups of the same group is a subgroup).

Let G be a group, H ≤ G and K ≤ G. Then H ∩ K ≤ G.

Proof. We have that e ∈ H and e ∈ K, so e ∈ H ∩ K, and thus H ∩ K 6= ∅. Hence, we

can apply the subgroup criterion to prove that H ∩ K is a subgroup.
To check condition (i), suppose that g, h ∈ H ∩ K. Then g ∈ H and h ∈ H. Since H ≤ G
we have gh ∈ H (because condition (i) of the subgroup criterion holds for subgroup H).
Likewise, g, h ∈ H ∩ K means that g ∈ K and h ∈ K. Since K ≤ G we have gh ∈ K
(because condition (i) of the subgroup criterion holds for subgroup K). Consequently,
gh ∈ H and gh ∈ K, and hence gh ∈ H ∩ K. This shows that condition (i) of the
subgroup criterion holds for the subset H ∩ K.
Now suppose g ∈ H ∩ K. We aim to show that g −1 ∈ H ∩ K too. Since g ∈ H we have
that g −1 ∈ H (because condition (ii) of the subgroup criterion holds for subgroup H).
Likewise, since g ∈ K we have that g −1 ∈ K (because condition (ii) of the subgroup
criterion holds for subgroup K). Hence, g −1 ∈ H and g −1 ∈ K, so g −1 is in H ∩ K. This
means that condition (ii) of the subgroup criterion is satisfied for H ∩ K. The Lemma
follows.

Remark 6.10. In the proof we have used the subgroup criterion in both directions.

Remark 6.11 (Union of subgroups of G need not be a subgroup of G). If H, K ≤ G,

then H ∪ K is not necessarily a subgroup of G. For example,

2Z = {2k | k ∈ Z} ≤ Z

and
3Z = {3k | k ∈ Z} ≤ Z,
but
2Z ∪ 3Z  Z
since, for example, 2 + 3 = 5 ∈
/ 2Z ∪ 3Z.
c University of Manchester Algebraic Structures 1 Week 3, Exercises, Page 1

Week 3: Exercises

Guidance: This week I would like you to practice determining whether a subset S of a
given group G is in fact a subgroup of G. In each case, the given group is one of our
favourite examples, and you should know what the binary operation of that group is.
Question 24 provides us with another interesting family of examples of groups: the
so-called alternating groups.

Subgroups
(requires material up to Lecture 5)

22. For the following subsets S of the given group (G, ∗) determine whether or not S is
a subgroup. Give reasons if your answer is NO.
(i) G = C, S = {n + mi | n, m ∈ Z} ,
(ii) G = C, S = {3n + mi | n, m ∈ Z} ,
(iii) G = Q∗ = Q\{0}, S = {1, −1} ,
(iv) G = Sn , S is the set of all transpositions,
(v) G = Z6 , S = {0, 2, 4} ,
(vi) G = Z6 , S = {0, 1, 3} .

23. Show that   

 1 a b 
H=  0 1 c  : a ∈ Z, b, c ∈ R
0 0 1
 

is a subgroup of GL(3, R).

24. Let An denote the subset of Sn consisting of all permutations of {1, 2, . . . , n} that
can be written as a composite of an even number of transpositions. Prove that An
is a subgroup of Sn .

Solutions to the exercises can be found here.

c University of Manchester Algebraic Structures 1 Week 4 Checklist

Resources for Week 4

Cyclic subgroups and order

(Click here to return to the front page navigation)

This week we will consider in more detail the idea of taking powers of a fixed element of a
group. The set of all powers of a fixed element of a group G turns out to be a subgroup
of G; subgroups created in this way are called cyclic subgroups. We also introduce the
idea of the order of an element, and explain its relationship to the cardinality of the
cyclic subgroup generated by that element. In the case of the symmetric groups, we shall
see that there is a simple way to compute the order of an element by using its cycle
structure. By the end of this week you should be able to explain the concepts of order
and cyclic subgroups (ILO2) and to compute examples of these.
Synchronous Sessions:[2 hours of timetabled study]
 REVIEW: Attend Review Session and Tutorial (Topic: Week 3 - Subgroups)
Asynchronous Tasks: [4 hours of directed private study]

 READ: Lecture 7 [∼ 2.5 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 7.1: Cyclic subgroups [14 mins 18s]
Lecture Video 7.2: Order of groups and their elements [9 mins 22s]

 CHECK UNDERSTANDING: Lecture 7 Quiz [Suggested study time = 10 minutes]

 READ: Lecture 8 [∼ 3 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 8.1: Infinite order and finite order [18 mins 22s]
Lecture Video 8.2: The order of a permutation [25 mins 37s]

 CHECK UNDERSTANDING: Lecture 8 Quiz [Suggested study time = 10 minutes]

 PRACTICE AND REFLECT: Do Exercises 25-28.

Take some time to consider which activities you found straightforward, and which
you found more challenging. Post any queries to the discussion board before the
next review session. [Suggested study time = 1 12 hours]
c University of Manchester Algebraic Structures 1 Week 4, Lecture 7, Page 1

Lecture 7
Cyclic subgroups and order

7.1 Cyclic subgroups

Definition 7.1. (The set of all powers of a fixed element)
For a group G and an element a ∈ G, let

hai = {ak | k ∈ Z} = {. . . a−2 , a−1 , a0 = e, a, a2 , a3 , . . .}.

denote the set of all powers of a in G.

Example 7.2. In R∗ we have

1 1 1
h−2i = {(−2)k | k ∈ Z} = {. . . − , , − , 1, −2, 4, −8, 16, . . .}
8 4 2
and
h−1i = {(−1)k | k ∈ Z} = {−1, 1}.

Example 7.3. In S3 , consider h(1 2 3)i. Here we have

(1 2 3)1 = (1 2 3)
(1 2 3)2 = (1 2 3)(1 2 3) = (1 3 2)
(1 2 3)3 = (1 2 3)2 (1 2 3) = (1 3 2)(1 2 3) = id3 .

For an arbitrary integer k, we use long division to write k = 3m + r with 0 ≤ r ≤ 2. Then

(1 2 3)k = (1 2 3)3m+r
= (1 2 3)3m (1 2 3)r
= ((1 2 3)3 )m (1 2 3)r
= idm
3 (1 2 3)
r

= (1 2 3)r .

Hence, 
 (1 2 3), if k ≡ 1 mod 3,
(1 2 3)k = (1 3 2), if k ≡ 2 mod 3,
id3 , if k ≡ 0 mod 3.


Consequently,
h(1 2 3)i = {id3 , (1 2 3), (1 3 2)}.
c University of Manchester Algebraic Structures 1 Week 4, Lecture 7, Page 2

Remark 7.4 (Powers in additive notation). If (G, +) is written additively, then of course
we write powers as multiples, as usual. Hence, for g ∈ G, we have

hgi = {ng : n ∈ Z} = {. . . − 2g, −g, e, g, 2g, 3g, . . .}.

Example 7.5. In Z, we have hmi = {. . . − 2m, −m, 0, m, 2m, . . .} = mZ, the set of all
multiples of m.

Lemma 7.6 (The set of all powers of a fixed element forms a subgroup).
Let G be a group and a ∈ G. Then hai is a subgroup of G. Moreover, the group hai is
abelian.

Proof. We use the subgroup criterion. Clearly, hai =

6 ∅ since a ∈ hai. Now, if x, y ∈ hai,
then x = ak and y = am for some integers k, m. But then

xy = ak am = ak+m ∈ hai.

Also
x−1 = (ak )−1 = a−k ∈ hai.
Hence, hai satisfies the two conditions of the subgroup criterion, so it is a subgroup.
Finally, ak am = ak+m = am+k = am ak , i.e. any two powers of a commute. Hence hai is
abelian.

Definition 7.7 (Cyclic subgroup). Let G be a group and a ∈ G. The subgroup hai of G
is called the cyclic subgroup generated by a.

7.2 Order of groups and their elements

Definition 7.8 (Order of a group). The order |G| of a group G is the number of
elements in the set G, or ∞ if G is infinite.

Example 7.9. |Z| = ∞, |R∗ | = ∞, |Zn | = n, | Sn | = n!, | GL(n, R)| = ∞.

We also define the order of an element of a group.

Definition 7.10 (Order of an element). Let G be a group and a ∈ G. The order |a| of a
is the smallest m ∈ N such that am = e. If there is no such m ∈ N, we say that a has
infinite order |a| = ∞.

Example 7.11.

1. In R∗ , | − 2| = ∞ and | − 1| = 2,

2. in S3 , (1, 2, 3) has order 3, i.e. |(1, 2, 3)| = 3,

√
3. in C∗ , i = −1 has order 4: i2 = −1, i3 = −i, i4 = 1.
c University of Manchester Algebraic Structures 1 Week 4, Lecture 7, Page 3

Example 7.12. In any group G, the identity element e has order 1, and it is the only
element of order 1.

Remark 7.13. Again, if (G, +) is written additively, we take multiples instead of powers:
the order of a ∈ G is the smallest m ∈ N such that ma = e, or ∞ if no such m exists.

Example 7.14. In the additive group Z6 , |a| is the smallest m ∈ N with ma = 0. E.g.

• 2 has order 3: 2, 2 ⊕ 2 = 4, 2 ⊕ 2 ⊕ 2 = 0,

• 3 has order 2: 3, 3 ⊕ 3 = 0,

• 5 has order 6: 5, 4, 3, 2, 1, 0.

Having defined the order of an element (Definition 7.10) and the order of a group
(Definition 7.8), we now investigate how these two notions are related.
c University of Manchester Algebraic Structures 1 Week 4, Lecture 8, Page 1

Lecture 8
The order of an element and the order of the group

8.1 Infinite order and finite order

Lemma 8.1 (All powers of an element of infinite order are distinct).
Let a be an element of infinite order in a group G. Then am 6= an for all m, n ∈ Z with
m 6= n.

Proof. Suppose am = an with m 6= n. We may assume that m > n; if not, then exchange
the roles of m and n. Hence, m − n > 0 and we have

am−n = am a−n = an a−n = an−n = a0 = e,

contradicting the assumption that a has infinite order.

Corollary 8.2 (Elements of infinite order generate infinite cyclic subgroups).

If a ∈ G has infinite order, then the cyclic subgroup hai generated by a has infinite order.

Proof. According to Definition 7.1, hai consists of all the powers of a. By Lemma 8.1,
they are all distinct, so hai is infinite.
Next we investigate the relation between the order |a| of an element a ∈ G and the order
|hai| of the cyclic subgroup generated by a in the case where these are finite.

Lemma 8.3 (Properties of powers of an element of order n).

Let G be a group and a ∈ G with |a| = n ∈ N. Then

1. all of the elements a, a2 , . . . , an−1 , an = e are distinct,

2. if s ∈ Z with s = kn + r, then as = ar ,

3. if s, t ∈ Z, then as = at if and only if s ≡ t mod n,

4. as = e if and only if n divides s.

Proof. (1): suppose as = at for some s, t with 0 < s < t ≤ n. Then n > t − s ≥ 1, and
at−s = at a−s = as a−s = a0 = e contradicting the assumption that n is the smallest
natural number such that an = e.
(2): we have
as = akn+r = (an )k ar = ar ,
since an = e.
c University of Manchester Algebraic Structures 1 Week 4, Lecture 8, Page 2

(3): if s ≡ t mod n, then n | s − t, so s = kn + t for some k ∈ N. Hence, as = at by (2).

Conversely, if as = at , use long division to write s = kn + r1 and t = mn + r2 with
0 ≤ r1 , r2 < n. Then as = ar1 and at = ar2 by (2), and since the elements
a, a2 , . . . , an−1 , an are distinct by (1), this implies r1 = r2 . Hence, s ≡ t mod n.
(4): This is the special case of (3) with t = 0.

Corollary 8.4 (Elements of order n generate cyclic subgroups of order n). Let a be an
element of order n in a group G. Then

hai = {e, a, a2 , . . . , an−1 },

In particular, the order of hai is n. In other words, |hai| = |a|.

Proof. The elements e, a, . . . , an−1 are all distinct by Lemma 8.3 (1), and every power of
a is equal to one of them by long division and Lemma 8.3 (2).

Example 8.5. In Z9 , we have |3| = 3 as 3 ⊕ 3 = 6 6= 0, but 3 ⊕ 3 ⊕ 3 = 0. Hence,

h3i = {0, 3, 6}. Compare this to Example 5.13.

8.2 The order of a permutation

We will determine completely the order |f | of any permutation f ∈ Sn . First, we consider
the case of a single cycle.

Lemma 8.6 (Cycles of length d have order d). Let f ∈ Sn be a d-cycle. Then |f | = d.

Proof. Write f = (α1 α2 · · · αd ). We need to find the smallest m ∈ N such that f m = idn .
By definition, f fixes all the elements of Nn r {α1 , . . . , αd }, whereas the αi are moved as
follows:
α1 7→ α2 7→ α3 7→ · · · αd−1 7→ αd 7→ α1 7→ α2 7→ α3 7→ · · · .
Hence, if f is applied d times to an arbitrary entry of the cycle, this entry is moved to
itself, i.e. fixed. Therefore, f d = idn . On the other hand, f k 6= idn for 1 ≤ k < d, because
then f k (α1 ) = αk+1 6= α1 . Hence, |f | = d.

Example 8.7. Consider the 5-cycle f = (1 2 3 4 5) ∈ S5 . Then

f2 = (1 3 5 2 4)
f3 = (1 4 2 5 3)
f4 = (1 5 4 3 2)
f5 = id5 .

Now we consider an arbitrary permutation f ∈ Sn . Recall that f can be expressed as a

composite of disjoint cycles.
c University of Manchester Algebraic Structures 1 Week 4, Lecture 8, Page 3

Theorem 8.8 (The order of a product of disjoint cycles is the l.c.m of cycle lengths).
Suppose that f ∈ Sn is a product
f = g1 g2 · · · gk ,
where g1 , g2 , . . . , gk are disjoint cycles of length d1 , d2 , . . . , dk , respectively. Then the order
of f is the least common multiple of d1 , d2 , . . . , dk , that is,
|f | = lcm(d1 , d2 , . . . , dk ).
Proof. Since disjoint cycles commute, we have
f m = (g1 · · · gk )m = g1m g2m · · · gkm .
Now note that f m = idn if and only if gim = idn for all i = 1, 2, . . . , k. Indeed, suppose
that gim 6= idn for some i, then gim (α) 6= α for some α ∈ Nn . This α must be one of the
entries of gi , as otherwise it would be fixed by gi and thus by gim . Similarly, gim (α)
appears in gi . Since the cycles are disjoint, we have gjm (α) = α and gjm (gim (α)) = gim (α)
for all j 6= i, and thus f m (α) = gim (α) 6= α. Hence, f m 6= idn .
Consequently, the smallest natural number m such that f m = idn is the smallest natural
number m such that gim = idn for i = 1, 2, . . . , k. Since the order of gi is di , the length of
the cycle gi , Lemma 8.3 shows that gim = idn if and only if di |m. Hence, the number we
are looking for is the smallest m ∈ N such that di |m for all i = 1, 2, . . . , k. This number is
the least common multiple of d1 , d2 , . . . , dk .
Example 8.9 (Finding the order of a permutation). Consider the permutation
 
1 2 3 4 5 6 7 8 9
f=
3 4 5 2 7 8 9 6 1
= (1 3 5 7 9)(2 4)(6 8) ∈ S9 .
Here we have one cycle of length 5 and two cycles of length 2. Hence,
|f | = lcm(5, 2, 2) = 10.
Example 8.10 (Carefully finding the order of a permutation...). Suppose that
f = (1 3 5)(2 4)(5 6).
Here f is a composite of one cycle of length 3 and two cycles of length 2. However, it is
not true that
|f | = lcm(3, 2, 2) = 6.
Theorem 8.8 does not apply here because the cycles in our example are not disjoint. In
order to apply our rule, we first need to rewrite f as a composite of disjoint cycles:
f = (1 3 5 6)(2 4).
Then we get
|f | = lcm(4, 2) = 4.
c University of Manchester Algebraic Structures 1 Week 4, Exercises, Page 1

Week 4: Exercises

Guidance: This week I would like you to practice computing the orders of elements in
some of our favourite examples of groups (as with last week: you should know what the
binary operation is in each case). These problems require some direct computation with
familiar objects (complex numbers, integers modulo n, permutations, matrices, etc.)
Question 28 is more challenging – make sure to give that one a go.
Cyclic subgroups and order
(requires material up to Lecture 7)
25. Work out the orders of the following elements.
√ √ √ √
(i) 2
2
− 2
2
i, 12 + 2
3
i, 2
3
+ 12 i ∈ C∗
(ii) 4, 15, 18, 33 ∈ Z36
(iii) f, g, h ∈ S9 , where
   
1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
f = ,g =
2 3 1 5 6 7 4 9 8 9 8 7 6 5 4 3 2 1
 
1 2 3 4 5 6 7 8 9
h =
3 4 5 6 7 8 9 2 1
     
0 1 1 1 0 −1
(iv) , , ∈ GL(2, R).
−1 1 0 1 1 0

26. (i) Compute h5i in Z.

(ii) Compute h(1 2 3 4)i in S4 .
(iii) Compute h1i , h2i and h3i in Z8 .
 
0 −1
(iv) Compute in GL(2, R).
1 0
27. Compute h(1 2 3)(4 5)i in S5 .

Order of permutations
(requires material up to Lecture 8)

28. What are the largest possible orders of elements in S3 , S5 and S7 ?

Solutions to the exercises can be found here.

c University of Manchester Algebraic Structures 1 Week 5 Checklist

Resources for Week 5

Centralisers and Centres

(Click here to return to the front page navigation)

This week we look at (i) the subset of a group G containing all elements which commute
with a fixed element a ∈ G (called the centraliser of a) and (ii) the subset of G
containing all elements which commute with every element of G (called the centre of G).
We show that these are subgroups of G. By the end of this week you should be able to
explain the concepts of centralisers and centres (ILO2), demonstrate that these are
subgroups (ILO3), and compute examples of these.
Synchronous Sessions:[2 hours of timetabled study]

 REVIEW: Attend Review Session and Tutorial (Topic: Week 4 - Cyclic subgroups
and order)

Asynchronous Tasks: [4 hours of directed private study]

 READ: Lecture 9 [∼ 3 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 9.1: What is the centraliser of an element? [12 mins 36s]
Lecture Video 9.2: Examples of centralisers [18 mins 34s]

 CHECK UNDERSTANDING: Lecture 9 Quiz [Suggested study time = 10 minutes]

 READ: Lecture 10 [∼ 3 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 10.1: What is the centre of a group? [13 mins 27s]
Lecture Video 10.2: Examples of the centre of a group [16 mins 34s]

 CHECK UNDERSTANDING: Lecture 10 Quiz [Suggested study time = 10 minutes]

 PRACTICE AND REFLECT: Do Exercises 29-34.

Take some time to consider which activities you found straightforward, and which
you found more challenging. Post any queries to the discussion board before the
next review session. [Suggested study time = 1 21 − 2 hours]
c University of Manchester Algebraic Structures 1 Week 5, Lecture 9, Page 1

Lecture 9
Centralisers

9.1 What is the centraliser of an element?

Definition 9.1 (The centraliser of an element). Let G be a group and a ∈ G. The
centraliser of a in G is the set

C(a) = {g ∈ G : ga = ag},

of all elements of G that commute with a.

Example 9.2. Let G = S4 and a = (1 2 3 4). Then (1 3)(2 4) ∈ C(a), since

(1 2 3 4)(1 3)(2 4) = (1 4 3 2) = (1 3)(2 4)(1 2 3 4).

On the other hand, (1 2 3) ∈/ C(a) since (1 2 3 4)(1 2 3) 6= (1 2 3)(1 2 3 4). Indeed, we

have (1 2 3 4)(1 2 3) = (1 3 2 4), but (1 2 3)(1 2 3 4) = (1 3 4 2).

Theorem 9.3 (The centraliser of an element of G is a subgroup of G).

For any group G and any a ∈ G, the centraliser C(a) is a subgroup of G.

Proof. We always have e ∈ C(a) by the group axiom (G2), so C(a) 6= ∅ and we can apply
the subgroup criterion (Theorem 5.7). Assume that g, h ∈ C(a), i.e. ga = ag and
ha = ah. Then
(gh)a = g(ha) = g(ah) = (ga)h = (ag)h = a(gh),
and thus gh ∈ C(a). Also, if g ∈ C(a), then g −1 ∈ C(a) since

g −1 a = g −1 ae = g −1 agg −1 = g −1 gag −1 = eag −1 = ag −1 .

