Chapter 1 1

A FIRST G RADUATE C OURSE IN A LGEBRA

ABSTRACT ALGEBRA

D R . M OHAMMED S ULIMAN E L -A TRASH

P ROFESSOR OF M ATHEMATICS
M ATH . D EPARTMENT
I SLAMIC U NIVERSITY OF G AZA
G AZA , P ALESTINE
Prof. Mohammed El-Atrash 2

T ABLE OF C ONTENTS
CHAPTER I 3
1.1 G ROUPS . 3

1.2 C YCLIC G ROUPS 6

1.3 COSETS AND NORMAL SUBGROUPS 7
1.4 PERMUTATION GROUPS 10
CHAPTER II 15
2.1 NORMAL SUBGROUPS 15
2.2 PRODUCT OF SETS 16
2.3 HOMOMORPHISM THEOREMS 19
2.4 GROUP ACTION 21
CHAPTER III 27
3.1 SYLOW THEORY 27
3.2 NILPOTENT GROUPS 31
3.3 DIRECT PRODUCT 34
3.4 PERMUTATION GROUPS 37
3.5 OPERATOR GROUPS 41
CHAPTER IV 45
4.1 RINGS 45
4.2 INTEGRAL DOMAINS 50
4.3 DEFINITION OF A MODULE 54
4.4 THE JACOBSON RADICAL 61
CHAPTER V 65
5.1 CHAIN CONDITIIONS 65
5.2 SEMIPRIMITIVE RINGS 69
5.3 COMPOSITION SERIES 73
5.4 SEMISIMPLE MODULES 76
 Abstract Algebra Notes 3

‫ﺑﺴﻢ ﺍﷲ ﺍﻟﺮﲪﻦ ﺍﻟﺮﺣﻴﻢ‬

Chapter I

1. Groups, definitions
2. Cyclic groups
3. Permutation groups
4. Cosets and normal subgroups
5. Factor groups
6. Group homomorphisms

1.1 Groups.

Modern definition of a group (Caley's definition).

1.1.1 D EFINITION . A group (G, .) is a nonempty set G together with a

binary operation . on G such that the following conditions hold:
(i) Closure: For all a, b ∈ G the element a.b is a uniquely defined element of G.
(ii) Associativity: For all a, b, c ∈ G, we have a.(b.c) = (a.b).c.
(iii) Identity: There exists an identity element e ∈ G such that e.a = a and a.e = a
for all a ∈ G.
(iv) Inverses: For each a ∈ G there exists an inverse element a -1 ∈ G such that: a.a -1
= e and a -1 .a = e.
We will usually simply write ab for the product a.b.

Remark: Identity e is unique, and inverse of any element a is unique.

1.1.2 P ROPOSITION . (Cancellation Property for Groups) Let G be a

group, and let a, b, c ∈ G.
(a) If ab = ac, then b = c.
(b) If ac = bc, then a = b.
 Prof. Mohammed El-Atrash 4

P ROOF . For (a) multiply by a -1 from the left.

For (b) multiply by c -1 from the right.

1.1.3 P ROPOSITION . Let G be a set with associative binary operation.

Assume that ∃ e ∈ G such that:
(a) Right identity : ae = a.
(b) Right inverse: for every a ∈ G; ∃ b ∈ G with ab = e.
Then G is a group.

P ROOF : For x, y, u ∈G, if xu = yu; we claim that x = y.

By (b), choose v ∈ G with uv = e. Now
x = xuv = yuv = ye = y.
Next, we want to show that e is left identity i.e., for every a ∈ G we have e.a =
a. Choose b ∈G with ab = e.
(ea)b = e(ab) = ee = e = ab.
Then, by our claim, ea = a.
It remains to show that if ab = e then ba = e.
(ba)b = b(ab) = be = b = eb,
therefore again by the claim ba = e.

1.1.4 D EFINITION . A group G is said to be abelian if ab = ba for all a, b ∈

G.

1.1.5 D EFINITION . A group G is said to be a finite group if the set G has

a finite number of elements. In this case, the number of elements is called the
order of G, denoted by |G|.

1.1.6 D EFINITION . Let a be an element of the group G. If there exists a

positive integer n such that a n = e, then a is said to have finite order, and the
smallest such positive integer is called the order of a, denoted by o(a).
If there does not exist a positive integer n such that a n = e, then a is said to have
infinite order.
Abstract Algebra Notes 5

1.1.7 D EFINITION . Let G be a group, and let H be a subset of G. Then H is

called a subgroup of G if H is itself a group, under the operation induced by
G. (we denote this by H ≤ G).

1.1.8 P ROPOSITION . Let G be a group with identity element e, and let H

be a non-empty subset of G. Then H is a subgroup of G if and only if the
following conditions hold:
(i) ab ∈ H for all a, b ∈ H;
(ii) a -1 ∈ H for all a ∈ H.

P ROOF : E XERCISE .

Theorem 1.1.8 states the sufficient conditions for a no-empty subset H to be a

subgroup. I.e., instead of checking all four conditions of the group we only
check two conditions.
In the following Exercise we can show even one condition is enough.

E XERCISE . Let G be a group with identity element e, and let H be a non-empty

subset of G. Then H ≤ G if and only if the following conditions hold:
ab -1 ∈ H for all a, b ∈ H;

R EMARK . Identity in H is the same as the identity in G.

1.1.9 D EFINITION . Let G be a group. Let x ∈ G. The centralizer of x is

the set C G (x) = { y ∈ G| xy = yx }.

E XERCISE . Show that C G (x) is a subgroup of G.

E XERCISE . Let π be a collection of subgroups of G. Show that the set H

= ∩{K| K ∈ π } is a subgroup of G.

1.1.10 D EFINITION . The center Z(G) = ∩{C G (x) for all x ∈ G}.

E XERCISE . Show that Z(G) is a subgroup of G.

 Prof. Mohammed El-Atrash 6

1.2 Cyclic Groups

1.2.1 D EFINITION . Let X be a non-empty set of G. The subgroup
generated by X is 〈 X 〉 = ∩ {H ≤ G | X ⊂ H}.
We write 〈 a 〉 for 〈 {a} 〉 . 〈 a 〉 is called the cyclic subgroup generated by a.
The group G is called a cyclic group if there exists an element a ∈ G such that G =
〈 a 〉 . In this case a is called a generator of G.

E XERCISE . Show that 〈a〉 = { a n | for some n ∈ Z}.

1.2.2 D EFINITION . For a ∈ G, the order of a (denoted by o(a)) is the least

positive integer n such that a n = e. If no such integer exist then we say that
the order of a is infinite.

1.2.3 P ROPOSITION . Let a be an element of the group G.

(a) If a has infinite order, and a k = a m for integers k, m, then k = m.
(b) If a has finite order and k is any integer, then a k = e if and only if o(a)|k.
(c) If a has finite order o(a) = n, then for all integers k, m, we have a k = a m
if and only if k ≡ m (mod n). Furthermore, | 〈 a 〉 |= o(a).

P ROOF . (a) Assume that k > m. Since a k = a m then

a k-m = a m-m = a 0 = e,
Since o(a) is infinite then k-m = 0. Hence k = m.
(b) Let n = o(a). By division algorithm, we have
k = qn + r with 0 ≤ r < n.
If r > 0, then

a k = a qn + r = a qn a r .

Therefore

a r = a k a -qn = ee = e.

This contradicts that the order of a is n.

Therefore r = 0. Thus n | k.
 Abstract Algebra Notes 7

(c) a k = a m iff a k-m = e.

Therefore by (b), n | k-m, then k ≡ m (mod n).

Conversely if k ≡ m (mod n) then n | k-m. Therefore k-m = qn for some integer
q, so
a k-m = (a n ) q = e q = e.

Furthermore, the only different elements are

{a 0 = e, a, a 2 , …, a n-1 }.

Hence |〈a〉| = o(a).

EXERCISE. Let G be a group. And let H ⊂ G with |H| < ∞.

Show that H is a subgroup of G iff xy ∈ H for all x, y ∈ H.

1.2.4 L EMMA . Let G be a group and let X ⊂ G. Assume that xy = yx for all
x, y ∈ X. Then 〈 X 〉 is abelian subgroup of G.

P ROOF . By hypothesis if x ∈ X then X ⊂ C(x).

Thus 〈X〉 ⊂ C(x) for all x ∈ X.
It follows that x ∈ C(〈X〉), for all x ∈ X.
Therefore X ⊂ C(〈X〉), and hence 〈X〉 ⊂ C(〈X〉). The proof is complete.

1.3 Cosets and Normal Subgroups

1.3.1 D EFINITION . Let H ≤ G. for a ∈ G the set Ha = {ha| h ∈ H}is called
right coset of a. The set aH = {ah| h ∈ H}is called left coset of a.

1.3.2 P ROPOSITION . Let H ≤ G. let a, b ∈ G. Then the following hold

(a) Ha = Hb iff ab -1 ∈ H.
(b) If Ha ∩ Hb ≠ ∅ then Ha = Hb.
(c) G = ∪ {Ha| a ∈ G}.
(d) |Ha| = |Hb| = |H|.

P ROOF . (a) Let a ∈ Ha then a = hb for some h ∈ H.

Thus ab -1 = h ∈ H.
 Prof. Mohammed El-Atrash 8

Conversely if ab -1 ∈ H, then ab -1 = h for some h ∈ H.

Therefore a = hb.
Now for x ∈ Ha ;
x = h'a for h' ∈ H,
thus x = h'hb ∈ H.
Hence Ha ⊂ Hb.
The converse is similar.
(b) Let z ∈ Ha ∩ Hb then
z = h'a = h''b for h', h'' ∈ H.
It follows that a = h' -1 h''b. Therefore
ha = hh' -1 h''b, so Ha ⊂ Hb.
The reverse inclusion is similar. Hence Ha = Hb.
(c) since a ∈ Ha then G ⊂ ∪{Ha| a ∈ G}.
The reverse inclusion is obvious.
(d) The map ψ: Ha → Hb that maps ha → hb is 1-1 and onto.

1.3.3 D EFINITION . The number of different cosets of a subgroup H of G is

called the index of H in G. and is denoted by [G:H] or |G:H|.

1.3.4 T HEOREM . (Lagrange's) If H is a subgroup of the finite group

G, then the order of H is a divisor of the order of G.

P ROOF . |G| = [G:H]|H|.

1.3.5 C OROLLARIES TO L AGRANGE ' S T HEOREM ( RESTATED ):

(a) For any a ∈ G, o(a) is a divisor of |G|.
(b) For any a ∈ G, a n = e, for n = |G|.
(c) Any group of prime order is cyclic.

P ROOF . E XERCISE .
Abstract Algebra Notes 9

1.3.6 T HEOREM . Every subgroup of a cyclic group is cyclic.

P ROOF . Let H be a subgroup of a cyclic group G. Let G = 〈a〉 with o(a) = n. Let
m be the smallest positive integer with a m ∈ H. We will show that b = a m is the
generator of H by showing that every element x of H; x is a power of b.
Now let x ∈ H, since H ⊂ G then x = a k for some integer k. Using division
algorithm, there are two integers q, r such that
k = qm + r with 0 ≤ r < m.
It follows that a k = a qm + r . Then
a r = a k - qm = a k (a m ) -q ∈ H.
Then a r ∈ H. but r < m. Therefore r = 0.
Hence x = a k = (a m ) q .
Hence H is cyclic with H = 〈a m 〉.
 Prof. Mohammed El-Atrash 10

1.4 Permutation Groups

1.4.1 D EFINITION . Let G 1 and G 2 be groups, and let ϕ :G 1 → G 2 be a
function. Then ϕ is said to be a group homomorphism if ϕ satisfies
ϕ (ab) = ϕ (a) ϕ (b) for all a, b ∈ G 1 .
If ϕ is one-to-one and onto ϕ is called isomorphism, in this case G 1 is said to be
isomorphic to G 2 , and this is denoted by G 1 ≈ G 2 .
And in the case that G 1 is the same as G 2 ; ϕ is called an automorphism of G 1 .

1.4.2 P ROPOSITION . Let ϕ :G 1 → G 2 be an isomorphism of groups.

(a) If a has order n in G 1 , then ϕ (a) has order n in G 2 .
(b) If G 1 is abelian, then so is G 2 .
(c) If G 1 is cyclic, then so is G 2 .

1.4.3 D EFINITION . A permutation of the set S is a one to one and onto

function. The set of all permutations of a set S is denoted by Sym(S). The set
of all permutations of the set {1,2,...,n} is denoted by S n .

1.4.4 P ROPOSITION . If S is any nonempty set, then Sym(S) is a group

under the operation of composition of functions.

1.4.5 D EFINITION . The set of all automorphisms of a group G is called the

automorphism group and is denoted by Aut(G) .

1.4.6 E XERCISE . Let G be a group. Let g ∈ G. define σ g : G → G by σ g (x) = g 1 xg

for x ∈ G. Show that σ g is an automorphism of G.

( σ g is called an inner automorphism of G).

The set of all inner automorphism of G is denoted by Inn(G).

Notation. Usually g -1 xg is written as x g because it follows the same rules of

exponentiation. i.e., h -1 (g -1 xg)h = (h -1 g -1 )x (gh) = x gh .
Abstract Algebra Notes 11

E XERCISE .
(a) Show that Aut(G) is a group.
(b) Show that Inn(G) ≤ Aut(G) ≤ Sym(G).
(c) Show that σ g is the identity automorphism iff g ∈ Z(G).

E XERCISE . Let S, T be two sets and let α : S → T be a bijection. Show that

Sym(S) ≅ Sym(T).

1.4.6 T HEOREM . Let G cyclic group.

(a) If G is infinite, then G ≈ Z.
(b) If |G| = n, then G ≈ Z n .

P ROOF . E XERCISE .

1.4.7 P ROPOSITION . Let G = 〈 a 〉 be a cyclic group with |G| = n.

(a) If m ∈ Z, then 〈 a m 〉 = 〈 a d 〉 , where d = gcd(m, n), and a m has order n/d.
(b) The element a k generates G if and only if gcd(k, n) = 1.
(c) The subgroups of G are in one-to-one correspondence with the positive divisors of
n. (i.e., if d | n then there is a subgroup of order d.)
(d) If m and k are divisors of n, then 〈 a m 〉 ⊂ 〈 a k 〉 if and only if k | m.

P ROOF . (a) Let x ∈ 〈a d 〉, then x = (a d ) k for some integer k.

We need to show that x is a power of a m . To do this we use the fact that d =
(m, n) that is; there are two integers s, t such that
d = sm + tn.
It follows that
a d = a sm + tn
= (a m ) s (a n ) t
= (a m ) s .
It follows that
x = (a d ) k ∈ 〈a m 〉.
Hence 〈a d 〉 ⊂ 〈a m 〉.
It follows that 〈a d 〉 = 〈a m 〉.
Prof. Mohammed El-Atrash 12

The reverse inclusion is easier because a m ∈ 〈a d 〉.

Thus 〈a m 〉 ⊂ 〈a d 〉.
To see that the order of a m is n/d note that
o(a m ) = o(a d ) = n/d.
(b) By part (a), a k generates G iff o(a k ) = n iff 1 = g.c.d(k, n).
Parts (c), (d) are left as an E XERCISE .

1.4.8 C OROLLARY . The number of generators of a cyclic group of order n

is
ϕ (n) = |{r| 1 ≤ r ≤ n-1,(r, n) = 1}|. Euler function ϕ (n).

1.4.9 A PPLICATION . Let n > 0. Then n = ∑ϕ (d)

d n

P ROOF . Let G be a cyclic group of order n. write α(k) = number of elements in

G of order k. Clearly ∑α ( k ) = n.
k

If x ∈ G has order k then 〈x〉 is the only subgroup of G of order k. Thus if α(k)

≠ 0 then k | n. So n = ∑α ( k ) .
kn

Now G = 〈g〉, If x, y ∈ G, |〈x〉| = |〈y〉| = k then o(x) = o(y) = k.

But ∃ only one subgroup of order k. So there are exactly ϕ(k) elements of order

k, i.e., α(k) = ϕ(k). Hence n = ∑ϕ( k ) .

kn
Abstract Algebra Notes 13

1.4.10 D EFINITION . Let G be a group. If there exists a positive integer N

such that a N = e for all a ∈ G , then the smallest such positive integer is called
the exponent of G.

1.4.11 P ROPOSITION . Let G be a group, and let a, b ∈ G be elements such

that ab = ba. If the orders of a and b are relatively prime, then o(ab)=o(a)o(b).

1.4.12 P ROPOSITION . Let G be a finite abelian group.

(a) The exponent of G is equal to the order of any element of G of maximal order.
(b) The group G is cyclic if and only if its exponent is equal to its order.

P ROOF . (a) E XERCISE .

(b) Let N be the exponent of G. By (a) there is an element g ∈ G with o(g) = N
= |G|, therefore G is cyclic.
Conversely if G = 〈g〉 is cyclic then o(g) = |G| = N.

