Naı̈ve Set Theory and Logic–A Review

EC-424
August 3, 2016

1 Basic Notation
We shall assume a naı̈ve point of view with regards to sets, that is to say what
is meant by a set is intuitively clear. The capital letters A, B, . . . shall be used
to denote sets, and the lower-case letters a, b, . . . shall be used to denote the
objects or elements belonging to these sets. If an object a belongs to the set
A, we shall express this fact by the notation
a∈A
If a doesn’t belong to A we express this fact as
a∈
/A
The equality symbol = is used throughout to mean logical identity. Thus, when
a = b is written, we mean that “a”, and “b” are symbols for the same object.
Similarly, the equation A = B means “A” and “B” are symbols for the same
set; that is, A and B consist of (own) precisely the same objects. If a and b
are different objects we write a 6= b; and if A and B are different sets we write
A 6= B. We say that A is a subset of B if every element of A is also an element
of B; and this is written as
A⊆B
Notice that if A ⊆ B and B ⊆ A, then we say that A = B. If however, A is
different from B, we say that A is a proper subset of B, and we write
A⊂B
The relations ⊆ and ⊂ are called inclusion and proper inclusion, respectively.
If the set has only few elements, then it can be listed simply as
A = {a, b, c}
where the braces are used to enclose the list of elements. However, the usual
way to specify a set is to take some set A of objects and some property that
the elements of A may or may not possess, and to form the set consisting of
all elements of A having that property. For instance, one can choose the set of
integers and form the subset X consisting of all even integers. In symbols, this
would be written as

1
 X = {k | k is an even integer}
Here, the braces stand for the words “the set of”, and the vertical bar stands
for the words“such that”. So the equation above is read as “X is the set of all
k such that k is an even integer.”

2 Some Basic Operations on Sets

Definition 2.1. Given, any two sets A and B, we can form a set from them
that consists of all elements of A and all the elements of B. This set is called
union of A and B and is denoted by, A ∪ B. Formally, we define
A ∪ B = {x | x ∈ A or x ∈ B}
In ordinary English, the word “or” is ambiguous, as sometimes the statement
“P or Q” means “P or Q, or both” and sometimes it means “P or Q, but not
both”. Throughout we shall use the word “or” in the first sense. So now the
definition of A∪B becomes unambiguous, it states that A∪B is the set consisting
of all elements x that belong to A or B or to both.
Definition 2.2. Given any two sets A and B, we can form another set by taking
only the elements common to both A and B. This set is called intersection of
A and B and is denoted by, A ∩ B. Formally, we define

A ∩ B = {x | x ∈ A and x ∈ B}
In this definition, a difficulty of another sort arises. It arises when the sets A
and B have no elements in common. In order to take care of such an eventuality,
we introduce a special set that we will call the empty set, denoted by ∅, which
we think of as “the set having no elements”. Using this convention, we shall
express the statement that A and B have no elements in common by saying
that A and B are disjoint and write it as the equation
A∩B =∅
Because, ∅ is thought of as “the set with no elements”, we make the convention
that for every object x, the relation x ∈ ∅ doesn’t hold. Similarly, the definitions
of union and intersection show that for every set A we should have the equations
A ∪ ∅ = A and A ∩ ∅ = ∅
Definition 2.3. Another operation on sets that would be occasionally useful is
the difference of two sets, denoted by A − B, and defined as the set consisting
of those elements of A that are not in B. Formally,
A r B = {x | x ∈ A and x ∈
/ B}
It is sometimes called the complement of B relative to A, or the complement
of B in A.

