Rings

Modern Algebra I is primarily the study of finite groups, and a general

theme is that of symmetry. If there is any theme to Modern Algebra II, it
is most likely that of factorization. Despite the formal similarity between
groups and rings, ring theory has a very different flavor than group theory.
Also, aside from a general level of conceptual sophistication, we shall use
very little of group theory in this course until the very end, when we study
Galois theory.

1 Basic definitions
Definition 1.1. A ring R = (R, +, ·) consists of a set R together with two
binary operations + and · on R such that:
1. The set R together with the binary operation +, i.e. the binary struc-
ture (R, +), is an abelian group.
2. The binary operation · is associative. We usually write rs for r · s.
3. The left and right distributive laws hold: for all r, r, t ∈ R,
(r + s)t = rt + st;
t(r + s) = tr + ts.

As with groups, we shall usually just write R instead of (R, +, ·), with the
operations + and · usually clear from the context. We write 0 for the additive
identity of R and −r for the additive inverse of r.
Before giving some of the very many examples of rings, we record some
easy consequences (mostly without proof) of the axioms for a ring R:

1. For all r ∈ R, 0r = r0 = 0. This follows by the usual argument, that

0r = (0 + 0)r = 0r + 0r,
and by (additive) cancellation, and similarly for r0.

1
 2. For all r, s ∈ R,
(−r)s = r(−s) = −rs,
and hence
(−r)(−s) = rs.
Pn
3. P
The generalized distributive law holds: given two sums i=1 ri and
m
j=1 sj , where the ri , sj ∈ R, then

n
! m 
X X X
ri  sj  = ri sj .
i=1 j=1 i,j

For example,

(r1 + r2 )(s1 + s2 ) = r1 s1 + r1 s2 + r2 s1 + r2 s2 .

4. The “laws of exponents” for the additive group (R, +) say that, for all
n, m ∈ Z and r ∈ R,

(n + m) · r = (n · r) + (m · r), and n · (m · r) = (nm) · r,

where n · r means r added to itself n times, for n > 0. More precisely,

we define n · r inductively by: 1 · r = r, and (n + 1) · r = (n · r) + r.
For n = 0, we set 0 · r = 0, and for n < 0, we set n · r = −((−n) · r).
Note that this · is not the same as multiplication in R. Then there is
an additional property for rings: for all n ∈ Z and r, s ∈ R,

(n · r)s = r(n · s) = n · (rs).

5. For n > 0, n ∈ Z, define rn as the product of r with itself n times. More

precisely, we define rn inductively as follows: r1 = r, and rn+1 = rn r.
Then we have the usual “laws of exponents:”

rn rm = rn+m and (rn )m = rnm .

These can be proved by induction.

We shall usually narrow the class of rings we consider as follows:

Definition 1.2. Let R be a ring.
1. The ring R is commutative if multiplication is commutative, i.e. if, for
all r, s ∈ R, rs = sr.

2
 2. The ring R is a ring with unity if there exists a multiplicative identity
in R, i.e. an element, almost always denoted by 1, such that, for all
r ∈ R, r1 = 1r = r. The usual argument shows that such an element
is unique: if 10 is another, then 1 = 10 1 = 10 . In this case, it is easy to
check that the element n · r defined above is actually equal to (n · 1)r.

3. If R is a ring with unity 1, then a unit of R is an element r ∈ R such

that r has a multiplicative inverse, i.e. there exists an r0 ∈ R such that
rr0 = r0 r = 1. (Unfortunately, it is easy to confuse the terms unity and
unit.) If r has a multiplicative inverse, then the inverse is unique, if it
exists, by the usual argument and using associativity: if r00 is another
such element, then

r0 rr00 = (r0 r)r00 = 1r00 = r00 = r0 (rr00 ) = r0 1 = r0 .

(This also shows as usual that if r has a left and a right multiplicative
inverse then they are equal and r is a unit.) We also say that r is
invertible and denote its unique multiplicative inverse by r−1 . An
argument which should be familiar from Modern Algebra I is that, if
we let
R∗ = {r ∈ R : R is a unit},
then (R∗ , ·) is a group. In particular, the product of two units is a unit
and the inverse of a unit is a unit (with (r−1 )−1 = r.)