Hence C(a) satisfies the two conditions of the subgroup criterion, so C(a) ≤ G.

Remark 9.4. Here are some obvious facts about centralisers:

• In any group G, C(e) = G.

• In any abelian group G, C(a) = G for all a ∈ G. In fact, a group G is abelian if and
only if C(a) = G for all a ∈ G.

• In any group G, hai ≤ C(a) for all a ∈ G, since a commutes with all of its powers:
aan = an+1 = an a.
c University of Manchester Algebraic Structures 1 Week 5, Lecture 9, Page 2

9.2 Examples of centralisers

Since groups of numbers Z, R, . . . and the residue groups Zn are abelian, centralisers are
not of interest in them. However, they are very interesting in symmetric groups and
matrix groups.
Example 9.5. Consider S3 = {e, (1 2), (1 3), (2 3), (1 2 3), (1 3 2)}. What is C((1 2 3))?
We can answer this by checking the six elements of this group one-by-one. Of course,

e ∈ C((1 2 3)) and (1 2 3) ∈ C((1 2 3)).

Also, (1 2 3)(1 3 2) = idn = (1 3 2)(1 2 3), so (1 3 2) ∈ C((1 2 3)). But the three
transpositions do not commute with (1 2 3): (1 2 3)(1 2) = (1 3) but (1 2)(1 2 3) = (2 3),
(1 2 3)(1 3) = (2 3) but (1 3)(1 2 3) = (1 2) and (1 2 3)(2 3) = (1 2) but
(2 3)(1 2 3) = (1 3). Hence

C((1 2 3)) = {e, (1 2 3), (1 3 2)} = h(1 2 3)i.

The last example is no coincidence, it holds more generally as follows.

Lemma 9.6 (Centraliser of (12 · · · n) in Sn ). In Sn , for any n ∈ N, we have

C((1 2 · · · n)) = h(1 2 · · · n)i.

Proof. Let g = (1 2 · · · n). We have already seen in Remark 9.4 that hgi ⊆ C(g), so we
need to show the inverse inclusion C(g) ⊆ hgi. Let f ∈ C(g) and write f (1) = k, for some
k ∈ Nn . We will show that f = g k−1 . For that we need to verify that f (m) = g k−1 (m) for
all m ∈ {1, 2, . . . , n}. In the proof, we will use the fact that

g m−1 (1) = m for all m ∈ Nn , so in particular g k−1 (1) = k.

Now, for any m ∈ Nn ,

f (m) = f (g m−1 (1))

= (f g m−1 )(1)
= (g m−1 f )(1) (since f ∈ C(g))
m−1
= g (f (1))
m−1
= g (k)
m−1 k−1
= g (g (1))
m−1 k−1
= (g g )(1)
k−1 m−1
= (g g )(1)
= g k−1 (g m−1 (1))
= g k−1 (m),

as required. Hence f ∈ hgi, so C(g) ⊆ hgi and hence C(g) = hgi.

c University of Manchester Algebraic Structures 1 Week 5, Lecture 9, Page 3

 
1 1
Example 9.7. Let A = ∈ GL(2, R). What is C(A)?
 1 0
a b
A 2 × 2 matrix B = is in C(A) if and only if B is invertible and AB = BA.
c d
The condition that AB = BA is the same as
     
1 1 a b a b 1 1
=
1 0 c d c d 1 0
We compute the left- and right-hand sides as
   
a+c b+d a+b a
LHS = RHS = .
a b c+d c
Hence, LHS = RHS if and only if
a+c = a+b
a = c+d
b+d = a
b = c.
This holds if and only if
b=c and a = b + d,
so AB = BA if and only if  
b+d b
B= .
b d
In conclusion,
    
1 1 b+d b 2
C = : b, d ∈ R, (b + d)d − b 6= 0 .
1 0 b d


1 0
Example 9.8. Let D = ∈ GL(2, R). What is C(D)? Again, equating
0 2
     
1 0 a b a b 1 0
=
0 2 c d c d 0 2
we get    
a b a 2b
LHS = RHS = .
2c 2d c 2d
Here we have LHS = RHS if and only if b = 2b and c = 2c, i.e. b = c = 0. Hence,
    
1 0 a 0
C = : a, d ∈ R, ad 6= 0 = D(2, R).
0 2 0 d
c University of Manchester Algebraic Structures 1 Week 5, Lecture 10, Page 1

Lecture 10
The centre

10.1 What is the centre of a group?

Definition 10.1 (The centre of a group). Let G be a group. The centre of G is the set
Z(G) = {g ∈ G : ∀x ∈ G, gx = xg}
of all elements that commute with all elements in G.
Example 10.2. Let (G, ∗) be the group with multiplication table:
∗ e v w x y z
e e v w x y z
v v e z y x w
w w x y z e v
x x w v e z y
y y z e v w x
z z y x w v e
Note that v does not commute with w, x, y or z. Thus Z(G) = {e}.
Theorem 10.3. (The centre of a group is a subgroup)
Let G be a group. The centre Z(G) is a subgroup of G.
Proof. Clearly, e ∈ Z(G), so Z(G) 6= ∅ and we can apply the subgroup criterion. Let
g, h ∈ Z(G), i.e. gx = xg and hx = xh for all x ∈ G. Then for all x ∈ G we have
(gh)x = g(hx) = g(xh) = (gx)h = (xg)h = x(gh),
so gh ∈ Z(G). Also, if g ∈ Z(G), then g −1 ∈ Z(G) since for all x ∈ G we have
g −1 x = g −1 xe = g −1 xgg −1 = g −1 gxg −1 = exg −1 = xg −1 .
Hence Z(G) satisfies the two conditions of the subgroup criterion, so Z(G) ≤ G.
Remark 10.4. Here are some obvious facts about the centre:
• In any abelian group G, we have Z(G) = G. In fact, a group G is abelian if and
only if Z(G) = G.
• For all a ∈ G, we have Z(G) ≤ C(a). In fact,
\
Z(G) = C(a).
a∈G
c University of Manchester Algebraic Structures 1 Week 5, Lecture 10, Page 2

10.2 Examples of the centre of a group

Again, since groups of numbers Z, R, . . . and the residue groups Zn are abelian, they just
coincide with their centres. Again, things are different for symmetric groups and matrix
groups.
Example 10.5 (Centre of S3 ). Consider S3 = {e, (1 2), (1 3), (2 3), (1 2 3), (1 3 2)}.
Here we have
(1 2)(1 3) = (1 3 2) 6= (1 2 3) = (1 3)(1 2) ⇒ (1 2), (1 3) ∈
/ Z(S3 )
(2 3)(1 2 3) = (1 3) =6 (1 2) = (1 2 3)(2 3) ⇒ (2 3), (1 2 3) ∈
/ Z(S3 )
(2 3)(1 3 2) = (1 2) = 6 (1 3) = (1 3 2)(2 3) ⇒ (1 3 2) ∈/ Z(S3 ).
Hence, Z(S3 ) = {idn }
Again, the last example holds in higher generality.
Lemma 10.6 (Symmetric groups of degree at least 3 have trivial centre).
For all n ≥ 3, we have
Z(Sn ) = {idn }.
Proof. We need to show that no element of Sn except idn is in the centre. In other words,
we need to verify that that for any f ∈ Sn with f 6= idn there exists a permutation g ∈ Sn
such that f g 6= gf . Let f 6= idn , then f moves at least one element of Nn = {1, 2, . . . , n},
so there are a, b ∈ Nn with a 6= b and f (a) = b. Since n ≥ 3, there is c ∈ Nn such that
c 6= a and c 6= b. Now define the transposition g = (a c). Then f ◦ g 6= g ◦ f . Indeed,
(f ◦ g)(a) = f (g(a)) = f (c),
(g ◦ f )(a) = g(f (a)) = g(b) = b.
So if f ◦ g = g ◦ f , then f (c) = b. But this is impossible since f (a) = b, a 6= c, and f is
bijective by definition. Hence f ◦ g 6= g ◦ f and so f ∈ / Z(Sn ).
Example 10.7 (The centre of GL(2, R)). For the general linear group we have
Z(GL(2, R)) = Scal(2, R).
Clearly, the scalar matrices are contained in the centre, since multiplying a matrix by a
scalar matrix (no matter if on the right or on the left) amounts to multiplying all entries
of the matrix by the scalar in question:
       
λ 0 a b λa λb a b λ 0
= = .
0 λ c d λc λd c d 0 λ
 
a b
Now suppose A = ∈ Z(GL(2, R)). Then
c d
     
a b 1 1 1 1 a b
= .
c d 0 1 0 1 c d
c University of Manchester Algebraic Structures 1 Week 5, Lecture 10, Page 3

The left- and right-hand sides in this equation are

   
a a+b a+c b+d
LHS = RHS =
c c+d c d

Hence, equality holds if and only if

a=a+c ∧ a+b=b+d ∧ c+d=d which is equivalent to c = 0 ∧ a = d.

 
a b
So if a matrix is in the centre, it must be of the form . But to be a central
0 a
matrix, it must commute with all matrices in GL(2, R). In particular, we must have
     
a b 1 0 1 0 a b
= .
0 a 1 1 1 1 0 a

Here,    
a+b b a b
LHS = RHS = .
a a a a+b
So we have equality
 if 
and only if a + b = a, and therefore b = 0. Hence a central matrix
a 0
is of the form , i.e. it is a scalar matrix. Of course we must have a 6= 0 for the
0 a
matrix to be invertible. In conclusion, we have proved that
  
a 0
Z(GL(2, R)) = : a 6= 0 = Scal(2, R).
0 a

Alternatively, we could have used our earlier examples of centralisers to show that
Z(GL(2, R)) = Scal(2, R). Recall from Example 9.7 and Example 9.8 that
    
1 1 b+d b 2
C = : b, d ∈ R, (b + d)d − b 6= 0
1 0 b d

and that     
1 0 a 0
C = : a, d ∈ R, ad 6= 0 = D(2, R).
0 2 0 d
Any matrix in the centre must belong to both of these centralisers, i.e. it must be in the
intersection of the two centralisers, which is the group of scalar matrices Scal(2, R).
Our first proof, however, can nicely be modified to establish a more general result, namely
that Z(GL(n, R)) = Scal(n, R) for all n ∈ N.
c University of Manchester Algebraic Structures 1 Week 5, Exercises, Page 1

Week 5: Exercises

Guidance: This week I would like you to practice computing elements that commute in
one of our favourite non-abelian groups (as ever: you should know what the binary
operation is in each case). These problems require some direct computation with familiar
objects (permutations and matrices). Questions 31 and 32 are more challenging.
Centralisers
(requires material up to Lecture 9)
29. Work out the centraliser C(g) in GL(2, R) in each of the following cases:
     
1 0 1 1 0 1
(i) g = , (ii) g = , (iii) g = .
0 2 0 1 1 0

30. Using inspection of all the 24 elements of S4 (or otherwise), work out the centraliser
of the transposition (1 2) in S4 .
31. Prove that the centraliser of the transposition (1 2) in Sn with n ≥ 3 is the
subgroup C((1 2)) = {f ∈ Sn | f (1), f (2) ∈ {1, 2}}.
32. Prove that in the symmetric group Sn ,
C((1 2 · · · n − 1)) = h(1 2 · · · n − 1)i ,
i.e. the centraliser of the cycle (1 2 · · · n − 1) (of length n − 1) in Sn coincides with
the cyclic subgroup generated by that cycle.

Centre
(requires material up to Lecture 10)

33. Work out the centre of

  
a b
T(2, R) = ; a, b, c ∈ R, ac 6= 0 .
0 c

34. Work out the centre of

  
 1 a b 
UT(3, R) =  0 1 c  ; a, b, c ∈ R .
0 0 1
 

Solutions to the exercises can be found here.

c University of Manchester Algebraic Structures 1 Week 7 Checklist

Resources for Week 7

Cyclic groups

(Click here to return to the front page navigation)

Note: There is a reduction in asynchronous material this week, to give you space to
prepare for and work on the test. The test covers material from weeks 1-5.
This week we consider in more detail the groups G with the property that G = hai for
some a ∈ G. By the end of this week you should be able to explain the concept of a cyclic
group (ILO2), give examples of cyclic groups, and identify the generators and subgroups
of cyclic groups (ILO5).
Synchronous Sessions:[2 hours of timetabled study]

 REVIEW: Attend Review Session and Tutorial (Topic: Week 5 - Centralisers and
Centres)

Asynchronous Tasks: [2.5 hours of directed private study]

 READ: Lecture 11 [∼ 2.5 pages; Suggested study time = 50 mins]

OR
WATCH: Lecture Video 11.1: What is a cyclic group? [13 mins 13s]
Lecture Video 11.2: Subgroups of finite cyclic groups [17 mins 36s]

 CHECK UNDERSTANDING: Lecture 11 Quiz [Suggested study time = 10 minutes]

 PRACTICE AND REFLECT: Do Exercises 35-40.

Take some time to consider which activities you found straightforward, and which
you found more challenging. Post any queries to the discussion board before the
next review session. [Suggested study time = 1 21 hours]
c University of Manchester Algebraic Structures 1 Week 7, Lecture 11, Page 1

Lecture 11
Cyclic groups

11.1 What is a cyclic group?

Let G be a group and recall from Definition 7.7 that we defined the cyclic subgroup
generated by an element a ∈ G as the subgroup of powers of a, i.e.

hai = {ak | k ∈ Z} = {. . . , a−2 , a−1 , a0 = e, a, a2 , a3 , . . .}.

Now we look at the situation where the whole group G is generated by an element a.

Definition 11.1 (Cyclic group; generator of a cyclic group). A group G is called cyclic,
if there exists an element a ∈ G such that G = hai. In other words, G is cyclic if all its
elements are powers of a single element a. The element a is called a generator of G.

Example 11.2. Z is an infinite cyclic group: Z = h1i = h−1i. Indeed, every n ∈ N has
· · + 1} = n1. Similarly, −n = (−1) + · · · + (−1) = (−n)1. This shows
the form n = |1 + ·{z
n times
that Z = h1i, as every integer is a multiple of 1. A similar argument shows that Z = h−1i:
n = (−(−1)) + · · · + (−(−1)) = (−n)(−1) and −n = (−1) + · · · + (−1) = n(−1).

Example 11.3. For any n ∈ N, the group Zn = h1i is a cyclic group of order n. The
argument is similar to the last example: any k ∈ Zn has the form k = |1 ⊕ ·{z
· · ⊕ 1} = k1.
k times
In particular, this example shows that for any n ∈ N there exists a finite cyclic group of
order n.

Theorem 11.4. Any subgroup of a cyclic group is itself cyclic.

Proof. Let G = hai be a cyclic group with generator a, and let H be a subgroup of G. If
H = {e}, then H = hei is cyclic with generator e. Now suppose H 6= {e}. Then H
contains some non-zero power of a, say ak . Being a subgroup, H also contains the inverse
of ak , that is a−k . Since one of k and −k is positive, we conclude that H contains a
positive power of a.
Now let m be the smallest positive integer such that am ∈ H. We will show that
H = ham i.
Clearly, since H is a group it contains all powers of am , so ham i ⊆ H. Hence, we need to
show that H ⊆ ham i. Let h be an arbitrary element of H. Then h = as for some s ∈ Z.
Using long division, write s = qm + r with 0 ≤ r < m. Then we have

h = as = aqm+r = (am )q ar .
c University of Manchester Algebraic Structures 1 Week 7, Lecture 11, Page 2

It follows that
ar = (am )−q h.
Consider the right-hand side of the last equation. Here am ∈ H, and hence (am )−q ∈ H.
Moreover, h ∈ H by hypothesis. Since H is a subgroup, it is closed under products, and
so (am )−q h ∈ H. But this means ar ∈ H. Now recall that r < m, and m was the smallest
positive integer such that am ∈ H. Hence, r = 0. But then h = (am )q , and thus h ∈ ham i.
We have shown that H ⊆ ham i, as required.

11.2 Subgroups of finite cyclic groups

Suppose the G = hai is a finite cyclic group of order n, so by Corollary 8.4, the order of
the generator a is n. Recall that this means that n is the smallest natural number with
an = e. By Theorem 11.4, any subgroup H of G is itself cyclic, so it is of the form
H = ham i for some integer m. In fact, we can be more precise.

Lemma 11.5 (Finding all cyclic subgroups of a finite cyclic group). Let G = hai be a
finite cyclic group of order n. Then, for any integer m, ham i = had i where d = gcd(m, n)
is the greatest common divisor of m and n.

Proof. Since d divides m, am is a power of ad , and we clearly have ham i ⊆ had i. It

remains to prove the inverse inclusion. Recall that you can use the Euclidean algorithm
to write the greatest common divisor d = gcd(m, n) as a linear combination of m and n.
Hence, there exist integers s and t such that d = sm + tn. But then

ad = asm+tn = (am )s (an )t = (am )s (e)t = (am )s ,

so ad is a power of am . Consequently, had i ⊆ ham i, and the result follows.

The following two corollaries use the same set-up as the lemma: G = hai is a finite cyclic
group of order n, we are moreover given a natural number m ∈ N and write
d = gcd(m, n).

Corollary 11.6 (Finding the order of an element in a finite cyclic group).

For m ∈ N, the order of am is equal to the order of ad , namely n/d.

Proof. We have |am | = |ham i| = |had i| = |ad | by Corollary 8.4. Since d is a divisor of n,
0
we can write n = dd0 for some d0 ∈ N. Then (ad )d = an = e and d0 is the smallest natural
number with this property. Hence, the order of ad is d0 = n/d.

Corollary 11.7 (Finding alternative generators of a finite cyclic group).

The element am ∈ G is a generator of G if and only if gcd(m, n) = 1.

Proof. By the last corollary, the subgroup ham i has order n/d. Hence, it coincides with G
if and only if d = 1.
c University of Manchester Algebraic Structures 1 Week 7, Lecture 11, Page 3

Another consequence of Lemma 11.5 is the following.

Theorem 11.8 (Finding all subgroups of a finite cyclic group).

The subgroups of a finite cyclic group G of order n are in one-to-one correspondence with
the divisors of n.
More precisely, if G = hai, then the subgroups of G are exactly the groups had i, for all
divisors d of n, and all of these groups are distinct.

Proof. Let a be a generator for G. Then, by Theorem 11.4 and Lemma 11.5, every
subgroup of G is of the form had i where d is a divisor of n. Also, if d1 and d2 are distinct
6 had2 i, as they have different orders, namely n/d1 and n/d2 ,
divisors of n, then had1 i =
respectively.
In view of Corollary 11.6, the order of a subgroup of a finite cyclic group G divides the
order of G. This is a special case of a very important theorem, called Lagrange’s
theorem, which states that if G is any finite group and H is a subgroup of G, then the
order of H divides the order of G. This theorem will be our next major goal for this
course. To understand the theorem, we need to introduce the concept of a coset. This is
the topic of our next lecture.
c University of Manchester Algebraic Structures 1 Week 7, Exercises, Page 1

Week 7: Exercises

Guidance: This week I would like you to practice computing the generators and
subgroups of some familiar cyclic subgroups. Questions 39 and 40 are more challenging.