2
Definition 2.4. If one collects all objects that are of interest to us in a given
context, then such a collection is called a universal set, or a universe of dis-
course, often written as U. Thus every set in this context would be a subset of
this U.
Definition 2.5. Consider some A ⊆ U. Then all members of U not in A form
0
a complement of A in U, denoted by A , or Ac . Formally,
Ac = {x | x ∈ U and x ∈
/ A}.
Operations on sets satisfy the following laws:
1. Commutative laws: (A ∪ B) = (B ∪ A); & (A ∩ B) = (B ∩ A).
2. Associative laws: (A ∪ B) ∪ C = A ∪ (B ∪ C); & (A ∩ B) ∩ C = A ∩ (B ∩ C).
3. Distributive laws: (A ∪ B) ∩ C = (A ∩ C) ∪ (B ∩ C); & (A ∩ B) ∪ C =
(A ∪ C) ∩ (B ∪ C).
4. (Ac )c = A.
5. A ∩ U = A; A ∪ U = U; A ∩ ∅ = ∅; A ∪ ∅ = A; U c = ∅; & ∅c = U.
6. DeMorgan’s laws: (A ∩ B)c = (Ac ∪ B c ); & (A ∪ B)c = (Ac ∩ B c ).
Definition 2.6. There is yet another way of forming new sets from old ones; it
involves the notion of an “ordered pair” of objects. Given any two sets A and
B, we define their Cartesian product A × B as the set of all ordered pairs
(a, b) for which a is an element of A and b is an element of B. Formally,

A × B = {(a, b) | a ∈ A and b ∈ B}.

Definition 2.7. Let A and B be some non-empty sets. Then a rule that
associates members of a set A with one or more members of set B is called a
mapping or a relation and is written as: r : A 7→ B.
Definition 2.8. A function f is a mapping, f : X 7→ Y such that the following
conditions are satisfied:
(i) ∀x ∈ X, ∃y ∈ Y, such that (x, y) ∈ f .
(ii) If (x, y) ∈ f and (x, y 0 ) ∈ f , then [y = y 0 ].
y is called the image of x, and x is called the pre-image or the inverse
image of y. X is called the domain of the function f , and a set S ⊆ Y , where
every element of the set S is image of some element of X, is called the range
of the function f . A function f where X ⊆ R and Y ⊆ R is called a function
of one variable. A function f where X ⊆ Rn and Y ⊆ R is called a function of
several variables, where n ≥ 2.

3
 A function f , where X ⊆ Rm and Y ⊆ Rn is called a vector-valued function,
if n ≥ 2. It is a vector valued function of a scalar, if m = 1 and of a vector,
if m ≥ 2. Any vector valued function is nothing but a collection of several
functions (order of functions held constant).
Definition 2.9. If f : X 7→ Y and if X0 ⊆ X, we define the restriction of f
to X0 to be the function mapping X0 into Y whose rule is

{(x, f (x)) | x ∈ X0 }

It is denoted by f |X0 , which is read “f restricted to X0 .”

Definition 2.10. Given functions f : X 7→ Y and g : Y 7→ Z, we define the
composite g ◦ f of f and g as the function g ◦ f : X 7→ Z defined by the
equation (g ◦ f )(x) = g(f (x)). Formally, g ◦ f : X 7→ Z is the function whose
rule is
{(x, z) |x ∈ X, and ∃y ∈ Y, f (x) = y and g(y) = z}.
Note that g ◦ f is defined only when the range of f is included in the domain
of g.
Definition 2.11. A function f : X 7→ Y is said to be injective (or one-to-
one) if for each pair of distinct points of X, their images under f are distinct.
It is said to be surjective (or f is said to map X onto Y ) if every element
of Y is the image of some element of X under the function f . If f is both
injective and surjective, it is said to be bijective (or is called a one-to-one
correspondence).
More formally, f is injective if
[f (x) = f (x0 )] =⇒ [x = x0 ]
and surjective if
[y ∈ Y ] =⇒ [y = f (x), ∃x ∈ X].
Injectivity of f depends only on the rule of f ; surjectivity depends on the
range of f as well.