4. The ring R is a division ring or skew field if R is a ring with unity

1, 1 6= 0 (this is easily seen to be equivalent to the hypothesis that
R 6= {0}), and R∗ = R − {0}, i.e. every nonzero element of R has a
multiplicative inverse. A field is a commutative division ring.

Let R be a ring. If we try to compute (r + s)2 , we don’t necessarily get

the “expected” answer. However, if R is commutative, then

(r + s)2 = r2 + rs + sr + s2 = r2 + rs + rs + s2 = r2 + 2 · rs + s2 .

More generally, for a commutative ring R and a positive integer n, we have:

Theorem 1.3 (Binomial Theorem). For all r, s ∈ R,

n  
n
X n
(r + s) = · rn−i si .
i
i=0

3
 This can be proved by adapting the usual inductive proof to our more
abstract setting. Here, the end terms are rn s0 and r0 sn , which we set
equal to rn and sn respectively. For a ring R with unity, not necessarily
commutative, we define r0 = 1 for all r ∈ R, although the binomial theorem
holds even if R does not have unity.

2 Examples
Rings are ubiquitous in mathematics. We list some important examples.

1. There are the familiar examples of numbers: Z, Q, R, C. These are

all commutative rings with unity. Here, Q, R, and C are fields, but
(Z)∗ = {±1}. A related example is nZ = hni, the cyclic subgroup
of Z generated by n. It is an additive group, and multiplication is a
well-defined binary operation since

(nk1 )(nk2 ) = n2 (k1 k2 ) = n(nk1 k2 ) ∈ nZ.

Note that, if n > 1, nZ is not a ring with unity.

2. As we saw in Modern Algebra I, Z/nZ is a finite commutative ring

with unity for all positive integers n. In the case, the group of units
is the multiplicative group (Z/nZ)∗ . The ring Z/nZ is a field ⇐⇒
n = p is a prime number. In this case, we will usually use the notation
Fp for the ring Z/pZ, thought of as a field. These are the first cases of
finite fields. Later in the course, we shall describe all finite fields.

3. Let Mn (R) be the set of n × n matrices with entries in R. Then we

can both add and multiply elements of Mn (R), and the left and right
distributive laws hold: for all A, B, C ∈ Mn (R),

(A + B)C = AC + BC and C(A + B) = CA + BA.

Thus Mn (R) is a ring, and it is not commutative as soon as n ≥ 2.

It does have unity, the identity matrix I. The units in Mn (R) form
the group GLn (R) of invertible n × n matrices with entries in R. We
can consider matrices with entries in other rings as well, for example
Mn (C), Mn (Q), or even Mn (Z). In fact, in order to be able to define
matrix multiplication, we just need to be able to add and multiply the
entries, and this will satisfy the usual properties (e.g. matrix multipli-
cation is associative and distributes over matrix addition) as long as

4
 addition and multiplication have these properties. So, for every ring
R, the set Mn (R) of n × n matrices with coefficients in R is a ring
under matrix addition and multiplication. If R is a ring with unity,
then so is Mn (R), where the unity is the n × n identity matrix. For
example, for every positive integer k, Mn (Z/kZ) is a finite ring (of
2
order k n ), and it is not commutative if n > 1 and k 6= 1.

4. There are trivial examples of rings. For example, the zero ring R is the
ring {0}, with the unique binary operations (0+0 = 0, 0·0 = 0). More
generally, if A is an abelian group, then we can define a multiplication
on A by the rule that a · b = 0 for all a, b ∈ A. Then it is easy to check
that A is a (commutative) ring with this definition of multiplication,
but it is not a ring with unity unless A = {0}.

5. Rings of functions arise in many areas of mathematics. For exam-

ple, the set RR of all real-valued functions f : R → R is a ring under
pointwise addition and multiplication: given two functions f and g,
we define the “pointwise sum” f + g and the “pointwise product” f g
by:
(f + g)(x) = f (x) + g(x); (f g)(x) = f (x)g(x).
Clearly, if X is any set and R is a ring, then the set RX of all functions
from X to R becomes a ring under pointwise addition and multiplica-
tion as well.
Often, we don’t look at all functions from, say, R to itself but at inter-
esting subsets. For example, let C 0 (R) be the set of all continuous
functions from R to R. Then C 0 (R) is a ring via pointwise addition
and multiplication. This fact doesn’t follow from pure thought: the
content of this statement is that the sum of two continuous functions is
continuous and the product of two continuous functions is continuous.
Similarly, the set C ∞ (R) of all functions from R to R with derivatives
of all orders is a ring via pointwise addition and multiplication.
Note that, if R is a commutative ring, then RX is commutative: the
pointwise product f g is equal to gf , since, for all x ∈ X, (f g)(x) =
f (x)g(x) = g(x)f (x) = (gf )(x). Also, if R is a ring with unity, then
so is RX : the constant function 1, i.e. the unique function from X
to R whose value at every x ∈ X is 1, is a unity under pointwise
multiplication.