1.4.13 C OROLLARY . Let G be a finite group of order n.

(a) For any a ∈ G, o(a) is a divisor of n.
(b) For any a ∈ G, a n = e.

E XAMPLE . (Euler's theorem) Let G be the multiplicative group of congruence

classes modulo n. G = { k | 1 ≤ k < n, (k, n) = 1}

The order of G is given by ϕ(n), and so by C OROLLARY ( ), raising any

congruence class to the power ϕ(n) must give the identity element.

1.4.14 C OROLLARY . Any group of prime order is cyclic.

P ROOF . Let G be a group of order p where p is prime. Let e ≠ a ∈ G. 〈a〉 is a

subgroup of G. By Lagrange's Theorem |〈a〉| divides the order of |G| = p.
It follows that |〈a〉| = p. Hence 〈a〉 = G, i.e., G is cyclic.

1.4.15 D EFINITION . Any subgroup of the symmetric group Sym(S) on a set

S is called a permutation group or group of permutations.
Prof. Mohammed El-Atrash 14

1.4.16 T HEOREM . (Cayley) Every group is isomorphic to a permutation

group.

P ROOF . Let G be a group. For every g ∈ G we will show that multiplication by

g is a permutaion. Let
π g : G Æ G,
be defined as follows
π g (x) = gx.
First π g is 1-1
since π g (x) = π g (y) iff gx = gy iff x = y.
Second, π g is onto,
since for every y ∈ G, π g (g -1 y) = y.
Hence π g ∈ Sym(G). Let
σ : G Æ Sym(G) be defined by
σ(g) = π g .
To see that σ is 1-1,
let σ(g) = σ(h), then π g = π h .
It follows that π g (e)= π h (e), i.e., ge = he, and thus g = h.
To see that σ is a homeomorphism;
note that σ(gh) = π gh = π g π h = σ(g) σ(h).
(Note that composition of permutations here is that we apply π g first and then
π h ). This completes the proof.
 Abstract Algebra Notes 15

Chapter II

1. Normal Subgroups
2. Product of Sets
3. Homomorphism Theorems
4. Group Action

2.1 Normal subgroups

2.1.1 D EFINITION . Let H ≤ G then H is normal subgroup of G if H g = H
for all g ∈ G. ( we denote this by H  G).

Example.
(1) {e}  G, G  G.
(2) If G is abelian then every subgroup of G is normal in G.

2.1.2 P ROPOSITION . Let H ≤ G then H  G iff H g ⊂ H for all g ∈ G.

P ROOF . We need to show that H g = H for all g ∈ G.

Since H g ⊂ H for all g ∈ G, take g -1 for g i.e., (H) g-1 ⊂ H. Then ((H) g-1 ) g ⊂ H g
It follows that H ⊂ H g . Hence H = H g .

1.56 E XERCISE . Show that for any group G; Inn(G)  Aut(G).

2.1.3 D EFINITION . Let H ≤ G, H is called characteristic in G if σ (H) = H

for all automorphisms σ ∈ Aut(G).

2.1.4 C OROLLARY . If H is characteristic in G then H  G.

2.1.5 D EFINITION . For x, y ∈ G,

we define the commutator [x, y] = x -1 y -1 xy.

2.1.6 C OROLLARY . xy = yx iff [x, y] = e.

 Prof. Mohammed El-Atrash 16

2.1.7 D EFINITION . The derived or the commutator subgroup of a group

G (denoted by G') is the smallest subgroup generated by all commutators.
i.e., G' = 〈 {[x, y] | x, y ∈ G} 〉 .

2.1.8 C OROLLARY . G is abelian iff G' = {e}.

E XERCISE . Show that G' is characteristic in G.

2.1.9 D EFINITION . Let G be a group. G (n) = the derived subgroup of G (n-1) .

By this definition we have :
G ≥ G' ≥ G'' ≥ … ≥ G (n) ≥ …

2.1.10 P ROPOSITION . If N  G, M characteristic in N then M  G.

P ROOF . For g ∈ G, let σ g ∈ Inn(G). We need to show that σ g (M) = M. By

normality of N we have: σ g (N) = N. But Inn(G) ≤ Aut(G). Therefore σ g is an
automorphism of G, it follows then that σ g an automorphism of N. Since M
characteristic in N, then σ g (M) = M.

E XERCISE . Let C  G, where C is cyclic subgroup of G. Suppose that |G| < ∞.

Show that if K ≤ C then K  G.

1.67 E XERCISE If M, N  G, M∩N = {e} then M ≤ C G (N) ( and N ≤ C G (M)).

2.2 Product of Sets

Let X, Y be two sets of a group G. Write XY = { xy | x ∈ X, y ∈ Y }. We write
Xy for X{y}, and if H ≤ G, (as we have seen before Hx is called right coset, xH
is called left coset).

2.2.1 T HEOREM . Let H, K ≤ G then HK is a subgroup of G iff HK = KH.

P ROOF . Assume HK ≤ G, then we have H ⊂ HK, K ⊂ HK and since HK is a

subgroup i.e., closed under multiplication then KH ⊂ HK.
For the reverse inclusion, let u ∈ HK, then u -1 ∈ HK. Write u -1 = hk. It follows
that u = (hk) -1 = k -1 h -1 ∈ KH. Thus HK = KH.
Abstract Algebra Notes 17

Conversely, Assume that HK = KH. Let u, v ∈ HK, write u = h 1 k 1 , v = h 2 k 2 . It

follows that uv = h 1 k 1 h 2 k 2 , but k 1 h 2 ∈ KH = HK, therefore k 1 h 2 = h 3 k 3 ∈ HK.
We say that X and Y permutes if XY = YX.

2.2.2 P ROPOSITION . Let N  G, and let X ⊂ G, then NX = XN.

P ROOF . Let u ∈ NX, then u = nx, n ∈ N, x ∈ X. write u = xx -1 nx = xn x ∈ xN ⊂

XN. Therefore, NX ⊂ XN. Similarly, XN ⊂ NX. Hence XN = NX.

2.2.3 C OROLLARY . (a) Let H ≤ G, N  G, then HN ≤ G.

(b) Let N  G, g ∈ G, then Ng = gN.

E XERCISE . Prove that the number of left cosets is the number of right cosets.

2.2.4 T HEOREM . Let N  G. Define a binary operation on the set of all

right cosets of N as follows: (Nx)(Ny) = Nxy. Then the set of all right cosets
with this operation is a group.

P ROOF . First we will show that this operation is well defined. This means that
if Nx = Nx', Ny = Ny' then (Nx)(Ny) = Nxy = (Nx')(Ny') = Nx'y'. Now since
(Nx) = (Ny) then by P ROPOSITION (), x'x -1 ∈ N, similarly y'y -1 ∈ N. It follows
that:
Nx'y' = Nx'x -1 xy'y -1 y = Nxy'y -1 y = (xN)y'y -1 y = x(Ny'y -1 y) = xNy = Nxy. Thus
this binary operation is well defined.
The set of all cosets is closed under this operation. Associativity is clear. The
identity element is Ne = N, since NNx = Nx. The inverse of Nx is Nx -1 .

2.2.5 D EFINITION . The group defined in the last theorem is called the
Factor group ( or quotient group). And is denoted by G/N.

Note. If |G| < ∞ then |G/N| = |G|/|N|.

E XERCISE . Let H ≤ G. Let S = {Hg | g ∈ G}. Show that for the multiplication
(Nx)(Ny) = Nxy to be defined H has to be normal in G.
Prof. Mohammed El-Atrash 18

E XAMPLE . For any group G, if N = {e} then G/N ≅ G. and if N = G then

G/N ≅ {e}.

E XAMPLE . For the group G = (Z, +), if N = nZ then G/N = Z/nZ = Z n (integers
modulo n).

2.2.6 D EFINITION . Let N  G, the map π : G Æ G/N defined by π (g) = Ng

is a surjective homomorphism called the canonical ( or natural)
homomorphism.

2.2.7 D EFINITION . If θ : G Æ K is a homomorphism, kernel of θ ( denoted

by ker( θ )) = { g ∈ G| θ (g) = e K }.

E XERCISE . Show that ker(θ)  G.

E XERCISE . A subgroup H is normal in G iff H is a kernel of some

homomorphism.

2.2.8 P ROPOSITION . Let N  G, then G/N is abelian iff G' ⊂ N.

P ROOF . Let π be the canonical homomorphism, π is surjective thus elements of

G/N are of the form π(g) for some g ∈ G. Now
G/N is abelian iff [π(g), π(h)] = e, for all g, h ∈ G.
iff π([g, h]) = e, for all g, h ∈ G.
iff [g, h] ∈ ker(π),for all g, h ∈ G.
iff [g, h] ∈ N, for all g, h ∈ G.
iff G' ⊂ N.

2.2.9 C OROLLARY . G/G' is abelian.

In fact G' is minimal among all normal subgroups with abelian factor group.
G/G' is called abelianization of G.

2.2.10 P ROPOSITION . Let ϕ : G Æ H be a homomorphism. Let N = ker(ϕ).

Then ϕ(x) = ϕ(y) iff Nx = Ny, for all x, y ∈ G.
 Abstract Algebra Notes 19

P ROOF . If ϕ(x) = ϕ(y) then ϕ(xy -1 ) = ϕ(x)ϕ(y) -1 = e. so xy -1 ∈ N, it follows that

Nx = Ny. Conversely, assume that Nx = Ny then y = nx for some n ∈ N. Thus
ϕ(y) = ϕ(nx) = ϕ(n)ϕ(x) = eϕ(x) = ϕ(x).

2.2.11 C OROLLARY . If ϕ is a homomorphism then ϕ is injective iff ker( θ ) =

e.
P ROOF : E XERCISE .

2.3 Homomorphism Theorems

2.3.1 T HEOREM . (First Homomorphism Theorem). Let ϕ : G → H be a
surjective homomorphism with N = ker( ϕ ). Then G/N ≅ H. In fact ∃ !
Surjective isomorphism θ : G/N Æ H such that πθ = ϕ .

ϕ
G H

π θ

G/N

P ROOF . If θ exist then for g ∈ G we have ((g)π)θ = (g)ϕ. i.e., (Ng)θ = (g)ϕ. This
proves that θ is unique, also shows us how to define θ. So we define
θ : G/N Æ H by (Ng)θ = (g)ϕ, we need to show that θ is well-defined. i.e., if
Nx = Ny then (Nx)θ = (Ny)θ, but by previous Proposition Nx = Ny then
(x)ϕ = (y)ϕ, since N = ker(ϕ). To show that θ is a homomorphism we note that
(NxNy)θ = (Nxy)θ = (xy)ϕ = (x)ϕ (y)ϕ = (Nx)θ(Ny)θ. Therefore θ is a
homomorphism. To show that θ is 1-1 note that Nx ∈ ker(θ) iff x ∈ ker(ϕ) i.e.,
iff x ∈ N. To show that θ is onto we note that if h ∈ H then there is x ∈ G such
that h = (x)ϕ. Thus h = (Nx)θ. This completes the proof.

E XERCISE . Prove that G/Z(G) ≅ Inn(G).

Note if ϕ is not surjective then we have this version G/ker(ϕ) ≅ ϕ(G).

Prof. Mohammed El-Atrash 20

2.3.2 C OROLLARY . Let N  G, H ≤ G then N ∩ H  H and NH/N ≅

G
PROOF. Let ϕ:H → G/N be defined by
ϕ(h) = Nh (i.e., ϕ = π|H). We need to NH
find ker(ϕ). h ∈ ker(ϕ) iff Nh = N i.e.,
h ∈ N. i.e., ker(ϕ) = N∩H. Now
Obviously ϕ(H) = NH/N. So by the N H
Theorem NH/N ≅ H/(H∩N).
N∩H
e
H/(H ∩ N).

2.3.3 T HEOREM . (Correspondence Theorem) Let ϕ : G → H be a surjective

homomorphism with ker( ϕ ) = N. Let S = { U| N ≤ U ≤ G}, T = { V| V ≤ H}.
Then the following hold:

(a) There is a bijective correspondence α : S → T given by α (U) = ϕ (U).

And α -1 (V) = { g ∈ G| ϕ (g) ∈ V}.
(b) If V = α (U) then U  G iff V  H.
(c) If V = α (U) then |G:U| = |H:V|
(d) If V = α (U), U  G, V  H then G/U ≅ H/V.

P ROOF . (a) Since U ≤ G then α(U) = ϕ(U) is a subgroup of H. Also if V ≤ H

then α -1 (V) = { g ∈ G| ϕ(g) ∈ V} is a subgroup of G containing N. Let β(V) = {
g ∈ G| ϕ(g) ∈ V}, V ≤ H. So β: T → S.
Now if we can show that β(α(U)) = U then α, β will be bijections and inverses
of each other. Let x ∈ β(α(U)) then ϕ(x) ∈ α(U) = ϕ(U), so ϕ(x) = ϕ(u) for some
u ∈ U. It follows that Nx = Nu. i.e., x ∈ Nu ⊂ U, therefore x ∈ U.

Conversely, Let u ∈ U then ϕ(u) ∈ ϕ(U) = α(U). Thus u ∈ β(α(U)). Similarly,

we can show that α(β(V)) = V.

(b) Let α(U) = V. Assume that U  G, let h ∈ H, therefore there is g ∈ G with

ϕ(g) = h. Now V h = ϕ(U) ϕ (g) = ϕ(U g ) = ϕ(U) = V. It follows that V  H.
 Abstract Algebra Notes 21

(c) Let α(U) = V. Let θ:{all cosets of U in G}Æ{set of all cosets of V in H},
defined by θ (Ux) = Vϕ(x). It easy to show that θ is a bijection. Thus |G:U| =
|H:V|.

(d) If U  G, V  H, ϕ(U) = V, the bijective map θ:G/U Æ H/V is an

isomorphism, since (UxUy)θ = (Uxy)θ = ϕ(U)ϕ(xy) = ϕ(U)ϕ(x)ϕ(y) =
ϕ(U)ϕ(x)ϕ(U)ϕ(y) = (Ux)θ(Uy)θ. The proof is complete.

2.3.4 C OROLLARY . Let N  G then every subgroup of G/N has the form
H/N for some subgroup H with N ≤ H ≤ G. Moreover H/N  G/N iff H  G
and (G/N)/(H/N) ≅ G/H.
P ROOF . E XERCISE .

2.4 Group Action

2.4.1 D EFINITION . Given a set Ω and a group assume we have a rule
which assigns an element of Ω for each α ∈ Ω , g ∈ G. we write α •g ∈ Ω . So
we have a function f: Ω × G Æ Ω , f( α , g) = α •g ∈ Ω , such that

1. ( α •g)•h = α •(gh) for all α ∈ Ω and g, h ∈ G.

2. α •e = α , for all α ∈ Ω .
We say that G acts on Ω (• is the action).

E XAMPLE . (1) Let G ≤ sym(Ω) and let α•g = (α)g for all α ∈ Ω.

(2) Let G be a group, Ω = G, define x•g = xg. (this is called the regular action,
or right multiplication action).
Note to define an action with left multiplication x•g = g -1 x.

(3) Let G be a group, Ω = G, define x•g = x g = g -1 xg (called conjugation action).

(4) Let G be a group, Ω = {H| H ≤ G}, define H•g = H g (called conjugation

action on subgroups).
 Prof. Mohammed El-Atrash 22

(5) Let G be a group, Ω = {Hg| g ≤ G}, define Hg• x = Hgx.

E XERCISE . Show that these are actions.

We use these actions for two reasons

(1) produce normal subgroups
(2) count things.

2.4.2 T HEOREM . Let G act on Ω . For any g ∈ G define the map π g : Ω Æ Ω

as follows ( α ) π g = α •g. Then π g ∈ sym( Ω ). Furthermore the map θ : G Æ
sym( Ω ) defined by θ (g) = π g is a homomorphism.

P ROOF . Let g, h ∈ G, α ∈ Ω. ((α)π g )π h = (α•g)π h = (α•g)h = (α)•(gh) = (α)π gh . So

π g π h = π gh .
Now π e is the identity permutation, since (α)π e = α•e = α.
Thus π g π g -1 = π e therefore π g is 1-1 and onto. Hence π g ∈ Sym(Ω).
To see that θ is a homomorphism we have θ(gh) = π gh = π g π h = θ(g)θ(h).

Note ker(θ) = { g ∈ G | α•g = α for all α ∈ Ω}. Is called the kernel of the
action.

2.4.3 T HEOREM . (n!) Let H ≤ G and assume |G:H| = n < ∞ . Then there
is a normal subgroup N ≤ H and |G:N| divides n!.

P ROOF . Let Ω = {Hx | x ∈ G} then | Ω| = n . G acts on Ω by right

multiplication. Let N = kernel of the action then N  G. To see that N ≤ H, for
x ∈ N then x fixes H i.e., Hx = H, therefore x ∈ H. Hence N ≤ H.
Now, G/N = G/ker(θ) ≅ θ(G) ≤ sym(Ω). Therefore by Lagrange's Theorem
|G/N| divides | sym(Ω)| = n!.

