Field Theory and Galois Theory

A Project Report

submitted by

Amit Kumar

in partial fulfilment of the requirements

for the award of the degree

of

MASTER OF SCIENCE
IN MATHEMATICS

2011

DEPARTMENT OF MATHEMATICS
NATIONAL INSTITUTE OF TECHNOLOGY ROURKELA
ROURKELA, ORISSA-769008
 THESIS CERTIFICATE

This is to certify that the Project Report entitled “Field Theory and

Galois Theory” submitted by Amit Kumar to National Institute of Tech-

nology Rourkela, Orissa for the partial fulfilment of the requirements of M.Sc.

degree in Mathematics is a bonafide record of review work carried out by him

under my supervision and guidance. The content of this project, in full or

in parts, has not been submitted to any other Institute or University for the

award of any degree or diploma.

(S.R.Pattanaik)

and

(R.S.Tungala)

Asst.Professor

Department of Mathematics

NIT Rourkela

ii
 DECLARATION

I declare that the topic ‘Field Theory and Galois Theory’ for my M.Sc.

degree has not been submitted in any other institution or university for the

award of any other degree or diploma.

Place: Amit Kumar

Date: Roll.No. 409MA2068

iii
 ACKNOWLEDGEMENT

I would like to warmly acknowledge and express my deep sense of

gratitude and indebtedness to my guides Dr S.R.Pattanaik and Dr.
R.S.Tungala,
Asst Professor, Department of Mathematics, NIT Rourkela, Orissa, for his
keen guidance, constant encouragement and prudent suggestions during the
course of my study and preparation of the final manuscript of this Project.

I would like to thank the faculty members of Department of Mathemat-

ics for allowing me to work for this Project in the computer laboratory and
for their cooperation.

My heartfelt thanks to all my friends for their invaluable co-operation

and constant inspiration during my Project work.

I owe a special debt gratitude to my revered parents, my brother, sister

for their blessings and inspirations.

Rourkela,769008
May,2011

(Amit Kumar)
M.Sc. Mathematics
NIT Rourkela

iv
No Contain page no

1 Chapter-1 1
Motivation 2
Euclidean Domains 2
Principle Ideal Domain 3
Unique Factorization Domain 6
Polynomial rings and irreducible criteria 7
Guess’s Lemma 8
Over Q implies over Z 9
Irreducibility Tests 10

2 Chapter-2 13
Characteristic of Field 14
Extension Field 14
Splitting Fields 16
Algebraic Extensions 18
Characteristics of Extension 19
Degree of Extension 20
Properties of Algebraic Extensions 25

3 Chapter- 3 28
Galois Theory 29
Basic Definitions 29
Fundamental Theorem of Galois Theory 34

4 Summary 41

5 References 42
Introduction

The general solutions of linear and quadratic polynomial in one variable were known
centuries before. For cubic and quartic equations also the general solutions are provided
by Cardano's and Ferrari's methods respectively. In 19th century a great work has been
done to find general solution of a general polynomial by radicals. However there was no
success even after efforts of many great mathematicians of that time. Eventually work by
Able and Galois gives satisfactory solution and complete understanding of this problem.

Galois Theory provides a connection between Field theory and Group theory, which
in turn useful to convert problems in field theory into Group theory, which are better
understood and easy to handle. Galois theory not only provide answer to the problem
discussed above but also explains why the general solution exists for polynomials with
degree less then or equal to 4.

In his original work, Galois used permutation groups to describe relations between roots of
the polynomial. In modern approach, developed by Artin, Dedekind etc., involves study of
automorphisms of field extensions.
 CHAPTER- 1

Euclidean Domain, Principal Ideal Domains and Unique Factorization

Domains. Polynomial Rings & Irreducibility Criteria.

1
Motivation:-
There are a number of classes of rings with more algebraic structure than generic
rings. Those considered in this chapter are rings with a division algorithm (Euclidean
Domains), rings in which every ideal is principal (Principal ideal Domains) and rings in
which element have factorization into primes (Unique Factorization Domains). The
principal examples of such rings are the ring Z of integers and polynomial rings F[x] with
coefficients in some field F.