Cyclic groups
(requires material up to Lecture 11)

35. Which elements of the cyclic group (Z30 , +) are generators?

36. Find all orders of subgroups of (i) Z31 , (ii) Z32 , (iii) Z33 .

37. Find all subgroups of Z15 .

38. Find all subgroups of the subgroup h(1 2 3)(4 5)i of S5 .

39. If G is a cyclic group of order n, and m divides n, show that G contains a subgroup
of order m.

40. Prove that (Q, +) is not a cyclic group.

Solutions to the exercises can be found here.

c University of Manchester Algebraic Structures 1 Week 8 Checklist

Resources for Week 8

Cosets, Lagrange’s Theorem

(Click here to return to the front page navigation)

This week we introduce the notion of cosets and use this idea to prove Lagrange’s
theorem, a fundamental result at the very heart of group theory and arguably the most
important theorem in this course. It is important to get a good understanding of cosets;
we will use this idea later on in the course when we come to define factor groups. By the
end of this week, you should be able to explain what a coset is (ILO2), describe
fundamental properties of cosets (ILO4), determine cosets in certain groups (ILO6), and
understand the statement and proof Lagrange’s theorem (ILO7).
Synchronous Sessions:[2 hours of timetabled study]
 REVIEW: Attend Review Session and Tutorial (Topic: Week 7 - Centralisers and
Centres)
Asynchronous Tasks: [4 hours of directed private study]

 READ: Lecture 12 [∼ 2 pages; Suggested study time = 50 mins]

OR
WATCH: Lecture Video 12.1: What is a coset? [22 mins 6s]
Lecture Video 12.2: Equal cosets [14 mins 6s]

 CHECK UNDERSTANDING: Lecture 12 Quiz [Suggested study time = 10 minutes]

 READ: Lecture 13 [∼ 2 pages; Suggested study time = 50 mins]

OR
WATCH: Lecture Video 13.1: Some facts about cosets [15 mins 14s]
Lecture Video 13.2: Lagrange’s theorem [14 mins 40s]

 CHECK UNDERSTANDING: Lecture 13 Quiz [Suggested study time = 10 minutes]

 PRACTICE AND REFLECT: Do Exercises 41-48.

Take some time to consider which activities you found straightforward, and which
you found more challenging. Post any queries to the discussion board before the
next review session. [Suggested study time = 1 21 hours]
c University of Manchester Algebraic Structures 1 Week 8, Lecture 12, Page 1

Lecture 12
Cosets

12.1 What is a coset?

Definition 12.1 (Coset). Let G be a group, let H ≤ G be a subgroup of G, and let g ∈ G
be an arbitrary element. The subset

gH = {gh : h ∈ H} ⊆ G

is called a left coset of H in G. That is, the left coset gH is the set of all products gh
where g is our fixed element and h runs over all elements of the subgroup H.

Remark 12.2.

1. If H = {h1 , h2 , h3 , . . .}, then the left coset of H in G with respect to g ∈ G is the set

gH = {gh1 , gh2 , gh3 , . . .}.

2. Similarly, one could define right cosets Hg = {hg : h ∈ H}, but we won’t use them
in this course. We will often just say “coset” instead of “left coset”.

3. For groups (G, +) in additive notation, a left coset of H in G takes the form

g + H = {g + h : h ∈ H}.

Example 12.3. Consider Z and its subgroup h3i = {. . . , −6, −3, 0, 3, 6, 9, . . .} = 3Z. For
a ∈ Z, the left coset a + h3i is

a + h3i = {. . . , a − 6, a − 3, a, a + 3, a + 6, a + 9, . . .}.

In particular, note that (a + 3) + h3i = a + h3i. More generally,

(a + q × 3) + h3i = a + h3i for all q ∈ Z. Hence, we have 3 distinct cosets of h3i in Z:

 {. . . , −6, −3, 0, 3, 6, 9, . . .} = 0 + h3i = h3i if a ≡ 0 mod 3,
a + h3i = {. . . , −5, −2, 1, 4, 7, 10, . . .} = 1 + h3i if a ≡ 1 mod 3,
{. . . , −4, −1, 2, 5, 8, 11, . . .} = 2 + h3i if a ≡ 2 mod 3.


Example 12.4. Consider the group Z15 and its subgroup h5i = {0, 5, 10}. For k ∈ Z15 ,
the left coset k ⊕ h5i is
k ⊕ h5i = {k, k ⊕ 5, k ⊕ 10}.
c University of Manchester Algebraic Structures 1 Week 8, Lecture 12, Page 2

So we have 5 distinct cosets of h5i in Z15 :



 {0, 5, 10} = 0 ⊕ h5i = h5i, if k ∈ {0, 5, 10},
 {1, 6, 11} = 1 ⊕ h5i, if k ∈ {1, 6, 11},


k ⊕ h5i = {2, 7, 12} = 2 ⊕ h5i, if k ∈ {2, 7, 12},
{3, 8, 13} = 3 ⊕ h5i, if k ∈ {3, 8, 13},




{4, 9, 14} = 4 ⊕ h5i, if k ∈ {4, 9, 14}.


Example 12.5. Consider the group S3 and its subgroup h(1 2)i = {e, (1 2)}. Here we get

eh(1 2)i = {ee, e(1 2)} = {e, (1 2)},

(1 2)h(1 2)i = {(1 2)e, (1 2)(1 2)} = {(1 2), e},
(1 3)h(1 2)i = {(1 3)e, (1 3)(1 2)} = {(1 3), (1 2 3)},
(2 3)h(1 2)i = {(2 3)e, (2 3)(1 2)} = {(2 3), (1 3 2)},
(1 2 3)h(1 2)i = {(1 2 3)e, (1 2 3)(1 2)} = {(1 2 3), (1 3)},
(1 3 2)h(1 2)i = {(1 3 2)e, (1 3 2)(1 2)} = {(1 3 2), (2 3)}.

Hence, we have three distinct cosets of h(1 2)i in S3 :

eh(1 2)i = (1 2)h(1 2)i = {e, (1 2)}

(1 3)h(1 2)i = (1 2 3)h(1 2)i = {(1 3), (1 2 3)}
(2 3)h(1 2)i = (1 3 2)h(1 2)i = {(2 3), (1 3 2)}

12.2 Equal cosets

Remark 12.6.

1. The subgroup H itself is always one of the cosets of H in G, since eH = H. In fact,

for all h ∈ H, we have hH = H. Indeed, of course we have hH ⊆ H, as H is closed
under multiplication. For the inverse inclusion, note that if k ∈ H, then h−1 k ∈ H
and thus k = h(h−1 k) ∈ hH.

2. For all g ∈ G, we have g ∈ gH, because g = ge ∈ gH.

The following lemma gives conditions under which two cosets are equal.

Lemma 12.7 (How to tell if two elements determine the same coset).
Let G be a group, H ≤ G and x, y ∈ G. Then

xH = yH ⇐⇒ y ∈ xH ⇐⇒ x−1 y ∈ H.

Proof. Let us start with the first equivalence: If xH = yH then clearly y ∈ yH = xH.
Now suppose that y ∈ xH, so y = xh1 for some h1 ∈ H. Let z ∈ yH, so z = yh2 for some
h2 ∈ H. In total, z = yh2 = xh1 h2 ∈ xH, as H is closed under multiplication. Hence,
c University of Manchester Algebraic Structures 1 Week 8, Lecture 12, Page 3

yH ⊆ xH. Moreover, let w ∈ xH, then w = xh3 for some h3 ∈ H. Then

w = xh3 = yh−1 1 h3 ∈ yH, and thus xH ⊆ yH. In total, we have shown that xH = yH, as
desired.
For the second equivalence, note that y ∈ xH if and only if y = xh for some h ∈ H, if and
only if x−1 y = h ∈ H.
c University of Manchester Algebraic Structures 1 Week 8, Lecture 13, Page 1

Lecture 13
Lagrange’s theorem

13.1 Some facts about cosets

We need two lemmas, and then the theorem will follow easily.
Lemma 13.1 (Cosets are equal or disjoint). Let G be a group, H ≤ G, and x, y ∈ G.
Then
either xH = yH or xH ∩ yH = ∅.
Proof. Assume that xH ∩ yH 6= ∅. We need to show that then xH = yH. Since
xH ∩ yH 6= ∅, there exists an element z ∈ xH ∩ yH. Then z ∈ xH and z ∈ yH, so
z = xh1 for some h1 ∈ H and z = yh2 for some h2 ∈ H.
Therefore, xh1 = yh2 , and hence y = xh1 h−1
2 ∈ xH. By Lemma 12.7, we get that
xH = yH.
Corollary 13.2 (The cosets of H in G form a partition of G).
If H ≤ G, then G is the disjoint union of the distinct left cosets of H in G.
Proof. Any element of G belongs to some left coset of H in G (for example, g ∈ gH), so G
is the union of those left cosets. Lemma 13.1 tells us that distinct cosets are disjoint.
Definition 13.3 (The index of a subgroup). The number of (distinct) left cosets of H in
G is called the index of H in G, and is denoted by [G : H].
Example 13.4. In Example 12.3, Example 12.4 and Example 12.5, we have seen that
[Z : h3i] = 3, [Z15 : h5i] = 5, [S3 : h(12)i] = 3.
The examples also illustrate what we have proved in Corollary 13.2, namely, that any
group G is the disjoint union of the left cosets of H in G. Moreover, they illustrate the
following lemma: all the cosets of a finite subgroup have the same number of elements.
Lemma 13.5 (Cosets of a subgroup H in a group G have the same cardinality).
Let H be a finite subgroup of a group G. Then |gH| = |H| for all g ∈ G.
Proof. We show that |gH| = |H| by defining a bijection between these two sets. Let
ϕ : H → gH be the map defined by ϕ(h) = gh for all h ∈ H.
This map is onto: Indeed, if x ∈ gH, then by definition x = gh for some h ∈ H and
hence x = ϕ(h).
This map is one-to-one: Suppose that ϕ(h1 ) = ϕ(h2 ) for some h1 , h2 ∈ H. We claim
that this can only happen if h1 = h2 . Indeed, if ϕ(h1 ) = ϕ(h2 ) then we have gh1 = gh2 ,
and so by cancellation, h1 = h2 . This completes the proof.
c University of Manchester Algebraic Structures 1 Week 8, Lecture 13, Page 2

13.2 Lagrange’s theorem

Now we are ready to state and prove Lagrange’s theorem.
Theorem 13.6 (Lagrange’s theorem). Let G be a finite group and H ≤ G. Then
|G| = [G : H]|H|
where [G : H] is the index of H in G, i.e. the number of distinct left-cosets of H in G.
Proof. Let [G : H] = r and let
g1 H, g2 H, . . . , gr H
be the distinct left cosets of H in G. By Corollary 13.2, the group G is the disjoint union
of those cosets:
G = g1 H ∪ g2 H ∪ · · · ∪ gr H.
Since the union is disjoint we have
|G| = |g1 H| + |g2 H| + · · · + |gr H|.
From Lemma 13.5, we know that |gi H| = |H| for i = 1, 2, . . . , r. Hence
|G| = |H| + |H| + · · · + |H| = r|H| = [G : H]|H|.
| {z }
r

So |G| = [G : H]|H| as required.

Corollary 13.7 (Order of a subgroup of a finite group divides the order of the group).
If G is a finite group and H ≤ G, then the order of the subgroup H divides the order of
the group G.
Corollary 13.8 (Order of an element of a finite group divides the order of the group).
If G is a finite group and g ∈ G, then |g|, the order of the element g, divides |G|, the
order of the group G.
Proof. This is because by Corollary 8.4 the order of the element g is equal to the order of
the cyclic subgroup hgi.
Corollary 13.9 (|G|-th powers in a finite group G are all equal).
Let G be a finite group and g ∈ G. Then g |G| = e.
Proof. This follows from Corollary 13.8 and Lemma 8.3 (4).
The following nice corollary shows in particular that every group of prime order is
abelian.
Corollary 13.10 (Every group of prime order is cyclic). If G is a group with |G| = p,
where p is a prime, then G is cyclic.
Proof. Let g ∈ G with g 6= e. By Lagrange’s theorem, |hgi| divides |G|. Since |G| = p, a
prime, this gives that |hgi| = 1 or |hgi| = p. But the former is impossible since hgi
contains at least two elements, namely g and e. Hence |hgi| = p = |G| and therefore
hgi = G.
c University of Manchester Algebraic Structures 1 Week 8, Exercises, Page 1

Week 8: Exercises

Guidance: This week I would like you to practice computing cosets of a subgroup in one
of our favourite groups (as ever: you should know what the binary operation is in each
case). Questions 41-45 are all of this kind; do as many of these exercises as you need to
become comfortable with the notion of cosets. Questions 46-48 involve abstract reasoning
– make sure to try all of these.

Cosets
(requires material up to Lecture 12)

41. For each pair G, H (where H ≤ G), determine [G : H] and list all left cosets of H in
G.
(i) G = Z15 , H = h12i ,
(ii) G = {e, (1 2), (3 4), (1 2)(3 4)} (a subgroup of the symmetric group S4 ),
H = h(1 2)(3 4)i ,
(iii) G = R∗ = (R\ {0} , ×), H = (R+ , ×).

42. What are the left cosets of H = {z ∈ C\ {0} ; |z| = 1} in C∗ = (C\ {0} , ×)?

43. For H ≤ G as specified below, determine the left cosets of H in G, and in each case
find the index of H in G.
(i) G = R∗ , H = h−1i
(ii) G = C∗ , H = R∗
(iii) G = C∗ , H = R+
(iv) G = Z36 , H = h30i
(v) G = T(2, R), H = UT(2, R)
  
λ 0
(vi) G = D(2, R), H = ; λ ∈ R, λ 6= 0
0 λ
(vii) G = GL(n, R), H = SL(n, R) (Hint: Show that the left coset determined by a
matrix A ∈ GL(n, R) with determinant det A = a is the set of all n × n matrices B
with determinant det B = a.)
c University of Manchester Algebraic Structures 1 Week 8, Exercises, Page 2

44. Prove that the left cosets of Z in R are in 1-1 correspondence with the set
[0, 1) = {x ∈ R | 0 ≤ x < 1} .

45. In the symmetric group S4 , work out the cosets

(i) (1 3 4)H, (ii) (2 3)H, (iii) (1 4 3 2)H,
where H = h(1 2 3 4)i is the cyclic subgroup generated by (1 2 3 4).

Lagrange’s theorem
(requires material up to Lecture 13)

46. Let G be a finite group and K ≤ H ≤ G. Prove that [G : K] = [G : H] [H : K] .

47. Let G be a finite group with |G| = n. Prove that g n = e for all g ∈ G.

48. Let G be a finite group and K, H ≤ G with gcd(|K| , |H|) = 1, i.e. the orders of the
subgroups K and H are coprime. Prove that H ∩ K = {e} .

Solutions to the exercises can be found here.

c University of Manchester Algebraic Structures 1 Week 9 Checklist

Resources for Week 9

Homomorphisms

(Click here to return to the front page navigation)

This week we introduce the notion of a group homomorphism. We pay special attention
to the bijective group homomorphisms; such maps are called isomorphisms. We show
that isomorphisms preserve many interesting properties of a group. By the end of this
week, you should be able to explain the concepts of homomorphism, isomorphism and
group theoretic properties (ILO2) and be able to give examples of these.
Synchronous Sessions:[2 hours of timetabled study]

 REVIEW: Attend Review Session and Tutorial (Topic: Week 8 - Cosets and
Lagrange’s Theorem)

Asynchronous Tasks: [4 hours of directed private study]

 READ: Lecture 14 [∼ 3 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 14.1: What is a homomorphism? [19 mins 22s]
Lecture Video 14.2: Elementary properties of homomorphisms [9 mins]

 CHECK UNDERSTANDING: Lecture 14 Quiz [Suggested study time = 10 minutes]

 READ: Lecture 15 [∼ 3.5 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 15.1: What is an isomorphism? [25 mins 50s]
Lecture Video 15.2: Group theoretic properties [16 mins 42s]

 CHECK UNDERSTANDING: Lecture 15 Quiz [Suggested study time = 10 minutes]

 PRACTICE AND REFLECT: Do Exercises 49-53.

Take some time to consider which activities you found straightforward, and which
you found more challenging. Post any queries to the discussion board before the
next review session. [Suggested study time = 1 21 hours]
c University of Manchester Algebraic Structures 1 Week 9, Lecture 14, Page 1

Lecture 14
Group homomorphisms

14.1 What is a homomorphism?

In general, homomorphisms of mathematical objects are functions between them that
behave nicely with respect to the intrinsic properties of these objects. For groups,
homomorphisms should respect the group operations.

Definition 14.1 (Group homomorphism). A (group) homomorphism from a group G to

a group H is a function ϕ : G → H such that

∀a, b ∈ G, ϕ(ab) = ϕ(a)ϕ(b).

Remark 14.2. The definition of a homomorphism involves two group operations, one in
G and one in H. If (just for the moment) the groups are (G, ∗) and (H, ⊗), the condition
reads
∀a, b ∈ G, ϕ(a ∗ b) = ϕ(a) ⊗ ϕ(b).

Example 14.3. Consider the map ϕ : R → GL(2, R) given by

 
1 x
ϕ(x) = .
0 1

We show that this map is a group homomorphism. The operations on R and GL(2, R) are
+ and ×, so a homomorphism must satisfy

∀x, y ∈ R, ϕ(x + y) = ϕ(x)ϕ(y).

We have:
 
1 x+y
ϕ(x + y) = ,
0 1
    
1 x 1 y 1 x+y
ϕ(x)ϕ(y) = = .
0 1 0 1 0 1

Hence, ϕ is a homomorphism.

Example 14.4. Consider the map ϕ : R → C∗ given by

ϕ(x) = cos x + i sin x.

c University of Manchester Algebraic Structures 1 Week 9, Lecture 14, Page 2

Here the homomorphism condition is

∀x, y ∈ R, ϕ(x + y) = ϕ(x)ϕ(y).

Using trigonometric identities, we get:

ϕ(x + y) = cos(x + y) + i sin(x + y),

ϕ(x)ϕ(y) = (cos x + i sin x)(cos y + i sin y)
= cos x cos y − sin x sin y + i(cos x sin y + sin x cos y)
= cos(x + y) + i sin(x + y).

Hence, ϕ is a homomorphism.

Example 14.5. Consider the map ϕ : GL(n, R) → R∗ given by

ϕ(A) = det A.

This is a homomorphism since for all A, B ∈ GL(n, R),

ϕ(AB) = det(AB) = (det A)(det B) = ϕ(A)ϕ(B).

Example 14.6. Let G be any group, let g ∈ G, and let ϕ : Z → G be given by ϕ(n) = g n .
Then ϕ is a homomorphism. Indeed, for all m, n ∈ Z we have

ϕ(m + n) = g m+n = g m g n = ϕ(m)ϕ(n).

Example 14.7. Let G be a group, n ∈ N, and let g ∈ G be an element such that g n = e

(for example, take G = Sn and g = (1 2 · · · n)). We show that the map ϕ : Zn → G
defined by
ϕ(m) = g m
is a homomorphism. The condition that we need to verify is that

∀a, b ∈ Zn , ϕ(a ⊕ b) = ϕ(a)ϕ(b).

Let a, b ∈ Zn and using long division write a + b = qn + r, with q ∈ Z and r ∈ Zn . Then

a ⊕ b = r, and we have

ϕ(a ⊕ b) = ϕ(r) = g r ,
ϕ(a)ϕ(b) = g a g b = g a+b = g qn+r = g qn g r = g r .

Hence, ϕ is a homomorphism.
c University of Manchester Algebraic Structures 1 Week 9, Lecture 14, Page 3

14.2 Elementary properties of homomorphisms

Recall that the image of a map ϕ : G → H is the set Im ϕ = ϕ(G) = {ϕ(g) | g ∈ G}.

Lemma 14.8 (Properties of homomorphisms). Let ϕ : G → H be a group

homomorphism. Then

1. ϕ maps the identity element of G to the identity element of H: ϕ(eG ) = eH .

2. For all a ∈ G, ϕ(a−1 ) = (ϕ(a))−1 .

3. The image of ϕ is a subgroup of H: Im ϕ ≤ H.

Proof. (1): We have

ϕ(eG )ϕ(eG ) = ϕ(eG eG ) = ϕ(eG ) = ϕ(eG )eH ,

and then cancellation on the left gives ϕ(eG ) = eH .

(2): We have (using (1))

ϕ(a−1 )ϕ(a) = ϕ(a−1 a) = ϕ(eG ) = eH .

Hence
ϕ(a−1 )ϕ(a) = eH .
Then multiplying by (ϕ(a))−1 on the right gives

ϕ(a−1 ) = (ϕ(a))−1 .

(3): Since G 6= ∅, we also have Im ϕ 6= ∅, so we can apply the subgroup criterion. Let
x, y ∈ Im ϕ, i.e. there exist a, b ∈ G such that x = ϕ(a) and y = ϕ(b). Then

xy = ϕ(a)ϕ(b) = ϕ(ab) ∈ Im ϕ

and
x−1 = (ϕ(a))−1 = ϕ(a−1 ) ∈ Im ϕ.
Hence, Im ϕ ≤ H.
c University of Manchester Algebraic Structures 1 Week 9, Lecture 15, Page 1

Lecture 15
Isomorphisms and group-theoretic properties

15.1 What is an isomorphism?

Building on our investigation of group homomorphisms from the last lecture, we now
define group isomorphisms.