E XERCISE . Prove that the ker(θ) = ∩{H g | g ∈ G}.

E XERCISE . Prove that if |G:H| = 2 then H  G.

Abstract Algebra Notes 23

E XERCISE . Prove that if |G:H| = p with p is the smallest prime dividing |G|
then H  G.

E XERCISE . Prove that if H ≤ G with |G:H| < ∞ then there is N  G with

|G:N| < ∞.

2.4.4 D EFINITION . Let G be a group acting on Ω , let α ∈ Ω then the orbit

of α under the given action is O α = { α •g | g ∈ G}.

2.4.5 T HEOREM . Let G acts on Ω , let O α be the orbit of α , α ∈ Ω . Then the

following hold:
(1) If β ∈ O α then β •g ∈ O α for all g ∈ G.
(2) If β , γ ∈ O α then γ = β •g for some g ∈ G.
(3) If β ∈ O α then O α = O β .
(4) If O α ∩ O β ≠ ∅ then O α = O β .

P ROOF . E XERCISE .

2.4.6 C OROLLARY . Ω is partitioned by the different orbits.

P ROOF . E XERCISE .

E XAMPLE .(1) Let H ≤ G, let H acts on G by right multiplication. i.e., g•h = gh

then the orbit O g = gH.
(2) If G acts on G by conjugation then for x ∈ G; O x = is the conjugacy class of
x = cl G (x) = { y ∈ G | y = x g for some g ∈ G}.

E XERCISE . Show that if | cl G (x)| = 1 iff x ∈ Z(G).

2.4.7 D EFINITION . An action is called transitive if there is only one

orbit.

E XAMPLE . The regular action is transitive, since for every pair of elements x, y
there is g ∈ G such that xg = y.

Note that elements of the same conjugacy class have the same order. Elements
of order 2 are called involutions.
Prof. Mohammed El-Atrash 24

2.4.8 D EFINITION . Let G act on Ω , let α ∈ Ω . The set G α = { g ∈ G| α •g

= α } is called the stabilizer of α .

E XERCISE . Show that G α ≤ G.

2.4.9 T HEOREM . (Fundamental Counting Principle) Let G act on Ω , let

O α be an orbit for α ∈ Ω . Then there is a bijection between {G α x | x ∈ G} and
O α (i.e., |O α | = |G:G α |).

P ROOF . Define the mapping

ϕ : {G α x | x ∈ G} → O α as follows
ϕ(G α x) = α•x.
We need to show that ϕ is well-defined, i.e., if G α x = G α y then αx = α•y.
But if G α x = G α y then y ∈ G α x i.e., y = gx for some g ∈ G α (i.e., g fixes α).
Therefore α•y = α•gx = (α•g)•x = α•x. To show that ϕ is 1-1, let α•x = α•y
then α = αxy -1 therefore xy -1 ∈ G α hence x ∈ G α y, thus G α x = G α y. To show that
ϕ is onto, let β ∈ O α we know that there is g ∈ G with β = α•x. So ϕ(G α g) =
α•g = β. Hence ϕ is a bijection.

2.4.10 C OROLLARY .
(1) |O α | = |G:G α |.
(2) If |G| < ∞ then |O α | = |G|/|G α |.
(3) Let x ∈ G, then |cl G (x)| = |G:C G (x)|.
(4) If |G| < ∞ and G has only two conjugacy classes then |G| = 2.

P ROOF . E XERCISE .

2.4.11 D EFINITION . Let G act on the set of subgroups of G by conjugation.

For H ≤ G, G H = N G (H). is called the normalizer of H in G.

E XERCISE . Show that

(1) H ≤ N G (H)
(2) Show that N G (H) is the largest subgroup in which H  N G (H).
Abstract Algebra Notes 25

2.4.12 T HEOREM . Let H, K ≤ G, |H|, |K| < ∞ then |HK| =

|H||K|/|H ∩ K|.

KH
|HK:K|

K H
|H:H∩K|

K∩H

P ROOF . HK = ∪{Hk| k ∈ K}, thus |HK| = |H||{Hk| k ∈ K}|. Let K act on Ω =

{Hg | g ∈ G} therefore the set {Hk| k ∈ K} forms one orbit of H under the
action. Therefore, by The Fundamental Counting Principle, |{Hk| k ∈ K}| =
|K:K H |, where K H is the stabilizer of H in K. Now given x ∈ K, we have x ∈
K H iff Hx = H iff x ∈ H. In other words K H = K∩H. Therefore |HK| =
|H||K:K∩H| = |H||K|/|H∩K|.

2.4.13 D EFINITION . Let p be a prime number, a finite p-group is a group G

with |G| = p n for some integer n ≥ 0.

2.4.14 P ROPOSITION . Let P be a p-group and assume that P acts on Ω . If p

does not divide | Ω | then P fixes some element of Ω .

P ROOF . Write |Ω| = |O 1 | + |O 2 | + …+ |O k |, where O i are the different

orbits. Thus ∃ an orbit O i such that p does not divide |O i |. By Fundamental
Counting Principle, |O i | divides |P|. Therefore |O i | = 1. Thus there is α ∈
Ω such that O α = {α}, and hence, α•g = α for all g ∈ P.
Prof. Mohammed El-Atrash 26

2.4.15 T HEOREM . Let P be a p-group and N  P, assume that |N| > 1.

Then |N ∩ Z(P)| > 1.

P ROOF . Let P act by conjugation on Ω = N\{e}, then |Ω| = |N| - 1. But |N| =
p α for some α > 0. So |Ω| = p α - 1, which is not divisible by p. therefore there
is some element x ∈ N fixed under P, i.e., x g = x for all g ∈ G. Thus x ∈ Z(P), it
follows that x ∈ N∩Z(P).

2.4.16 C OROLLARY . Let P be a p-group, |P| > 1. Then |Z(P)| > 1.

P ROOF . E XERCISE .

2.4.17 D EFINITION . A group G is called simple if G has no proper normal

subgroups.

2.4.18 C OROLLARY . If |G| = p α , p prime, G is simple then α = 1.

P ROOF . E XERCISE .
 Abstract Algebra Notes 27

Chapter III
1. Sylow Theory
2. Nilpotent Groups
3. Direct Product
4. Permutation Groups

3.1 Sylow Theory

3.1.1 D EFINITION . Let |G| = n = p α k where p does not divide k. p α is
called the p-part of n. A subgroup H ≤ G is called a sylow p-subgroup if
|H| = p α . The set of all sylow p-subgroups of G is denoted by syl p (G).

Remark: (1) If H ≤ G, H ∈ syl p (G) iff |H| is a power of p and p does not
divide |G:H|.
(2) If H ∈ syl p (G), g ∈ G then H g ∈ syl p (G).

3.1.2 P ROPOSITION . If p is a prime number, then

⎛ p⎞
⎜⎜ ⎟⎟ ≡ 0 (mod p), for all 1 ≤ i ≤ p.
⎝i⎠

3.1.3 C OROLLARY . For every integer x, (x + 1) p ≡ (x p + 1) mod p.

⎛ pα k ⎞
3.1.4 P ROPOSITION . ⎜ ⎟ ≡ k (mod p), for prime p.
⎜ pα ⎟
⎝ ⎠

3.1.5 T HEOREM . (Sylow Existence) For a prime p, if G is finite then

syl p (G) ≠ ∅ .

P ROOF . Let n = |G| = p α k , p does not divide k. Let Ω = {X ⊂ G| |X| = p α }.

⎛ pα k ⎞
Then |Ω| = ⎜ ⎟ ≡ k (mod p) ≠ 0 mod p. Let G act on Ω by right
⎜ pα ⎟
⎝ ⎠
multiplication, i.e., X•g = Xg, for all g ∈ G.
So, there must be an orbit O X with |O X | is not congruent to 0 (mod p). Then
by The Fundamental Counting Principle, |O X | = |G:G X |, and p does not
Prof. Mohammed El-Atrash 28

divide |G:G X |. Therefore p does not divide |G|/|G X |, it follows that p α |

|G X |. Therefore p α ≤ |G X |. To get the reverse inequality we know that Xh = X
for h ∈ G X . Fix x ∈ X, then xh ∈ X for all h ∈ G X . It follows that xG X ⊂ X. Thus
|G X | ≤ |X|. Hence |G X | = |X|, G X ∈ syl p (G).

3.1.6 T HEOREM . (Sylow Conjugacy and Development Theorem). Let G be

a finite group. Let P ≤ G be a p-group and let S ∈ syl p (G). Then P ≤ S x for
some x ∈ G.

P ROOF : Let Ω = {Sx| x ∈ G} and let P act on Ω by right multiplication. Then

we have |Ω| = |G:S| which is not divisible by p, since S is a sylow p-
subgroup. Therefore there is an orbit O α with p does not divide |O α |. But
since P is a p-group then all orbits must divide |P|. It follows that there is an
orbit with |O α | = 1. So P stabilizes Sx for some x ∈ G, i.e.,if y ∈ P then Sx =
Sxy and hence x -1 Sx = x -1 Sxy i.e., S x = S x y. Therefore y ∈ S x i.e., P ⊂ S x .

3.1.7 C OROLLARY . (Sylow conjugacy Theorem) Let P, Q be two sylow p-

group, for a prime number p. Then there is an element x ∈ G such that P x =
Q.

P ROOF . Take S in the Theorem to be Q. then Q ≤ P x . Now since both have the
same cardinality |Q| = |P x |, then Q = P x .

3.1.8 C OROLLARY . (Sylow development Theorem) Let P be a p-group, for a

prime number p. Then there is a sylow p-group Q such that P ≤ Q.

P ROOF . Take S x in the Theorem to be Q. then P ≤ Q.

3.1.9 C OROLLARY . Let G be a finite group, let P ∈ syl p (G).

Then | syl p (G)| = |G:N G (P)|. In particular | syl p (G)| divides |G:P|.
P ROOF : E XERCISE .
Abstract Algebra Notes 29

3.1.10 C OROLLARY . Let S ∈ syl p (G). Then the following are equivalent:
(i) S  G.
(ii) S is the unique sylow p-subgroup of G.
(iii) Every p-subgroup of G is contained in S.
(iv) S is characteristic in G.

P ROOF : E XERCISE .

Write n p = | syl p (G)|, now if | G| = p α m. Then n p | m.

3.1.11 T HEOREM . (Sylow Counting) Let G be a finite group. Then n p ≡ 1

mod p.

P ROOF : Let P ∈ syl p (G). Let P act by conjugation on Ω = syl p (G). Then {P}
forms one orbit by itself. Now we claim that every other orbit has size strictly
bigger than one. To see this, let S = S x , x ∈ P then x ∈ N G (S), So P ≤ N G (S).
Since P ∈ syl p (G) then syl p (N G (S)). S  N G (S) implies that S is the unique
sylow p-subgroup of N G (S). Therefore P = S. Therefore n p = 1 + |O 1 | + |O 2 | +
… |O k |. where p divides |O i | since |O i | divides |P|. Hence n p ≡ 1 mod p.

3.1.12 C OROLLARY . Let Q ∈ syl p (G), P any p-subgroup of G. Suppose that

P ≤ N G (Q) then P ≤ Q.
P ROOF . E XERCISE .

3.1.13 C OROLLARY . If |G| = 72 then G is not simple.

P ROOF . 72 = 2 3 3 2 . We will compute n 3 (G). Since n 3 must divide 2 3 , then n 3 ∈ {1,

2, 4, 8}. Since n 3 ≡ 1 mod 3, then n 3 ∈ {1, 4}.
If n 3 = 1 then G is not simple since it contains a normal subgroup of order 9. If
n 3 = 4 then |G:N(S)| = 4 for some S ∈ syl 3 (G). Therefore by (n!) Theorem
there is a normal subgroup N ⊂ N(S). If |N| > 1 then G is not simple. So,
assume that |N| = 1, then |G| = |G/N| | 4!, but 72 does not divide 24. Thus
n 3 ≠ 4. Therefore G is not simple.
Prof. Mohammed El-Atrash 30

3.1.14 C OROLLARY . If |G| = pq where p, q are primes with p > q, then

(a) G has a normal sylow p-subgroup
(b) G is cyclic unless q | (p-1).

P ROOF : n p | q so n p = 1 or n p = q. But if n p = q then q ≡ 1(mod p), i.e., q ≥ p + 1

which contradicts the hypothesis that q < p. Thus n p = 1.
Now assume that q does not divide p - 1. i.e., p is not congruent to 1 mod q. n q
= 1mod q, therefore n q = 1. Let P ∈ syl p (G), Q ∈ syl q (G). Then P  G, Q  G,
P∩Q = {e}. Therefore elements of P commutes with elements of Q. Let x ∈ P, y
∈ Q with x ≠ e ≠ y therefore xy = yx, therefore o(xy) = pq, Thus G = 〈xy〉.

3.1.15 C OROLLARY . Let |G| = p 2 q, q ≠ p be primes. Then G has a normal

subgroup.

P ROOF : Assume n p > 1, n q > 1 then n p | q and therefore n p = q. it follows that q

≡ 1 mod p and this gives that q = 1 + kp. Hence n q = p or n q = p 2 .
If n q = p then p ≡ 1 mod q, therefore p > q, contradicting the fact that q > p. If
n q = p 2 , then the number of elements of order q is p 2 (q-1), so the rest of the
elements is p 2 q - p 2 (q -1) = p 2 , but this is only the number of elements in one
sylow p-subgroup, i.e., n p = 1 contradicting our assumption that n p > 1.
Therefore either n p = 1 or n q = 1.

3.1.16 C OROLLARY . Let |G| = p 3 q, p ≠ q primes. Then G has a normal

sylow subgroup except when |G| = 24.

P ROOF : E XERCISE .

3.1.17 T HEOREM . (Burnside) If |G| = p α q β , where p, q are primes, then α ,

β ≥ 1, |G| is not simple.

P ROOF : Omitted.

3.1.18 T HEOREM . (Frattini Argument) Let N  G, with |N| < ∞ . Let P

∈ syl p (N). Then G = N G (P)N.
 Abstract Algebra Notes 31

P ROOF : Let g ∈ G. Then P g ⊂ N g = N. But |P g | = |P|. Therefore P g ∈ syl p (N).

It follows by Sylow Conjugacy Theorem that P g , P are conjugate in N. Thus
there is an element n ∈ N such that P g = P n . It follows that gn -1 ∈ N G (P). Hence
g ∈ N G (P)N, i.e., G = N G (P)N.

3.1.19 D EFINITION . Let ϕ (G) = ∩ { H ≤ G| H is maximal subgroup in G}.

ϕ (G) is called Frattini subgroup.

3.1.20 P ROPOSITION . Let |G| < ∞ and let H ≤ G. If ϕ (G)H = G then H =

G.

P ROOF : E XERCISE .

3.1.21 T HEOREM . If |G| < ∞ then every sylow subgroup of ϕ (G) is

normal.
P ROOF : Let F = ϕ(G) and let P ∈ syl p (F). we know that F  G (in fact F is
characteristic in G). By Frattini argument N G (P)F = G, and then by the
proposition N G (P) = G, therefore P  G.

3.2 Nilpotent Groups

Let P be a finite p-group, |P| > 1 the P has a non-trivial center, let Z 1 = Z(P),
Z 1 is characteristic in P. If P is not abelian then Z(P) ≠ P, and hence P/Z(P) is
again a p-group and then we do the same, i.e., we find a subgroup Z 2 with
Z 2 /Z 1 = Z(P/Z 1 ). Therefore we have the following series of subgroups

{e} = Z 0 ≤ Z 1 ≤ Z 2 ≤ …≤ Z k = P.

with each Z i is called ith center, and (Z i + 1 )/Z i = Z(P/ Z i ).

Note. In p-groups we always get Z k = P for some k. But in general it might not
be true.
Prof. Mohammed El-Atrash 32

3.2.1 D EFINITION . A set of subgroups G 0 , G 1 , …, G k in G is a central

series if {e} = G 0 ⊂ G 1 ⊂ … ⊂ G k = G

where each G i  G, and (G i + 1 )/G i ⊂ Z(G/ G i ), for 0 ≤ i ≤ k, k is finite.

3.2.2 C OROLLARY . A finite p-group has a central series.

3.2.3 D EFINITION . A group is called nilpotent if it has a central series.

3.2.4 C OROLLARY . A finite p-group is nilpotent.

R EMARK . Every abelian group is nilpotent.

3.2.5 P ROPOSITION . If 1 < |G| < ∞ and all its sylow subgroups are
normal then |Z(G)| > 1.

P ROOF . If p | |G|, p is prime. Let P ∈ syl p (G). Let Z = Z(P). we know that |Z|
> 1. We claim that Z ⊂ Z(G). To prove this claim, let C = C G (Z), we will show
that C = G, by showing that |G:C| = 1. So assume |G:C| > 1. Let q be a prime
number with q | |G:C|. Let Q ∈ syl q (G). Since q | |G:C| we have Q ≠ C. If q
= p then P = Q and because n p = 1, Z = Z(P) thus P ⊂ C(Z) = C which is a
contradiction since P = Q ⊄ C. Thus q ≠ p, and therefore P ∩ Q = {e}, P  G, Q
 G, thus Q ⊂ C(P) ⊂ C another contradiction.