6. Given two rings R1 and R2 , the Cartesian product R1 ×R2 is a ring un-
der componentwise addition and multiplication: given (r1 , r2 ), (s1 , s2 ) ∈

5
 R1 × R2 , we define

(r1 , r2 ) + (s1 , s2 ) = (r1 + s1 , r2 + s2 );

(r1 , r2 ) · (s1 , s2 ) = (r1 s1 , r2 s2 ).

Then R1 × R2 is commutative if R1 and R2 are commutative, and it

is a ring with unity if R1 and R2 both have unity; in fact, the unity in
R1 × R2 is then necessarily (1, 1).

7. There are many interesting rings which are subsets of C defined by

special numbers. For example, define the Gaussian integers Z[i] by

Z[i] = {a + bi : a, b ∈ Z}.

Addition and multiplication are given by the usual addition and mul-
tiplication of complex numbers, and multiplication defines a binary
operation on Z[i] because Z[i] is closed under multiplication:

(a + bi)(c + di) = (ac − bd) + (ad + bc)i.

The ring Z[i] is a commutative ring with unity, but is not a field. In
fact (Z[i])∗ = {±1, ±i} and hence is a cyclic group of order 4. There
is no reason to just look at integer coefficient a, b. We could also take
a, b to be rational. In this case, for reasons we will explain later, we
use the notation Q(i) instead:

Q(i) = {a + bi : a, b ∈ Q}.

In this case, many of you have probably seen in high school that Q(i)
is a field. In fact, given a+bi ∈ Q(i), where not both a, b are 0, we find
a multiplicative inverse for a + bi by “rationalizing the denominator:”
1 1 a − bi 1 a b
= · = 2 2
· (a − bi) = 2 2
− 2 i.
a + bi a + bi a − bi a +b a +b a + b2
√
A similar construction works with 2: define
√ √
Z[ 2] = {a + b 2 : a, b ∈ Z}.
√
As before, Z[ 2] is closed under multiplication:
√ √ √
(a + b 2)(c + d 2) = (ac + 2bd) + (ad + bc) 2.

6
The ring is a commutative
√ ring with unity, but is still not a field. For
example, 2, 3, 2 have no multiplicative inverses in this ring. However,
there are lots of units! For example, since
√ √
(1 + 2)(1 − 2) = 1 − 2 = −1,
√ √ √ √
we see that (1√+ 2)−1 = −(1 − 2) = −1 √ + 2 ∈ Z[ 2]. It is easy to
see that
√ 1 + 2 has infinite order in Z[ 2]: since √ absolute value
the
|1 +√ 2| is greater than 1, no positive power of 1 + 2 can be 1. Thus
(Z[ 2])∗ contains the infinite cyclic subgroup
√ √
h1 + 2i = {(1 + 2)n : n ∈ Z}.

As before we can consider Q coefficients and define

√ √
Q( 2) = {a + b 2 : a, b ∈ Q}.

This is again a field by rationalizing denominators:

√
1 1 a−b 2 1 √ a b √
√ = √ · √ = 2 2
·(a−b 2) = 2 2
− 2 2
2.
a+b 2 a+b 2 a−b 2 a − 2b a − 2b a − 2b

Note that the denominator a2 − 2b2 is nonzero as long as at √least

one of a, b are nonzero; this is equivalent to the statement that 2 is
irrational.
√
For a final example of this type, consider instead
√ Z[ 3 2]. We can’t
3
just take complex numbers of the form a + b√ 2, because (as we will
see more carefully much later) the number ( 3
2)2 is not of the form
√
3
a + b 2. Thus we need three coefficients, and define
√ √ √
Z[ 2] = {a + b 2 + c( 2)2 : a, b, c ∈ Z}.
3 3 3

