Lecture 1, March 15, 2011

1 Equivalence Relations
Let Ω1 and Ω2 be any two sets. We then define the “Cartesian product”
of Ω1 with Ω2 , (denoted by Ω1 × Ω2 ), as

Ω1 × Ω2 = {(ω1 , ω2 ) : ω1 ∈ Ω1 , ω2 ∈ Ω2 } (1.1)

Any subset of Ω1 × Ω2 is called a “Relation” from Ω1 to Ω2 . In particular, if

we take Ω1 = Ω2 , say = Ω, we then get the notion of the Cartesian product
of a set with itself,

Ω × Ω = {(ω1 , ω2 ) : ω1 , ω2 ∈ Ω} (1.2)

Then any subset R of Ω × Ω is called a relation on Ω. We now look at some

special types of relations. If (x, y) ∈ R we say that x is R related to y and
write xRy
Definition 1.1 A relation R on Ω is said to be,
1. “Reflexive” if xRx for every x ∈ Ω,

2. “Symmetric” if x, y ∈ Ω, xRy =⇒ yRx,

3. “Transitive” if x, y, z ∈ Ω, xRy and yRz =⇒ xRz,

4. “Antisymmetric” if x, y ∈ Ω, xRy and yRx =⇒ x = y

We next have the notions of two important types of relations which possess
combiations of the above properties. We have
Definition 1.2 We have
1. A relation R on Ω is said to be an “Equivalence relation on Ω” if
it is Reflexive, Symmetric and Transitive. (This means that Properties
(1) to (3) above hold).

1
 2. A realtion R on Ω is said to be a “Partial order on Ω” if it is
Reflexive, Antisymmetric and Transitive. (This means that Properties
(1), (4) and (3) above hold).

Remark 1.1 An Equivalence relation is generally denoted by ∼ and a Parial

order is usually denoted by 

We shall first study equivalence relations.

2 Equivalence Realtions and Partitions

We first begin with the notion of a partition of a set. We have

Definition 2.1 Consider any nonempty set Ω. A collection Π of subsets of

Ω is said to be a Partition of Ω if

1. P ∈ Π =⇒ P is nonempty,

2. P1 , P2 ∈ Π =⇒ P1
T
P2 = φ,

[
3. P =Ω
P ∈Π

(This means that the collection Π is a collection of nonempty disjoint subsets

of Ω whose union is Ω)

We shall now see that every equivalence relation gives rise to a partition of
Ω. Consider any equivalence relation ∼ on Ω. For any x ∈ Ω we define the
“equivalence class generated by x”, (which we denote by [x]), as
def
[x] = {y ∈ Ω : x ∼ y} (2.1)

We now observe the following properties of equivalence classes:

1. By the reflexivity property of ∼ we have x ∼ x and hence we have

x ∈ [x] for every x ∈ Ω (2.2)

2
 2. x, y ∈ Ω and y ∈ [x] =⇒ x ∼ y
Hence we have
z ∈ [y] =⇒ y ∼ z
=⇒ x ∼ z (by transitivity of ∼ since we have x ∼ y and y ∼ z)
=⇒ z ∈ [y]
Thus we have

y ∈ [x] =⇒ [y] ⊆ [x] (2.3)

Now clearly y ∈ [x] means x ∼ y and hence by symmetry of ∼ we have

y ∼ x and hence x ∈ [y]. Hence by above argument we get

[x] ⊆ [y] (2.4)

Thus we get

y ∈ [x] =⇒ [y] = [x] (2.5)

3. Suppose x, y ∈ Ω and we consider the corresponding equivalence classes

[x] and [y]. Now

\
[x] [y] 6= φ
\
=⇒ ∃z ∈ [x] [y]
=⇒ [x] = [z] = [y] (by (2.5) above)

Thus
\
[x] [y] 6= φ =⇒ [x] = [y]

Hence we get
\
x, y ∈ Ω =⇒ [x] = [y] OR [x] [y] = φ (2.6)

From the above properties we see that the distinct equivalence classes give
rise to a partition of Ω. We denote the partition of Ω generated by an equiv-
alence relation by Ω/ ∼.
Conversely if Π is any partition of Ω then we define a relation R on Ω as

3
follows:

xRy ⇐⇒ ∃P ∈ Π 3 x and y ∈ P (2.7)

(that is xRy if and only if both belong to the same part P of the partition
Π). It is easy to see that this relation R is indeed an equivalence relation
∼ on Ω an that the partition Ω/ ∼ generated by this equivalence relation is
precisely Π. Thus
Every equivalence relation gives rise to a partition and every par-
tition gives rise to an equivalence relation. We shall now look at some
simple examples.