3.2.6 T HEOREM . Let G be a finite group. Then the following are

equivalent:
(i) G is nilpotent.
(ii) If H < G then N G (H) > H. (normalizers grow)
(iii) If M is maximal in G then M  G.
(iv) Each sylow of G is normal in G { ∃ ! Sylow subgroups}
(v) If N  G with N ≠ G then |Z(G/N)| > 1. { if N = {e} then |Z(G)| > 1}.

P ROOF . (i) → (ii) Given H < G, we have a central series

{e} = G 0 ≤ G 1 ≤ G 2 ≤ … ≤ G k = G.
Choose G i ⊂ H such that G i+ 1 ⊄ H.
Abstract Algebra Notes 33

Now H/G i ⊂ G/G i , G (i + 1) /G i ⊂ Z(G/G i ) ⊂ N(H/G i ) = N/G i for some N ≤ G.

and H/G i  N/G i then by correspondence Theorem H  N. Therefore G i+1 ⊂
N ⊂ N G (H) > H.
(ii) → (iii) Let M be maximal in G, then M < G and by (ii) N G (M) > M, thus
N G (M) = G by maximality. Hence M  G.

(iii) → (iv) Let P ∈ syl p (G) for some p, p prime. Let N = N G (P). If N < G then by
finiteness, ∃ a maximal subgroup M with N ⊂ M then M  G. Note P ⊂ M
thus P ∈ syl p (M), by Frattini Argument G = N G (P)M = NM = M, this of course
a contradiction. So N = G, so P  G.

(iv) → (v) Given N  G, to show that Z(G/N) > {e}, it suffices by the
Proposition to show that each sylow of G/N is normal. We will use property
(iv) to prove the same condition on G/N.
Let p be prime, let P ∈ syl p (G). Look at (PN)/N ⊂ G/N. P  G by (iv), N  G
implies that PN  G. Therefore PN/N  G/N.
Now we will show that PN/N ∈ syl p (G/N). | PN/N| is a p-power, |G:PN| is
not divisible by p. Therefore G/N has a normal sylow p-subgroup.

(v) → (i) Let G 0 = {e}, by induction define G i , i > 0, by the formula G i /G i-1 =
Z(G/G i-1 ), not all G i  G. If G i-1 < G then by (v) Z(G/G i-1 ) > {e}. i.e., G i /G i-1 >
{e} implies that G i > G i-1 , and by finiteness of G we have G k = G for some k.
Thus G is nilpotent.

3.2.7 C OROLLARY . Let P ≠ {e} be a finite p-group. Let M be any maximal

subgroup of P. Then M  P and P/M is cyclic of order p.

P ROOF : We have M  P since P is nilpotent. Subgroup of P/M are in bijection

correspondence with S = {U| M ≤ U ≤ P}. Maximality implies that S = {M, P}.
Thus P/M has just two subgroups; itself and the identity. Therefore P/M is
cyclic of prime order p.

3.2.8 C OROLLARY . Last corollary states that if M is maximal than |P:M|

= p.
 Prof. Mohammed El-Atrash 34

3.2.9 C OROLLARY .If |P| = p α and 0 ≤ β ≤ α , then ∃ subgroup Q ⊂ P

with |Q| = p β .

P ROOF . E XERCISE .

3.2.10 C OROLLARY . If |G| < ∞ , and p β | |G|, where p is prime, then

there is a subgroup Q ⊂ G with |Q| = p β .

P ROOF : E XERCISE .

E XERCISE . If |G| < ∞, show that ϕ(G) is nilpotent.

E XERCISE . Let G be a finite group.

(i) If G/ϕ(G) is nilpotent show that G is nilpotent.
(ii) If G is nolpotent, N  G, show that G/N is nilpotent.

3.3 Direct Product

Give two groups U, V. Let G = {(u, v)| u ∈ U, v ∈ V}, i.e., G = U × V. Define
the multiplication on G component wise by
(u, v)(u', v') = (uu', vv').
Identity of G is (e 1 , e 2 ), where e 1 , e 2 is the identity of U, V respectively. And
inverse of (u, v) -1 is (u -1 , v -1 ). G is a group called the external direct product of
U, V. Of course |G| = |U||V|.

Let U = {(u, e 2 )| u ∈ U}, V = {( e 1 , v)| v ∈ V}. Easy to see that U, V are

subgroups of G. In fact U, V  G, and U ≅ U, V ≅ V.
Similarly, if U 1 , U 2 , … U n are n groups. Then the external direct product
G = {(u 1 , u 2 , … u n )| u i ∈ U i , i = 1, 2, …, n}.
n n
This product is denoted by ∏i =1
U i . or ∑
i =1
Ui.

Let U i = {(e 1 , e 2 , …, u i , …, e n )| u i ∈ U i ,}, i = 1, 2, …, n.

E XERCISE . (i) Show that U i  G for i = 1, 2, …, n.

Abstract Algebra Notes 35

(ii) Show that U i ≅ U i for i = 1, 2, …, n.

3.3.1 T HEOREM . Let U 1 , U 2 , … U n be finite p-groups (Possibly for

n
different primes). Let G = ∏i =1
U i . Then G is nilpotent.

P ROOF : We may assume that U i is a p i -group with p i ≠ p j , for i ≠ j.

n
Note |G| = ∏ i =1
|U i |. Thus |U i | is the order of a sylow p i -subgroup of G. But

U i has the order equal to U i and so U i ∈ syl p (G). Therefore for each prime p i |
|G| a sylow p i -subgroup is normal in G. Hence G is nilpotent.

3.3.2 D EFINITION . Given a group G, suppose M 1 , M 2 , …, M n  G.

Assume G = M 1 M 2 … M n . Assume also that each g ∈ G the decomposition g =
x 1 x 2 … x n with x i ∈ M i is unique. Then we say that G is the internal direct
n
product of M i ., i = 1, 2, 3, …, n. and this product is denoted by ∏ • Mi or
i =1

∑ • Mi if the notation is addition.

i =1

n
3.3.3 T HEOREM . Let G = ∏ • Mi, then
i =1

n
(i) ( ∏ M i ∩ ) M j = {e}.
i≠ j

(ii) M i ⊂ C G (M j ), if i ≠ j.
n
(iii) G ≅ ∏
i =1
Mi.

n
P ROOF : (i) Let g ∈ ∏i≠ j
M i ∩ M j , then g = x 1 x 2 … x n , x j = e i . and since g ∈ M i

thus g = ee …g…e. Therefore g = x i = e. by uniqueness of the decomposition.

(ii) By (i) M i ∩ M j = {e} for i ≠ j. Normality of M i implies that M i ⊂ C(M j ).
n
(iii) Let θ: ∏ M i → G, defined by θ((x 1 , x 2 , … , x n )) = x 1 x 2 … x n . Clearly θ is an
i =1

isomorphism.
Prof. Mohammed El-Atrash 36

E XAMPLE . Let G = Z 30 . Let M 1 = {0, 15}, M 2 = {0, 10, 20}, M 3 = {0, 6, 12, 18, 24}.
Then by the Theorem, we have G = M 1 ⊕M 2 ⊕M 3 , Since M 1 ⊕M 2 ∩M 3 = {e},
M 1 ∩M 2 = {e}.

3.3.4 C OROLLARY . If G is the direct product of M, N. Then G/N ≅ M.

P ROOF : E XERCISE .

E XERCISE . If G is the direct product M × N and M × L, then N ≅ L.

(Cancelation).

r −1
3.3.5 T HEOREM . Let M 1 , M 2 , … M n  G. Assume that ( ∏ M i ∩ ) M r =
i =1

{e}. Then the product G = M 1 M 2 … M n is direct.

P ROOF : Suppose that x 1 x 2 … x n = y 1 y 2 … y n , where x i , y i ∈ M i , i = 1, 2, …, n. To

show that x i = y i for all i = 1, 2, …, n. Assume that x r ≠ y r for some r, choose r
as larg as possible so that x i = y i for i > r. Use cancelation to get x 1 x 2 … x r = y 1
y 2 … y r . Let u = x 1 x 2 … x r-1 , v = y 1 y 2 … y r-1 then ux r = vy r . Now v -1 u = y r x r -1 ∈
r −1 r −1 r −1
M r , u, v ∈ ∏i =1
M i thus v -1 u ∈ ∏
i =1
M i . Therefore y r x r -1 ∈ ( ∏ M i ∩) M r = {e}.
i =1

Hence x r = y r and this is a contradiction.

n
E XERCISE . Let G be a finite group. G = ∏ i =1
M i , M i  G. Show that |G| ≤

∏i =1
|U i | with equality holds iff G is the direct product .

3.3.6 C OROLLARY . Assume that G is a finite nilpotent group. Let P 1 , P 2 ,

…, P n ∈ syl pi (G) for different primes p i . then G is the direct product G =
n

∏i =1
Pi.
 Abstract Algebra Notes 37

n n
P ROOF : we know that P i  G, |G| = ∏i =1
|P i | also | ∏ P i | is divisible by
i =1

n
|P i |, thus G = ∏ i =1
P i and this product is direct.

E XERCISE . Prove that every finite group G is nilpotent iff G is the (direct)
product of p-groups.

E XERCISE . Prove that every p-group G is isomorphic to a (direct) product of

cyclic groups.

3.3.7 T HEOREM . (Fundamental Theorem of Finite Abelian Groups). If

G is finite abelian group then G is isomorphic to the direct product of cyclic
ν
groups of prime power order i.e., G ≅ Π C i , |C i | = p i α i , α i ≥ 0.
ι =1

3.4 Permutation groups

3.4.1 D EFINITION . The set of all permutations of a set S is denoted by
Sym(S).
The set of all permutations of the set {1,2,...,n} is denoted by S n .

3.4.2 P ROPOSITION . If S is any nonempty set, then Sym(S) is a group

under the operation of composition of functions.

3.4.3 D EFINITION . A k-cycle (or a cycle of length k) is a permutation π =

(a 1 a 2 … a k ) where π (a i ) = a i+1 , for 1 ≤ i < k, and π (a k ) = a 1. π fixes every other
element of S.

E XERCISE . Show that the order of a cycle of length k is k.

E XERCISE . Show that the order of disjoint cycles is the least common multiple
of their lengths.

E XERCISE . If C = (a 1 , a 2 , …, a k ), π ∈ S n . Prove that

π (a 1 , a 2 , …, a k )π -1 = (π(a 1 ), π(a 2 ), …, π(a k )).
Prof. Mohammed El-Atrash 38

E XERCISE . Let x, y ∈ S n . Show that x, y are conjugate iff x, y, have the same
cycle structure.

3.4.4 D EFINITION . A transposition is a cycle of length 2.

E XERCISE . Prove that a k-cycle is the product of k-1 transpositions

3.4.5 T HEOREM . Every permutation in S n can be written as a product of

disjoint cycles. The cycles that appear in the product are unique.

3.4.6 P ROPOSITION . If a permutation in S n is written as a product of

disjoint cycles, then its order is the least common multiple of the lengths of its
cycles.

3.4.7 D EFINITION . Any subgroup of the symmetric group Sym(S) on a set

S is called a permutation group or group of permutations.

3.4.8 T HEOREM . (Cayley) Every group is isomorphic to a permutation

group.

3.4.9 D EFINITION . Let n > 2 be an integer. The group of rigid motions of a

regular n-gon is called the n th dihedral group, denoted by D n .

We can describe the n th dihedral group as

D n = { a k , a k b | 0 ≤ k < n },

subject to the relations o(a) = n, o(b) = 2, and ba = a -1 b.

3.4.10 T HEOREM . If a permutation is written as a product of transpositions

in two ways, then the number of transpositions is either even in both cases or
odd in both cases.

3.4.11 D EFINITION . A permutation is called even if it can be written as a

product of an even number of transpositions, and odd if it can be written as a
product of an odd number of transpositions.
Abstract Algebra Notes 39

3.4.12 P ROPOSITION . The set of all even permutations of S n is a subgroup

of S n .

3.4.13 D EFINITION . The set of all even permutations of Sn is called the

alternating group on n elements, and will be denoted by A n .

E XERCISE . Show that |S n :A n | = 2.

3.4.14 T HEOREM . Let G be a group, Ω be a finite set. Assume that there is g

∈ G that acts " oddly" on Ω . Then ∃ N  G with |G:N| = 2.

P ROOF : The action gives a homeomorphism θ: G → Sym(Ω).

θ(g) is an odd permutation.
Therefore, θ(G) ∩ Alt(Ω) < θ(G), since θ(G) has an odd permutation. Then we
have | θ(G):θ(G) ∩ Alt(Ω)| = 2. θ is surjective, therefore there is a normal
subgroup N = θ -1 (θ(G) ∩ Alt(Ω)) of index 2.

C OROLLARY . Let G be simple, with |G| > 2. Let H ≤ G such that |G:H| = n > 1.
Then |G| | n!/2.

3.4.15 C OROLLARY . Let |G| = 2n with n odd. Then G has a normal

subgroup of order n.

P ROOF : Let x ∈ G with o(x) = 2. We claim that x acts oddly on G. To prove

this, let π be the permutation of G induced by x. Since x 2 = e, we have π 2 = e.
in fact π fixes no element of G, since If π(g) = g then gx = g and this would
imply that x = e , this is a contradiction. The cycle structure of π consists just
of 2-cycles. Thus π is n different cycles, then π is an odd permutation. So N
exist.

3.4.16 T HEOREM . A 5 is simple.

P ROOF : If N  G, assume N < A 5 . We can show that N = {e}. Choose N as large

as possible. If 3| |N| then if P ∈ syl 3 (N) we have P ∈ syl 3 (A 5 ). Thus by sylow
conjugacy N contains all syl 3 (G) subgroups of A 5 and therfore it contains all
Prof. Mohammed El-Atrash 40

elements of order 3 in A 5 . Now if x ∈ A 5 , o(x) = 3, then the cycle structure of x

is 1 2 3. But we have 20 elements of those 3-cycles, it follows that |N| ≥ 21,
hence |N| = 30. And similarly if 5| |N| we get |N| = 30. It follows that in
either case |N| = 30. This would imply that |N| > 1 + 20 + 24 > 30, which is
a contradiction. If |N| = 2, or 4, then |G/N| = 30 or 15. But G/N is simple
by maximality of N, however there is no such group. Thus N = {e}.

3.4.17 T HEOREM . A n is simple for all n ≥ 5.

P ROOF : Omitted.

E XERCISE . Show that if |G| = 180 then G is not simple.

E XERCISE . Show that if |G| = 396 then G is not simple.

 Abstract Algebra Notes 41

3.5 Operator Groups

3.5.1 D EFINITION . Given a set S ( may be empty), and given a group G
(may be infinite). Assume that for every s ∈ S and for every g ∈ G there is an
element g s ∈ G that satisfies (gh) s = g s h s . Then G is called a group with
operator set S.

Note. Each s ∈ S induces an indomorphism (a homomorphism of G into G).

E XAMPLE . Let S = G, action is conjugation.

E XAMPLE . If V is a vector space over a field F. V is a group with operator set F.

3.5.2 D EFINITION . H is called an s-subgroup of a group G with operator

set S (denoted by H ≤ s G) if H ≤ G and h s ∈ H for all h ∈ H.

3.5.3 D EFINITION . H is called s-normal in G (denoted by H  s G) if H 

G and H ≤ s G.

3.5.4 L EMMA . If G is a group with operator set S, H  s G, then G/H is a

group with operator set S.

P ROOF . Define the action of S on G/H by (Hg) s = H g s . We need to show that

(Hx) s = (Hy) s if Hx = Hy. Now x ∈ H y implies that x = hy. Then x s = (hy) s =
h s y s ∈ Hy s . Therefore (Hx) s = (Hy) s .

3.5.5 D EFINITION . A homomorphism θ : G 1 → G 2 is called an s-

homomorphism of groups G 1 , G 2 with operator set S if θ (g s ) = θ (g) s . If θ is
onto and 1-1 then θ is called an s-isomorphism.

3.5.6 L EMMA . If θ : G 1 → G 2 is a surjective s-homomorphism then

G 1 /ker( θ ) ≅ s G 2 ( G 1 is s-isomorphic to G 2 ).

3.5.7 D EFINITION . A group G is called s-simple if the only s-normal

subgroups of G are {e}, G.
Prof. Mohammed El-Atrash 42

3.5.8 D EFINITION . The series {G 0 , G 1 , G 2 , …, G n } is called s-composition

series for G if {e} = G 0  s G 1  s …  s G n = G, with G i +1 /G i are s-simple, i =
0, 1, …, n-1.

3.5.9 L EMMA . Let G be a group that have an s-composition series. Let N

 s G then N has an s-composition series.