3 Collection of Sets
The objects belonging to a set may themselves be sets. Given a set A, one can
consider sets whose elements are subsets of A. In particular, one can consider
the set of all subsets of A. This set is sometimes denoted by the symbol P(A)
[also written as 2A ] and is called the power set of A.
When we have a set whose elements are sets, we shall refer to it as a collection
of sets and denote it by a script letter such as A or B.
Given a collection A of sets, the union of the elements of A is defined by the
equation

4
 [
A = {x | x ∈ A for at least one A ∈ A}.
A∈A

The intersection of the elements of A is defined by the equation

\
A = {x | x ∈ A for every A ∈ A}.
A∈A

4 Sequences of Numbers
Let X be any set of numbers, which is non-null, that is, let X ⊆ R. A sequence
is formed by writing numbers drawn from a set X in a particular order. It
is permissible to repeat a number in a sequence at a later stage, even if it
has occurred earlier in that sequence. Thus, {x1 , x2 , x3 , . . .} is a sequence of
numbers, where x1 ∈ X, x2 ∈ X, x3 ∈ X, and so on.

A sequence may have only finite numbers in it, then it is called a finite se-
quence. We will generally be interested in infinite sequences. An infinite se-
quence of numbers, {x1 , x2 , x3 , . . .} is represented as {xn }.
Definition 4.1. Let {x1 , x2 , x3 , . . .} be a sequence of real numbers and let r
be a real number. We say that r is the limit of this sequence if for any (small)
positive number , there is a positive integer N such that for all n ≥ N , xn is
in the -interval about r; that is,
|xn − r| < .
In this case, we say that the sequence converges to r and we write
lim xn = r or limn→∞ xn = r or simply xn → r.
This definition states that xn converges to r if no matter how small an
interval one chooses about r, from some point on (n ≥ N in the above definition)
the entries of the sequence get into and stay in that interval. Of course, how
soon the sequence gets into the interval will, in general, depend on the size of
the interval; in other words, N depends on the size of .
Remark. A sequence can have at most one limit.
To define the concept of a subsequence carefully, think of a sequence as the
image of a function F from the natural numbers to the real numbers, where we
write F (n) as xn . Now, let M be any infinite subset of the natural numbers.
Write M as {n1 , n2 , n3 , . . .}, where
n1 < n2 < n3 . . ..
Create a new sequence {yn }, where

5
 yj = xnj , for j = 1, 2, 3, . . .
This new sequence {yj }∞
j=1 is called a subsequence of the original sequence
{xn }.
Definition 4.2. A sequence is called bounded if there exists a lower bound
{x} and an upper bound {x} of {xn } for the sequence such that x ≤ xk ≤ x,
for all k = 1, 2, 3, . . .
Definition 4.3. A sequence {xn } is monotone increasing sequence, if xk+1 ≥
xk for k = 1, 2, 3, . . . and is a monotone decreasing sequence, if xk+1 ≤ xk
for k = 1, 2, 3, . . ., such a sequence is called a monotone sequence.
Remark. (a) If a sequence {xn } is both bounded and monotone, then a sequence
is a convergent sequence.
(b) Let {xn }∞ ∞
n=1 and {yn }n=1 be sequences with limits x and y, respectively.
Then the sequence {xn + yn }∞
n=1 converges to the limit x + y.

(c) Let {xn }∞ ∞

n=1 and {yn }n=1 be sequences with limits x and y, respectively.
Then the sequence {xn − yn }∞
n=1 converges to the limit x − y.

(d) Let {xn }∞ n=1 be a sequence with limit x. Then the sequence of products
{l · xn }∞
n=1 converges to the limit l · x, where l is any real number.

(e) Let {xn }∞ ∞

n=1 and {yn }n=1 be sequences with limits x and y, respectively.
Then the sequence of products {xn yn }∞
n=1 converges to the limit xy.

(f) Let {xn }∞ ∞

n=1 and {yn }n=1 be sequences with limits x and y 6= 0, respectively.
Then the sequence of division { xynn }∞ x
n=1 converges to the limit y .

(g) Let {xn }∞

n=1 be a convergent sequence with limit x, and let b be a number
such that xn ≤ b ∀n. Then, x ≤ b. If xn ≥ b ∀n, then x ≥ b.