Example 2.1 Let Ω = N, the set of natural nimbers. Define a relation R

on N as follows:

For x, y ∈ N, xRy ⇐⇒ x − y is a multiple of 5 (2.8)

Then it is easy to see that this is an equivalence relation on N. This is called

congruence modulo 5. If xRy we write

x ≡ ymod(5)

The distinct equivalence classes corresponding to this equivalence relation

are given by

[1] = {1, 6, 11, 16, · · ·} = {n ∈ N : n = 5k + 1, k = 0, 1, 2 · · ·}

[2] = {2, 7, 12, 17, · · ·} = {n ∈ N : n = 5k + 2, k = 0, 1, 2 · · ·}

[3] = {3, 8, 13, 18, · · ·} = {n ∈ N : n = 5k + 3, k = 0, 1, 2 · · ·}

[4] = {4, 9, 14, 19, · · ·} = {n ∈ N : n = 5k + 4, k = 0, 1, 2, · · ·}

[5] = {5, 10, 15, 20 · · ·} = {n ∈ N : n = 5k, k = 1, 2, 3, 4, 5 · · ·}

It is easy to see that these distinct equivalence classes partition the set N

4
Example 2.2 Now consider the set N2 and the following relation R on N2 .

For (a, b), (c, d) ∈ N2 , we say (a, b)R(c, d) ⇐⇒ a + d = b + c (2.9)

It is easy to verify that this is an equivalence relation on N2 . Consider all the

points in the plane on the line making an angle of 45◦ with the horizontal and
passing through the point (1, 1). All points on this line whose coordinates
are positive integers form an equivalence class. Similarly every line parallel
to this defines an equivalence class with the points having positive integer
coordinates. The collecion of all such equivalence classes defined by these
parallel lines give us the collection of all distinct equivalence classes.

Figure 1
We can list the distinct equivalence classes as follows:

E = {(n, n) : n ∈ N} 

E (k) = {(n, n + k) : n ∈ N} , k = 1, 2, 3, · · ·  (2.10)
E(k) = {(n + k, n) : n ∈ N} , k = 1, 2, 3, · · · 

Example 2.3 Analogous to the Example 2 we have the following equivaence

relation on N2 :

5
 For (a, b), (c, d) ∈ N2 , we say (a, b)R(c, d) ⇐⇒ a × d = b × c (2.11)

It is easy to check that this is an equivalence relation on N2 . The distinct

equivalence classes in this case can be listed as follows:

E = {(n, n) : n ∈ N} 

(k)
E = {(n, kn) : n ∈ N} , k = 1, 2, 3, · · · (2.12)
E(k) = {(kn, n) : n ∈ N} , k = 1, 2, 3, · · ·



We shall be looking at the above two equivalence relations on N2 later.

3 Equivalence of Sets
We now look at the notion of equivalence of sets. Let A and B be any two
sets. Then a map
f : A −→ B
is said to be,

1. “one-one” (also known as “injective”) if

a1 , a2 ∈ A, and a1 6= a2 =⇒ f (a1 ) 6= f (a2 ) (3.1)

(This means that distinct elements in A get mapped under f to distinct

elements in B)

2. “onto” (also known as “surjective”) if

∀ b ∈ B ∃ a ∈ A 3 f (a) = b (3.2)

(This means that every element of B is the image of some element in A)

3. “isomorphism” (also known as “bijective”) if it is both one-one and

onto

These lead us to the notion of equivalence of sets. We have

6
Definition 3.1 A set A is said to be equivalent to a set B (and we then
write A ∼ B) if there exists an isomorphism

f : A −→ B

If A and B are any two sets then there arise, at the outset, the following four
possibilities:

1. There exists an isomorphism from A to a subset B1 of B, but there

does not exist any isomorphism from B to any subset of A,
(This means that there exists a one-one map from A to B and there
does not exist any one-one map from B to A)

2. There exists an isomorphism from B to a subset A1 of A, but there

does not exist any isomorphism from A to any subset of B,
(This means that there exists a one-one map from B to A and there
does not exist any one-one map from A to B)

3. There exists an isomorphism from A to a subset B1 of B and there

exists any isomorphism from B to a subset A1 of A,
(This means that there exists a one-one map from A to B and a one-
one map from B to A)

4. There does not exist an isomorphism from A to any subset of B and

there does not exist any isomorphism from B to any subset of A,
(This means that there does not exist a one-one map either from A to
B or from B to A )

It can be shown that the fourth possibility can never occur. Further we have

Theorem 3.1 Schroeder-Bernstein:

If A is equivalent to a subset B1 of B and B is equivalent to a subset A1 of
A then A is equivalent to B.
(This is the same as saying that the possibility (3) above implies there exists
an isomorphism from A to B)