P ROOF : Let {e} = G 0  s G 1  s …  s G n = G be an s-composition series for G, let

N i = N ∩ G i , Then {e} = N 0  s N 1  s …  s N n = N, note that N i+1  G i+1 , since
N  G. Also G i  G i+1 . Thus G i  G i N i+1  G i+1 , it follows that either G i N i+1 =
G i or G i N i+1 = G i+1 .
Case 1.
N i+1 ⊂ G i N i+1 = G i , So N i+1 ⊂ G i , Thus N i = N ∩ G i ⊇ N∩N i+1 = N i+1 .
Case 2.
G i N i+1 = G i+1 therefore, G i+1 /G i = G i N i+1 /G i ≅ N i+1 /( G i ∩N i+1 ) = N i+1 /N i .
Therefore, N i+1 /N i is s-simple. From case 1 and case 2 we have either N i+1 = N i
or N i+1 /N i is s-simple. Delete repeats to get an s-composition series for N.

3.5.10 T HEOREM . (Jordan Hölder) Let G be an s-group, suppose that

{e} = N 0  s N 1  s …  s N n = G
{e} = M 0  s M 1  s …  s M m = G, be two s-composition series for G. Then
(i) m = n.
(ii) Up to possible rearrangements the two series have s-isomorphic factors.

P ROOF : We may assume that n < m, we will pursue the proof by induction on
n. If n = 1 then G is s-simple, and so M m-1 = {e}. Thus m = 1.
If N n-1 = M m-1 = K we may assume that n > 1.
Case 1. We get
{e} = M 0  s M 1  s …  s M m-1 = K
{e} = N 0  s N 1  s …  s N m-1 = K.
Then by inductive hypothesis in K we get n-1 = m-1, so n = m.
Case 2.
Abstract Algebra Notes 43

If N n-1 ≠ M m-1 but N n-1 M m-1 is s-normal in G and G/ N n-1 ,G/M m-1 are s-simple,
then N n-1 M m-1 = G. Let D = N n-1 ∩M m-1 , then D has an s-composition series by
the lemma (say)
{e} = D 0  s D 1  s …  s D k = D
Now inductive hypothesis in N n-1 we get n-1 = k +1, and M m-1 has one of
length k+1 and the other one of length m-1. Hence m-1 = k + 1. It follows that

G= Mm-1Nn-1

Nn-1 Mm-1

m = n.

3.5.11 D EFINITION . A group G is called solvable if there is a series

{e} = N 0  s N 1  s …  s N n = G, where each N i  G and each N i+1 /N i is abelian.

E XAMPLE . Abelian groups are solvable, nilpotent groups are solvable too.

3.5.12 L EMMA . A group G is solvable iff G (n) = {e} for some n < ∞ .

P ROOF : If G (n) = {e} then we have

{e} = G (n) ≤ G (n-1) ≤ … ≤ G' ≤ G, where each one is normal in G and factors are
abelian. Conversely, suppose that we have a series
{e} = N 0  s N 1  s …  s N m = G with N i+1 /N i abelian thus N m-1  G, G/N m-1
abelian implies that G' ⊂ N m-1 . G' ⊂ N m-1 , thus G'' ⊂ (N m-1 )' ⊂ N m-2 (since N m-
2 /N m-1 is abelian). Continue to get G (k) ⊂ N m-k . Thus G (m) ⊂ N 0 .
Prof. Mohammed El-Atrash 44

3.5.13 T HEOREM . The following are equivalent for a finite group G.

(i) G is solvable
(ii) Every composition factor is of prime order
(iii) G (n) = {e}, for some finite n.

P ROOF : (i) → (ii) Let {e} = N 0 ≤ N 1 ≤ …≤ N m = G, where N i  G, N i+1 /N i abelian.

Now insert groups between the N i 's so this series can be refined to be a
composition series, say,
{e} = G 0  G 1  …  G n = G,
with G i +1 /G i are abelian, i = 0, 1, …, n-1, and it is a composition factor, so it is
simple. Simple and abelian implies that they have prime order.
(ii) → (iii), and (iii) → (i) are done before.

E XERCISES . 1. If G is solvable, H ≤ G then H is solvable.

E XERCISE s. 2. If G is solvable, N  G then G/N is solvable.

E XERCISE s. 3. N  G, N is solvable, G/N is solvable then G is solvable.

 Abstract Algebra Notes 45

Chapter IV

4.1 Rings
4.1.1 D EFINITION . A ring R is a non-empty set with two binary
operations, denoted by addition and multiplication "+", ".", such that the
following properties hold:
(i) For all a, b, c ∈ R, a + (b + c) = (a + b) + c and a(bc) = (ab)c.
(ii) For all a, b ∈ R, a + b = b + a.
(iii) For all a, b, c ∈ R, a(b + c) = ab + ac and (a + b)c = ac + bc.
(iv) The set R contains an additive identity element, denoted by 0, and a
multiplicative identity element, denoted by 1, such that a + 0 = a, 1a = a, and a1 = a,
for all a ∈ R.
(v) For each a ∈ R, the equation a + x = 0 has a solution x = -a in R, the additive
inverse of a.

A ring R is called commutative if ab = ba for all elements a, b ∈ R.

commutative examples

E XAMPLE . 1 (Z, +, .) is a ring.

E XAMPLE . 2 (Z n , ⊕, ⊗) is a ring for any positive integer n ≥ 2.

Non-commutative examples

We want to include, among other examples, the study of n×n matrices. Recall
that if F is a field, then the set of n×n matrices M n (F) corresponds to the set of
linear transformations of an n-dimensional
vector space over F. This is a special case of the most general example of a
ring. Just as permutation groups are the generic groups (as shown by Cayley's
theorem), the generic examples of rings are found in studying
endomorphisms of abelian groups.
Prof. Mohammed El-Atrash 46

E XAMPLE . (Endomorphisms of abelian groups) Let A be an abelian group,

with its operation denoted by +. Let R be the set of all endomorphisms of A.
That is, R is the set of all group homomorphisms f : A → A. We can define
addition and multiplication of elements of R as follows: if f, g ∈ R, then (f +
g)(x) = f(x) + g(x) and (f.g) (x) = f(g(x)) for all x ∈ A.
R forms a ring and is denoted by End(A).

E XAMPLE . (Polynomial Rings) Let R be any ring. We let R[x] denote the set of
infinite tuples (a 0 , a 1 , a 2 , ...) such that a i ∈ R for all i, and a i ≠ 0 for only finitely
many terms a i . Two sequences are equal if and only if all corresponding terms
are equal. We introduce addition and multiplication as follows:
(a 0 , a 1 , a 2 , ...) + (b 0 , b 1 , b 2 , ...) = (a 0 + b 0 , a 1 + b 1 , a 2 + b 2 , ...)
(a 0 , a 1 , a 2 , ...).(b 0 , b 1 , b 2 , ...) = (c 0 , c 1 , c 2 , ...),
n
where c n = ∑i =0
a i b n-i .

With these operations it can be shown that R[x] is a ring.

We can identify a ∈ R with (a, 0, 0,...) ∈ R[x], then (1, 0, 0, ...) is an identity for
R[x]. If we let x = (0, 1, 0,...), then the elements of R[x] can be expressed in the
form
a 0 + a 1 x + … + a m-1 x m-1 + a m x m ,
allowing us to use our previous notation for the ring of polynomials over R in
the indeterminate x.
Note that although the elements of R need not commute with each other, they
do commute with the indeterminate x.
If n is the largest nonnegative integer such that a n ≠ 0, then we say that the
polynomial has degree n, and a n is called the leading coefficient of the
polynomial.

E XAMPLE . (Differential operator rings) Consider the homogeneous linear

differential equation a n (x)D n y + … + a 1 (x)Dy + a 0 (x)y = 0, where the solution
y(x) is a polynomial with complex coefficients, and the terms a i (x) also belong
Abstract Algebra Notes 47

to C[x]. The equation can be written in compact form as L(y) = 0, where L is

the differential operator
a n (x)D n + … + a 1 (x)D + a 0 (x) = 0,
with D = d/dx. Thus the differential operator can be thought of as a
polynomial in the two indeterminates x and D, but in this case the
indeterminates do not commute, since D(xy(x)) = y(x) + x D(y(x)),
yielding the identity Dx = 1 + xD.
Repeated use of this identity makes it possible to write the composition of
two differential operators in the standard form
a 0 (x) + a 1 (x)D + … + a n (x)D n ,
and we denote the resulting ring by C[x][D].

E XAMPLE . (Group algebras) Let K be a field, and let G be a finite group of

order n, with elements 1 = g 1 , g 2 , … , g n . The group algebra KG is defined to
be the n-dimensional vector space over K with the elements of G as a basis.
Vector addition is used as the addition in the ring. Elements of KG can be
n
described as sums of the form ∑
i =0
c i g i and multiplication is defined as for

polynomials, where the product g i g j is given by the product in G.

E XAMPLE . (Matrix rings) Let R be a ring. We let M n (R) denote the set of all n×n
matrices with entries in R. For [a ij ] and [b ij ] in M n (R), we use componentwise
addition [a ij ] + [b ij ] = [a ij + b ij ] and the multiplication is given by [a ij ].[b ij ] = [c ij ]
n
where [c ij ] is the matrix whose j, k-entry is c jk = ∑
i =0
a ji b ik .

4.1.2 D EFINITION . Let R be a ring, and let a ∈ R. If ab = 0 for some

nonzero b ∈ R, then a is called a left zero divisor. Similarly, if ba = 0 for
some nonzero b ∈ R, then a is called a right zero divisor. If a is neither a left
zero divisor nor a right zero divisor, then a is called a regular element.
The element a ∈ R is said to be invertible if there exists an element b ∈ R such that
ab = 1 and ba = 1. The element a is also called a unit of R, and its multiplicative
inverse is usually denoted by a -1 . The set of all units of R is denoted by U(R).
Prof. Mohammed El-Atrash 48

E XERCISE . In any ring R, show that the following is true for all a, b ∈ R:
(a) 0.a = 0.
(b) (-1)a = - a.
(c) (-a)b = a(-b) = - ab.
(d) if u is a unit then u is not a zero divisor.

4.1.3 P ROPOSITION . Let R be a ring. Then the set U(R) of units of R is a

group under the multiplication of R.

P ROOF . E XERCISE .

4.1.4 D EFINITION . A ring R in which each nonzero element is a unit is

called a division ring or skew field.

4.1.5 D EFINITION . A commutative ring R in which each nonzero element

is a unit is called a field.

E XAMPLE . Let R 1 , R 2 be two rings. Let R = R 1 ⊕ R 2 = {(r 1 , r 2 )| r 1 ∈R 1 , r 2 ∈ R 2 }.

Then R is a ring with addition and multiplication defined componentwise.

4.1.6 D EFINITION . Let R be a ring. A nonempty subgroup I of R under

addition is called an ideal of R if ra, ar ∈ I, for all a ∈ I and r ∈ R. I is called
left ideal if only ra ∈ I, and is called right ideal if only ar ∈ I.

4.1.7 P ROPOSITION . Let R be a commutative ring. Then R is a field if and

only if it has no proper nontrivial ideals.

P ROOF . Assume that R is a field. Let I be an ideal, if there is a ∈ I, a ≠ 0. Then a

has an inverse a -1 . Therefore aa -1 ∈ I, by definition of I. Therefore 1 ∈ I, it
follows that for every b ∈ R, b = b.1 ∈ I. Thus R ⊂ I.
Conversely, If R is a ring with no proper ideals then for every a ∈ R, a ≠ 0 the
ideal Ra = R, thus there is an element b ∈ R, such that ba = 1. Therefore, a is a
unit. Hence R is a field.

4.1.8 D EFINITION . A ring R with no proper ideals is called simple.

Abstract Algebra Notes 49

E XAMPLE . Let R = Z, nZ = {nz| z ∈ Z} is an ideal for every n ∈ Z.

4.1.9 D EFINITION . Let I be a proper ideal of the commutative ring R. Then

I is said to be a prime ideal of R if for all a, b ∈ R it is true that ab ∈ I
implies a ∈ I or b ∈ I.

4.1.10 D EFINITION . The ideal I is said to be a maximal ideal of R if for all

ideals J of R such that I ⊂ J ⊂ R, either J = I or J = R.

4.1.11 D EFINITION . For an ideal I of a commutative ring R, the set {a + I |

a ∈ R } of cosets of I in R (under addition) is denoted by R/I. The set R/I forms
a group under addition.

The next theorem justifies calling R/I the factor ring of R modulo I.

4.1.12 T HEOREM . If I is an ideal of the commutative ring R, then R/I is a

commutative ring, under the operations
(a + I) + (b + I) = (a + b) + I and (a + I)(b + I) = ab + I, for all a, b ∈ R.
P ROOF . E XERCISE .

4.1.13 D EFINITION . Let R, S be rings. A function ϕ : R → S is called a ring

homomorphism from R into S if the following two conditions hold:
(a) ϕ (a + b) = ϕ (a) + ϕ (b)
(b) ϕ (ab) = ϕ (a) ϕ (b), for all a, b ∈ R.
We denote ker ϕ = {a ∈ R| ϕ (a) = 0}.

A ring homomorphism that is one to one and onto is called isomorphism, and
in this case R and S are called isomorphic and denoted by R ≅ S. If R = S then
it is called an automorphism of R.
of course (a) says that ϕ is a group homomorphism. This implies that all
homomorphism theorems for groups hold for rings.

E XERCISE . Show that if ϕ is a homomorphism then kerϕ is an ideal.

 Prof. Mohammed El-Atrash 50

4.2 Integral domains

4.2.1 D EFINITION . A commutative ring R is called an integral domain if
for all a, b ∈ R, ab = 0 implies a = 0 or b = 0.

The ring of integers Z is the most fundamental example of an integral

domain. The ring of all polynomials with real coefficients is also an integral
domain, but the larger ring of all real valued functions is not an integral
domain. The cancellation law for multiplication holds in R if and only if R has
no nonzero divisors of zero. One way in which the cancellation law holds in R
is if nonzero elements have inverses in a larger ring; the next two results
characterize integral domains as subrings of fields (that contain the identity
1).

4.2.2 D EFINITION . A subset S of a ring R is called a subring if

(a) (S, +) is a subgroup of R.
(b) multiplication is a binary operation on S.
i.e., a subring is a subset of R that is a ring under the same operations of R.
Subrings of R do not have to have the same multiplicative identity of R. We
can see this clear in the following example.

E XAMPLE . Let R = M 2×2 (R). Let S = { [a ij ] | a 12 = 0, a 21 = 0, a 22 = 0}. S is a subring

under matrix addition and multiplication with multiplicative identity [a ij ]
with a 11 = 1, a 12 = 0, a 21 = 0, a 22 = 0. This is of course different than the identity
of R.

Subrings with the same multiplicative identity is called unital subrings.

4.2.3 T HEOREM . Let F be a field . Any unital subring of F is an integral

domain.

P ROOF . E XERCISE .
Abstract Algebra Notes 51

4.2.4 T HEOREM . Any finite integral domain must be a field.

P ROOF . E XERCISE .

4.2.5 D EFINITION . Let R be a ring. An integer n is called the

characteristic of R if n is the smallest integer such that na = 0, for all a ∈ R.

E XERCISE . Show that n is the characteristic of R iff n1 = 0.

4.2.6 P ROPOSITION . An integral domain has characteristic 0 or p, for

some prime number p.

4.2.7 P ROPOSITION . Let I be a proper ideal of the commutative ring R.

(a) The factor ring R/I is a field if and only if I is a maximal ideal of R.
(b) The factor ring R/I is a integral domain if and only if I is a prime ideal of R.
(c) If I is maximal, then it is a prime ideal.

P ROOF . (a) Let R/I be a field. Let J be an ideal with I ⊂ J ⊂ R. Then J/I is an
ideal of R/I, however R/I is a field, therefore either J/I = I or J/I = R/I. It
follows that either J = I or J = R. Thus I is maximal. Conversely, If I is maximal
then R/I has no proper ideals therefore by E XERCISE () R/I is a field.
(b) Assume that R/I is an integral domain. Let ab ∈ I, we would like to show
that either a ∈ I or b ∈ I. Note that (I + a)(I + b) = (I + ab) = I, but R/I is an
integral domain i.e., it has no zero-divisors, it follows that either (I + a) = I, or
(I + b) = I, i.e., a ∈ I or b ∈ I. i.e., I is prime. Conversely, assume that I is
prime. To show that R/I does not have zero-divisors. Let (I + a)(I + b) = (I +
ab) = I, this implies that ab ∈ I, and since I is prime then either a ∈ I or b ∈ I.
Thus either (I + a) = I, or (I + b) = I, i.e., R/I has no zero-divisors.
(c) I is maximal implies, by (a), that R/I is a field and, by (b), every field is an
integral domain, so R/I is an integral domain, therefore , by (b), I is prime.

4.2.8 D EFINITION . Let R be a commutative ring, and let a ∈ R. The ideal

Ra = {x ∈ R | x = ra for some r ∈ R }
Prof. Mohammed El-Atrash 52

is called the principal ideal generated by a. An integral domain in which every

ideal is a principal ideal is called a principal ideal domain denoted by PID.