Definition 15.1 (Group isomorphism). A bijective group homomorphism ϕ : G → H is

called a (group) isomorphism.
We say that a group G is isomorphic to a group H (notation: G ∼
= H) if there exists an
isomorphism ϕ : G → H.

If ϕ : G → H is an isomorphism, then one can think of the groups G and H as “the

same” for all group theoretic purposes. Indeed, you can get H just by “renaming” the
elements of G using ϕ.

Example 15.2. Consider the group UT(2, R) of upper unitriangular (2 × 2)-matrices.

We show that UT(2, R) ∼
= R. The map ϕ : UT(2, R) → R given by
 
1 x
ϕ =x
0 1

is an isomorphism. Indeed, ϕ is obviously 1-1 and onto, hence bijective, and it is a

homomorphism because
    
1 x 1 y 1 x+y
ϕ = ϕ
0 1 0 1 0 1
= x+y
   
1 x 1 y
= ϕ +ϕ .
0 1 0 1

Example 15.3. We show that S2 ∼ = Z2 . The map ϕ : S2 → Z2 given by ϕ(id2 ) = 0,

ϕ((1 2)) = 1 is an isomorphism.
Indeed, this is of course 1-1 and onto, hence bijective. It is a homomorphism, because

ϕ(id2 id2 ) = ϕ(id2 ) = 0 = 0 ⊕ 0 = ϕ(id2 ) ⊕ ϕ(id2 ),

ϕ(id2 (1 2)) = ϕ((1 2)) = 1 = 0 ⊕ 1 = ϕ(id2 ) ⊕ ϕ((1 2)),
ϕ((1 2) id2 ) = ϕ((1 2)) = 1 = 1 ⊕ 0 = ϕ((1 2)) ⊕ ϕ(id2 ),
ϕ((1 2)(1 2)) = ϕ(id2 ) = 0 = 1 ⊕ 1 = ϕ((1 2)) ⊕ ϕ((1 2)).
c University of Manchester Algebraic Structures 1 Week 9, Lecture 15, Page 2

Example 15.4. The group R of real numbers under addition is isomorphic to the group
R+ of positive real numbers under multiplication: R ∼
= R+ . The map ϕ : R → R+ given
x
by ϕ(x) = e for all x ∈ R is an isomorphism.
Indeed, this map is 1-1 and onto, and it is a homomorphism since

ϕ(x + y) = ex+y = ex ey = ϕ(x)ϕ(y).

Lemma 15.5 (The inverse of an isomorphism is an isomorphism). If ϕ : G → H is a

group isomorphism, then the unique inverse map ϕ−1 : H → G with

ϕ(a) = x ⇐⇒ ϕ−1 (x) = a ∀a ∈ G, x ∈ H

is also an isomorphism.
Proof. Since ϕ is bijective, the same holds for ϕ−1 . Let x, y ∈ H. Since ϕ is onto, there
exist a, b ∈ G such that x = ϕ(a), y = ϕ(b). Then

ϕ−1 (xy) = ϕ−1 (ϕ(a)ϕ(b)) = ϕ−1 (ϕ(ab)) = ab = ϕ−1 (x)ϕ−1 (y).

Hence, ϕ−1 is a homomorphism and therefore an isomorphism.

The lemma shows that being isomorphic is symmetric: G ∼ = H if and only if H ∼ = G.
Next, we show that every cyclic group is isomorphic to Z (if it is infinite) or to Zn (if it is
finite of order n).
Theorem 15.6 (Classification of cyclic groups up to isomorphism). Let G = hai be a
cyclic group with generator a.
1. If |G| = ∞ then ϕ : Z → G given by ϕ(m) = am is an isomorphism.

2. If |G| = n ∈ N then ϕ : Zn → G given by ϕ(m) = am is an isomorphism.

Proof. (1): ϕ is a homomorphism by Example 14.6. Since G = hai, every element of G is
a power of a, hence ϕ is onto. The element a must have infinite order, as otherwise
|G| < ∞. By Lemma 8.1, all powers of a are distinct, so ϕ is 1-1, hence bijective and an
isomorphism.
(2): Since |a| = |hai| = |G| = n, Example 14.7 shows that ϕ is a group homomorphism.
Moreover, by Corollary 8.4, we have G = {a, a2 , . . . , an = e}, and hence ϕ is onto. Since
all these powers of a are distinct, ϕ is 1-1, hence bijective and an isomorphism.

15.2 Group theoretic properties

Definition 15.7 (Group theoretic property). A property P of groups is called a group
theoretic property if it is common to all isomorphic groups, in other words, whenever a
group G has the property P and G ∼ = H, then H too has the property P.
Example 15.8. All of the following properties are group-theoretic.
c University of Manchester Algebraic Structures 1 Week 9, Lecture 15, Page 3

1. being finite

2. being infinite

3. being abelian

4. being cyclic

5. having an element of infinite order

6. having an element of given order n

7. having a trivial centre

8. having a non-trivial centre.

Let us prove for some of these properties that they are indeed group-theoretic, the
arguments for the other properties are similar.
Proof. Suppose G ∼ = H and ϕ : G → H is an isomorphism.
(1) and (2): If G is finite (infinite), then H must be finite (infinite) since ϕ is a bijection.
(3): Suppose G is abelian. Let x, y ∈ H. Since ϕ is onto, there exist a, b ∈ G such that
x = ϕ(a), y = ϕ(b). Then xy = ϕ(a)ϕ(b) = ϕ(ab) = ϕ(ba) = ϕ(b)ϕ(a) = yx, hence H is
abelian.
(7): Suppose Z(G) = {e}. Let x ∈ H with x 6= e. Then x = ϕ(a) for some a ∈ G with
a 6= e. Since Z(G) = {e}, there exists an element b ∈ G such that ab 6= ba. Let y = ϕ(b).
Then xy 6= yx. Indeed, otherwise we would have

ϕ(ab) = ϕ(a)ϕ(b) = xy = yx = ϕ(b)ϕ(a) = ϕ(ba),

and since ϕ is 1-1, this gives ab = ba, a contradiction. Hence xy 6= yx, and since x was an
arbitrary non-identity element, this gives Z(H) = {e}.

Example 15.9. Here are some examples of properties that are not group theoretic:

• being a group consisting of matrices

• being a group of numbers

• being a group of permutations

In algebra, isomorphic objects are generally considered the same. The ideal goal of group
theory is to classify all groups “up to isomorphism”, meaning that we would have a list of
groups that we know very well, and any group in the world is isomorphic to one of the
groups in our list. Unfortunately (or fortunately, depending on your point of view) this
can never be achieved.
How can we prove that two groups G and H are isomorphic? We need to find an
isomorphism from G to H.
c University of Manchester Algebraic Structures 1 Week 9, Lecture 15, Page 4

However, to prove that G and H are not isomorphic, we can use group theoretic
properties. If we can find a group theoretic property that G possesses, but H not, then
we know that G and H are not isomorphic.

Example 15.10.

• Z8  Z10 , because they have different orders,

• S3  Z6 , because Z6 is abelian but S3 is not,

• Z4  {e, (12), (34), (12)(34)} ≤ S4 , because Z4 has an element of order 4, and the
other group doesn’t.
c University of Manchester Algebraic Structures 1 Week 9, Exercises, Page 1

Week 9: Exercises

Guidance: This week I would like you to practice determining whether a given function
is a homomorphism of groups, and (by using group-theoretic properties) whether two
groups are isomorphic. The problems in Questions 49 and 53 are all of this kind.
Questions 50-52 involve more abstract reasoning – make sure to try each of these.

Group homomorphisms
(requires material up to Lecture 14)

49. Which of the following maps are group homomorphisms?

(a) ϕ : C∗ → R, ϕ(z) = log |z| (z ∈ C∗ )

(b) ϕ : C → R, ϕ(a + bi) = b (a, b ∈ R)
 
a b
(c) ϕ : GL(2, R) → R, ϕ = a − d, a, b, c, d ∈ R.
c d
   
a b a 0
(d) ϕ : T(2, R) → D(2, R), ϕ = , a, b, d ∈ R, ad 6= 0
0 d 0 d
 
1 2 ... n
(e) ϕ : Sn → Sn+1 , where for f = ∈ Sn ,
f (1) f (2) ... f (n)
 
1 2 ... n n+1
ϕ (f ) = ∈ Sn+1
f (1) f (2) ... f (n) n + 1

(f) ϕ : Z → Sn , ϕ (k) = (1 2 3 · · · n)k ∈ Sn , k ∈ Z

50. Let ϕ : G → H be a homomorphism of groups. Show that if a ∈ G has order n,

then ϕ (a) ∈ H has order dividing n.

51. Let G be a group. Show that the map ϕ : G → G defined by ϕ (a) = a−1 is a
homomorphism if and only if G is abelian.

Isomorphisms and group-theoretic properties

(requires material up to Lecture 15)

52. Let G be a group and x ∈ G. Show that the map ϕx : G → G defined by

ϕx (a) = xax−1 for all a ∈ G (with x fixed) is an isomorphism.
c University of Manchester Algebraic Structures 1 Week 9, Exercises, Page 2

53. By using group-theoretic properties, show that the following statements are true.

(a) S4  Z24 .
(b) Z  Q.
(c) Q  Q∗ .
(d) GL(2, R)  UT(3, R).

Solutions to the exercises can be found here.

c University of Manchester Algebraic Structures 1 Week 10 Checklist

Resources for Week 10

Conjugacy

(Click here to return to the front page navigation)

This week we introduce the notion of conjugacy classes, and prove a very useful result,
known as the class formula. We describe the conjugacy classes in the symmetric group,
proving that two permutations are conjugate if and only if they have the same cycle type.
(Conjugation plays a key role in understanding normal subgroups – the second key
ingredient needed to define factor groups. We shall see this next week.) By the end of
this week, you should be able to explain the concept of conjugacy (ILO2) and determine
conjugacy classes in certain groups (ILO6).
Synchronous Sessions:[2 hours of timetabled study]
 REVIEW: Attend Review Session and Tutorial (Topic: Week 9 - Cosets and
Lagrange’s Theorem)
Asynchronous Tasks: [4 hours of directed private study]

 READ: Lecture 16 [∼ 3 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 16.1: What is conjugacy? [21 mins 32s]
Lecture Video 16.2: Conjugacy and the class formula [14 mins 55s]

 CHECK UNDERSTANDING: Lecture 16 Quiz [Suggested study time = 10 minutes]

 READ: Lecture 17 [∼ 4 pages; Suggested study time = 1 hour]

OR
WATCH: Lecture Video 17.1: How to conjugate in Sn [21 mins 17s]
Lecture Video 17.2: Conjugacy classes in Sn [19 mins 53s]

 CHECK UNDERSTANDING: Lecture 17 Quiz [Suggested study time = 10 minutes]

 PRACTICE AND REFLECT: Do Exercises 54-59.

Take some time to consider which activities you found straightforward, and which
you found more challenging. Post any queries to the discussion board before the
next review session. [Suggested study time = 1 21 hours]
c University of Manchester Algebraic Structures 1 Week 10, Lecture 16, Page 1

Lecture 16
Conjugacy

16.1 What is conjugacy?

Definition 16.1 (Conjugate). Let G be a group and let a, b ∈ G. We say that b is a
conjugate of a, and write b ∼ a, if

b = xax−1 for some x ∈ G.

We also say that b is conjugate to a by x.

Remark 16.2. Note that ∼ is an equivalence relation on G:

reflexive ∀a ∈ G, a ∼ a, as a = eae−1 ,

symmetric if a ∼ b, then a = xbx−1 for some x ∈ G. Hence,

b = x−1 ax = (x−1 )a(x−1 )−1 , so b ∼ a.

transitive if a ∼ b and b ∼ c, then a = xbx−1 and b = ycy −1 for some x, y ∈ G. Then

a = xbx−1 = x(ycy −1 )x−1 = (xy)c(xy)−1 , so a ∼ c.

An equivalence class of ∼ is a subset of G whose elements are all conjugate to each other.
Recall that, as ∼ is an equivalence relation, G is the disjoint union of all the distinct
equivalence classes of ∼.

Definition 16.3 (Conjugacy classes). The equivalence classes of the equivalence relation
∼ are called the conjugacy classes of G. We denote the conjugacy class of an element
a ∈ G by
aG = {xax−1 | x ∈ G}.

Example 16.4 (Conjugacy in abelian groups). Let G be an abelian group and a ∈ G.

Then xax−1 = a for all x ∈ G. Hence, the only element conjugate to a is a itself, and
therefore
aG = {a} for all a ∈ G.

Example 16.5 (Conjugacy class of the identity element). In any group G, we have
xex−1 = xx−1 = e for all x ∈ G. Hence, we always have eG = {e}.
c University of Manchester Algebraic Structures 1 Week 10, Lecture 16, Page 2

Example 16.6 (Conjugacy in S3 ). Consider

S3 = {id3 , (1 2), (1 3), (2 3), (1 2 3), (1 3 2)}. By the last example,

idS3 3 = {id3 }.

Now let us work out the conjucacy class of the 3-cycle (1 2 3):

id3 (1 2 3) id3−1 = (1 2 3),

(1 2 3)(1 2 3)(1 2 3)−1 = (1 2 3),
(1 3 2)(1 2 3)(1 3 2)−1 = (1 2 3)2 (1 2 3)(1 2 3)−2 = (1 2 3),
(1 2)(1 2 3)(1 2)−1 = (1 2)(1 2 3)(1 2) = (1 3 2),
(1 3)(1 2 3)(1 3)−1 = (1 3)(1 2 3)(1 3) = (1 3 2),
(2 3)(1 2 3)(2 3)−1 = (2 3)(1 2 3)(2 3) = (1 3 2).

Hence,
(1 2 3)S3 = {(1 2 3), (1 3 2)}.
What about the 2-cycle (1 2)? Here we have

id3 (1 2) id−1
3 = (1 2)
(1 2 3)(1 2)(1 2 3)−1 = (1 2 3)(1 2)(1 3 2) = (2 3),
(1 3 2)(1 2)(1 3 2)−1 = (1 3 2)(1 2)(1 2 3) = (1 3),
(1 2)(1 2)(1 2)−1 = (1 2),
(1 3)(1 2)(1 3)−1 = (1 3)(1 2)(1 3) = (2 3),
(2 3)(1 2)(2 3)−1 = (2 3)(1 2)(2 3) = (1 3).

Hence,
(1 2)S3 = {(1 2), (1 3), (2 3)}.
Our calculations confirm that S3 is indeed the disjoint union of its conjugacy classes:

S3 = 1S3 ∪ (1 2 3)S3 ∪ (1 2)S3 .

Example 16.7 (Conjugacy in the general linear group). Conjugacy in GL(n, R) appeared
in Linear Algebra, where two square matrices A and B were called similar, if there is an
invertible matrix C such that B = CAC −1 . Hence, conjugacy in GL(n, R) is the same as
similarity of invertible matrices. Results from Linear Algebra can be translated into
conjugacy language. For example, we have the following result:

Theorem (rephrasing of a result from last year’s Linear Algebra course).

An element A ∈ GL(n, R) is conjugate to a diagonal matrix if and only if the vector
space Rn has a basis consisting of eigenvectors of A.
c University of Manchester Algebraic Structures 1 Week 10, Lecture 16, Page 3

16.2 Conjugacy and the class formula

Remark 16.8. Here are some simple observations.

• For any a ∈ G, we have a ∈ aG , since a = eae−1 .

• aG = {a} if and only if a ∈ Z(G). Indeed, aG = {a} means that xax−1 = a for all x,
i.e. xa = ax for all x ∈ G.

• Like cosets, the conjugacy classes form a partition of the group.

• Unlike cosets, conjugacy classes can contain different numbers of elements. E.g. in
G = S3 we have seen that |eG | = 1, |(1 2)G | = 3, |(1 2 3)G | = 2.

What can we say about the number of elements in a given conjugacy class? There is a
very beautiful answer to this question, the so-called class formula. Recall the the
centraliser of an element a ∈ G is the subgroup defined as

C(a) = {x ∈ G : xa = ax}.

Theorem 16.9 (Class formula). Let G be a group and a ∈ G. Then

|aG | = [G : C(a)],

i.e. the number of conjugates of a is equal to the index of the centraliser of a in G.

Proof. We will show that gag −1 = hah−1 with g, h ∈ G if and only if g and h are in the
same left coset of C(a) in G. Once this is proved we can conclude that the number of
conjugates of a is equal to the number of left cosets of C(a) in G, which is the desired
result. Now,

gag −1 = hah−1 ⇔ h−1 ga = ah−1 g

⇔ h−1 g ∈ C(a)
⇔ g C(a) = h C(a), (by Lemma 12.7)

and the latter means that g and h are in the same left coset of C(a) in G.

Corollary 16.10 (Size of a conjugacy class in a finite group divides the group’s order).
Let G be a finite group. Then, for any a ∈ G, the number |aG | of conjugates of a divides
the order of G.

Proof. By Lagrange’s theorem, |G| = [G : C(a)]| C(a)|, hence |G| = |aG || C(a)|.
c University of Manchester Algebraic Structures 1 Week 10, Lecture 17, Page 1

Lecture 17
Conjugacy in symmetric groups

Now we take a closer look at conjugacy in Sn . First we obtain an instant method for
calculating conjugates in Sn .

17.1 How to conjugate in Sn

We start by showing that a conjugate of an m-cycle is again an m-cycle.
Lemma 17.1 (The conjugate of a cycle). Let f, g ∈ Sn where

g = (α1 α2 · · · αm )

is a cycle of length m (1 ≤ m ≤ n). Then

f gf −1 = (f (α1 ) f (α2 ) · · · f (αm )). (17.1)

That is, the conjugate of a cycle g of length m by an arbitrary permutation f is again a

cycle of length m with entries obtained by applying f to all entries of the original cycle g.
Example 17.2. Let g = (1 2 3 4), f = (1 4 2) ∈ S4 . Then

f gf −1 = (f (1) f (2) f (3) f (4)) = (4 1 3 2) = (1 3 2 4),

gf g −1 = (g(1) g(4) g(2)) = (2 1 3) = (1 3 2).

Proof of Lemma 17.1. We will show that the permutations on the RHS and LHS of
(17.1) have the same effect on all elements of Nn = {1, 2, . . . , n}, so they are equal. Let
a ∈ Nn . We distinguish between two cases.
Case 1: a is an entry in the cycle on the RHS of (17.1). Then a = f (αk ) for some k with
1 ≤ k ≤ m, and the RHS of (17.1) maps
(
f (αk+1 ), if k < m
a = f (αk ) 7→ = f (g(αk )).
f (α1 ), if k = m

Using that a = f (αk ), we find that the LHS of (17.1) maps

a = f (αk ) 7→ f gf −1 (a) = f gf −1 f (αk ) = f (g(αk )).

Hence, LHS and RHS of (17.1) have the same effect on a.

Case 2: a is not an entry in the cycle on the RHS of (17.1). Then a is fixed by the RHS
of (17.1). Write a = f (b). Then b is fixed by g, since the elements which are moved by g
c University of Manchester Algebraic Structures 1 Week 10, Lecture 17, Page 2

are exactly the αi , but a is not one of the f (αi ) so b is not one of the αi . Now the LHS of
(17.1) maps
a = f (b) 7→ f gf −1 (a) = f gf −1 f (b) = f (g(b)) = f (b) = a.
Hence again LHS and RHS have the same effect on a. This proves the lemma.
The lemma yields an instant method for computing conjugates in Sn : Let f, g ∈ Sn where
g is a product of (not necessarily disjoint) cycles, say

g = (α1 · · · αk )(β1 · · · βm ) · · · (γ1 · · · γr ). (17.2)

Then

f gf −1 = f (α1 · · · αk )(β1 · · · βm ) · · · (γ1 · · · γr )f −1

= f (α1 · · · αk )f −1 f (β1 · · · βm )f −1 · · · f (γ1 · · · γr )f −1
= (f (α1 ) · · · f (αk ))(f (β1 ) · · · f (βm )) · · · (f (γ1 ) · · · f (γr )),

where Lemma 17.1 was used for the last equality. Hence,

f gf −1 = (f (α1 ) · · · f (αk ))(f (β1 ) · · · f (βm )) · · · (f (γ1 ) · · · f (γr )). (17.3)

Example 17.3. Let f = (1 2 3)(4 5 6), g = (1 2 3 4)(5 6). Then

f gf −1 = (2 3 1 5)(6 4) = (1 5 2 3)(4 6)
gf g −1 = (2 3 4)(1 6 5).