[2]EUCLIDEAN DOMAINS
In Z , the absolute value or |n| ∊ Z, is non-negative number. Given a, b∊ Z, b≠0
such that
a= bq + r, where 0≤|r|<|b|
Definition:- An integer domain R is called Euclidean Domain if a function
Φ : R | {0} Z ≥ 0 satisfying
[non-zero element of R to the non-negative integers, Φ is norm ]
1. Ф(a)≤ Ф(b) for all element a, b in R
2. Given a, b ∊ R , b ≠ 0 q, r ∊R such that a= bq + r and either r=0 or Ф(r)≤ Ф(b).

Examples :-
1. The ring Z is Euclidean domain with Ф(n)=|n|
Ans :-
i . Let m,n ∊Z Ф(n)=|n| Ф(m)=|m|
Then Ф(nm)=|nm|
Now we have
Ф(n)=|n| ≤ |n||m|= Ф(nm)
Ф(n) ≤ Ф(nm).
ii. m, n ∊Z , n ≠ 0 q , r such that m= n q + r and |r| < |n|.

2
 (2) Let F be a field ,then F[x] for f[x] ≠ 0∊ F[x]
Define Ф(f) = deg f
Ans:-
i. Let f, g ∊ F[x] such that
Ф(f) = deg f
Ф(f) = deg g
Ф(fg) = deg( fg)
Ф(f) = deg f ≤ deg( fg) = Ф(fg)

ii. 2nd condition is follows from the division algorithm for F[x].

Principle Ideal Domain:-

An integer domain R is called P.I.D if every ideal of R is principal. [Principal Ideal: -
Ideal generated by one element]
i.e. every ideal has the form < a > = { ra | r∊ R} for some a ∊R

Example :-

i. The ring Z is a P.I.D generated by < n >

ii. A field F is P.I.D. The only ideal of F are {0} and F itself.
iii. Let F be a field, then F[ x ] is a P.I.D
Ans:-
We know that F[x] is integral domain.
Let us consider I be an ideal in F[x]
If I = {0} then I = < 0>
If I ≠ {0} then among all the elements of I,

3
 Again let g( x) be one of the minimum degree.
Claim: I = < g(x) >
As g(x) ∊ I
We have < g(x) > I
Only to show I < g(x) >

Let f(x) ∊I, then by division algorithm, we can write

f( x ) = g( x ) q( x ) + r( x )
Where r(x) = 0 or deg r(x) < deg g(x)
⟹ r( x ) = f( x ) – g( x )q( x ) ∊ I [ As f( x ), g( x )∊I ]
so, r( x ) = 0 [ As deg r( x ) < deg g( x ) which is not possible as
the minimality of deg g( x ) ]

⟹f(x) = g(x) q(x)

⟹f(x) ∊ < g(x) >
⟹I < g(x) >

iv. The ring Z[ x ] is an integral domain but not P.I.D

Solution:
Let us consider the ideal ‘S’ of Z[x] generated by the elements 2 and x of
Z[x], then
S = 2 f( x ) + xg( x ) ; f( x ), g( x ) ∊ Z [ x ]
Assume Z[x] is P.I.D and generated by < h(x) >
Say h(x) ∊ Z(x)
i.e. Z( x ) = { < h( x ) > : h( x ) ∊ Z[ x ] }

2 ∊ S ⟹2 ∊ < h(x)>

4
 ⟹2= h(x) h1(x) for some h1(x ) ∊ Z[ x ]
x ∊ S ⟹ x ∊ < h( x ) >

⟹ x = h( x ) h2( x ) for some h2 ( x ) ∊ Z[ x]

Therefore 2h2(x) = xh1( x )

⟹ Each co-efficient of h1(x) is an even-integer. So h1(x) = 2 p(x) for some p(x) ∊ Z[ x ]

So, 2 = 2h(x) p(x)

⟹ h( x ) p( x ) = 1
⟹ 1 ∊ < h(x) > i.e. 1∊ S

Now 1∊ S

⟹ 1 = 2q( x ) + xr( x ) for some q( x ), r( x ) ∊ Z[ x ]

Let q( x ) = a0 + a1 x + a2 x2 + . . .
r( x ) = b0 + b1 x + b2 x2 + . . .
Then 1 = 2(a0 + a1x + a2x2 + . . .) + x(b0 + b1x + b2x2 + . . .)
⟹ 1 = 2a0
Which is not possible as a0 is an integer. Contradicts the assumption, so
Z [x] is not a P.I.D.