E XAMPLE . (Z is a principal ideal domain) Theorem [] shows that the ring of

integers Z is a principal ideal domain. Moreover, given any nonzero ideal I of
Z, the smallest positive integer in I is a generator for the ideal.

4.2.9 T HEOREM . Every nonzero prime ideal of a principal ideal domain is

maximal ideal.

P ROOF . E XERCISE .

E XAMPLE . (Ideals of F[x]) Let F be any field. Then F[x] is a principal ideal
domain, since by Theorem [] the ideals of F[x] have the form I = 〈f(x)〉, where
f(x) is the unique monic polynomial of minimal degree in the ideal. The ideal I
is prime (and hence maximal) if and only if f(x) is irreducible. If p(x) is
irreducible, then the factor ring F[x]/〈p(x)〉 is a field.

For any ring R, it is clear that the set {0} is an ideal, which we will refer to as
the trivial ideal. Another ideal of R is the ring R itself.

4.2.10 D EFINITION . Let R be a ring, and let a ∈ R. The left ideal

Ra = {x ∈ R | x = ra for some r ∈ R}
is called the principal left ideal generated by a.

4.2.11 P ROPOSITION . Let R be a ring, and let I, J be left ideals of R. The

following subsets of R are left ideals.
(a) I ∩ J;
(b) I + J = {x ∈ R | x = a + b for some a ∈ I, b ∈ J};
n
(c) IJ = { ∑
i =1
a i b i | a i ∈ I, b i ∈ J, n ∈ Z}.

P ROOF . E XERCISE .

E XAMPLE . (Ideals of M n×n (R))

Abstract Algebra Notes 53

Let R be a ring, and let M n×n (R) be the ring of matrices over R. If I is an ideal
of R, then the set M n×n (I) of all matrices with entries in I is an ideal of S.
Conversely, every ideal of S is of this type.

4.2.12 P ROPOSITION . Any ring R is isomorphic to a subring of an

endomorphism ring End(A), for some abelian group A.

P ROOF . For a ∈ R, Let r a : R → R be defined by r a (x) = xa. r a is an

endomorphism of abelian group (R, +), since (x + y)r a = (x + y)a = xa + ya =
(x)r a + (y)r a . Let θ: R → End(R) defined by θ(a) = r a . θ is a ring isomorphism.
To see that we need to show that (1) θ(ab) = θ(a)θ(b). (2) θ(a + b) = θ(a) + θ(b).
(3) ker(θ) = {0}.
For (1) we need to show that r ab = r a r b , but this means that, for x ∈ R (x)ab =
(xa)b, and this is the associativity in R. For (2) we use the left distributive law
in R. For (3) a ∈ ker(θ) iff r a = 0, i.e., 0 = (1)r a = 1a = a. So ker(θ) = 0. This
completes the proof.

4.2.13 T HEOREM . (Fundamental Homomorphism Theorem for Rings)

Let ϕ :R → S be a ring homomorphism. Then ϕ (R) is a subring of S, R/ker( ϕ )
is a ring, and R/ker( ϕ ) ≅ ϕ (R).

P ROOF . Let N = ker(θ). Let θ be the homomorphism θ : R/N → ϕ(R), defined

by θ(Nx) = ϕ(x). we have seen before that this is a group isomorphism. So we
have R/N ≅ ϕ(R) as abelian groups under +. To show that it is a ring
homomorphism we need to show that θ( NaNb) = θ(Na)θ(Nb), i.e., but this is
true since θ(NaNb) = θ(Nab) = ϕ(ab) = ϕ(a)ϕ(b) = θ(Na)θ(Nb).
 Prof. Mohammed El-Atrash 54

4.2.14 P ROPOSITION . Let I be an ideal of the ring R.

(a) The natural projection mapping π :R → R/I defined by π (a) = a + I for all a ∈ R is
a ring homomorphism, and ker( π ) = I.
(b) There is a one-to-one correspondence between the ideals of R/I and ideals of R
that contain I.
(c) If K is an ideal of R with I ≤ K ≤ R, then (R/I)/(K/I) ≅ R/K.

P ROOF . E XERCISE .

4.2.15 T HEOREM . (Chinese Remainder Theorem) Let R be a ring, and let

I 1 , I 2 be ideals of R such I 1 + I 2 = R. Then
( R / I 1 ) ⊕ ( R / I 2 ) ≅ R / (I 1 ∩ I 2 ).

P ROOF . Let θ : R → ( R / I 1 )⊕( R / I 2 ) be a function defined as follows

θ(r) = (r + I 1 , r + I 2 )
To see that θ is a homomorphism θ( a + b) = (a + b + I 1 , a + b + I 2 ) = (a + I 1 + b +
I 1 , a + I 2 + b + I 2 ) = (a + I 1 , a + I 2 ) + (b + I 1 , b + I 2 ) = θ(a) + θ(b).
θ(ab) = (ab + I 1 , ab + I 2 ) = ((a + I 1 )(b + I 1 ), (a + I 2 )(b + I 2 )) = (a + I 1 , a + I 2 )(b + I 1 , b
+ I 2 ) = θ(a)θ(b).
ker θ = { r ∈ R | (r + I 1 , r + I 2 ) = (I 1 , I 2 )}
= { r ∈ R | (r ∈ I 1 and r ∈ I 2 )}
= { r ∈ R | (r ∈ I 1 ∩ I 2 )}.
θ is surjective, since if (a + I 1 , b + I 2 ) ∈ ( R / I 1 )⊕( R / I 2 ), we need to find r ∈ R
with r + I 1 = a +I 1 , r + I 1 = b +I 2 . But since I 1 + I 2 = R then a -b = r 2 - r 1 where
r 1 ∈ I 1 , r 2 ∈ I 2 and then a = b + r 2 - r 1 . Let r = a + r 1 = b + r 2 . Then r ∈ a + r 1 + I 1 =
a + I 1 and r = b + r 2 ∈ b + r 2 + I 2 = b + I 2 . Then by Fundamental theorem ( R /
I 1 )⊕( R / I 2 ) ≅ R / (I 1 ∩I 2 ).

4.3 Definition of a module

4.3.1 D EFINITION . Let R be a ring, and let M be an abelian group. Then
M is called a left R-module if there exists a scalar multiplication
Abstract Algebra Notes 55

ψ :R × M → M denoted by ψ (r, m) = r m, for all r ∈ R and all m ∈ M, such that for

all r, r 1 , r 2 ∈ R and all m, m 1 , m 2 ∈ M,
(i) r(m 1 + m 2 ) = rm 1 + rm 2
(ii) (r 1 + r 2 ) m = r 1 m + r 2 m
(iii) r 1 (r 2 m) = (r 1 r 2 ) m
(iv) 1 m = m.

To denote that M is a left R-module we write R M.

E XAMPLE . If R is a ring then R itself is an R-Module, Left R-module and right

R-module. So when we want to stress the fact that R is a left R-module we
write R R.

E XAMPLE . (Vector spaces over F are F-modules) If V is a vector space over a

field F, then it is an abelian group under addition of vectors. The familiar
rules for scalar multiplication are precisely those needed to show that V is a
module over the ring F.

E XAMPLE . (Abelian groups are Z-modules) If A is an abelian group with its

operation denoted additively, then for any element x ∈ Z and any positive
integer n, we have defined nx to be the sum of x with itself n times. This is
extended to negative integers by taking sums of -x. With this familiar
multiplication, it is easy to check that A becomes a Z-module.

Another way to show that A is a Z-module is to define a ring homomorphism

ϕ :Z → End(A) by letting ϕ (n) = n1, for all n ∈ Z. This is the familiar mapping
that is used to determine the characteristic of the ring End(A). The action of Z
on A determined by this mapping is the same one used in the previous
paragraph.

If M is a left R-module, then there is an obvious definition of a submodule of

M: any subset of M that is a left R-module under the operations induced from
M. The subset {0} is called the trivial submodule, and is denoted by (0). The
module M is a submodule of itself, an improper submodule. It can be shown
Prof. Mohammed El-Atrash 56

that if M is a left R-module, then a subset N ⊂ M is a submodule if and only if

it is nonempty, closed under sums, and closed under multiplication by
elements of R.

If N is a submodule of R M, then we can form the factor group M/N. There is a

natural multiplication defined on the cosets of N: for any r ∈ R and any x ∈
M, let r(x + N) = rx + N. If x + N = y + N, then x-y ∈ N, and so rx-ry = r(x-y)N,
and this shows that scalar multiplication is well-defined. It follows that M/N
is a left R-module, called left factor R-module

Any submodule of R R is a left ideal of R. A submodule of R R is called a ideal

of R, and it is clear that a subset of R is an ideal if and only if it is both a left
ideal and a right ideal of R.

For any element m of the module M, we can construct the submodule

Rm = { x ∈ M | x = rm for some r ∈ R }.

This is the smallest submodule of M that contains m, so it is called the cyclic

submodule generated by m. More generally, if X is any subset of M, then the
intersection of all submodules of M which contain X is the smallest
submodule of M which contains X. We will use the notation 〈X〉 for this
submodule, and call it the submodule generated by X. We must have Rx ⊂ 〈X〉
for all x ∈ X, and then it is not difficult to show that

〈X〉 = ∑a
x∈ X
X x.

4.3.2 D EFINITION . The left R-module M is said to be finitely generated

if there exist m 1 , m 2 , . . . , m n ∈ M such that

n
M= ∑
i =1
Rm i .
Abstract Algebra Notes 57

In this case, we say that { m 1 , m 2 , . . . , m n } is a set of generators for M. The module

M is called a free module if there exists a subset X ⊂ M such that each element m ∈
M can be expressed uniquely as a finite sum

n
m= ∑
i =1
a i x i , with a 1 , . . . , a n ∈ R and x 1 , . . . , x n ∈ X.

We note that if N is a submodule of M such that N and M/N are finitely

generated, then M is finitely generated. In fact, if x 1 , . . . , x n generate N and y 1
+ N, y 2 + N, . . . , y m + N generate M/N, then x 1 , . . . , x n , y 1 , . . . , y m generate
M.

The module R R is the prototype of a free module, with generating set {1}. If
RM is a module, and X ⊂ M, we say that the set X is linearly independent if
n

∑
i =1
a i x i = 0 implies a i = 0 for i = 1, ..., n, for any distinct x 1 , . . . , x n ∈ X and any

a 1 , a 2 , . . . , a n ∈ R. Then a linearly independent generating set for M is called a

basis for M, and so M is a free module if and only if it has a basis.

4.3.3 D EFINITION . Let M and N be left R-modules. A function f :M → N

is called an R-homomorphism if
f(m 1 + m 2 ) = f(m 1 ) + f(m 2 ) and f(rm) = rf(m)

for all r ∈ R and all m, m 1 , m 2 ∈ M. The set of all R-homomorphisms from M

into N is denoted by
Hom R (M, N) or Hom( R M, R N).
For an R-homomorphism f ∈ Hom R (M, N) we define its kernel as
ker(f) = { m ∈ M | f(m) = 0 }.
We say that f is an isomorphism if it is both one-to-one and onto. Elements of
Hom R (M, M) are called endomorphisms, and isomorphisms in Hom R (M, M)
are called automorphisms. The set of endomorphisms of R M will be denoted
by End R (M).
 Prof. Mohammed El-Atrash 58

4.3.4 P ROPOSITION . Let M be a free left R-module, with basis X. For any
left R-module N and any function ϕ :X → N there exists a unique R-
homomorphism

f :M → N with f(x) = ϕ (x), for all x ∈ X.

i
X M
f
ϕ

P ROOF . Since X is a basis for M, then every element of M can be written as a

linear combination of elements of the basis i.e., for m ∈ M there is a 1 , a k ∈ R
such that m = ∑a i x i , with x i ∈ X. Define f : M → N by the rule f(m) = ∑a i ϕ(x i ),
then by definition the right hand side belongs to N. It is easy to see that f is a
R-homomorphism that satisfies the given equation f(x) = ϕ(x), for all x ∈ X.

4.3.5 T HEOREM . Let N, N 0 , M 0 be submodules of R M.

(a) N 0 / (N 0 ∩ M 0 ) ≅ (N 0 + M 0 ) / M 0 .
(b) If N 0 ⊂ N, then (M / N 0 ) / (N / N 0 ) ≅ M / N.
(c) If N 0 ⊂ N, then N ∩ (N 0 + M 0 ) = N 0 + (N ∩ M 0 ).

P ROOF . E XERCISE .

4.3.6 D EFINITION . A non-empty set P is called a poset (Partially ordered

set), if there is a relation ≤ on the elements of P satisfies the following
conditions:
(i) ≤ is reflexive, i.e., for every x ∈ P; x ≤ x.
(ii) ≤ is antisymmetric, i.e., if x ≤ y and y ≤ x then x = y.
(iii) ≤ is transitive, i.e., if x ≤ y and y ≤ z then x ≤ z.

4.3.7 L EMMA . (Zorn) Given a poset P ≠ ∅ . Assume that for every linearly
ordered subset L ⊂ P, there is b ∈ P such that b ≥ x for all x ∈ L. Then there
exists m ∈ P such that m is maximal in P.
Abstract Algebra Notes 59

4.3.8 L EMMA . Let X be any subset of the module R M. Any submodule N

with N ∩ X ⊂ (0) is contained in a submodule maximal with respect to this
property.

P ROOF . Let H = { N ≤ M| N ∩ X = (0)}. H ≠ φ, since (0) ∈ H. We would like to

show that H has a maximal submodule. By Zorn's Lemma, it is enough to
show that every linearly ordered set of H has a maximal submodule. So let N 1
⊂ N 2 ⊂ N 3 ⊂ …, be a series of submodules in H. Let N = ∪ i N i . N is a
submodule (prove !) in H, since N i ∩ X = (0) for all i. So every linearly
ordered set of H has a maximal submodule. Thus H has a maximal submodule
N.

R EMARK . Here, I want to remark that the maximal submodule that we have
proved to exist in the last lemma is not maximal in the sense that there is no
submodule larger than it, except the whole module, but it is maximal having
the property that it has no elements common with the set X. i.e., If L is
another submodule with this property N ∩ X ⊂ (0) then it must contain L ⊂ N.

A submodule N of the left R-module M is called a maximal submodule if N ≠

M, and for any submodule K with N ⊂ K ⊂ M, either N = K or K = M.
Consistent with this terminology, a left ideal A of R is called a maximal left
ideal if A ⊂ R and for any left ideal B with A ⊂ B ⊂ R, either A = B or B = R.
Thus A is maximal precisely when it is a maximal element in the set of proper
left ideals of R, ordered by inclusion. It is an immediate consequence of
Lemma ( ) that every left ideal of the ring R is contained in a maximal left
ideal, by applying the proposition to the set X = {1}. Furthermore, any left
ideal maximal with respect to not including 1 is in fact a maximal left ideal.

4.3.9 P ROPOSITION . For any nonzero element m of the module R M and

any submodule N of M with m not in N, there exists a submodule N* maximal
Prof. Mohammed El-Atrash 60

with respect to N* containing N with m not in N*. Moreover, M/N* has a

minimal submodule contained in every nonzero submodule.

P ROOF . E XERCISE .

4.3.10 C OROLLARY . Any proper submodule of a finitely generated module

is contained in a maximal submodule.

4.3.11 D EFINITION . Let R be a ring, and let M be a left R-module. For any
element m ∈ M, the left ideal
Ann(m) = { r ∈ R | rm = 0 }
is called the annihilator of m. The ideal
Ann(M) = { r ∈ R | rm = 0 for all m ∈ M }.
is called the annihilator of M.
The module M is called faithful if Ann(M) = (0).

4.3.12 D EFINITION . A nonzero module RM is called simple (or

irreducible) if its only submodules are (0) and M.

We first note that a submodule N ⊂ M is maximal if and only if M/N is a

simple module. A submodule N ⊂ M is called a minimal submodule if N ≠ (0)
and for any submodule K with (0) ⊂ K ⊂ N, either N = K or K=(0). With this
terminology, a submodule N is minimal if and only if it is simple when
considered as a module in its own right.

4.3.13 L EMMA . (Schur) If R M is simple, then End R (M) is a division ring.

P ROOF . End R (M) has ring structure under addition and composition of maps,
defined as follows: Let ϕ, θ ∈ End R (M), let x ∈ M, then
(ϕ + θ)(x) = ϕ(x) + θ(x),
(ϕ . θ)(x) = ϕ(θ(x)).

It is easy to see that End R (M) has ring structure with the 0, 1 as the zero
endomorphism and identity endomorphism.
 Abstract Algebra Notes 61

The only thing we need to prove is that every non-zero endomorphism ϕ has
an inverse ϕ -1 . But this holds since kerϕ is an R-submodule of M, therefore by
simplicity of M kerϕ = 0. Thus ϕ is one to one. Since ϕ(M) is also an R-
submodule of M, therefore by simplicity of ϕ(M) = M provided that ϕ is not
the zero endomorphism.