5 Sequence of Points
A sequence in Rm is an assignment of a vector in Rm to each natural number
n : {x1 , x2 , x3 , . . .}. For such sequences we need to keep track of two different
indices: one for the m coordinates of each m-vector, and the other to indicate
which element in the sequence is under consideration.
2 1/2
Pn

Definition 5.1. Let a ∈ Rn . Then kak = i=1 ai

Definition 5.2. Let r be a vector in Rm and let  be a positive number. The

-ball about r is
B (r) ≡ {x ∈ Rm : kx − rk < }
Intuitively, a vector x in Rm is close to r if x is in some B (r) for a small but
positive . The smaller  is, the closer x is to r.

6
Definition 5.3. A sequence of vectors {x1 , x2 , x3 , . . .}, is said to converge to
the vector x if for any choice of a positive real number , there is an integer N
such that, for all n ≥ N , xn ∈ B (x): that is,
kxn − xk < .
The vector x is called the limit of the sequence.
In other words, a sequence of vectors {xn }∞
n=1 converges to a limit vector x
iff the sequence of distances from the vector xn to x, {kxn − xk}∞n=1 , converges
to 0 in R1
Remark. (a) A sequence of vectors in Rm converges iff all m sequences of its
components converge in R1 .
(b) Let {xn }∞ ∞ n
n=1 and {yn }n=1 be convergent sequences of vectors in R with
limits x and y, respectively; and let {cn }∞
n=1 be a convergent sequence of
real numbers with limit c. Then the sequence {cn xn + yn }∞
n=1 converges to
the limit cx + y.
Definition 5.4. The vector x is an accumulation point of the sequence
{xn }∞
n=1 , if for any given  > 0 there are infinitely many integers n such that
kxn − xk < .
Definition 5.5. A sequence {yj }∞ j=1 of vectors in R
m
is a subsequence of se-
∞ m
quence {xi }i=1 in R if there exists an infinite increasing set of natural numbers
{nj } with
n1 < n2 < n3 < n4 < · · ·
such that y1 = xn1 , y2 = xn2 , y3 = xn3 , and so on.

6 Binary Relations on a Set of Points

Let X ⊆ Rn be a non-empty set of points. A binary relation designated as R,
can be defined over this set X. If x ∈ X and y ∈ X, then we write xRy if x is
related to y in the manner described by the relation R. For example, 5 is one
such binary relation among the members of a set of points having two or more
co-ordinates, and ≤ is another binary relation among members of a set of real
numbers. So, let x ∈ Rn and y ∈ Rn . We say,
1. x 5 y iff xi ≤ yi , ∀i = 1, 2, . . . , n.
2. x ≤ y iff xi ≤ yi , ∀i = 1, 2, . . . , n, and xj < yj , for some j.
3. x < y iff xi < yi , ∀i = 1, 2, . . . , n.
Some properties of interest can be defined on binary relation R. Let R be a
binary relation defined for X ⊆ Rn . Then, R is said to be:
(a) Reflexive iff (∀x ∈ X)[xRx].

7
(b) Transitive iff (∀x, y, z ∈ X)[(xRy ∧ yRz) =⇒ (xRz)].
(c) Complete iff (∀x, y ∈ X, x 6= y)[xRy ∨ yRx].
(d) Symmetric iff (∀x, y ∈ X)[xRy =⇒ yRx].
(e) Asymmetric iff (∀x, y ∈ X)[xRy =⇒∼ yRx].
(f) Anti-symmetric iff (∀x, y ∈ X)[(xRy ∧ yRx) =⇒ (x = y)].

7 Lower and Upper Bounds of a Set

Consider set Rn be ordered by a binary relation 5, and let S ⊆ R and let S be
non-empty.
Definition 7.1. Let b ∈ R. Then the number b is called an upper bound for
S if a ≤ b, ∀a ∈ S; the number b is called a lower bound for S if b ≤ a, ∀a ∈ S.
Definition 7.2. If b is an upper bound for S and no element smaller than b
is an upper bound for S, then b is called a least upper bound (lub) for S.
Similarly, if b is a lower bound for S and no number larger than b is a lower
bound of S, then b is called a greatest lower bound (glb) for S.
A special case occurs when n = 1, i.e. in this case, we are dealing with sets of
real numbers.