5
Unique Factorization Domain :-
An integer domain R is called an U.F.D if every non-zero elements of r ∊ R which
is not an unit has the following two properties:-
i) r can be written as the products of irreducible in R (not necessary distinct)
i.e. r = p1, p2, . . . pn and
ii) The decomposition of ---- (1) is unique up to associates.
i.e. if r = q1, q2, . . . qm is another factorization of r into irreducible
then m= n and there is some renumbering of the factors so that p i is associates to qi for
i=1, 2, . . . , n
Example:- (i) The integral domain Z is U.F.D . Every non-zero elements other than 1
and -1 in Z can be expressed as the product of a finite number of irreducible elements
in Z and the factorization is unique except for the order of the factors.
12= 2·2 · 3 = (-2) · 2 · (-3) = (-2).(-3) ·2
These factorization are some except for the order and the associates of the irreducible.
(ii) Z is U.F.D so integral domain Z[x] is a U.F.D.
(iii) The domain D= Z[ ] is not a U.F.D
Solution:
6 has two different factorization into irreducible as 6 = 2·3 = (1 + ) (1 - )
In D, each 2 · 3 , 1 + ,1 - is irreducible in D .
6 = 2 · 3 = (1 + ) (1 - )
None of the factors (1 + ) and (1 - ) associates of 2 or 3 . Since 1 and -
1 are only units in D.
[2]Conclusion: - Finally as above our discussion we find the relation as
Fields E.D P.I.D U.F.D I.D
But in each of cases converse part is not true. Which are shown by the examples:-
i. Z is E.D but not field

ii. Z[ ] is P.I.D but not E.D

iii. Z [x] is U.F.D but not P.I.D

iv. Z[ ] is I.D but not U.F.D

6
[1]Polynomial rings and irreducible criteria

Let R be a commulative Ring. The set of formal symbols.

R[ x ]={ an xn +a n-1 xn-1 + . . . + a1 x + a0 | ai ∊R, n is an nonnegative integer }
is called the ring of polynomials over R in the indeterminant x.
If an ≠ 0 then the polynomial degree ‘n’ and an xn is the leading term .
The polynomial is monic if an = 1.

Content of polynomial, primitive polynomials:

The content of a nonzero polynomial an xn + an xn + . . . + a1 x + a0, where a’s
are integers, is the greatest common divisor of the integers a n , an-1 , . . . ,a0 . A
primitive polynomial is element of Z[ x ] with content 1.

Irreducible polynomial, reducible polynomial

A non-constant polynomial f(x) is irreducible over F[x], if f( x) cannot be expressed
as a product of g( x ) h( x ) of two polynomials g( x ) and h( x ) in F[ x ] both of lower
degree than the degree of f( x ).
If f(x) ∊F[x] is a non-constant polynomial that is not irreducible over F then f( x ) is
reducible over F.
Theorem: let f(x) ∊ F[x]and let f(x) be of degree 2 or 3 then f(x) is reducible over F if
and only if it has a zero in F.

Proof:
Let f(x) is reducible over F
To show : f(x) has a zero in F
As f(x) is reducible so we can write
f( x ) = g( x ) h( x ), where degree of g x) and h( x )are both less than the degree
of f( x ), then since f( x ) is either quadratic or cubic, either g( x ) or h( x ) is of degree 1.
If say g( x ) is of degree 1 , then expect for a possible factor in F , g( x ) is the form x-a

7
Then g( a ) = 0
⟹ f( a) = 0
So, f(x) has a zero in F
Conversely from the factor theorem we know that , if f(a ) = 0 for a ∊ F , then
x-a is a factor of f( x ).
So f( x ) is reducible.
Example : i) X2 - 2 ∊ Q[ x ] has no zero in Q . This shows that x2- 2 is irreducible
over Q.
ii) In particular it is easy to use when the field is Zp .
As in Z2 polynomial x2 + x + 1 is irreducible. As it has no root .
i.e. f(0) = 0+0+1 = 1
and f( 1) =1+1+1 =3 = 1