4.3.14 P ROPOSITION . The following conditions hold for a left R-module M.

(a) The module M is simple if and only if Rm = M, for each nonzero m ∈ M.
(b) If M is simple, then Ann(m) is a maximal left ideal, for each nonzero m ∈ M.
(c) If M is simple, then it has the structure of a left vector space over a division ring.

P ROOF . (a) Let M be a simple R-module. The set Rm is a submodule of M. It

follows by simplicity of M that Rm = M.
Conversely, if Rm = M for all 0 ≠ m ∈ M then M has no proper submodules.
I.e., M is simple.

(b) Let M be a simple R-module. Let m ∈ M. Let θ : R → M be defined by

θ(r) = rm. θ is a R-homomorphism,
since θ(s + r ) = (s + r)m = sm + rm = θ(s) + θ(r) and
θ(sr) = (sr)m = s(rm) =sθ(r). Then by homomorphism theorem we have
R R/ker(θ) ≅ M ( Note that θ is onto by simplicity of M).
Now ker(θ) = { r ∈ R | rm = 0} = ann(m). Thus R/ker(θ) ≅ M. Simplicity of M
now implies that ann(m) is maximal.

(c) E XERCISE .

4.4 The Jacobson Radical

4.4.1 D EFINITION . Let M be a left R-module. The intersection of all

maximal submodules of M is called the Jacobson radical of M, and is
denoted by J(M).

This would make J(R) = ∩{I | I is maximal left ideal of R}.

Prof. Mohammed El-Atrash 62

4.4.2 P ROPOSITION . J(R) = ∩ {ann(M)| M is simple left R-module}.

P ROOF . By Proposition () ann(M) is maximal left ideal of R. It follows that

J(R) ⊂ ∩{ann(M)| M is simple left R-module}. For the reverse inclusion, let I
be a maximal left ideal of R. Let u ∈ ∩{ann(M)| M is simple left R-module}.
We will show that u ∈ I. Since I is maximal left ideal then R/I is a simple left
R- module. It follows that u(R/I) = I (0 in R/I). So, u(a + I) = I for all a ∈ R.
Take a = 1, then u + I = I. Thus u ∈ I. Hence ∩{ann(M)| M is simple left R-
module} ⊂ ∩{I | I is maximal left ideal of R}. Hence they are equal.

4.4.3 L EMMA . (Nakayama) If R M is finitely generated and J(R)M = M,

then M = (0).

P ROOF . E XERCISE .

4.4.4 L EMMA . Let U be a proper left ideal of the ring R, then there exits a
maximal left ideal V of R such that U ⊂ V.

P ROOF . Let P = { I ⊂ R | I is left ideal of R , U ⊂ I ≠ R}. P ≠ ∅ since U ∈ P. P is

a poset ordered by inclusion. Let L be a set of linearly ordered ideals of P. we
need to find I ∈ P such that J ⊂ I for all J ∈ L. Let I = ∪ { J | J ∈ L}. Show that I
is an ideal containing U. Then by Zorn's Lemma there is an ideal V that
satisfies the lemma.

4.4.5 P ROPOSITION . For any ring R, J(R) is two sided ideal.

P ROOF . Let r ∈ R, let x ∈ J(R) and Let M be a simple left R-module. Since xM =
0 then rxM = 0. Since rM = M by simplicity of M then x(rM) = (xr)M = xM = 0,
it follows that xr ∈ J(R). Thus J(R) is a right ideal of R. Whence it is two sided
ideal.
Abstract Algebra Notes 63

4.4.6 D EFINITION . If R is a ring, x ∈ R is called right-quasi-regular if 1 -

x has a right inverse (denoted rqr). Similarly x is left-quasi-regular if 1 - x
has a left inverse (denoted lqr) and x is called quasi-regular if 1 - x is both rqr
and lqr ( denoted qr).

Note x is qr iff x is a unit in R.

4.4.7 P ROPOSITION . Let x ∈ J(R) then x is lqr.

P ROOF . R(1 - x) is a left ideal of R. If R(1 - x) = R, then 1∈ R(1 - x), so there is

r∈R such that 1 = r(1 - x), so r is the right inverse of (1 - x). Thus x is lqr.
Now assume that R(1 - x) < R. By Zorn's Lemma there is a maximal ideal I
with R(1 - x) ⊂ I. Thus 1 - x ∈ I, but x ∈ I then 1 ∈ I implying that I = R
contradicting the maximality of I.

4.4.8 C OROLLARY . Let x ∈ K(R) = ∩ {I | I is maximal right ideals of R}

then x is rqr.

P ROOF . E XERCISE .

4.4.9 T HEOREM . Let I be any left ideal of R with the property that every
element of I is lqr then I ⊂ J(R).

P ROOF . Let M be a maximal left ideal of R. We will show that I ⊂ M. If not ,

i.e., if I is not contained in M then there is an ideal I + M which is a left ideal
of R containing M properly, therefore I + M = R. Thus there is u ∈ I, m ∈ M
such that u + m = 1. Then m = 1 - u. Since u is lqr then there is a left inverse of
1- u = m. It follows that 1 ∈ Rm ⊂ M, contradicting the maximality of M.
Hence I ⊂ M.

4.4.10 T HEOREM . For a ring R, J(R) = ∩ {I | I is maximal right ideals of

R}.
Prof. Mohammed El-Atrash 64

P ROOF . Let K(R) = ∩{I | I is maximal right ideals of R}. Let u ∈ J(R), to show
that u ∈ I for every maximal right ideal I of R. It is enough to show that every
element u ∈ J(R) is rqr. Since u is lqr then there is r∈R such that r(1 - u) =
1. Let z = 1 - r. Then (1 - z)(1 - u) = 1. It follows that 1 - z - u + zu = 1. Thus z =
zu -u. This implies that z ∈ J(R) i.e., z is lqr. Hence 1 - z has a left inverse. I.e.,
y has a right inverse and it must be 1 - u. So, y(1 - u) = (1 - u)y = 1. Thus u is a
rqr. So u ∈ K(R). The reverse inclusion is similar.

4.4.11 C OROLLARY . (Jacobson-Perlis Condition). x ∈ J(R) iff 1 -rx has

left inverse for all r ∈ R.

P ROOF . x ∈ J(R) then rx ∈ J(R) since J(R) is left ideal of R. Therefore rx is lqr,
i.e., 1 - rx has left inverse. Conversely, suppose that 1 - rx has left inverse for
all r ∈ R. Therefore all the elements of the ideal Rx are lqr then by Theorem ( )
Rx ⊂ J(R), so x ∈ J(R).

4.4.12 D EFINITION . An element x is called nilpotent if x n = 0, for some n ≥

0. An additive subgroup U is called nil if each element of U is nilpotent.

E XERCISE . Let I be a nil left ideal of R. Show that I ⊂ J(R).

4.4.13 T HEOREM . The Jacobson radical J(R) of the ring R is equal to each of
the following sets:
(1) The intersection of all maximal left ideals of R;
(2) The intersection of all maximal right ideals of R;
(3) {x ∈ R | rx is lqr for all r ∈ R };
(4) {x ∈ R | xr is rqr for all r ∈ R };
(5) The largest ideal J of R such that 1 - x is invertible in R for all x ∈ J.
(6) The largest ideal J of R such that J containing nil left ideals of R.

4.4.14 D EFINITION . The ring R is said to be semiprimitive if J(R) = (0).

 Abstract Algebra Notes 65

Chapter V

5.1 Chain Conditiions

5.1.1 D EFINITION . Let P be a poset with order relation ≤ . We say that P
satisfies the ascending chain condition (ACC) if for every chain x 1 ≤ x 2 ≤ x 3
≤ … ≤ x n ≤ … .there is an integer n such that x n = x n + 1 = x n + 2 = x n + 3 = … .

And We say that P satisfies the descending chain condition (DCC) if for every
chain x 1 ≥ x 2 ≥ x 3 ≥ … ≥ x n ≥ … .there is an integer n such that x n = x n + 1 = x n + 2 =
x n + 3 = ….

P is said to satisfy the maximal condition (MaxC) if for every non-empty set S ⊂ P
there is a maximal element x ∈ S, such that if y ∈ S then y ≤ x.

P is said to satisfy the minimal condition (MinC) if for every non-empty set S ⊂ P
there is a minimal element x ∈ S, such that if y ∈ S then y ≥ x.

5.1.2 D EFINITION . An s-group G is said to be Noetherian if the poset of

all s-subgroups of G satisfies the (ACC). Similarly, G is said to be Artinian if
the poset of all s-subgroups of G satisfies the (DCC).

5.1.3 D EFINITION . A module R M is said to be Noetherian if the poset of

all sumodules of R M satisfies the (ACC). Similarly, M is said to be Artinian
if the poset of all sumodules of R M satisfies the (DCC).

E XAMPLE . Z is Notherian but not Artinian, since

〈2〉 ⊃ 〈4〉 ⊃ 〈8〉 ⊃ … .

5.1.4 D EFINITION . A ring R is said to be left Noetherian if the module

RR is Noetherian. A ring R is said to be left Artinian if the module R R is
Artinian. If R satisfies the conditions for both right and left ideals, then it is
simply said to be Noetherian or Artinian.
Prof. Mohammed El-Atrash 66

5.1.5 T HEOREM . Let P be a poset. P satisfies (ACC) iff P satisfies the

(MaxC). And P satisfies (DCC) iff P satisfies the (MinC).

P ROOF . Assume that P satisfies the (MaxC), to show that P satisfies the (ACC),
let x 1 ≤ x 2 ≤ … be an ascending chain in P. Let S = { x i | i ≥ 1}. S is non-empty.
Therefore by the (MaxC) there is x ∈ S with x maximal. I.e., x = x n for some n.
Then x n = x n + 1 = x n + 2 = x n + 3 = ….
Conversely, assume that P satisfies the (ACC). Let S ⊂ P, S ≠ ∅. Assume that
there is no maximal element in S. Then for x ∈ S the set S x = { y ∈ S | y > x } ≠
∅. So by the axiom of choice there is a function ϕ : S → S such that ϕ(x) ∈ S x .
i.e., x < ϕ(x) < ϕ(ϕ(x) < ϕ 3 (x) < … < ϕ n (x) < … which is a chain that is not
eventually contant. Contradicting that P satisfies the (ACC).
Similarly for the (DCC).

5.1.6 C OROLLARY . The following conditions are equivalent for a module

R M:

(1) M is Noetherian;
(2) every nonempty set of submodules of M has a maximal member.

E XERCISE . Let A, B, K be left submodules of a left R-module M. If A⊂ B and

A +K = B + K and A∩K = B∩K then A = B.

5.1.7 P ROPOSITION . The following conditions hold for a module R M and

any submodule N.
(a) M is Noetherian if and only if N and M/N are Noetherian.
(b) M is Artinian if and only if N and M/N are Artinian.

P ROOF . (a) Assume that M is Noetherian. Let N 1 ≤ N 2 ≤ … , be an ascending

chain of submodules of N, K 1 /N ≤ K 2 /N ≤ … be an ascending chain of
submodules of M/N. Then both N 1 ≤ N 2 ≤ … and K 1 ≤ K 2 ≤ … are ascending
chains of submodules of M. Therefore by the (ACC) of M both are eventually
constant. I.e., N, M/N satisfy the (ACC). Conversely, assume that both N,
M/N satisfy the (ACC).
Abstract Algebra Notes 67

Let M 1 ≤ M 2 ≤ … , be an ascending chain in M. Consider the ascending chains

N∩M 1 ≤ N∩M 2 ≤ … ,
(N + M 1 )/N ≤ ( N + M 2 )/N ≤ …
in N, M/N respectively. Since both N, M/N satisfy the (ACC) then there is a
finite integer m such that
N∩M m = N∩M m+1 = … ,
(N + M m )/N = ( N + M m+1 )/N = …
Thus N + M m = N + M m+1 and N∩M m = N∩M m+1 . Hence by the E XERCISE M m
= M m+1 = … . whence M satisfies the (ACC).

(b) The proof is similar.

E XERCISE . For an R-Module M, show that M is finitely generated iff M is

Noetherian

5.1.8 C OROLLARY . A finite direct sum of modules is Noetherian if and

only if each summand is Noetherian; it is Artinian if and only if each
summand is Artinian.

5.1.9 P ROPOSITION . A ring R is left Noetherian if and only if every

finitely generated left R-module is Noetherian; it is left Artinian if and only if
every finitely generated left R-module is Artinian.

5.1.10 T HEOREM . (Hilbert basis theorem) If R is a left Noetherian ring,

then so is the polynomial ring R[x].

P ROOF . Suppose that R is Noetherian. Let I ⊂ R[x] be a left ideal. It is enough

to show that I has a finite generating set. For n ≥ 0, define A n = { a ∈ R | a is a
leading coefficient of some polynomial f ∈ I with degree of f = n}∪ {0}. We
claim that A n is an ideal of R. To see this, let a, b ∈ A n , we want to show that a
- b ∈ A n , we may assume that a ≠ b and a, b ≠ 0. So there are two polynomials
f, g ∈ I with leading coefficient of f = a, and leading coefficient of g = b. Thus a
- b is the leading coefficient of f - g and so is all R multiples of a - b. Hence A n
is a left ideal for all n.
Prof. Mohammed El-Atrash 68

Note that A 1 ⊂ A 2 ⊂ … ⊂ A n ⊂ … is a chain of left ideals. Since if a ∈ A n then

there is a polynomial f for which a is the leading coefficient, then xf ∈ A n+ 1 is
a polnomial in I with leading coefficient a, so a ∈ A n + 1 .
Since R is Noetherian then there is N such that A N = A N + 1 = … . Choose a
finite generating set S i for A i with 1 ≤ i ≤ N. Let S i = { a i1 , a i2 , a i3 , …, a ik }. Let f ij
be polynomials in I with leading coefficients a ij and degree i. We claim now
that I is generated by all these polynomials f ij . To prove that, let J be the ideal
generated by all f ij . Since f ij ∈ I then J ⊂ I. Assume that there is f ∈ I, f ∉ J.
Assume also that the degree of f is minimal. Let m = degree of f. We have two
cases:
Case 1. If m ≤ N, let a = leading coefficient of f thus a ∈ A m and threfore a is an
R-linear combination of a mj . Let g be the R-linear of f mj with same coefficients.
Then g ∈ J thus a is a leading coefficient of g ∈ J, degree of g = m. Now f - g ∉
J, degree (f - g) < m thus f - g ∈ I. A contradiction because f was such example.
Case 2. M > N. a is a leading coefficient of g, a ∈ A m = A N . Therefore a is an R-
linear combination of a Nj . Get g as an R-linear combination of f Nj . Thus g ∈ J,
degree of g = N and the leading coefficient og g is a. Now x m-N g ∉ J and f - x m-
Ng ∈ I. But degree of f - x m-N g < m. a contradiction.C

We can now give some fairly wide classes of examples of Noetherian and
Artinian rings. If D is a principal ideal domain, then D is Noetherian since
each ideal is generated by a single element. It follows that the polynomial
ring D[x 1 , x 2 , ..., x n ] is also Noetherian. If F is a field, then F[x]/I is Artinian,
for any nonzero ideal I of F[x], since F[x] is a principal ideal domain. This
allows the construction of many interesting examples. Note that D[x]/I need
not be Artinian when D is assumed to be a principal ideal domain rather than
a field, since Z[x]/〈x〉 is isomorphic to Z, which is not Artinian.
 Abstract Algebra Notes 69

5.2 Semiprimitive Rings

5.2.1 D EFINITION . Let A, B be additive subgroups of a ring R. AB = the
additive subgroup generated by all products a b i.e.,
AB = 〈 {a b| a∈ A, b ∈ B}〉
Note it can be shown easily that
AB = { a 1 b 1 + a 2 b 2 + … + a n b n | a i ∈ A, b i ∈ B}
E XERCISE . Show that if A is left ideal then AB is left ideal, and if B is right
ideal then AB is right ideal.

5.2.2 D EFINITION . Let A ⊂ R be an additive subgroup. A is called

nilpotent if A n = 0, for some n > 0.

5.2.3 T HEOREM . If R is right Artinian then J(R) is nilpotent.

P ROOF . Let J = J(R). Consider the descending chain

J ⊇ J2 ⊇ J3 ⊇ … .
Since R is Artinian then ∃ integer n such that J n = J n+1 =… .
To show that J n = 0, let I = J n . Assume I ≠ 0. Note I 2 = J 2n = J n = I.
Let S = { N ⊂ I | N is right ideal of R and NI ≠ 0}. Since I ∈ S then S ≠ ∅.
Since R is Artinian then S has a minimal ideal, say N. Note that N ⊂ I, NI ≠ 0.
So there is x ∈ N such that xI ≠ 0.
Now (xI)I = xI ≠ 0 therefore xI ⊇ N by minimality of N. But x ∈ N thus xI = N.
therefore, ∃ y ∈ I such that x y = x. It follows that x(1 - y) = 0.
Since y ∈ I ⊂ J, then ∃ z ∈ R such that (1 - y)z = 1.
It follows that 0 = 0.z = x(1 - y) z = x 1 = x, a contradiction.