We can now state the least upper bound property and greatest lower bound
properties on R. Let S be any subset of R. If S has an upper bound, it has a
least upper bound; if S has a lower bound, it has a greatest lower bound.

8 Open, Close, and Compact Sets

Definition 8.1. For a vector z ∈ Rm and a positive number , the -ball about
z is B (z) = {x ∈ Rm : kx − zk < }. Sometimes B (z) is called the open -ball
to distinguish it from the closed ball B (z) = {x ∈ Rm : kx − zk ≤ } which
includes the boundary.
Definition 8.2. A set S in Rm is open if for each x ∈ S, there exists an open
-ball around x contained in S:
x ∈ S =⇒ there is an  > 0 such that B (x) ⊂ S.
An open set T containing the point x is called open neighbourhood of x.
Remark. (a) Open balls are open sets.
(b) Any union of open sets is open.
(c) Any finite intersection of open sets is open.

8
Definition 8.3. Let S ⊆ Rn . Then a point a ∈ Rn is said to be an adherent
point of S, iff there exists a convergent sequence of points drawn from S, i.e.
a sequence (xn ), xk ∈ S ∀k = 1, 2, 3, ..., such that xn → a.
Definition 8.4. A set S in Rm is said to be closed closed iff it owns all its
adherent points. Thus, whenever {xn }∞ n=1 is a convergent sequence completely
contained in S, its limit is also contained in S.
Remark. (a) A set S in Rm is closed iff its complement, S c = Rm r S, is open.
(b) Any intersection of closed sets is closed.
(c) The finite union of closed sets is closed.
Definition 8.5. Suppose that S ⊆ Rm . Let clS (or sometimes S) denote the
intersection of all closed sets containing S. The closed set clS is called the
closure of S. A closure of any set S is collection of all its adherent points.
Definition 8.6. A set S ∈ Rn is compact iff it is both closed and bounded.
Remark. (a) Any sequence contained in the closed and bounded interval [0, 1]
has a convergent subsequence.
(b) Let C be a compact subset in Rn and let {xn }∞
n=1 be any sequence in C.
Then {xn }∞
n=1 has a convergent subsequence whose limit lies in C.

Definition 8.7. Suppose that S is a subset of Rn . Let intS denote the union
of all open sets contained in S. The open set intS is called the interior of S.
Definition 8.8. A point x is in the boundary of a set S if every open ball
about x contains both points in S and points in the complement of S. So, the
set of boundary points of a set S equals clS ∩ clS c .
Definition 8.9. A point x is said to be in the exterior of set S, denoted as
extS, iff there exists an open ball around x which is contained in S c . extS =
(S)c .

9 Convex Sets
Definition 9.1. A set C ⊆ Rm is convex if for every x, y ∈ C and λ ∈
1
[0, 1], [λx + (1 − λ)y] Pn∈ C. For vectors x P , . . . , xn and non-negative scalars
n
λ1 , . . . , λn satisfying i=1 λi , a vector, xλ = i=1 λi xi is called a (finite) convex
1 n
combination of x , . . . , x . A strictly positive convex combination is a convex
combination where each scalar λi > 0.
Definition 9.2. For A ⊆ Rm , the convex hull of A, denoted coA (or some-
times, Ȧ), is the set of all finite convex combinations from A, i.e., coA is the set
of all the vectors x of the form
Pn
x = i=1 λi xi