[1]Gauss’s Lemma:
Statement → The product of two primitive polynomials is primitive.
Proof :
let f( x ) and g ( x )be primitive polynomials .To prove f ( x ) g( x ) is primitive
If possible, let f( x )g( x ) is not primitive
So let p be a primitive divisor of the content of f( x )g( x )
And let the polynomials f’( x ) and g’( x ) and ( f ( x ) g( x ))’ obtained from f(x), g(x)
and f(x)g(x) respectively, by reducing the co-efficients modulo p. Then f’(x) and g’(x)
belongs to integral domain Zp[ x ]and
f’( x ) g’( x ) = ( f ( x ) g( x) )’ = 0, the zero elements of Zp[x]
So, either f’( x ) = 0 or g’( x ) = 0
[Since these are in integral domain]
This means that either p divides every coefficient of f(x) or p divides every coefficient
of g(x).
Therefore either f(x) is not primitive or g(x) in not primitive, which is contradiction the
assumption.
So f(x) g(x) is primitive .Hence the statement.

8
[1] Over Q implies over Z
Let f(x) ∊ Z[ x ] . If f(x) is reducible over Q, then it is reducible over Z
Proof:
Given f(x) is reducible over Q, so we can write f( x ) = g( x ) h( x), where g( x ) and
h( x ) ∊ Q( x ) we may assume that f( x ) is primitive because we can divide both f( x )
and g( x ) h( x ) by the content of f( x ).
Let a be the l.c.m of the denominators of the coefficients of g(x) and b be the l.c.m of
the denominators of the coefficient of h( x ).
Then abf( x ) = ag( x ) . bh( x ) , where ag( x ) and bh( x ) ∊ Z[ x ]
Let c1 be the content of ag( x ) and
c2 be the content of bh( x )
Then ag( x ) = c1g1( x ) and
bh( x ) = c2h1( x )
Where g1( x ) and h1( x ) both are primitive
and ab f( x ) = c1c2g1( x )h1( x ) ……………………………….(1)

f( x ) is primitive so content of abf( x ) is ab and g1( x ) h1( x ) is primitive [ product of

two primitive polynomials is primitive]
So content of c1c2g1(x)h1(x ) is c1c2
Thus from ………………………………..(i)
ab =c1c2
f( x ) = g1( x ) h1( x )
Where g1( x ) and h1( x ) ∊ Z[ x ] and
deg g1( x ) = deg g( x ) and deg h1( x ) = deg h( x )

9
IRREDUCIBILITY TESTS:
1) Mod p irreducibility test
Statement: let p be a prime and suppose that f( x ) ∊ Z[ x ] with deg f( x ) ≥ 1.
Let f’(x) be the polynomial in Zp [x]obtained from f( x )modulo p. If f’(x) is irreducible over
Zp and deg f’(x) = deg f( x ), then f(x)is irreducible over Q.
Proof:
Let f (x) ∊ Z [x]
If possible let f(x) is reducible over Q then we have
f ( x ) = g ( x ) h ( x ) with g( x ), h( x ) ∊ Z[ x ]
and both g( x ) and h( x ) have degree less than that of f( x )
let f’( x ), g’( x )and h’( x ) be the polynomials obtain from f( x ), g( x ) and h( x )by
reducing all the co-efficient modulo p .
Since deg f( x ) = deg f’( x )
we have deg g’( x ) ≤ deg g( x ) < deg f’( x )
again deg h’( x ) ≤ deg h( x ) < deg f’( x )
but f’(x) = g’( x ) h’( x )
 f’( x ) is reducible over Zp ,

which is contradiction
Hence f(x) is irreducible over Q.
Example: f(x) = 11x4 + 8x3 + 5x2 + 5. Then over Z2 , we have
f’(x) = x4 + x2 + 1 and
since deg f( x ) = deg f’( x )
f’( 0 ) = 1 and f’( 1 ) = 1 + 1 + 1 = 3 = 1
we find that f’( x ) is irreducible over Z2
Thus f( x ) is irreducible over Q.