These observations allow us to decide questions about conjugacy of permutations f by

looking at their cycle structure, i.e. how many cycles of any given length appear if f is
written as a product of disjoint cycles. Here, and whenever the cycle structure plays a
role, we always include all the elements fixed by f as 1-cycles in the decomposition of f
into disjoint cycles.
Definition 17.4 (Cycle structure of a permutation). Let n1 , . . . , nk be a sequence of
non-negative integers with nk > 0. A permutation f ∈ Sn has cycle structure
n1 , n2 , . . . , nk if the decomposition of f as a product of disjoint cycles, including 1-cycles
for all fixed elements, contains exactly ni cycles of length i (for all 1 ≤ i ≤ k).
Two permutations f, g ∈ Sn have the same cycle structure if, when f and g are written as
products of disjoint cycles, including 1-cycles for fixed elements, the number ni of cycles
of any given length i is the same in both decompositions.
Example 17.5. Consider the following permutations in S9 :

f = (1 2 3)(4 5)(6 7)(8 9), g = (1 2)(3 4)(5 6 7)(8 9), h = (1)(2 3)(4 5 6 7)(8 9).

Then f and g have the same cycle structure 0, 3, 1 (no 1-cycles, three 2-cycles and one
3-cycle), but h has a different cycle structure 1, 2, 0, 1 (a 1-cycle, two 2-cycles, no 3-cycle
and a 4-cycle).
c University of Manchester Algebraic Structures 1 Week 10, Lecture 17, Page 3

Lemma 17.6 (A property of conjugate permutations). Conjugate permutations in Sn

have the same cycle structure.
Proof. If the cycles on the RHS of (17.2) are disjoint, then the cycles on the RHS of
(17.3) are disjoint too, since f is one-to one. Hence, g and its conjugate f gf −1 have the
same cycle structure.

17.2 Conjugacy classes in Sn

We have seen that if permutations are conjugate, then they must have the same cycle
structure. In fact, the inverse implication is also true, yielding the following powerful
theorem.
Theorem 17.7 (How to tell if two elements of Sn are conjugate). Two permutations f
and g in Sn are conjugate if and only if they have the same cycle structure.
Proof. “only if”: This is Lemma 17.6.
“if”: Let f and g be permutations with the same cycle structure. Write
f = (α1 · · · αk ) · · · (β1 · · · βm )
g = (γ1 · · · γk ) · · · (δ1 · · · δm )
in such a way that cycles of equal length (including trivial cycles of length one for fixed
elements) are written above each other (this can be done since disjoint cycles commute).
Then let h be the permutation defined by
h(α1 ) = γ1 , h(α2 ) = γ2 , . . . , h(αk ) = γk , . . . , h(β1 ) = δ1 , . . . , h(βm ) = δm ,
i.e. h maps the entries of the cycles of f to the entries of the cycles of g written below
them. Then we have (using (17.3))
g = (h(α1 ) · · · h(αk )) · · · (h(β1 ) · · · h(βm )) = hf h−1 .
In other words, f ∼ g.
Example 17.8. Consider the following permutations in S9 :
f = (1 2)(3 4 5)(6)(7 8 9) and g = (1)(2 3 4)(5 6 7)(8 9).
These permutations have the same cycle structure, so they are conjugate by Theorem
17.7. Let us find a permutation h such that g = hf h−1 . As in the proof of Theorem 17.7,
we write f above g, with matching lengths of cycles:
f = (3 4 5)(7 8 9)(1 2)(6)
g = (2 3 4)(5 6 7)(8 9)(1)
Then h maps each α ∈ Nn to the number below it, giving
 
1 2 3 4 5 6 7 8 9
h= .
8 9 2 3 4 1 5 6 7
This h satisfies g = hf h−1 (check it!).
c University of Manchester Algebraic Structures 1 Week 10, Lecture 17, Page 4

Remark 17.9. In general, if f ∼ g then the permutation h with g = hf h−1 is not

unique: if k ∈ C(f ), then f = kf k −1 , so also g = hkf (hk)−1 .

Definition 17.10 (Partition of a natural number). A partition of a natural number n is

a sequence of natural numbers m1 , m2 , . . . , mk with m1 ≥ m2 ≥ · · · ≥ mk ≥ 1 such that
n = m1 + m2 + · · · + mk .

Example 17.11. The number 3 has three partitions, and the number 4 has five partitions
as follows:

4 = 4
3 = 3
= 3 + 1
= 2 + 1
= 2 + 2
= 1 + 1 + 1
= 2 + 1 + 1
= 1 + 1 + 1 + 1

Corollary 17.12. The conjugacy classes of Sn are in 1-1 correspondence with the
partitions of n.

Proof. By Theorem 17.7, the conjugacy classes of Sn are in 1-1 correspondence with cycle
structures, and the latter are in 1-1 correspondence with partitions of n. Indeed, each
number in Nn appears exactly once in the disjoint cycle decomposition of a permutation
in Sn (with 1-cycles for fixed elements), so the lengths of all cycles sum up to n. Hence,
any given cycle structure n1 , . . . , nk corresponds to the partition

n=k · · + k} + · · · + 2| + ·{z
| + ·{z · · + 2} + |1 + ·{z
· · + 1} .
nk times n2 times n1 times

Example 17.13. The partitions of the number 3 from Example 17.11 correspond to the
following cycle structures:
0, 0, 1 1, 1 3,
i.e. one 3-cycle, a 1-cycle and a 2-cycle, and three 1-cycles. The second cycle structure is
that of transpositions, and the third one is the identity idn . These are all cycle structures
of permutations in S3 , corresponding to the three conjugacy classes from example 16.6.
The partitions of the number 4 from Example 17.11 correspond to the following cycle
structures:
0, 0, 0, 1 1, 0, 1 0, 2 2, 1 4,
i.e. one 4-cycle, a 1-cycle and a 3-cycle, two 2-cycles, two 1-cycles and a 2-cycle, four
1-cycles. These are all the cycle structures of permutations in S4 , and hence there are 5
conjugacy classes.
c University of Manchester Algebraic Structures 1 Week 10, Exercises, Page 1

Week 10: Exercises

Guidance: This week I would like you to practice (a) determining whether two elements
of a given symmetric group are conjugate, finding a conjugating element where possible,
and (b) applying the class formula. Questions 55-57 can be solved by direct computation;
the remaining questions require some reasoning.

Conjugacy
(requires material up to Lecture 16)

54. Show that in any group conjugate elements have the same order. (That is, if
g, h ∈ G with |g| = |h|, then g ∼ h. )

Conjugacy in symmetric groups

(requires material up to Lecture 17)

55. For f, g ∈ S6 as specified below, decide whether of not g is a conjugate of f. If your

answer is ”yes”, find an element h ∈ S6 such that g = hf h−1 .

(a) f = (1 2 3 4) (5 6) , g = (1 2) (3 4 5 6) ,
(b) f = (1 2 3 4 5) , g = (1 2 3 4 5 6) ,
(c) f = (4 5) (3 6) (1 2) , g = (1 6) (2 5) (3 4) ,
(d) f = (1 2)(3 4 6), g = (1 2 3) (5 6) .

56. Work out the number of conjugacy classes in S5 and S6 .

57. Work out the order of the centraliser of (1 2) (3 4) in S4 and find all its elements.
(Hint: Apart from obvious members of the centraliser, consider (1 3 2 4) .)

58. Determine the number of cycles of length n in Sn . (Hint: Use the class formula,
and the fact that C ((1 2 · · · n)) = h(1 2 · · · n)i .)

59. Use the formula for conjugates in Sn to prove that Z(Sn ) = {e} for n ≥ 3.

Solutions to the exercises can be found here.

c University of Manchester Algebraic Structures 1 Week 11 Checklist

Resources for Week 11

Factor groups

(Click here to return to the front page navigation)

This week we introduce the notion of a factor group. There are two key ingredients to
understanding this – cosets (Lecture 12) and normal subgroups (Lecture 18, below). Our
main aim this week is to show that we can define a group structure on the set of all
cosets of a fixed normal subgroup N of a group G. For this construction to work, it is
really important that the subgroup N is normal. By the end of this week, you should
be able to explain the concepts of normal subgroups and factor groups (ILO2), know
some sources of examples of these, be able to describe fundamental properties of factor
groups (ILO4) and determine factor groups in certain groups (ILO6).
Synchronous Sessions:[2 hours of timetabled study]
 REVIEW: Attend Review Session and Tutorial (Topic: Week 10 - Conjugacy)
Asynchronous Tasks: [4 hours of directed private study]
 READ: Lecture 18 [∼ 2 pages; Suggested study time = 1 hour]
OR
WATCH: Lecture Video 18.1: What is a normal subgroup? [17 mins 47s]
Lecture Video 18.2: Examples of normal subgroups [17 mins 40s]
 CHECK UNDERSTANDING: Lecture 18 Quiz [Suggested study time = 10 minutes]
 READ: Lecture 19 [∼ 3 pages; Suggested study time = 1 hour]
OR
WATCH: Lecture Video 19.1: What is a factor group? [24 mins 11s]
Lecture Video 19.2: Examples and properties of factor groups [22 mins
36S]
 CHECK UNDERSTANDING: Lecture 19 Quiz [Suggested study time = 10 minutes]
 PRACTICE AND REFLECT: Do Exercises 60-66.
Take some time to consider which activities you found straightforward, and which
you found more challenging. Post any queries to the discussion board before the
next review session. [Suggested study time = 1 12 hours]
c University of Manchester Algebraic Structures 1 Week 11, Lecture 18, Page 1

Lecture 18
Normal subgroups

18.1 What is a normal subgroup?

Definition 18.1 (Normal subgroup). Let G be a group. A subgroup H ≤ G is normal in
G, if for all h ∈ H and all g ∈ G, we have ghg −1 ∈ H.
In other words, any conjugate of an element of H is again in H. We also say that “H is
closed under conjugation” and write H
G if H is a normal subgroup of G.

Remark 18.2 (Normal subgroups = subgroups that are unions of conjugacy classes).
Notice that H
G if and only if H ≤ G and hG ⊆ H for all h ∈ H. That is H
G if and
only if H ≤ G and H is the union of some of the conjugacy classes of G.

Example 18.3. For any group G, we have {e}
G and G
G. Indeed {e} is clearly a
subgroup, and it is a conjugacy class by Example 16.5.

Example 18.4. If G is abelian, then every subgroup H of G is normal since

ghg −1 = h ∈ H for all g ∈ G, h ∈ H.

Example 18.5. Let us show that h(1 2 3)i
S3 . We have

h(1 2 3)i = {idn , (1 2 3), (1 3 2)}. We need to check that any conjugate of an element of
h(1 2 3)i is itself an element of h(1 2 3)i. This is obvious for the identity element idn .
For the two 3-cycles (1 2 3) and (1 3 2), we know from Example 16.6 that they form a
conjugacy class in S3 . Hence h(123)i
S3 .

Example 18.6. The subgroup h(12)i ≤ S3 is not normal. Indeed, h(12)i = {e, (12)} and
(123)(12)(123)−1 = (23) ∈
/ h(12)i.

Example 18.7. We show that SL(n, R)
GL(n, R). Indeed, for any A ∈ SL(n, R) and
any B ∈ GL(n, R), we have BAB −1 ∈ SL(n, R) since

det(BAB −1 ) = det(B) det(A) det(B −1 ) = det(A) det(B) det(B)−1 = det(A) = 1.

    
0 1 0 1
Example 18.8. The subgroup = I, ≤ GL(2, R) is not normal.
1 0 1 0
Indeed, we have
   −1       
1 1 0 1 1 1 1 1 1 −1 1 0 0 1
= = ∈
/ .
0 1 1 0 0 1 1 0 0 1 1 −1 1 0
c University of Manchester Algebraic Structures 1 Week 11, Lecture 18, Page 2

18.2 Examples of normal subgroups

A great source of normal subgroups are group homomorphisms. The kernel of a
homomorphism ϕ : G → H, i.e. the set of all elements of G that are mapped to the
identity element of H, will turn out to be a normal subgroup of G.
Definition 18.9 (Kernel of a homomorphism). The kernel of a homomorphism
ϕ : G → H from a group G to a group H is the subset of G defined by
Ker ϕ = {x ∈ G | ϕ(x) = e}.
Example 18.10. Consider the homomorphism ϕ : GL(n, R) → R∗ given by
ϕ(A) = det(A). Note that

Ker ϕ = {A ∈ GL(n, R) | ϕ(A) = 1} = SL(n, R).

We have seen that SL(n, R)
GL(n, R).
Theorem 18.11 (Kernels are normal subgroups). For any group homomorphism
ϕ : G → H, the kernel of ϕ is a normal subgroup of G, in symbols Ker ϕ
G.
Proof. We have e ∈ Ker ϕ since ϕ(e) = e. Hence, Ker ϕ 6= ∅ and we can apply the
subgroup criterion. Let x, y ∈ Ker ϕ, i.e. ϕ(x) = ϕ(y) = e. Then
ϕ(xy) = ϕ(x)ϕ(y) = ee = e. Hence xy ∈ Ker ϕ. Also, x−1 ∈ Ker ϕ since
ϕ(x−1 ) = (ϕ(x))−1 = e−1 = e. Hence, we have shown that Ker ϕ is a subgroup of G by
the subgroup criterion. To verify that this is a normal subgroup, let x ∈ Ker ϕ and let
g ∈ G. Then
ϕ(gxg −1 ) = ϕ(g)ϕ(x)ϕ(g −1 ) = ϕ(g)eϕ(g)−1 = ϕ(g)ϕ(g)−1 = e,
so gxg −1 ∈ Ker ϕ. Hence Ker ϕ is closed under conjugation and therefore normal in
G.
Here are some more constructions of normal subgroups.
Lemma 18.12 (Centres are normal subrgoups). For any group G, we have Z(G)
G.
Proof. We know from Theorem 10.3 that Z(G) ≤ G. Let h ∈ Z(G), g ∈ G. Then
ghg −1 = hgg −1 = h ∈ Z(G), so Z(G) is indeed normal.
Lemma 18.13 (Any subgroup of index 2 is normal). Let G be a group and H ≤ G with
[G : H] = 2, i.e. the index of H in G is 2. Then H
G.
Proof. Since G is the disjoint union of the two left cosets of H in G, one of which is H
itself, the other one must be G r H. In particular, for all g ∈ G r H, we have
gH = G r H.
Let h ∈ H and g ∈ G. We need to show that ghg −1 ∈ H. If g ∈ H this is clear, so
suppose g ∈ G r H. If ghg −1 ∈ / H, then ghg −1 ∈ G r H = gH, so ghg −1 = gh1 for some
h1 ∈ H. Cancellation on the left yields hg −1 = h1 ∈ H, so g = h−11 h ∈ H, contradicting
our assumption that g ∈ G r H. The lemma follows.
c University of Manchester Algebraic Structures 1 Week 11, Lecture 19, Page 1

Lecture 19
Factor groups

From a group G and a normal subgroup N
G, we can construct a new group called
G/N . If G = Z and N = hni = nZ, it will turn out that Z/nZ is “the same as Zn ”.

19.1 What is a factor group?

Definition 19.1 (The set of cosets of N in G). Let G be a group and N a normal
subgroup of G. By G/N (in words: “G modulo N ”) we denote the set of all left cosets of
N in G:
G/N = {xN : x ∈ G}.
Example 19.2. If G = Z and N = h3i = 3Z, then G/N = {3Z, 1 + 3Z, 2 + 3Z} by
Example 12.3.
We want to define a binary operation on the cosets of N by setting xN ∗ yN = xyN . For
this to make sense, it has to be well defined, i.e., the coset xyN should depend only on
the cosets xN and yN , but not on the elements x, y chosen from these cosets. This is
verified in the following lemma.
Lemma 19.3 (Multiplication of cosets of a normal subgroup is well defined). Let G be a
group, N
G, and x, y, a, b ∈ G. Then
xN = aN and yN = bN implies that xyN = abN.
Proof. Suppose that xN = aN and yN = bN . Then x ∈ aN and y ∈ bN , so x = ah1 ,
y = bh2 for some h1 , h2 ∈ N . To verify the equality xyN = abN , it is sufficient to show
that xy ∈ abN (by Lemma 12.7). We have
xy = ah1 bh2 = abb−1 h1 bh2 .
Since N is normal and h1 ∈ N , we also have b−1 h1 b ∈ N , say b−1 h1 b = h3 with h3 ∈ N .
Then
xy = abh3 h2
and h3 h2 ∈ N . Hence xy ∈ abN . Consequently, we have shown that xyN = abN as
required.
Theorem 19.4 (Cosets of a normal subgroup form a group under coset multiplication).
Let G = (G, ∗) be a group and N
G. The set of left cosets G/N is a group with respect
to the binary operation, denoted again by ∗, and defined by
xN ∗ yN = xyN. (19.1)
c University of Manchester Algebraic Structures 1 Week 11, Lecture 19, Page 2

Proof. By Lemma 19.3, (19.1) determines a well-defined binary operation. It remains to

check that the group axioms are satisfied. For associativity, let x, y, z ∈ G. Then, using
(19.1) and associativity of the operation in G, we find

(xN ∗ yN ) ∗ zN = xyN ∗ zN = (xy)zN = x(yz)N = xN ∗ yzN = xN ∗ (yN ∗ zN ).

Hence (19.1) is associative. The identity element for (19.1) is the coset N = eN . Indeed,
for all x ∈ G,

eN ∗ xN = exN = xN, and xN ∗ eN = xeN = xN.

Finally, the inverse of the coset xN is the coset x−1 N . Indeed, one has

xN ∗ x−1 N = xx−1 N = eN = N, and x−1 N ∗ xN = x−1 xN = eN = N.

Definition 19.5 (Factor group). Let G be a group and N
G. The group G/N is called
the factor group of G by N , or the quotient group of G by N .

Remark 19.6.

1. The hypothesis that N is a normal subgroup of G is used in the proof of Lemma

19.3 It is absolutely essential for the construction of the factor group. This
construction does not work for subgroups that are not normal.

2. For multiplicative groups (G, ∗), we continue our previous convention and suppress
the notation ∗ also for the multiplication of cosets. Hence, for (19.1) we just write
xN yN = xyN .

3. For an additive group (G, +), cosets are of the form g + N , and (19.1) takes the
form
(x + N ) + (y + N ) = (x + y) + N.