9
 Pn
for some n, where each xi ∈ A, λ1 , . . . , λn ∈ R+ and i=1 λi = 1.
Alternatively, a convex hull of a set S is intersection of all convex sets con-
taining that set S.
Definition 9.3. The convex hull of a finite number of points is called the
convex polyhedron spanned by these points.
Remark. (a) Let E ⊆ Rm . If x ∈ coE, then x can be written as a convex
combination of no more than Pm+1 points in E, i.e., there are z 0 , . . . , z m ∈ E
i=0
and λ0 , . . . , λm ∈ R+ with m λi = 1 such that
Pm
x= i=0 λi z i .
T
(b) If for all i in some index set I, Ci is convex, then i∈I Ci is convex.
(c) If K is compact, then coK is compact.
(d) If A is convex, then intA and clA are convex.
Definition 9.4. A point x ∈ X is an extreme point of the convex set X ⊆ Rm
iff there do not exist points x1 , x2 (x1 6= x2 ) such that
x = (1 − λ)x1 + λx2 , where λ ∈]0, 1[.
Definition 9.5. Let x and y be two points from Rm . Then the scalar Pn or dot
product of these two points denoted as x · y [or simply xy] is xT y = i=1 xi .yi ,
where xT is transpose of the column vector x.
Definition 9.6. A hyperplane in Rm is a set of the form {x ∈ Rm : p · x = c}
where 0 6= p ∈ Rm and c ∈ R.
Definition 9.7. Given a boundary point w of a convex set X, then cx = z is
called a supporting hyperplane at w if cw = z and if all of X lies in one
closed half-space produced by the hyperplane, that is, either cu ≥ z ∀u ∈ X or
cu ≤ z ∀u ∈ X.
Definition 9.8. A closed half-space in Rm is a set of the form {x ∈ Rm :
p · x ≤ c}. An open half-space in Rm is a set of the form {x ∈ Rm : p · x < c}.
Remark. (a) Given any closed convex set X, a point y either belongs to the
set X or there exists a hyperplane which contains y such that all of X is
contained in one open half-space produced by that hyperplane.
(b) If w is a boundary point of a closed convex set, then there is at least one
supporting hyperplane at w.
(c) A closed convex set which is bounded from below has extreme points in
every supporting hyperplane.
(d) If a closed strictly bounded convex set X has a finite number of extreme
points, any point in the set can be written as a convex combination of the
extreme points, that is, set X is the convex hull of its extreme points.

10
Definition 9.9. Let x∗1 , x∗2 be distinct extreme points of the convex set X.
The line segment joining them is called an edge of the convex set if it is the
intersection of X with a supporting hyperplane. If x∗1 is an extreme point of
X, and if there exists another point x in X such that x = x∗1 + λ(x − x∗1 ) is in
X for every λ ≥ 0, and if, in addition, the set L = {x | x = x∗1 + λ(x − x∗1 ),
∀λ ≥ 0} is the intersection of X with a supporting hyperplane, then the set L
is said to be an edge of X which extends to infinity.
Definition 9.10. Two distinct extreme points x∗1 , x∗2 of the convex set X are
called adjacent if the line segment joining them is an edge of the convex set.

Definition 9.11. A cone is a non-empty subset of Rm closed under multipli-

cation by non-negative scalars. That is, C is a cone if whenever x ∈ C and
λ ∈ R+ , then λx ∈ C.
Definition 9.12. A cone is a convex cone if it is a convex set.

Definition 9.13. The point 0 is an element of any cone and is called the vertex
of the cone.
Definition 9.14. If C ⊆ Rm is a convex cone, the dual cone of C, denoted
C ∗ , is {p ∈ Rm |∀x ∈ C, p · x ≤ 0}.
Remark. (a) The intersection of cones is a cone.

(b) If C is a cone, then 0 ∈ C.

(c) Any set E ⊂ Rm generates a cone, {λx|x ∈ E, λ ∈ R+ }.
(d) The cone generated by E is the intersection of all cones containing E.

(e) A cone is convex iff it is closed under addition, i.e., a cone C is convex iff
x, y ∈ C implies (x + y) ∈ C.

Prepared by:
Shobhit Mohan (with some editing inputs from Naresh Kumar Sharma)

11