10
2) Eisenstein’s Criterion:-
In 1980 Ferdin and Eisenstein, a student of Gauss, was found another important

irreducibility test . [1]

Statement :-
let
f( x ) = an xn +a n-1 xn-1 + . . . + a1 x + a0∊ Z[x]
If there is a prime p such that p∤an ,p |an-1, …, p | ao and p2 ∤ a0 then f(x) is irreducible
over Q

Proof :-
If possible let f(x) is reducible over Q.
Then we know that ∃ elements g(x) and h(x) in Z[x] such that
f(x)=g(x)h(x) and
deg g(x) ≥ 1 ,deg h(x) < n
Say g(x)= brxr+…..+ b0 and
h(x)= csxs+….+ c0
Then since p | a0 , p2 ∤ a0
and a0=b0c0,
It follows that p divides one of b0 and c0 but not the other
Let us consider the case p |a0but p ∤ c0
..
. p ∤ an => p ∤ brcs
⟹p ∤ br or p ∤cs
p ∤ br so there is a least ineger ’t’ such that p ∤ bt
Now consider
at= bt co + bt-1 c1 + . . . + b0 ct .
By assumption, p divides at and by choice of t every summand on the right hand side
after the first one is divisible by p .
Then it is true that p to divides bt c0 , this is impossible .

11
 p is prime and p divides neither bt nor c0 which gives contradiction
Hence the statement .

Example :-
Taking p = 5 , f(x) = 18 x5 – 5x4-10x2-102 is irreducible over Q by Eisenstein’s
criterion,
because 5 ∤ 18 and 25∤102 but 5 does divide -5 and -10.

Important Corollary :-
Let F be a field and p(x) an irreducible polynomial over F.
Then F[x] | < p(x)> is a field .

12
CHAPTER-2

Fields Theory-Extension Fields, Algebraic Extentions.

13
Characteristic of field:-

Is denoted by char (F) and defined as smallest positive integer P such that

P.1F = 0, where 1F is the identity of F

 Characteristic of a field is either ‘0’ or a prime ‘P’.

Extension Field: - A Field E is an extension field of a field F if F⊆E and the operation
of F are those of E restricted to F. Denoted by E|F

Ex: - C E (Extension Field)

Q F (Base Field)

FUNDAMENTAL THEOREM OF FIELD THEORY (KRONECKER’S

THEOREM, 1887) [1]

Let F be a field and f(x) a non-constant polynomial in F[x]. Then there is an

extension field E of F in which f(x) has a zero.

Proof: - As F[x] is a U.F.D., hence by definition of U.F.D f(x) has an irreducible factor,
say it p(x).

Now, it is sufficient to construct an extension field E of F in which p(x) has a zero.

Hence, E= F[x]|<p(x)>

14
Also, since the mapping φ : F E, given by φ (a) =a+<p(x)> is one-one and
preserves both operations.

E has a subfield isomorphic to F.

Enough to show that p(x) has zero in E.

Let p(x) = anxn + an-1xn-1 + . . . + a0

Then, in E, x= x + <p(x)> is a zero of p(x).

p(x+<p(x)>) = an (x+<p(x)>) n + an-1 (x+<p(x)>) n-1 + . . . +a0

= an (x n +<p(x)>) + an-1 (x n-1+<p(x)>) + . . . +a0

= an xn + an-1 xn-1 +. . . . . . . . +a0 +<p(x)>

=P(x) + <p(x)>

=0+<p(x)>

=<p(x)>

Example:-

Let F=R and let f(x) = x2 +1 clearly x2 +1 is irreducible over R. Then < x2 +1> is a
maximal ideal in R[x], so R[x]/< x2 +1> is a field. Let r ∊ R with r + < x2 +1> in R[x] / < x2
+1>,

Here, R is a subfield of E = R[x]/< x2 +1>

Let α= x+ < x2 +1>

Now,

α2 +1 = (x+< x2 +1>) 2 + 1

= (x2 +1) +< x2 +1>

=0

Thus α is a zero of x2 +1

15
SPLITTING FIELDS:-

The extension field E of F is called splitting field for the polynomial f(x) ∊ F[x], if f(x)
factors completely into linear factor in E[x] and f(x) does not factor completely into linear
factor over any proper subfield of E containing F.

Examples:-

(i) The splitting field for x2 -2 over Q is just Q (√2) as x= ±√2 and -√2∊ Q (√2)

(ii) The splitting field of (x2 -2) (x2 -3) is field Q (√2, √3)

(iii) Let f(x) = x2 +1 ∊ Q[x]

x2 +1 = (x + √-1) (x -√-1), here f(x) splits in ₵, but a splitting over Q is

Q (i) ={r+si / r, s ∊Q}

But x2 +1 ∊ R [x], here splitting field for (x2 +1) over R is ₵.