5.2.4 C OROLLARY . If R is right Artinian then any nil right ideal is

nilpotent.

P ROOF . If I is nil right ideal then I ⊂ J(R), but J(R) is nilpotent, then I n ⊂ J n = 0,
for some n > 0.

Also any nil left ideal of R is nilpotent.

Prof. Mohammed El-Atrash 70

5.2.5 C OROLLARY . Let R be left Artinian then TFAE.

(i) J(R) = 0.( R is semiprimitive)

(ii) If I is left ideal and I 2 = 0 then I = 0.( I is semiprime)
(iii) If I is an ideal and I 2 = 0 then I = 0.

P ROOF . (i) → (ii) I 2 = 0 implies that I is nil, therefore I ⊂ J(R) = 0, thus I = 0.

(ii) → (iii) straightforward.
(iii) → (i) we know that J(R) n = 0 for some n ≥ 1. Take smallest such n, then J n-1
≠ 0.
Let I = J(R) n -1 . It follows that I 2 = J(R) 2n -2 .= 0, then I = 0 by (iii), a
contradiction.

5.2.6 D EFINITION . A ring R is called a Wedderburn ring if it is Artinian

and semiprimitive.

5.2.7 P ROPOSITION . Let I be a minimal left ideal of R. Assume that I 2 ≠ 0

then I = Re for some e ∈ R, e 2 = e (e is called idempotent).

P ROOF . ∃ a ∈ I, Ia ≠ 0. Ia ⊂ I since a ∈ I and I is left ideal. Ia is a left ideal then

by minimality of I Ia = I. Thus ∃ e ∈ I, such that ea = a. Therefore e 2 a = ea =a. it
follows that a(e 2 - e) = 0.
Let S = { x ∈ I | xa = 0}. S is a left ideal, S ⊂ I. Thus S = 0 or S = I. but I ≠ 0, so
S = 0. e 2 - e ∈ S. then e 2 - e = 0. Hence e 2 = e. Since e ∈ I then Re ⊂ I, then Re = I
by minimality of I.

5.2.8 P ROPOSITION . (Pierce Decomposition). Let I be a left ideal of R,

let e ∈ I with e 2 = e then I = Ie ⊕ I(1 - e).

P ROOF . x ∈ I then x(1 - e) ∈ I and Ie ⊂ I. Hence I ⊇ Ie ⊕ I(1 - e). we need to

show that Ie ∩ I(1 - e) = 0. If x ∈ Ie ∩ I(1 - e), then x = ye for some y ∈ I, and x
= (1 - e)z, z ∈ I.
x = ey = e(ez) = e(1 - e)z = (e - e)z = 0z = 0.
Abstract Algebra Notes 71

5.2.9 T HEOREM . Every Wedderburn ring is the direct sum of finitely

many minimal left ideals.

P ROOF . we will prove a more general assertion. We will show that every left
ideal is the sum of finitely many minimal left ideals. Suppose false. Let S = { I
⊂ R| I is not the direct sum of finitely many minimal left ideals}. S ≠ ∅. Since
R is Artinian ring then S has a minimal element I. I ≠ 0. Let M be minimal left
ideal with M ⊂ I, such ideal exists, again because R is Artinian and the set of
all ideals contained in I has a minimal one. M ≠ 0, M 2 ≠ 0, since R is
semiprimitive. Thus M = Re, for some idempotent element e ∈ M, with e ≠ 0.
Now using Pierce decomposition then I = Ie ⊕ I(1 - e) = M ⊕ I(1 - e). Let K =
I(1 - e), then I = M ⊕ K. Since M ≠ 0, K < I. then K ∉ S. therefore K is the direct
sum of finitely many minimal left ideals of R. It follows that I is the direct
sum of minimal left ideals of R, contradicting our assumption. Therefore S =
∅. Hence R is the direct sum of minimal left ideals.

Note this theorem could have been stated as follows: if R is Artinian ring then
RR is the dircet sum of simple left R-modules.

5.2.10 C OROLLARY . If RR is the finite direct sum of minimal left ideals

{M i } m i = 1 then every minimal left ideal of R is isomorphic to one of the M i 's.

5.2.11 C OROLLARY . If R R is the direct sum of simple left R-modules {M i } m i

= 1 then every simple left simple R-module of R is isomorphic to one of the
M i 's.

P ROOF . Let S be any simple left R-module, let s ∈ S, s ≠ [Link] 1 = e 1 + e 2 + …

+ e n , for idempotent elements e i ∈ M i , i = 1, 2, …, n.
0 ≠ s = 1.s = (e 1 + e 2 + … + e n )s = e 1 s+ e 2 s+ … + e n s. thus e i s ≠ 0 for some i.
Fix such i. we will show that S ≅ M i .
Let θ : M i → S be defined by θ(x) = sx, for all x ∈ M i . It is easy to show that θ
is an R- homomorphism. θ(e i ) = e i s ≠ [Link] by simplicity of S, θ(M i )= S. again
by simplicity of M i , kernel of θ is 0. Thus they are isomorphic.
 Prof. Mohammed El-Atrash 72

E XERCISES
1. An element γ of a ring R, is called central if γ r = r γ for all elements
r ∈ R. Prove that the set of all central elements Z(R) forms a ring.
2. Let ϕ be an R-homomorphism of a left R-module M into a left R-
module N; i.e., ϕ : M → N. Let U be a subset of M and L be the left
submodule of M generated by U. Prove that the left submodule of N
generated by ϕ(U) is ϕ(L).
3. Let {A i } i∈I be a family of subsets of a left R-module M, and let N i
be the left submodule generated by A i . Show that Σ i N i is generated by ∪ i A i
.
4. Let N be a left submodule of a left R-module M. Let {A i } i∈I be a
family of submodules of M with N⊂ A i ⊂ M for all i ∈ I. Show that
(Σ i A i )/N = Σ i (A i /N) and (∩ i A i )/N = ∩ i (A i /N).
5. Let m be a positive integer and p be prime. Prove that mZ/mpZ is
a simple Z-module.
6. Let Ω denote the set of all rational numbers that can be expressed
in the form m/2 k , where m, k are integers. Show that Ω is a Z-module
having Z itself as a submodule. Also show that
(i) each proper submodule of Ω/Z contains only a finite number
of elements.
(ii) Ω/Z satisfies the (MinC) but not the (MaxC)
(iii) Ω/Z as a Z-module is not finitely generated.

7. Show that an integral domain that satisfies the (DCC) is a field.

8. Two ideals A, B are called comaximal if A + B = R. If A, B are

comaximal show that AB = A∩B.

9. If the rings R 1 , R 2 , …, R n are left noetherian show that the direct

sum R = R 1 ⊕ R 2 ⊕ ….⊕ R n is left noetherian.
 Abstract Algebra Notes 73

10. Let M be a left R-module. Let M[X] denotes the formal set of "
polynomials" elements of the form m 0 + m 1 X + m 2 X 2 + …+ m k X k , for some
integer k. Show that M[X] is an R[X]-module and if M satisfies the (ACC)
then so is M[X].

5.3 Composition series

5.3.1 D EFINITION . A composition series of length n for a nonzero module
M is a chain of n + 1 submodules
M = M 0 ⊇ M 1 ⊇ . . . ⊇ M n = (0)
such that M i-1 /M i is a simple module for i = 1, 2, ..., n. These simple modules are
called the composition factors of the series.

5.3.2 T HEOREM . (Jordan-Holder) If a module M has a composition

series, then any other composition series for M is equivalent to it.

P ROOF . E XERCISE .

As an immediate consequence of the Jordan-Holder theorem, if a module R M

has a composition series, then all composition series for M must have the
same length, which we denote by λ(M). This is called the length of the
module, and we simply say that the module has finite length. Since any
ascending chain of submodules can be refined to a composition series, λ(M)
gives a uniform bound on the number of terms in any properly ascending
chain of submodules. We also note that if M 1 and M 2 have finite length, then
λ(M 1 ⊕ M 2 ) = λ(M 1 ) + λ(M 2 ).

5.3.3 P ROPOSITION . A module has finite length if and only if it is both

Artinian and Noetherian.

A module R M is said to be indecomposable if its only direct summands are

(0) and M. As our first example, we note that Z is indecomposable as a
module over itself, since the intersection of any two nonzero ideals is again
nonzero. To give additional examples of indecomposable Z-modules, recall
Prof. Mohammed El-Atrash 74

any finite abelian group is isomorphic to a direct sum of cyclic groups of

prime power order. Using this result, we see that a finite Z-module is
indecomposable if and only if it is isomorphic to Z n , where n = p k for some
prime p.

5.3.4 P ROPOSITION . If R M has finite length, then there exist finitely

many indecomposable submodules M 1 , M 2 , . . . , M n such that

M = M1 ⊕ M2 ⊕ . . . ⊕ Mn.

5.3.5 L EMMA . [Fitting] Let M be a module with length n, and let f be

an endomorphism of M. Then

M = Im(f n ) ⊕ ker(f n ) .

5.3.6 P ROPOSITION . Let M be an indecomposable module of finite length.

Then for any endomorphism f of M the following conditions are equivalent.
(1) f is one-to-one;
(2) f is onto;
(3) f is an automorphism;
(4) f is not nilpotent.

5.3.7 P ROPOSITION . Let M be an indecomposable module of finite length,

and let f 1 , f 2 be endomorphisms of M. If f 1 + f 2 is an automorphism, then either
f 1 or f 2 is an automorphism.

5.3.8 L EMMA . Let X 1 , X 2 , Y 1 , Y 2 be left R-modules, and let

f: X 1 ⊕ X 2 → Y 1 ⊕ Y 2
be an isomorphism. Let
i1 : X1 → X1 ⊕ X2
and
i2 : X2 → X1 ⊕ X2
be the natural inclusion maps, and let
p1 : Y1 ⊕ Y2 → Y1
and
Abstract Algebra Notes 75

p2 : Y1 ⊕ Y2 → Y2
be the natural projections. If
p 1º f º i 1 : X 1 -> Y 1
is an isomorphism, then
p 2º f º i 2 : X 2 -> Y 2
is an isomorphism.

5.3.9 T HEOREM . (Krull-Schmidt) Let {X j , j = 1, 2, …, m} and { Y i , i = 1,

2, …, n } be indecomposable left R-modules of finite length. If

X1 ⊕ . . . ⊕ Xm ≅ Y1 ⊕ . . . ⊕ Yn,

then m = n and there exists a permutation π ∈ S n with π (j) = i and X j ≅ Y i , for 1 ≤ j

≤ m.
 Prof. Mohammed El-Atrash 76

5.4 Semisimple Modules

5.4.1 D EFINITION . Let M be a left R-module. The sum of all minimal
submodules of M is called the socle of M, and is denoted by Soc(M). The
module M is called semisimple if it can be expressed as a sum of minimal
submodules.

A semisimple module R M behaves like a vector space in that any submodule

splits off, or equivalently, that any submodule N has a complement N' such
that N + N' = M and N ∩ N' = 0.

5.4.2 T HEOREM . Any submodule of a semisimple module has a

complement that is a direct sum of minimal submodules.

5.4.3 C OROLLARY . The following conditions are equivalent for a module

R M.

(1) M is semisimple;
(2) Soc(M) = M.
(3) M is completely reducible;
(4) M is isomorphic to a direct sum of simple modules.

5.4.4 C OROLLARY . Every vector space over a division ring has a basis.

5.4.5 D EFINITION . The module R Q is said to be injective if for each one-to-

one R-homomorphism i: R M → R N and each R-homomorphism f:M → Q there
exists an R-homomorphism f*:N → Q such that f*i = f.
Abstract Algebra Notes 77

5.4.6 T HEOREM . The following conditions are equivalent for the ring R.
(1) R is a direct sum of finitely many minimal left ideals;
(2) R R is a semisimple module;
(3) every left R-module is semisimple;
(4) every left R-module is projective;
(5) every left R-module is injective;
(6) every left R-module is completely reducible.

5.4.7 C OROLLARY . Let D be a division ring, and let R be the ring M n (D)
of all n × n matrices over D. Then every left R-module is completely reducible.

Let R be a ring, and let G be a group. The group ring RG is defined to be a

free left R-module with the elements of G as a basis. The multiplication on RG
is defined by

( ∑ w ∈ G a w w )( ∑ x ∈ G b x x ) = ∑ z ∈ G c z z where c z = ∑ z=wx a w b x .

The crucial property of a group ring is that it converts group homomorphisms

from G into the group of units of a ring into ring homomorphisms. To be
more precise, let S be a ring, let ϕ :G → S x be a group homomorphism, and let
θ:R → Z(S) be any ring homomorphism. (Recall that S x denotes the group of
invertible elements of S and Z(S) denotes the center of S.) Then there is a
unique ring homomorphism ψ :RG → S such that
ψ(g) = ϕ(g) for all g ∈ G and ψ(r) = θ(r) for all r ∈ R.

5.4.8 T HEOREM . (Maschke) Let G be a finite group and let K be a field

such that |G| is not divisible by chr(K). Then every KG-module is completely
reducible.
Prof. Mohammed El-Atrash 78

5.4.9 T HEOREM . (Baer's criterion) For any left R-module Q, the

following conditions are equivalent.
(1) The module Q is injective;
(2) for each left ideal A of R and each R-homomorphism f :A → Q there exists an
extension f*:R → Q such that f*(a) = f(a) for all a ∈ A;
(3) for each left ideal A of R and each R-homomorphism f :A → Q there exists q ∈ Q
such that f(a) = aq, for all a ∈ A.

5.4.10 P ROPOSITION . Let D be a principal ideal domain, with quotient field

Q.
(a) The module D Q is injective.
(b) Let I be any nonzero ideal of D, and let R be the ring D/I. Then R is an injective
module, when regarded as an R-module.
Abstract Algebra Notes 79

Index
45 · Group algebras 22 · class conjugacy
6 · group cyclic
6 · cyclic subgroup
#
H
70 · ( M ) λ
64 · Hilbert basis theorem
D
9 · homomorphism
62 · DCC A
14 · derived
I 4 · abelian
· descending chain condition
62 · ACC
46 · ideal 62
27 · Development 20 · action
3 · identity element
Differential operator rings 20 · acts
70 · indecomposable
44 · 37 · alternating group
8 · index
57 · annihilator
10 · ( G ) Inn
E 62 · Artinian
10 · automorphism inner
· ascending chain condition
48 · integral domain
44 · Endomorphisms 62
3 · inverse element
12 · Euler 9 · ( G ) Aut
45 · invertible
11 · ( n ) ϕ Euler function 47 ,9 · automorphism
57 · irreducible
37 · even 9 · group automorphism
47 ,9 · isomorphism
33 · external direct product
B
J
F
73 · criterion Baer's
59 · Jacobson radical
16 · Factor group 29 · Burnside
· Jacobson-Perlis Condition
57 · faithful
61
40 · lder ö Jordan H 4 · group finite C
69 · Holder -Jordan 54 · generated finitely
Cancellation Property for
70 · Fitting
3 · Groups
29 · Frattini Argument
K 17 · canonical
54 · module free
36 ,13 · Cayley
Fundamental
21 · kernel 5 · center
Homomorphism Theorem
71 · Schmidt -Krull 51 · for Rings 68 · central
Fundamental Theorem of 5 · centralizer
· Finite Abelian Groups
L 14 · characteristic
35
Chinese Remainder Theorem
8 · Lagrange's 51 ·
62 · Artinian left G 69 · comaximal
7 · coset left 43 · commutative
6 · generate
53 · left factor R-module 14 · commutator
6 · generator
62 · Noetherian left 14 · commutator subgrou
3 · group
52 · left R-module 69 · factors composition
Dr. Mohammed El-Atrash 80

57 ,46 ,25 · simple P 45 · left zero divisor

72 · ( Soc(M
72 · socle 36 ,12 · permutation group
M
41 · solvable 16 · permutes
22 · stabilizer 67 · Pierce Decomposition 73 · Maschke
4 · subgroup 44 · Polynomial Rings 62 · MaxC
53 · submodule 56 · poset 62 · maximal condition
48 · subring 47 · prime ideal 47 · ideal maximal
26 · sylow 49 · principal ideal 56 · ideal left maximal
27 · Sylow Conjugacy 49 · principal ideal domain 56 · submodule maximal
26 · Sylow Existence 50 · principal left ideal 62 · MinC
9 · ( Sym(S 26 · subgroup -p 62 · minimal condition
57 · submodule minimal

T Q
N
· the length of the module 16 · quotient group
70 59 · Nakayama
22 · transitive 17 · natural
R
66 ,30 · nilpotent
U 7 · coset right 62 · Noetherian
46 · right ideal 23 · normalizer
48 · subrings unital 4 · Right identity 31 · normalizers grow
4 · Right inverse
W 45 · right zero divisor
O
43 · ring
67 · Wedderburn ring 47 · ring homomorphism 37 · odd
21 · orbit
Z S 6 ,4 · order

56 · Zorn 57 · Schur
72 · semisimple