19.2 Examples and properties of factor groups

Example 19.7 (The factor group Z/nZ). Consider the factor group Z/nZ. Here, G = Z
and N = nZ = hni is the (normal) subgroup of all integers divisible by n, for some fixed
n ∈ N.
The elements of Z/nZ are the cosets k + nZ with k ∈ Z. In other words, the elements are
equivalence classes of congruent modulo n integers. The binary operation is, for k, m ∈ Z,
given by
(k + nZ) + (m + nZ) = (k + m) + nZ.
Since k + nZ = (k + qn) + nZ for all k, q ∈ Z, all the distinct cosets are given as follows:

0 + nZ = nZ, 1 + nZ, 2 + nZ, ..., (n − 1) + nZ.

c University of Manchester Algebraic Structures 1 Week 11, Lecture 19, Page 3

Hence, Z/nZ = {k + nZ : k ∈ Zn }, and in particular |Z/nZ| = n. If k, l ∈ Zn and

k + l = qn + r with r ∈ Zn , then
(k + nZ) + (l + nZ) = (k + l) + nZ = (r + qn) + nZ = r + nZ = (k ⊕ l) + nZ.
In this way, we can identify Z/nZ with Zn . More precisely, the map ϕ : Zn → Z/nZ
defined by ϕ(k) = k + nZ is an isomorphism. The factor group Z/nZ is the proper
mathematical framework for studying integer congruences, and mathematicians usually
prefer to work with this over Zn . But in the end it doesn’t matter since they are
isomorphic and easily identified.
Example 19.8 (The factor group Z/5Z). As a concrete example, take n = 5 and look at
the factor group
Z/5Z = {k + 5Z : k ∈ Z} = {5Z, 1 + 5Z, 2 + 5Z, 3 + 5Z, 4 + 5Z}.
Here we have, for example,
(2 + 5Z) + (3 + 5Z) = 5 + 5Z = 5Z,
(2 + 5Z) + (4 + 5Z) = 6 + 5Z = 1 + 5Z.
Example 19.9 (The factor group S3 /h(1 2 3)i). Consider the factor group S3 /h(1 2 3)i.
Here, G = S3 and the normal subgroup is N = h(1 2 3)i = {idn , (1 2 3), (1 3 2)}. This is
indeed a normal subgroup by Example 18.5.
By Lagrange’s theorem, we have [G : N ] = |G|/|N | = 6/3 = 2, so there are two distinct
left cosets. One is N = {idn , (1 2 3), (1 3 2)}, so the other one has to be
(1 2)N = {(1 2), (1 3), (2 3)}. We have
(1 2)N (1 2)N = (1 2)(1 2)N = idn N = N,
so
S3 /h(1 2 3)i = {N, (1 2)N } = h(1 2)N i
is a cyclic group of order 2.
For each group G and normal subgroup N
G, there is a natural way to define a
homomorphism G → G/N .
Lemma 19.10 (A homomorphism from G to G/N ). Let G be a group and N
G. The
map ν : G → G/N given by ν(x) = xN is a homomorphism of groups. It is onto and
Ker ν = N .
Proof. Let x, y ∈ G. Then, by definition of the binary operation on G/N in (19.1),
ν(xy) = xyN = xN yN = ν(x)ν(y).
Hence, ν is a homomorphism. Every left coset xN ∈ G/N is in the image of ν, since
xN = ν(x), so ν is onto. Finally, an element x ∈ G is in the kernel of ν if and only if
xN = ν(x) = N , the identity element of G/N . But xN = N if and only if x ∈ N , and
hence Ker ν = N .
c University of Manchester Algebraic Structures 1 Week 11, Lecture 19, Page 4

Definition 19.11 (Natural homomorphism). The homomorphism ν : G → G/N is called

the natural homomorphism from G to G/N .

Remark 19.12. Lemma 19.10 shows that every normal subgroup N of a group G is the
kernel of at least one homomorphism, namely the natural homomorphism. On the other
hand, we have shown in Theorem 18.11 that the kernel of every homomorphism is a
normal subgroup. Consequently, normal subgroups of G are exactly the same as kernels of
group homomorphisms with domain G.
c University of Manchester Algebraic Structures 1 Week 11, Exercises, Page 1

Week 11: Exercises

Guidance: This week I would like you to practice determining whether a given subgroup
of a group G is normal in G. Questions 60-63 are all of this type. Questions 64-66
require you to practice reasoning with abstract groups; make sure to try each of these!

Normal subgroups
(requires material up to Lecture 18)

60. For a group G and a subgroup H ≤ G as specified below, decide if H is normal in G.

(a) G = T(2, R), H = UT(2, R);

(b) G = GL(2, R), H = the subgroup of scalar matrices in GL(2, R);
 
−1 0
(c) G = SL(2, R), H = .
0 −1
61. Decide whether the following subgroups H of S4 are normal in S4 :

(a) H = h(1 2 3 4)i,

(b) H = {id4 , (1 2) (3 4) , (1 3) (2 4) , (1 4) (2 3)} .

62. Suppose H C G and K C G. Show that H ∩ K C G.

63. Suppose that H C G and K C G. The product of H and K in G is the set

HK = {hk : h ∈ H, k ∈ K}. Show that HK C G.

64. Let H ≤ K ≤ G.

(a) Suppose H C G. Show that H C K.

(b) Suppose that H C K and K
G. Does this imply that H C G?

65. Let N and M be normal subgroups of G such that N ∩ M = {e} . Show that for all
x ∈ N and all y ∈ M one has xy = yx. (Hint: Show that xyx−1 y −1 = e.)

66. Let G be a group, and let H C G be a cyclic group of order two which is normal in
G. Prove that H ⊆ Z(G).

Solutions to the exercises can be found here.

c University of Manchester Algebraic Structures 1 Week 12 Checklist

Resources for Week 12

First Isomorphism Theorem

(Click here to return to the front page navigation)

This week we shall put our understanding of factor groups to work to prove the
interesting fact that any group of order p2 , where p is a prime, is abelian. We then
conclude the course by stating and proving an extremely useful result: the First
Isomorphism Theorem. This result draws together several previous topics
(homomorphisms, normal subgroups, factor groups, and isomorphisms). By the end of
this week, you should be able to state and apply key results from the course such as
Lagrange’s theorem (ILO7) and the First Isomorphism Theorem for groups (ILO8).
Synchronous Sessions:[2 hours of timetabled study]

 REVIEW: Attend Review Session and Tutorial (Topic: Weeks 11 + 12 - Factor

groups and the First Isomorphism Theorem)

Asynchronous Tasks: [4 hours of directed private study]

 READ: Lecture 20 [∼ 3 pages; Suggested study time = 1 12 hours]

OR
WATCH: Lecture Video 20.1: An application of factor groups [23 mins 8s]
Lecture Video 20.2: The first isomorphism theorem [25 mins 18s]

 CHECK UNDERSTANDING: Lecture 20 Quiz [Suggested study time = 10 minutes]

 PRACTICE AND REFLECT: Do Exercises 67-72, and revisit earlier exercises.

Take some time to consider which activities you found straightforward, and which
you found more challenging. Post any queries to the discussion board. [Suggested
study time = 2 21 hours]
c University of Manchester Algebraic Structures 1 Week 12, Lecture 20, Page 1

Lecture 20
Factor groups and the First Isomorphism Theorem

20.1 An application of factor groups

Our goal here is to prove the stunning fact that every group of order p2 , where p is a
prime, is abelian. We start with the following remarkable consequence of the class
formula concerning groups of order pn , so called p-groups.
Theorem 20.1 (Groups of prime power order have non-trivial centre). Let G be a group
with |G| = pn where p is a prime and n ∈ N. Then Z(G) 6= {e}.
Proof. Recall from Remark 16.8 that a ∈ Z(G) if and only if its conjugacy class consists
of just a, i.e. aG = {a}. Let C1 , C2 , . . . , Cq be the conjugacy classes of G, and let
mi = |Ci | (1 ≤ i ≤ q). By Corollary 16.10, each mi divides |G| = pn . Hence, each mi is
either 1 or a power of p. Since G is the disjoint union if its conjugacy classes, we have
pn = |G| = |C1 | + |C2 | + · · · + |Cq | = m1 + m2 + · · · + mq .
Since the identity element of G is a conjugacy class on its own, say C1 = {e}, we may
assume m1 = 1. Then
pn = 1 + m2 + · · · + mq .
In this equation it is not possible that all the mi with i ≥ 2 are powers of p, since
otherwise the LHS is divisible by p, but the RHS is not. Hence, there is more than one
mi with mi = 1. In other words, there is more than one conjugacy class consisting of a
single element. But these elements are contained in the centre. Thus Z(G) 6= {e}.
Combining the previous theorem with the notion of factor groups, we can prove that
every group of order p2 , where p is a prime, is abelian. We first prove a lemma. Recall
from Lemma 18.12 that the centre Z(G) of G is a normal subgroup of G.
Lemma 20.2 (Quotient of a non-abelian group by its centre is not cyclic). If G is a
non-abelian group, then the factor group G/ Z(G) is not cyclic.
Proof. We prove the contrapositive: we assume that G/ Z(G) is cyclic with generator,
say, x Z(G) and show that G is abelian. Under our assumption, any coset of Z(G) in G is
of the form (x Z(G))k for some integer k. Since G is the union of its cosets and
(x Z(G))k = xk Z(G), any element in G is of the form xk z where z ∈ Z(G).
Let g and h be arbitrary elements of G. Then g = xk1 z1 and h = xk2 z2 for some integers
k1 , k2 and some elements z1 , z2 ∈ Z(G). But then,
gh = xk1 z1 xk2 z2 = xk1 +k2 z1 z2 = xk2 z2 xk1 z1 = hg,
i.e. any two elements in G commute. Hence G is abelian.
c University of Manchester Algebraic Structures 1 Week 12, Lecture 20, Page 2

Theorem 20.3. Every group of order p2 , where p is a prime, is abelian.

Proof. Let G be a group with |G| = p2 and consider Z(G). In view of Lagrange’s
Theorem, there are three possibilities for the order of Z(G):

| Z(G)| = 1 or p or p2 .

The case | Z(G)| = 1 is impossible since any group of prime-power order has a non-trivial
centre (by Theorem 20.1). The case | Z(G)| = p is also impossible. Indeed, if | Z(G)| = p
then |G/ Z(G)| = p by Lagrange’s theorem, but every group of prime order is cyclic (by
Corollary 13.10). However, if Z(G) is a proper subgroup of G, then G is not abelian, and
hence G/ Z(G) cannot be cyclic by Lemma 20.2. The only remaining case is | Z(G)| = p2 .
But then Z(G) = G, and hence G is abelian, as required.

20.2 The first isomorphism theorem

The concluding result of this course is the first isomorphism theorem for groups. Recall
that if N
G, then we have the natural homomorphism ν : G → G/N with Im ν = G/N
and Ker ν = N . In other words, Im ν = G/ Ker ν. This is a special case of the following
theorem.
Theorem 20.4 (First isomorphism theorem). Let ϕ : G → H be a homomorphism of
groups. Then there is an isomorphism

ϕ̃ : G/ Ker(ϕ) → Im ϕ defined by
ϕ̃(x Ker(ϕ)) = ϕ(x).

We say that the isomorphism ϕ̃ is “induced by ϕ”. In particular,

Im ϕ ∼
= G/ Ker ϕ.
Proof. Note that Ker ϕ
G, so the factor group G/ Ker ϕ is defined. We write K = Ker ϕ
for brevity and start by showing that ϕ̃ is a well-defined map: if xK = yK for x, y ∈ G,
then ϕ(x) = ϕ(y). Indeed, if xK = yK, then y = xh for some h ∈ K, and hence

ϕ(x) = ϕ(x)ϕ(h) = ϕ(xh) = ϕ(y).

Next, we show that ϕ̃ is a homomorphism: for all x, y ∈ G we have

ϕ̃(xKyK) = ϕ̃(xyK) = ϕ(xy) = ϕ(x)ϕ(y) = ϕ̃(xK)ϕ̃(yK).

It remains to show that ϕ̃ is 1-1 and onto. Assume ϕ̃(xK) = ϕ̃(yK), i.e. ϕ(x) = ϕ(y).
Then ϕ(x)−1 ϕ(y) = e, hence ϕ(x−1 y) = e, hence x−1 y ∈ K, and thus xK = yK by
Lemma 12.7. This proves that ϕ̃ is 1-1. Finally, if g ∈ Im ϕ, then g = ϕ(x) for some
x ∈ G. But then g = ϕ̃(xK), and hence g ∈ Im ϕ̃. This proves that ϕ̃ is onto. Hence, ϕ̃ is
indeed an isomorphism and G/ Ker ϕ ∼ = Im ϕ. To conclude the proof, recall that being
isomorphic is symmetric by Lemma 15.5.
c University of Manchester Algebraic Structures 1 Week 12, Lecture 20, Page 3

Example 20.5. What is R/Z? Consider the homomorphism ϕ : R → C∗ given by

ϕ(x) = cos 2πx + i sin 2πx.

This is indeed a homomorphism as, similar to Example 14.4,

ϕ(x + y) = cos(2πx + 2πy) + i sin(2πx + 2πy),

ϕ(x)ϕ(y) = (cos 2πx + i sin 2πx)(cos 2πy + i sin 2πy)
= cos 2πx cos 2πy − sin 2πx sin 2πy + i(cos 2πx sin 2πy + sin 2πx cos 2πy)
= cos(2πx + 2πy) + i sin(2πx + 2πy).

We have Im ϕ = {z ∈ C∗ : |z| = 1} and

Ker ϕ = {x ∈ R : cos(2πx) = 1 and sin(2πx) = 0} = Z.

Hence, by the first isomorphism theorem,

R/Z ∼
= {z ∈ C∗ : |z| = 1}.

Example 20.6. What is GL(n, R)/ SL(n, R)?

Consider the homomorphism ϕ : GL(n, R) → R∗ given by ϕ(A) = det A. Then Im ϕ = R∗
and Ker ϕ = SL(n, R). Hence, by the first isomorphism theorem,

GL(n, R)/ SL(n, R) ∼

= R∗ .
c University of Manchester Algebraic Structures 1 Week 12, Exercises, Page 1

Week 12: Exercises

Guidance: This week I would like you to practice applying the first isomorphism
theorem. To apply this theorem you will first need a homomorphism with certain
properties; in some cases I have suggested a map – check that this map has all of the
correct properties; in other cases, you will need to find a map with the right properties.

Factor groups and first isomorphism theorem

(requires material up to Lecture 20)

67. Show that R∗ /R+ ∼

= {±1} , where R+ = {r ∈ R∗ : r > 0} and {±1} is regarded as
a group under multiplication.

68. Show that C∗ / {z ∈ C : |z| = 1} ∼

= R+ , where R+ = {r ∈ R∗ ; r > 0} .
69. Show that T(2, R)/ UT(2, R) ∼
= D(2, R).
70. Let   
a b
N= : a, b ∈ R, a 6= 0 ≤ T(2, R).
0 a−1
Show that normal in T(2, R), and prove that T(2, R)/N ∼
 N is  = R∗ . (Hint: Consider
a b
the map → ac.)
0 c

71. Let   
 1 a b 
N=  0 1 0  : a, b ∈ R ≤ UT(3, R).
0 0 1
 

Show that N is normal in UT(3, R), and prove that UT(3, R)/N ∼
= R. (Hint:
Consider the map  
1 a b
 0 1 c  → c.)
0 0 1

72. For each of the maps ϕ : G → H in Exercise 49, if they are homomorphisms, find
the kernel and the image. In each case, what conclusion can be drawn by applying
the first isomorphism theorem?

Solutions to the exercises can be found here.

Solutions to Exercises
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 1

(Click here to return to the front page navigation)

Week 0: Warm-up exercises

Back to the warm-up exercises.

1. The answer is YES for (i), (ii), (iv), (v), (vi), (ix). and (xi).
(iii) The answer is NO since the operation is not well-defined (for example the given
rule doesn’t give a definitive answer for 2 ∗ 1),
(vii)
 The answer  is NO since,
 for example,
1 2 3 1 2 3 1 2 3
◦ = ∈
/ S.
2 3 1 3 2 1 1 3 2
   
1 2 3 4 1 2 3 4
(viii) The answer is NO since, for example, and are
2 1 3 4 1 2 4 3
elements of S,
  but
   
1 2 3 4 1 2 3 4 1 2 3 4
◦ = ∈/ S.
2 1 3 4 1 2 4 3 2 1 4 3
(x) The answer is NO since A ∗ B is not always defined: the matrices A and B may
not be invertible.

2. There are n2 pairs (g, h) ∈ G × G, and for each one there are n possible values for
g ∗ h ∈ G. Hence, the number of binary operations on G is n(n ) .
2

3. (i) not associative, e.g. (1 ∗ 1) ∗ 1 = (1 − 1) − 1 = −1, but

1 ∗ (1 ∗ 1) = 1 − (1 − 1) = 1; not commutative, e.g. 2 ∗ 3 = 2 − 3 = −1, but
3 ∗ 2 = 3 − 2 = 1; no identity element (if e was an identity element for ∗, then we’d
have e ∗ a = a for all a ∈ Z, i.e. e − a = a for all a ∈ Z, which is the same as to say
that e = 2a for all a ∈ Z, but this is of course impossible).
(ii) not associative, e.g. (2 ∗ 2) ∗ 4 = 3 6= 25 = 2 ∗ (2 ∗ 4), commutative as
a ∗ b = a+b
2
= b+a
2
= b ∗ a, but no identity element (e ∗ a = a for all a ∈ Z, means
e+a
2
= a for all a ∈ Z, which is the same as to say that e = a for all a ∈ Z, but this
is of course impossible).
(iii) associative as (a ∗ b) ∗ c = b ∗ c = c = a ∗ c = a ∗ (b ∗ c), not commutative as for
example 1 ∗ 2 = 2 6= 1 = 2 ∗ 1, no identity element (if e was an identity element for
∗ we must have a ∗ e = a ∀a ∈ N, whereas the definition of ∗ implies that a ∗ e = e,
so e = a ∀a ∈ N, which is impossible).
(iv) associative and commutative since multiplication of integers is associative and
commutative, no identity element since ae = a implies e = 1, but 1 ∈ / S.
(v) not associative, since (a ∗ a) ∗ b = b ∗ b = a whereas a ∗ (a ∗ b) = a ∗ b = b;
commutative, since the multiplication table is symmetric; no identity element, since
for an identity element one has e ∗ e = e, which holds for neither a nor b.
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 2

(vi) to verify associativity for ∗, we need to check that the equality

(a ∗ b) ∗ c = a ∗ (b ∗ c) (+)

holds whenever we replace a, b and c by elements of S. Since S has two elements, we

need to check 8 equations: For a = b = c = u , we get (u ∗ u) ∗ u = u ∗ u = u on the
left-hand side of (+), and u ∗ (u ∗ u) = u ∗ u = u on the right-hand side, as required.
For a = b = u and c = v, we get (u ∗ u) ∗ v = u ∗ v = v on the LHS and
u ∗ (u ∗ v) = u ∗ v = v on the RHS. Again we have equality, as required. The
remaining 6 possibilities are checked similarly. So ∗ is associative. It is also
commutative, since the multiplication table is symmetric, and u is an identity
element, as u ∗ u = u and u ∗ v = v ∗ u = v.
(vii) associative, since composition of function from a given set to itself is always
associative: f ◦ (g ◦ h)(x) = f (g(h(x))) = (f ◦ g) ◦ h(x). It is not commutative, e.g.
if f (n) = 1 ∀n ∈ N and g(n) = 2 ∀n ∈ N, then (f ◦ g) (n) = f (g(n)) = f (2) = 1
∀n ∈ N, but (g ◦ f ) (n) = g(f (n)) = g(1) = 2 ∀n ∈ N. Hence f ◦ g 6= g ◦ f. Finally,
there is no identity element. Indeed, if e with e(n) = c (c ∈ N) was an identity
element for S, we’d have e ◦ f = f and e ◦ g = g. Since (e ◦ f ) (n) = c ∀n ∈ N, the
first equation implies c = 1. Likewise, the second equation implies c = 2,
contradiction.

4.
assoc. comm. identity el.
(i) NO NO NO
(ii) YES YES e=1
(iv) YES YES e=X
(v) YES YES e=∅
assoc. comm. identity el.
(vi) NO NO NO
(ix) NO NO NO
(xi) YES YES e = O (zero matrix)

Week 1: Groups

Back to Week 1 exercises.

 
1 1
5. (i) Matrix multiplication is associative, is the identity element,
0 0
 −1  
a a 1/a 1/a
= .
0 0 0 0

Hence (i) is a group.

c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 3

 
a b
(ii) This is not a group as it has no identity element. Indeed, suppose is
     0 0
1 1 a b 1 1
the identity element. Then = , which implies
  0  0 0 0  0 0 
1 2 1 1 1 2 1 1
a = b = 1. But 6= . Hence is not the
0 0 0 0 0 0 0 0
identity element. Hence there is no identity element.
(iii) The operation is associative as for all a, b, c, d, e, f ∈ Z we have

((a, b)(c, d))(e, f ) = (a + c, b + d)(e, f ) = (a + c + e, b + d + f ),

(a, b)((c, d)(e, f )) = (a, b)(c + e, d + f ) = (a + c + e, b + d + f ),

(0, 0) is the identity element and (a, b)−1 = (−a, −b). Hence (iii) is a group.
(iv) For a, b, c, d, e, f ∈ Z we have

((a, b)(c, d))(e, f ) = (a + c, (−1)c b + d)(e, f ) = (a + c + e, (−1)e ((−1)c b + d) + f ),

(a, b)((c, d)(e, f )) = (a, b)(c + e, (−1)e d + f ) = (a + c + e, (−1)c+e b + (−1)e d + f ),

and both these expressions are equal to (a + c + e, (−1)c+e b + (−1)e d + f ). Hence
the operation is associative. The identity element is (0, 0), and
(a, b)−1 = (−a, (−1)a (−b)). Hence (iv) is a group.
(v) This operation is associative as for all A, B, C ⊆ X we have
(A ∩ B) ∩ C = A ∩ (B ∩ C). It has an identity element, namely X. Indeed, for all
A ⊆ X we have A ∩ X = X ∩ A = A. However, the third condition is violated: As
X is non-empty, ∅ =6 X, and as ∅ ∩ A = ∅ for all A ⊆ X, there is no inverse for ∅.
Hence (v) is not a group.
(vi) We can assume associativity, the empty set ∅ is the identity element
(∅A = (∅ ∪ A) \ (∅ ∩ A) = A \ ∅ = A and likewise A∅ = A for all A ⊆ Ω) and
A−1 = A for all A ⊆ Ω, i.e. every element is its own inverse, since

AA = (A ∪ A) \ (A ∩ A) = A \ A = ∅.