Existence of splitting fields

Let F be a field and let f(x) be a non-constant element of F[x]. Then there exists a
splitting field E for f(x) over F.

Proof:-

We prove it by induction on deg f(x).

If deg f(x) = 1, then f(x) is already linear and E = F.

Now suppose that the statement is true for all fields and all polynomial of degree less
than that of f(x).

Hence by Fundamental Theorem of Field Theory, there is an extension E of F in which

f(x) has a zero, say a1

f(x) = (x- a1 ) g(x), where g(x) ∊ E[x].

16
 since deg g(x) < deg f(x), by induction, there is a field M that contains E and all the
zeros of g(x), say, a2 , a3 , . . . an . Clearly, then, a splitting field for f(x) over F is F (a 1 ,
a2, . . . , an)

Splitting Fields are unique:-

Let Ф be an isomorphism from a field F to a field F’ and let f(x) ∊ F[x]. If E is a splitting
field for f(x) over F and E’ is a splitting field for Ф (f(x)) over F, then there is an
isomorphism from E to E’ that agrees with Ф on F [i.e. splitting fields are unique] .

Proof: We prove it by induction on deg f(x). If deg f (x) =1, then E=F and E’ = F’. So that
Ф is itself the required mapping. If deg f(x)>1, let p(x) be an irreducible factor of f(x), let
a be a zero of p(x) in E, and let b be a zero of Ф (p(x)) in E’. By known lemma, there is
an isomorphism α from F (a) to F’ (b) that agrees with Ф on F and carries a to b.

Now, f(x) = (x-a) g(x), where g(x) ∊ F (a)[x]

Then E be a splitting field of g(x) over F (a)

E’ be a splitting field of α (g(x)) over F’(b)

Since deg g(x) < deg f(x), there is an isomorphism from E to E’ that agrees with α on
F(a) and therefore with Ф on F.

Theorem:-

Let F be a field and let p(x) ∊ F[x] be irreducible over F. If a is a zero of p(x) in some
extension E of F, then F (a) is isomorphic to F[x] / <p(x)>. Furthermore, if deg p(x) =n,
then every member of F (a) can be uniquely expressed in the form

Cn-1 an-1 + Cn-2 an-2 + . . . + c1a+c0,

Where c0, c1 , . . . cn-1∊ F.

17
Examples:-

In a language of vector space,if ‘a’ is a zero of an irreducible polynomial over F of

degree n, then the set {1,a,a2,. . . , an-1} is a basis for F (a) over F. Consider the
irreducible polynomial f(x) = x6-2 over Q. Since is a zero of f(x), so as above

theorem, we have the set {1, 21/6, 22/6, 23/6 , 24/6, 25/6} is a basis for Q ( ) over Q.

Thus,

Q( ) = {a0 +a1+21/6 +a2 22/6 +a3 23/6 + a4 24/6 + a5 25/6 | ai ∊Q}

This Field is isomorphic to Q[x] |< x6-2>.

[1]Algebraic Extensions

Let F be a field and let E be an extension of F. An elements α ∊ E is said to be algebraic

over F if α is a root of some non-zero polynomial f(x) ∊ F[x]. [I.e. f (α) =0].

If α is not algebraic over F, then ‘α’ is said to be transcendental over F.

An extension E of F is called algebraic extension of F if every elements of E is algebraic

over F. If E is not an algebraic extension of F, then it is called a transcendental
extension of F.

e.g.: e is transcendental over Q,

π is transcendental over Q.

But it is not known about (π+e)

18
[3]Characteristics of Extension:

Let E be an extension field of a field F and let a є E be algebraic over F. Let p(x) ∊F[x]
be a polynomial of least degree such that p(a) =0. Then

I. P(x) is irreducible over F.

II. If g(x) ∊ F[x] is such that g (a) =0, then p(x)| g(x).

III. There is exactly one monic polynomial p(x) ∊ F[x] of least degree such that
p(a)=0

Proof:

(i) Let p(x) is reducible over F. p(x) = p1(x) p2(x) and degree of p1(x), degree of p2 (x)
be less than that of p(x).