With this definition,√it should be intuitively clear (if a little tedious

to write out) that Z[ 3 2] is closed under multiplication and hence is a
ring. For example,
√ √ √ √
( 2)2 ( 2)2 = ( 2)4 = 2 2.
3 3 3 3

√
Likewise we define Q( 3 2) via
√ √ √
Q( 2) = {a + b 2 + c( 2)2 : a, b, c ∈ Q}.
3 3 3

7
 √
In this case, it is still true that Q( 3 2) is a field, but it is far from
obvious: given a, b, c ∈ Q, not all 0, we have to find a way to rationalize
the denominator of the expression
1
√ √ .
a + b 2 + c( 3 2)2
3

We will describe several different ways to think about this over the
course of the semester.

8. The quaternions are an interesting example of a division ring which

is not a field. Let H = R4 , with basis 1, i, j, k. Thus an element of
H is uniquely written as α = x0 + x1 i + x2 j + x3 k. We view R as a
subset of H by identifying t with t · 1. Two quaternions are multiplied
by the following rules: if t = t1, then it = ti and similarly for j, k.
Otherwise, i2 = j 2 = k 2 = −1, and ij = k = −ji, jk = i = −kj, and
ki = j = −ik. Then we expand out by the usual rules for distributivity.
It is a little painful to check that multiplication is associative and left
and right distributes over addition, but one can identify H with a
subset of M2 (C), or M4 (R), with the inherited operations of addition
and matrix multiplication, and use this to prove associativity and left
and right distributivity. Then (homework) H is a division ring, but it
is clearly not commutative. It is interesting for many reasons, but one
of the most important ones is the following fact: if R is a division ring
containing the real numbers R, and R is finite dimensional over R in
a natural way, then R is essentially R, C, or H.

9. A fundamental class of rings are polynomial rings. We will use the

notation R[x] to denote the set of all polynomials with real coefficients:
(N )
X
R[x] = ai xi : ai ∈ R .
i=0

Here, however, we write 1x = x, and we agree that adding terms of the

form 0xk does not affect the polynomial. The sets C[x], Q[x], Z[x] of
polynomials with coefficients in C, Q, Z respectively are defined sim-
ilarly. More generally, let R be a commutative ring with unity. Then
a polynomial with coefficients in R is an expression f (x) of the form
P N i
i=0 ai x with ai ∈ R. We define R[x] to be the set of all polynomials
with coefficient in R. To avoid the issue with terms of the form 0xk ,
we can simply identify the polynomial f (x) with the infinite sequence

8
 of coefficients (a0 , a1 , . . . ), where ai ∈ R and ai = 0 for all i > N . The
largest d (possibly 0) such that ad 6= 0 is called the degree of f (x) and
written deg f (x). Here, the degree of a “constant polynomial” a0 is 0,
but the degree of the zero polynomial 0 is undefined (some people
define it to be −∞).
Addition of polynomials is defined by adding coefficients, so that
X X X
ai xi + bi xi = (ai + bi )xi .
i i i

Multiplication of polynomials is given by the formula

! !
X X X
ai xi bi xi = ci xi ,
i i i
P
where ck = i+j=k ai bj . It is the formula forced on us by requiring
that xi a = axi , xi xj = xi+j , and distributivity. With these definitions,
a somewhat tedious argument (which we shall discuss in more detail
shortly) shows that R[x] is a commutative ring with unity.
In precalculus and calculus, polynomials with real coefficients define
functions (polynomial functions) from R to R and we can identify the
polynomial with the function that it defines. For a general ring, it
turns out that this is not quite the case. We will discuss later the
many ways in which polynomials can be used to define functions, but
for the moment we can just think of them as formal objects.

3 A few more general definitions

Many of the basic definitions of group theory have straightforward general-
izations to rings.

Definition 3.1. Let R1 and R2 be two rings. A homomorphism f : R1 →

R2 is a function (not necessarily injective or surjective) such that, for all
r, s ∈ R1 ,

f (r + s) = f (r) + f (s);
f (rs) = f (r)f (s).

An isomorphism f : R1 → R2 is a homomorphism which is a bijection. If

R1 = R2 , then an isomorphism from R1 to itself is called an automorphism

9
of R1 . For example, the identity is an automorphism. Two rings R1 and R2
are isomorphic if there exists an isomorphism f : R1 → R2 . In this course,
unless otherwise specified, a homomorphism is always understood
to mean a ring homomorphism and similarly for isomorphisms.