Hence (vi) is a group.

6. YES for (i), (ii), (iv), (v), (vi), NO for

(iii) since, e.g., 2 ∈ Z does not have an inverse in Z,
     
1 2 3 1 2 3 1 2 3
(vii) since, e.g., ◦ = ∈
/ G, i.e. the operation
2 3 1 2 3 1 3 1 2
is NOT a binary operation on the set G,
(viii) since there is no inverse for 0,
(ix) since, e.g., there is no inverse for 3.
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 4

7. For (a, b), (c, d), (e, f ) ∈ R × R \ (0, 0) we have

(a, b)((c, d)(e, f )) = (a, b)(ce − df, de + cf )
= (a(ec − df ) − b(de + cf ), a(de + cf ) + b(ce − df )
= (ace − adf − bde − bcf, ade + acf + bce − bdf )
and
((a, b)(c, d))(e, f ) = (ac − bd, bc + ad)(e, f )
= ((ac − bd)e − (bc + ad)f, (bc + ad)e + (ac − bd)f
= (ace − bde − bcf − adf, bce + ade + acf − bdf ).
This proves that the binary operation is associative. The identity element is (1, 0),
as (1, 0)(a, b) = (a, b) = (a, b)(1, 0). The inverse of (a, b) is
a b
(a, b)−1 = ( ,− 2 ),
a2 +b 2 a + b2
as
a b a b
(a, b)( ,− 2 ) = (1, 0) = ( 2 ,− 2 )(a, b).
+b 2a2 a +b 2 a +b 2 a + b2
Hence we have got a group.

8. First note that a + b − ab = 1 implies that a − 1 = (a − 1)b, and thus a = 1 or

b = 1. Hence a ∗ b ∈ G for all a, b ∈ G, so ∗ is indeed a binary operation on G. Now
check the group axioms: (G1) holds as for arbitrary a, b, c ∈ G one has
(a ∗ b) ∗ c = (a + b − ab) + c − (a + b − ab)c
= a + b + c − ab − ac − bc + abc
and
a ∗ (b ∗ c) = a + (b + c − bc) − a(b + c − bc)
= a + b + c − ab − ac − bc + abc
Hence (a ∗ b) ∗ c = a ∗ (b ∗ c) ∀a, b, c ∈ G, i.e. (G1) is satisfied. The element e = 0 is
the identity element for (G, ∗). Indeed, for all a ∈ G one has
0 ∗ a = 0 + a − 0 × a = a and a ∗ 0 = a + 0 + a × 0 = a. This proves that (G2) holds.
Finally, (G3) holds as for each a ∈ G the element
−a
a−1 =
1−a
is the inverse of a. Indeed, note that a−1 is correctly defined since a 6= 1 by
assumption, and that
−a a a 1−a
a∗ =a− +a =a−a = 0,
1−a 1−a 1−a 1−a
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 5

and
−a a a 1−a
∗a=− +a+a =a−a = 0.
1−a 1−a 1−a 1−a

 
a b
9. The elements of GL(2, Z2 ) are all those matrices with a, b ∈ Z2 = {0, 1}
c d
such that ad − bc ≡ 1 mod 2. These are
           
1 0 1 1 1 0 0 1 0 1 1 1
, , , , , .
0 1 0 1 1 1 1 0 1 1 1 0

10. We have (gh)(h−1 g −1 ) = g(hh−1 )g −1 = gg −1 = e and

(h−1 g −1 )(gh) = h−1 (g −1 g)h = h−1 h = e, and therefore h−1 g −1 is the inverse of gh.

Week 2: The symmetric group

Back to Week 2 exercises.

   
1 2 3 4 5 1 2 3 4 5
11. (i) g ◦ f = , f ◦g = ,
3 4 2 1 5 4 5 3 2 1
   
1 2 3 4 5 1 2 3 4 5
g◦g = = id5 , f ◦ f =
1 2 3 4 5 3 2 5 4 1
(ii) f = (1 5 3) ◦ (2 4) , g = (1 2) ◦ (3 5) , g ◦ f = (1 3 2 4) , f ◦ g = (1 4 2 5)
g ◦ g = id5 , f ◦ f = (1 3 5)
   
1 2 3 4 5 1 2 3 4 5
12. (i) , (ii) id5 , (iii) id5 , (iv)
3 5 4 1 2 5 1 3 4 2

13. (i) and (iii) are true, (ii) and (iv) are false.

14. f = (1 4 5 6) ◦ (2 9) ◦ (3 8 7) , g = (1 5 4 3) ◦ (2 9 6 8)

15. (i) (1 3) ◦ (2 4) (ii) (2 4 5 3) (iii) (1 2 3 4 5)

16. (a) (1 5 3 2 7 8 4)(6),

(b) (1 4 5 6 3)(2)(7 8 9).
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 6

17. The order of S4 is 24, and its elements (written as composites of disjoint cycles) are
the identity element e = id24 ,
the six transpositions (1 2), (1 3), (1 4), (2 3), (2 4), (3 4),
the three double transpositions (1 2)(3 4), (1 3)(2 4), (1 4)(2 3),
the eight 3-cycles (1 2 3), (1 3 2), (1 2 4), (1 4 2), (1 3 4), (1 4 3), (2 3 4), (2 4 3),
and the six 4-cycles (1 2 3 4), (1 2 4 3), (1 3 2 4), (1 3 4 2), (1 4 2 3), (1 4 3 2).
18. Every non-trivial permutation f is a composite of non-trivial disjoint cycles, say
f = g1 g2 · · · gk
with k ≥ 1. Let ri denote the length of gi . Clearly, r1 + r2 + ... + rk ≤ n. In view of
(α1 α2 · · · αl ) = (α1 αl ) (α1 αl−1 ) · · · (α1 α3 ) (α1 α2 ) ,
every cycle of length l is a composite of l − 1 transpositions. Hence f is a composite
of (r1 − 1) + (r2 − 1) + ... + (rk − 1) transpositions. But
(r1 − 1) + (r2 − 1) + ... + (rk − 1) = (r1 + r2 + ... + rk ) − k ≤ n − k < n
since k ≥ 1.

Week 2: Multiplication tables

19. ∗ e a b c
e e a b c
a a e c b
b b c a e
c c b e a
Clearly, ∗ is a correctly defined binary operation on G (as we can see from the
table, ∗ assigns to any ordered pair of elements of G a well defined element of G). It
remains to check the group axioms (G1)-(G3). (G1) comes for free since matrix
multiplication is associative. The existence of an identity element, i.e. (G2) can be
read off the table: An element is an identity element if and only if the row and
column labeled by it coincide with the row and column labeled by ∗, respectively.
In the above table, e has this property, and hence it is the identity element. The
existence of an inverse for each element can also be read off the table: one has
e−1 = e, a−1 = a, b−1 = c and c−1 = b. Hence G is a group.

20. ⊕ 0 1 2 3 ⊗ 0 1 2 3
0 0 1 2 3 0 0 0 0 0
1 1 2 3 0 1 0 1 2 3
2 2 3 0 1 2 0 2 0 2
3 3 0 1 2 3 0 3 2 1
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 7

21. ∗ e a b c
e e a b c
a a e c b
b b c e a
c c b a e
The proof that G is a group is similar to that in Exercise 19. Clearly, ∗ is a
correctly defined binary operation on G (as we can see from the table, ∗ assigns to
any ordered pair of elements of G a well defined element of G). It remains to check
the group axioms (G1)-(G3). (G1) comes for free since composition of functions
from one set to itself, in particular, composition of permutations, is associative.
The existence of an identity element, i.e. (G2), can be read off the table: An
element is an identity element if and only if the row and column labeled by it
coincide with the row and column labeled by ∗, respectively. In the above table, e
has this property, and hence it is the identity element. The existence of an inverse
for each element can also be read off the table: Here each element is its own inverse
(e2 = a2 = b2 = c2 = e). Hence G is a group.

Week 3: Subgroups

Back to Week 3 exercises.

22. YES for (i), (ii), (iii), and (v),

NO for (iv) (e.g. since e = idn ∈
/ S) and (vi) (e.g. since 1 ⊕ 1 = 2 ∈
/ S).

23. For n = 1, we have S1 = {id1 } = A1 , as id1 is a product of zero transpositions. If

n ≥ 2, we have idn = (1 2) (1 2) ∈ An and so An is a non-empty subset of Sn . Now
if f, g ∈ An , i.e. f = h1 h2 · · · h2k and g = j1 j2 · · · j2m where
h1 , h2 , . . . , h2k , j1 , j2 , . . . , j2m are transpositions, then

f g = h1 h2 · · · h2k j1 j2 · · · j2m ,

i.e. f g is a composite of an even number of transpositions (namely 2k + 2m

transpositions). Hence f g ∈ An . Also, for f as above,

f −1 = h−1 −1 −1 −1
2k h2k−1 · · · h2 h1 = h2k h2k−1 · · · h2 h1 ,

since any transposition is its own inverse. Consequently, f −1 is a composite of an

even number of transpositions and hence f −1 ∈ An . So An satisfies the subgroup
criterion and hence it is a subgroup of Sn . (The subgroup An is a very important
group. It is called the alternating group of degree n. It plays a crucial part in the
proof of the famous result of Galois theory that there is no general formula for the
zeros of a polynomial of degree > 4.)
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 8

Week 3: Subgroups of GL(n, R)

24. We have I ∈ H, so H 6= ∅. Hence, we can apply the subgroup criterion. Let

1 a0 b 0
   
1 a b
A = 0 1 c  , B = 0 1 c0  ∈ H,
0 0 1 0 0 1

with a, a0 ∈ Z and b, b0 , c, c0 ∈ R. Then

1 a + a0 b + b0 + ac0
 

AB = 0 1 c + c0  ∈ H,
0 0 1

as a + a0 ∈ Z. Moreover,
 
1 −a ac − b
A−1 = 0 1 −c  ∈ H,
0 0 1

so H is a subgroup by the subgroup criterion.

Week 4: Cyclic subgroups and order

Back to Week 4 exercises.

25. (i) 8, 6, 12
(ii) 9, 12, 2, 12
(iii) 12, 2, 20
(iv) 6, ∞, 4

26. (i) h5i = {5n | n ∈ Z} ,

(ii) h(1234)i = {(1 2 3 4) , (1 3) (2 4) , (1 4 3 2) , id4 } ,
(iii) h1i = Z8 , h2i = {2, 4, 6, 0} , h3i = {3, 6, 1, 4, 7, 2, 5, 0} = Z8 ,
(iv)
         
0 −1 0 −1 −1 0 0 1 1 0
= , , ,
1 0 1 0 0 −1 −1 0 0 1

27. {(1 2 3)(4 5), (1 3 2), (4 5), (1 2 3), (1 3 2)(4 5), id5 }

Week 4: Order of permutations

c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 9

28. Every permutation in Sn can be written as a product of disjoint cycles, and we

know that the order of the permutation depends only on the lengths of those cycles;
it is the least common multiple of those lengths. If we include cycles of length one
for all elements that are fixed by the permutation, then every element of Nn
appears in exactly one cycle, and hence the lengths of the disjoint cycles making up
a particular permutation in Sn must add up to n. Hence the possible orders of
elements in Sn correspond to the different ways in which the number n can be
written as a sum of positive integers. For n = 3 there are 3 different ways,

3 = 3
= 2+1
= 1 + 1 + 1,

and the corresponding orders are 3, 2 and 1. For n = 5 there are 7 different ways,

5 = 5
= 4+1
= 3+2
= 3+1+1
= 2+2+1
= 2+1+1+1
= 1 + 1 + 1 + 1 + 1,

and the corresponding orders are 5, 4, 6, 3, 2, 2 and 1. Finally, for n = 7 there are
15 different ways,

7 = 7=6+1=5+2=5+1+1=4+3
= 4+2+1=4+1+1+1=3+3+1
= 3+2+2=3+2+1+1=3+1+1+1+1
= 2+2+2+1=2+2+1+1+1
= 2 + 1 + 1 + 1 + 1 + 1 = 1 + 1 + 1 + 1 + 1 + 1 + 1,

and the corresponding orders are 7, 6, 10, 5, 12, 4, 4, 3, 6, 6, 3, 2, 2, 2 and 1. Hence

the largest possible orders of elements in S3 , S5 , and S7 are 3, 6 and 12, respectively.

Week 5: Centralisers

Back to Week 5 exercises.

29. We give details only for (ii), the other problems being analogous but slightly
simpler.
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 10

    
1 0 a 0
(i) C = : ad 6= 0 = D(2, R)
0 2 0 d
 
a b
(ii) For a matrix ∈ GL(2, R), we have
c d
    
a b 1 1 a a+b
= ,
c d 0 1 c c+d
    
1 1 a b a+c b+d
= .
0 1 c d c d
The two matrices on the right-hand sides of the above equations are equal if and
only if a = d and c = 0. Hence,
    
1 1 a b
C = : a 6= 0 .
0 1 0 a
    
0 1 a b 2 2
(iii) C = : a 6= b .
1 0 b a

30. Direct inspection of the 24 elements of S4 gives

C((1 2)) = {e, (1 2), (3 4), (1 2)(3 4)}.

31. We need to show that the centraliser C((1 2)) in Sn coincides with the subgroup
H = {f ∈ Sn | f (1), f (2) ∈ {1, 2}}.
In other words, H consists of all permutations f of Nn = {1, 2, ..., n} such that, if
written as a product of disjoint cycles, then f involves either both the cycles (1)
and (2) of length 1, i.e. both 1 and 2 are fixed by f, or f involves the transposition
(1 2), i.e. 1 and 2 are swapped by f. All such permutations commute with (1 2), so
H ⊆ C((1 2)). On the other hand, if f ∈ Sn \H, then either f (1) ∈ / {1, 2} or
f (2) ∈
/ {1, 2}. WLOG assume that f (1) = k > 2. Then f ∈ / C((1 2)). Indeed,
evaluating at 1 gives (1 2)f (1) = k and f (1 2)(1) = f (2), so (1 2)f = f (1 2) implies
that f (2) = k, contradicting our assumption that f (1) = k. This proves that
C((1 2)) = H.

32. Let g = (1 2 · · · n − 1) ∈ Sn . Clearly, hgi ⊆ C(g). It therefore remains to show that

C(g) ⊆ hgi . Let f ∈ C(g). Then f (1) 6= n. Indeed, if f (1) = n, then f g(1) = gf (1)
yields f (2) = n, contradicting the assumption that f (1) = n. Hence f (1) = k for
some k with 1 ≤ k ≤ n − 1. Note that m = g m−1 (1) for all m ∈ {1, 2, ...n − 1}. For
any such m, we have
f (m) = f g m−1 (1) = g m−1 f (1) = g m−1 (k) = g m−1 g k−1 (1)
= g m−1+k−1 (1) = g k−1 g m−1 (1) = g k−1 (m).
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 11

Hence f (m) = g k−1 (m) for all m ∈ {1, 2, ..., n − 1}, and hence f (n) = g k−1 (n) (= n)
as well. Hence f = g k−1 ∈ hgi . Consequently, C(g) ⊆ hgi , as required.

Week 5: Centre
 
a b
33. If ∈ Z(T(2, R)), then in particular
0 c
           
a b 1 0 1 0 a b a b 1 1 1 1 a b
= and = .
0 c 0 2 0 2 0 c 0 c 0 1 0 1 0 c
The first equality implies that b = 0, the second that a = c. Hence,
  
a 0
Z(T(2, R)) ⊆ : a ∈ R \ {0} = Scal(2, R).
0 a
The reverse inclusion is obvious, so Z(T(2, R)) = Scal(2, R).
34. Suppose that  
1 a b
A = 0 1 c  ∈ Z(UT(3, R)).
0 0 1
Then in particular
     
1 a b 1 1 0 1 1 0 1 a b
0 1 c  0 1 0 = 0 1 0 0 1 c  ,
0 0 1 0 0 1 0 0 1 0 0 1
which implies that c = 0. Moreover,
     
1 a b 1 0 0 1 0 0 1 a b
0 1 c  0 1 1 = 0 1 1 0 1 c  ,
0 0 1 0 0 1 0 0 1 0 0 1
which implies that a = 0. This shows that
  
 1 0 b 
Z(UT(3, R)) ⊆  0 1 0  : b ∈ R .
0 0 1
 

For the other inclusion, note that

       
1 0 b 1 x y 1 x y+b 1 x y 1 0 b
0 1 0 0 1 z  = 0 1 z  = 0 1 z  0 1 0
0 0 1 0 0 1 0 0 1 0 0 1 0 0 1
holds for all x, y, z ∈ R, thus showing that
  
 1 0 b 
Z(UT(3, R)) =  0 1 0  : b ∈ R .
0 0 1
 
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 12

Week 7: Cyclic groups

Back to Week 7 exercises.

35. Note that 1 is a generator, and that the multiple n1 = n mod 30 ∈ Z30 for all
n ∈ N. Using Corollary 11.7, the generators of Z30 are all the numbers in a ∈ Z30
with gcd(a, 30) = 1. These are the numbers 1, 7, 11, 13, 17, 19, 23, 29.
36. The subgroups of a cyclic group G = hai of order n are in one-to-one correspondence
with the divisors of n by Theorem 11.8: for each divisor d, the subgroup had i has
order n/d. Hence, the orders of subgroups are the divisors of n. We get:
(i) 1, 31; (ii) 1, 2, 4, 8, 16, 32; (iii) 1, 3, 11, 33.

37. {0} , h3i , h5i , Z15 .

38. Write f = (1 2 3)(4 5). Then |f | = 6, so the subgroups of hf i are generated by

f, f 2 , f 3 , f 6 . We have
hf i = {(1 2 3)(4 5), (1 3 2), (4 5), (1 2 3), (1 3 2)(4 5), id5 },
hf 2 i = {(1 3 2), (1 2 3), id5 },
hf 3 i = {(4 5), id5 },
hf 6 i = {id5 }.

39. Since m divides n, n = ms for some s ∈ N. Let g be a generator for G. But then hg s i
has order n/(n, s) = n/s = m, and hence it generates a cyclic subgroup of order m.
40. Suppose Q = hri for some r ∈ Q. This means that Q = {mr | m ∈ Z} . But clearly
r/2 6= mr ∀m ∈ Z. Hence r/2 ∈/ hri , but r/2 ∈ Q, so Q 6= hri .

Week 8: Cosets

Back to Week 8 exercises.

41. (i) the left cosets are {0, 3, 6, 9, 12} , {1, 4, 7, 10, 13} and {2, 5, 8, 11, 14} ,
(ii) the left cosets are {e, (1 2)(3 4)} and {(1 2), (3 4)} ,
(iii) the left cosets are R+ and R− = {x ∈ R |x < 0} .

42. For complex numbers α, β, we have |αβ| = |α||β|, so if |α| = r then

αH = {αz : |z| = 1} = {αz : |αz| = r} = {z : |z| = r}.
Hence αH consists of all complex numbers with modulus r, that is all complex
numbers of the same modulus as α. Consequently, the left cosets of H in C\ {0} are
the sets {z ∈ C; |z| = r}, where r ∈ R+ , and there is one coset for each r ∈ R+ .
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 13

43. (i) {a, −a}, a ∈ (0, ∞)

Thus [G : H] = ∞.
(ii) the coset of z ∈ C∗ has the form zR∗ = {rz : r ∈ R∗ }, i.e. the line going
through the origin and z, excluding the origin itself. Each such line contains exactly
one number on, say, the upper half unit circle U = {cos θ + i sin θ : θ ∈ [0, π)}, so
all the distinct cosets are the sets zR∗ for each z ∈ U .
Thus [G : H] = ∞
(iii) the coset of z ∈ C∗ has the form zR+ = {rz : r > 0}, i.e. the half-line
extending from the origin through z, excluding the origin itself. Each such half-line
contains exactly one number on the unit circle S = {cos θ + i sin θ : θ ∈ [0, 2π)}, so
all the distinct cosets are the sets zR+ for each z ∈ S.
Thus [G : H] = ∞.
(iv) 0 + h30i = h30i = {0, 6, 12, 18, 24, 30}, 1 + h30i = {1, 7, 13, ..., 31},
2 + h30i = {2, 8, ..., 32}, 3 + h30i , 4 + h30i , 5 + h30i .
Thus [G : H] = 6.
   
a b 1 x
(v) For an arbitrary ∈ T(2, R) and an arbitrary ∈ UT(2, R), we
0 d 0 1
have     
a b 1 x a ax + b
= .
0 d 0 1 0 d
Since a 6= 0, every y ∈ R is of the form y = ax + b for the right x = (y − b)/a ∈ R.
This shows that
    
a b a y
UT(2, R) = : y∈R ,
0 d 0 d

so there is precisely one coset for each choice of a, d with ad 6= 0.

Thus [G : H =]∞
   
a 0 λ 0
(vi) For an arbitrary ∈ D(2, R) and an arbitrary ∈ H, we have
0 b 0 λ
    
a 0 λ 0 λa 0
= .
0 b 0 λ 0 λb

Since a 6= 0, we can always get λa = 1 for the right choice of λ = 1/a. This shows
that     
a 0 x 0 ∗
H= : x∈R .
0 b 0 xb/a

Thus [G : H] = ∞.
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 14

(vii) Let A, B ∈ GL(n, R). Then B ∈ A SL(n, R) if and only if B −1 A ∈ SL(n, R),
i.e. if and only if det(B −1 A) = 1. But this is the case if and only if
det(A) = det(B). Hence
A SL(n, R) = {B ∈ GL(n, R); det(B) = det(A)} .
Hence there is one coset of SL(n, R) in GL(2, R) for every possible value of the
determinant of an invertible matrix. In other words, these cosets are in 1-1
correspondence with the nonzero real numbers. Thus [G : H] = ∞

44. Every real number r can be written as r = [r] + r0 , where [r] denotes the integral
part of r and 0 ≤ r0 < 1. Let r, s ∈ R with r = [r] + r0 , s = [s] + s0 . By Lemma 12.7,
r and s belong to the same coset of Z in R if and only if s − r ∈ Z. Clearly, this is
the case if and only if r0 = s0 . Hence the cosets of Z in R are in 1-1 correspondence
with [0, 1).

45. First of all we compute the cyclic subgroup H:

(1 2 3 4)2 = (1 3)(2 4), (1 2 3 4)3 = (1 4 3 2), (1 2 3 4)4 = e.
Hence
H = h(1 2 3 4)i = {e, (1 2 3 4), (1 3)(2 4), (1 4 3 2)}.
Then
(1 3 4)H = {(1 3 4)e, (1 3 4)(1 2 3 4), (1 3 4)(1 3)(2 4), (1 3 4)(1 4 3 2)}
= {(1 3 4), (1 2 4 3), (1 4 2), (2 3)}.
Now we could compute the other two cosets in the same way by working out the
relevant four products, but on noting that (2 3) ∈ (1 3 4)H we can say immediately
that
(2 3)H = (1 3 4)H
(since any two cosets either coincide or are disjoint) and on noting that
(1 4 3 2) ∈ H we can say immediately that
(1 4 3 2)H = H.

Week 8: Lagrange’s theorem

46. By Lagrange’s Theorem, applied to K ≤ G, H ≤ G and K ≤ H, we have that

|G| = [G : K] |K| , |G| = [G : H] |H| , |H| = [H : K] |K| .
Substituting |H| from the third equation into the second gives
|G| = [G : H] [H : K] |K| ,
and comparing this with the first equation gives the required result.
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 15

47. By Corollary 13.8 to Lagrange’s Theorem, the order of every element in G divides
the order of G. Hence g n = e for all g ∈ G.

48. The intersection H ∩ K is a subgroup of G. Moreover, H ∩ K is a subgroup of both

H and K. Hence, by Lagrange’s Theorem, |H ∩ K| divides both |H| and |K| . Since
(|H| , |K|) = 1, this yields |H ∩ K| = 1, and hence H ∩ K = {e} .

Week 9: Group homomorphisms

Back to Week 9 exercises.

49. The maps (a), (b), (d), (e), and (f) are homomorphisms, (c) is not a
homomorphism.

(a) ϕ : C∗ → R, ϕ(z) = log |z| (z ∈ C∗ )

Solution: First notice that the binary operation of the group C∗ is
multiplication, and the binary operation of the group R is addition. Thus the
question asks:
Is it true that ϕ(z1 z2 ) = ϕ(z1 ) + ϕ(z2 ) for all z1 , z2 ∈ C∗ ?
(This is precisely what it means for ϕ is homomorphism of groups.)
Using properties of modulus and logarithms, it is easy to see that for all
z1 , z2 ∈ C∗ we have:
ϕ(z1 z2 ) = log(|z1 z2 |) = log(|z1 | · |z2 |) = log(|z1 |) + log(|z2 |) = ϕ(z1 ) + ϕ(z2 ), so
we can conclude that ϕ is a homomorphism of groups.
(b) ϕ : C → R, ϕ(a + bi) = b (a, b ∈ R)
Solution: The binary operation of both C and of R is addition. Thus the
question asks:
Is it true that ϕ((a + bi) + (c + di)) = ϕ(a + bi) + ϕ(c + di) for all
a + bi, c + di ∈ C?
Using basic arithmetic it is easy to see that for all a + bi, c + di ∈ C we have:
ϕ((a + bi) + (c + di)) = ϕ((a + c) + (b + d)i) = b + d = ϕ(a + bi) + ϕ(c + di), so
we can conclude that ϕ is a homomorphism of groups.
 
a b
(c) ϕ : GL(2, R) → R, ϕ = a − d, a, b, c, d ∈ R.
c d
Solution: The binary operation of GL(2, R) is matrix multiplication, whilst the
binary operation of R is addition. Thus the question asks:
Is it true that ϕ(AB)
 = ϕ(A) + ϕ(B)  for all A, B ∈ GL(2, R)?
a b e f
Let’s write A = and B = for two elements of GL(2, R).
c d g h
Then using the definition of matrix multiplication we have:
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 16

    
a b e f ae + bg af + bh
ϕ(AB) = ϕ = ϕ
c d g h ce + dg cf + dh
= (ae + bg) − (cf + dh)

On the other hand:    

a b e f
ϕ(A) + ϕ(B) = ϕ +ϕ = (a − d) + (e − h) The two
c d g h
expressionswe found  for allchoices of A, B ∈ GL(2, R). Indeed,
 are not equal
2 0 1/2 0
when A = and B = for example, we get
0 1 0 1
ϕ(AB) = ϕ(I2 ) = 1 − 1 = 0 whilst ϕ(A) + ϕ(B) = (2 − 1) + (1/2 − 1) = 1/2.
So this is not a homomorphism.
   
a b a 0
(d) ϕ : T(2, R) → D(2, R), ϕ = , a, b, d ∈ R, ad 6= 0
0 d 0 d
Solution: The binary operation of both T(2, R) and D(2, R) is matrix
multiplication. Thus the question asks:
Is it true that ϕ(AB)
 =ϕ(A)ϕ(B) for
 all A,B ∈ T(2, R)?
a b e f
Let’s write A = and B = for two arbitrary elements of
0 d 0 h
T(2, R). Then using the definition of matrix multiplication we have:

     
e f
a b ae af + bh ae 0
ϕ(AB) = ϕ =ϕ =
0 d
0 h 0 dh 0 dh
        
a b e f a 0 e 0 ae 0
ϕ(A)ϕ(B) = ϕ ϕ = =
0 d 0 h 0 d 0 h 0 dh

Since A and B were arbitrary, this shows that ϕ is a homomorphism.

 
1 2 ... n
(e) ϕ : Sn → Sn+1 , where for f = ∈ Sn ,
f (1) f (2) ... f (n)
 
1 2 ... n n+1
ϕ (f ) = ∈ Sn+1
f (1) f (2) ... f (n) n+1

Solution: The binary operation of both Sn and Sn+1 is function composition.

Thus the question asks:
Is it true that ϕ(f ◦ g) = ϕ(f ) ◦ ϕ(g) for all f, g ∈ Sn ?
Let f, g be arbitrary elements of Sn . Then ϕ(f ◦ g) and ϕ(f ) ◦ ϕ(g) are
elements of Sn+1 , i.e. bijective functions from the set {1, . . . , n + 1} to itself.
To determine whether these two functions are equal, we need to see if they
agree when we evaluate them at each element of {1, . . . , n + 1}.
On the one hand we have:
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 17

 
1 2 ... n n+1
ϕ (f ◦ g) =
f ◦ g (1) f ◦ g (2) ... f ◦ g (n) n + 1
 
1 2 ... n n+1
=
f (g (1)) f (g (2)) ... f (g (n)) n + 1

On the other hand we have:

   
1 2 ... n n+1 1 2 ... n n+1
ϕ(f ) ◦ ϕ(g) = ◦
f (1) f (2) ... f (n) n + 1 g (1) g (2) ... g (n) n + 1
 
1 2 ... n n+1
= ,
f (g (1)) f (g (2)) ... f (g (n)) n + 1

where the final equality follows by observing that each of the values
g(1), . . . , g(n) lie in the set {1, . . . , n}, and so composing these two functions
from right to left in effect applies g and then f to each element of {1, . . . , n},
whilst n + 1 remains fixed. This shows that ϕ is a homomorphism.
(f) ϕ : Z → Sn , ϕ (k) = (1 2 3 · · · n)k ∈ Sn , k ∈ Z
Solution: The binary operation of Z is addition and the binary operation of Sn
is function composition. Thus the question asks:
Is it true that ϕ(j + k) = ϕ(j) ◦ ϕ(k) for all j, k ∈ Sn ?
Using properties of powers it is easy to see that for all j, k ∈ Z we have

ϕ(j + k) = (1 2 3 · · · n)j+k = (1 2 3 · · · n)j (1 2 3 · · · n)k = ϕ(j) ◦ ϕ(k)

and so ϕ is a homomorphism.

50. Since an = eG , we have ϕ(a)n = ϕ(an ) = ϕ(eG ) = eH . Hence ϕ(a)n = eH , and so the
order of ϕ(a) divides n.
In more detail: Using the facts that an = eG and that ϕ : G → H is a
homomorphism we have ϕ(a)n = ϕ(a) · · · ϕ(a) = ϕ(a · · · a) = ϕ(an ) = ϕ(eG ) = eH .
Hence ϕ(a)n = eH , and so the order of ϕ(a) divides n. (Take a look at Lemma 8.3 if
you don’t see this immediately.)

51. If ϕ is a homomorphism we have ϕ(ab) = ϕ(a)ϕ(b) for all a, b ∈ G, i.e.

(ab)−1 = a−1 b−1 . Taking inverses on both sides gives ab = ba, i.e. G is abelian.
Conversely, if G is abelian we have ϕ(ab) = (ab)−1 = b−1 a−1 = a−1 b−1 = ϕ(a)ϕ(b).
Hence ϕ is a homomorphism.

Week 9: Isomorphisms and group-theoretic properties

c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 18

52. For a, b ∈ G we have

ϕx (ab) = x (ab) x−1 = xax−1 xbx−1 = ϕx (a) ϕx (b) .

Therefore, ϕx is a homomorphism. Finally, ϕ is one-to-one as xax−1 = xbx−1

implies a = b, and ϕ is onto as for all a ∈ G, a = ϕx (x−1 ax) . Hence ϕ is an
isomorphism.
53. (a) S4 is not abelian, but Z24 is abelian.
(b) Z is cyclic, but Q is not cyclic (see Ex. 40).
(c) In Q all non-identity elements have infinite order, but Q∗ has an element of
order 2, namely −1.
(d) GL(2, R) has an element of order 2, e.g. the scalar matrix with −1 on the main
diagonal, but in UT(3, R) all elements except the identity have infinite order:
 n    n  
1 a ∗ 1 na ∗ 1 0 c 1 0 nc
 0 1 b  =  0 1 nb  ,  0 1 0  =  0 1 0  .
0 0 1 0 0 1 0 0 1 0 0 1

Week 10: Conjugacy

Back to Week 10 exercises.

54. In a group G let b = xax−1 . If a has order n, then an = e and am 6= e for all m ∈ N
with m < n. But then
n
bn = (xax−1 ) = xax−1 xax−1 xax−1 ...xax−1 = xan x−1 = xex−1 = e, and for m < n
we have bm 6= e as bm = e implies xam x−1 = e. This implies that am = e,
contradicting the assumption that the order of a is n. If a has infinite order, the
same argument shows that b has infinite order as well.

Week 10: Conjugacy in symmetric groups

 
1 2 3 4 5 6
55. (a) Yes, e.g. h = . Note that h is not unique.
3 4 5 6 1 2
(b) No.
 
1 2 3 4 5 6
(c) Yes, e.g. h = .
1 6 3 2 5 4
 
1 2 3 4 5 6
(d) Yes, e.g. h = .
5 6 1 2 4 3
56. The number of conjugacy classes in Sn is equal to the number of partitions of the
number n. Counting the partitions of 5 and 6 gives that S5 has 7, and S6 has 11
conjugacy classes.
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 19

57. By Theorem 17.7, the conjugacy class (1 2) (3 4)S4 consists of all elements of S4
with the same cycle structure as (1 2) (3 4) . Hence

(1 2) (3 4)S4 = {(1 2) (3 4) , (1 3) (2 4) , (1 4) (2 3)},

and then the class formula gives

|C((1 2) (3 4))| = |S4 | / (1 2) (3 4)S4 = 4!/3 = 8.

Finally,

C((1 2) (3 4)) = {e, (1 2) , (3 4) , (1 2) (3 4) , (1 3 2 4) , (1 4 2 3), (1 3) (2 4) , (1 4) (2 3)}.

58. Let f = (1 2 · · · n) ∈ Sn . By Theorem 17.7, the conjugacy class f Sn consists

precisely of all cycles of length n in Sn . By the class formula,
f Sn =[Sn : C(f )] = |Sn | / |C(f )| . But |C(f )| = n. Hence f Sn = n!/n = (n − 1)!.

59. For an arbitrary permutation f 6= e in Sn we will find a permutation g such that

gf g −1 6= f, i.e. f and g don’t commute. Let f be written as a composite of disjoint
cycles: f = (a b · · · ) .... By Lemma 17.1, we then have

gf g −1 = (g(a) g(b) · · · ) ...

Hence g = (b c) with c 6= a, b does the job as f moves a to b, whereas gf g −1 moves

a to c.

Week 11: Normal subgroups

Back to Week 11 exercises.

60. (a) normal as UT(2, R) is the kernel of the homomorphism from Ex. 49 (d),
(b) normal as H = Z(G),
(c) normal as H ⊆ Z(G), in fact, one can show that H = Z(G).

61. (a) not normal as, for example, (1 2) (1 2 3 4) (1 2)−1 = (2 1 3 4) ∈

/ h(1 2 3 4)i,
(b) normal as H is the union of two conjugacy classes, namely eS4 and (1 2) (3 4)S4 .

62. We know that H ∩ K ≤ G. It remains to check that it is normal. Let h ∈ H ∩ K

and let g ∈ G. Then h ∈ H and ghg −1 ∈ H since H C G. Likewise, h ∈ K and
ghg −1 ∈ K since K C G. Hence ghg −1 ∈ H ∩ K, so H ∩ K C G.
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 20

63. First we show that HK is a subgroup. Since HK 6= ∅, we can apply the subgroup
criterion. Let x, y ∈ HK, i.e. x = h1 k1 and y = h2 k2 for some
h1 , h2 ∈ H, k1 , k2 ∈ K. Now
xy = h1 k1 h2 k2 = h1 h2 h−1 −1
2 k1 h2 k2 = (h1 h2 )(h2 k1 h2 k2 ),

where h1 h2 ∈ H as H ≤ G and h−1 −1

2 k1 h2 k2 ∈ K as h2 k1 h2 ∈ K (because K C G)
and k2 ∈ K. Hence xy ∈ K. Also,
x−1 = (h1 k1 )−1 = k1−1 h−1 −1
h1 k1 h−1


1 = h1 1 ∈ HK.
So HK ≤ G. Finally, let g ∈ G, x as before, then
gxg −1 = gh1 k1 g −1 = gh1 g −1 gk1 g −1 ∈ HK
as gh1 g −1 ∈ H and gk1 g −1 ∈ K since H, K C G. (Note: For the proof that HK is a
subgroup, we have only used that K is normal in G, and not that H is normal in G.
For the proof that HK is normal, however, we need that both H and K are
normal.)
64. (a) Since H C G, we have ghg −1 ∈ H for all h ∈ H and all g ∈ G, and since
K ⊆ G, we have it in particular for all g ∈ K. Hence H C K.
(b) No. For example, take G = S4 ,
K = {e, (1 2) (3 4) , (1 3) (2 4) , (1 4) (2 3)} ; and H = h(1 2)(3 4)i. Then
H
K, since K is abelian. Moreover, K
S4 by Exercise 60 (d), but H 6 S4
since, e.g., (1 2 3)(1 2)(3 4)(1 2 3)−1 = (1 4)(2 3) ∈/ h(1 2)(3 4)i.
65. Let x ∈ N, y ∈ M and consider the element xyx−1 y −1 . Since M is normal,
xyx−1 ∈ M, and hence xyx−1 y −1 ∈ M. Since N is normal, yx−1 y −1 ∈ N, and hence
xyx−1 y −1 ∈ N. But then xyx−1 y −1 ∈ M ∩ N = {e}, so xyx−1 y −1 = e. But this
implies xy = yx (multiply both sides of xyx−1 y −1 = e by yx).

66. Let a be a generator of H, i.e. H = {e, a} . We need to show that a ∈ Z(G), i.e.
that ag = ga for all g ∈ G. This is the same as to say that gag −1 = a for all g ∈ G.
But this is the case: Since H is normal in G we have gag −1 ∈ H for all g ∈ G, i.e.
either gag −1 = a or gag −1 = e. But the latter is impossible as it implies a = e
contradicting the assumption that a has order 2.

Week 12: Factor groups and first isomorphism theorem

Back to Week 12 exercises.

67. The map ϕ : R∗ → {±1} given by ϕ(x) = |x|
x
is a homomorphism. It is obvious that
Im ϕ = {±1} and Ker ϕ = R+ . Hence R /R+ ∼
∗
= {±1} by the first isomorphism
theorem.
c University of Manchester Algebraic Structures 1 Solutions to Exercises, Page 21

68. The map ϕ : C∗ → R∗ given by ϕ(z) = |z| is a homomorphism with Im ϕ = R+ and

Ker ϕ = {z ∈ C; |z| = 1} . Hence C∗ / {z ∈ C; |z| = 1} ∼
= R+ .

69. Apply the first isomorphism theorem to the homomorphism from Exercise 49 (d).

70. The hinted map is a homomorphism from T (2, R) to R∗ with kernel N and image
R∗ . Then the first isomorphism theorem applies.

71. The hinted map is a homomorphism from UT (3, R) to R with kernel N and image
R. Then the first isomorphism theorem applies.

72. (a) This repeats Q68. Ker ϕ = {z ∈ C∗ : |z| = 1}, Im ϕ = R. By the first
isomorphism theorem, we have C∗ /{z ∈ C∗ : |z| = 1} ∼
= R.
(b) Ker ϕ = {a + 0i ∈ C : a ∈ R} = R, Im ϕ = R. By the first isomorphism
theorem, we have C/R ∼= R.
(c) This map was not a homomorphism.
(d) Ker ϕ = UT(2, R), Im ϕ = D(2, R). By the first isomorphism theorem, we have
T(2, R)/ UT(2, R) ∼
= D(2, R).
(e) Ker ϕ = {e}, Im ϕ = {g ∈ Sn+1 : g(n + 1) = n + 1}. By the first isomorphism
theorem, we have §n /{e} ∼
= {g ∈ Sn+1 : g(n + 1) = n + 1}.
(f) Ker ϕ = hni , Im ϕ = h(1 2 3 · · · n)i. By the first isomorphism theorem, we have
Z/nZ ∼= h(1 2 3 · · · n)i.