Example 3.2. If R1 and R2 are any rings and we define f (r) = 0 for all r,
then f is a homomorphism under the above definition. However, we shall
shortly modify the definition so that such an f is not allowed.
For a more interesting example, the natural projection homomorphism
π : Z → Z/nZ is a homomorphism. For R = C, complex conjugation is an
automorphism: if we set f (z) = z̄, then the well-known identities z + w =
z̄ + w̄, zw = z̄ w̄ imply that f is an automorphism (it is a bijection with
inverse equal to f ). Conjugation also defines automorphisms
√ Z[i]√→ Z[i]
and Q(i) → Q(i). A similar construction works for Z[ 2] and Q( 2), by
defining √ √
f (a + b 2) = a − b 2.
(However,
√ aside from
√ the zero homomorphism, the only other automorphism
3 3
of Z[ 2] or of Q( 2) is the identity.)

Definition 3.3. Let S be a ring. A subring of S is a subset R such that

1. R is an (additive) subgroup of S, i.e. a subgroup of (S, +).

2. R is closed under multiplication.

In this case, R is a ring as well, with the inherited operations (since mul-
tiplication becomes a well-defined binary operation which is automatically
associative and left and right distributes over addition). We write R ≤ S is
R is a subring of S.

As has already been implicit in our list of examples of rings, many rings
arise as subrings of other rings. Here are some examples:
√
1. Z ≤
√ Q ≤ R ≤ C. But also Z ≤ Z[i] ≤ Q(i) ≤ C, and Z ≤ Z[ 2] ≤
Q( 2) ≤ R.

2. nZ ≤ Z. (We will modify the definition shortly to disallow this exam-

ple.)

3. The trivial ring {0} is a subring of every ring R. (As with homomor-
phisms, we will shortly disallow this example.)

10
 4. Given two rings R1 , R2 , the subsets R1 ×{0} and {0}×R2 are subrings
of R1 × R2 .
5. R ≤ H.
6. If R is a commutative ring with unity, then R ≤ R[x] in the obvious
way, as the subset of “constant polynomials,” i.e. the set of polynomials
of degree 0 union {0}.
Basic Conventions: (i) If R1 and R2 are two rings with unity, and f : R1 →
R2 is a homomorphism, we also require that f (1) = 1. Thus, with this
convention, the zero homomorphism is no longer allowed to be a homo-
morphism from one ring with unity to another. Note that, in this case,
if r is a unit of R, so that there exists an r−1 ∈ R with rr−1 = 1, then
f (1) = 1 = f (rr−1 ) = f (r)f (r−1 ). Thus, if r is a unit, then so is f (r), and
in fact (f (r))−1 = f (r−1 ).
(ii) If S is a ring with unity 1 and R is a subring of S, then we also require
that 1 ∈ R. Thus, R is also a ring with unity and it has the same unity as
S. For example, {0} is no longer allowed to be a subring of a ring S with
unity unless S = {0} also. Likewise, nZ is not allowed to be a subring of
Z for n > 1. For another, slightly more complicated example, if R1 and R2
are two nonzero rings with unity, then the “subring” R1 × {0} is not, in this
new sense, a subring of R1 × R2 : it has the unity (1, 0), which is not the
same as the unity (1, 1) of R1 × R2 . However, most of the examples in our
list of subrings still have the property that they contain the unity, and so
are subrings in this new sense.
One simple remark that we shall use frequently is the following:
Proposition 3.4. Ler R and S be rings and let f : R → S be a homomor-
phism. Then Im R = f (R) is a subring of S.
Proof. From Modern Algebra I, we know that f (R) is a subgroup of (S, +).
It is closed under multiplication since, if f (r1 ), f (r2 ) ∈ f (R), then
f (r1 )f (r2 ) = f (r1 (r2 ) ∈ f (R).
Finally, if both of R, S are rings with unity, then the assumption that
f (1) = 1 implies that 1 ∈ f (R).

For the rest of this course, unless otherwise stated, all rings
R will be commutative rings with unity 1, all subrings of R have
to contain the unity 1, and all homomorphisms f : R → S satisfy
f (1) = 1.

11