Then 0=p (a) = p1(a). p2(a)

=> Either p1 (a) = 0 [As field]

or p2(a)=0

i.e. ‘a’ satisfy a polynomial of degree less than p(x), a contradiction

So p(x) is irreducible over F.

(ii) Since g(x) ∊ F[x],

By division algorithm

g(x) = p(x) q(x) + r(x), where r(x)=0 or deg r(x)< deg p(x)

Then,

g(a) = p(a). q (a) + r (a)

i.e. r(a)=0

As p(x) is least degree among the polynomial satisfying by a,

19
 i.e. r(x) =0

Thus p(x) |g(x)

(iii) Let g(x) be a monic polynomial of least degree such that g(a)=0

Then by (ii) p(x)| g(x) and g(x) |p(x)

⟹ p(x) = g(x) (since both are monic)

Degree of Extension:

Let E be a field extension of a field F, we may view that E is vector space over F. The
degree of field extension E|F is denoted by [E: F] and defined as-

deg [E: F] = dimension of E over F

If [E: F] =n

E has a degree n over F.

If E has dimension n as a vector space over F.

If [E: F] is finite, E is called a finite extension of F; otherwise, we say that E is an infinite

extension of F.

Example:

1. The field of complex numbers has degree 2 over the reals since {1, i} is a basis.
2. If a is algebraic over F and its minimal polynomial over F has degree n, then we
have {1, a . . . an-1} is a basis for F (a) over F; and therefore [F(a):F]=n. In this
case, we say that a has degree n over F.

20
Q( ) Q( ) Q( )

2 3 6

Q Q Q

[Q ( ): Q] =2 [Q( ) :Q] =3 [Q( ) :Q] =6

[E: F]= n

Theorem:-

Finite Implies Algebraic

If E is a finite extension of F, then E is algebraic extension of F.

Proof: Suppose that [E: F] = n and a є E. Then the set {1, a, . . . ,an} is linearly
dependent over F; so there are elements c0, c1, . . . , cn in F, not all zero, such that cn an
+ cn-1 an-1 + cn-2 an-2 +. . . + c1a+c0 = 0

So ’a’ satisfy cn an + cn-1 an-1 + cn-2 an-2 +. . . .+ c1a+c0 = 0

so ‘a’ is algebraic over F.

21
(*) But the converse of the above theorem is not true which is shown in the following
example

e.g. Q (21/2, 21/3, 21/4 ,…, 21/n,…)

Let X= 21/n

=> Xn – 2=0

So algebraic extension as (an – 2=0)

Clearly it is not finite extension.

Theorem:-

[K: F] = [K: E] [E: F]

Let K be a finite extension field of the field E and let E be a finite extension field of the
field F. Then K is finite extension of the field F and [K: F] = [K: E] [E:F]

Proof: Let X = { x1, x2,. . . ,xn } be a basis for K over E and

Let Y = {y1, y2,. . . ,yn} be a basis for E over F.

It remains to prove that

YX= {yixi | 1≤j≤m, 1≤i≤n} is a basis for K over F.

For this, let a ∊ k

Then there are elements b1, b2, . . . , bn ∊ E such that

22
a = b1x1 + b2x2 + . . . +bnxn

and, for each i= 1,. . . , n, there are elements ci1, ci2,. . . ,cim ∊ F such that

bi = ci1y1 + ci2y2 + . . . + cimym

Thus,

n
a= =Ʃ i=1 ( xi

= )

This proves that YX spans K over F. Now suppose there are elements c ij in F such that
0= Ʃi,j cij(yjxi) = Ʃi Ʃj(cijyj)xi

Then since each cijyj єE and X is a basis for k over E, we have

Ʃjcijyj =0 for each i

But each cij ∊F and Y is a basis for E over F, so each cij =0.

This proves that the set YX is linearly independent over F.

Example: since {1, √3} is a basis for Q(√3 ,√5) over Q(√5) and {1,√5} is a basis for Q(√5)
over Q, then as above theorem shows that {1, √3,√5,√15} is a basis for Q(√3, √5) over
Q.

Q(√3, √5)

2 2

Q(√3) 4 Q(√5)

2 2

23
Consider Q ( ).

Then [Q ( ):Q].=12

For clearly,

[Q ( ):Q] = [Q ( ):Q( ) [Q( ):Q]

And

[Q ( ):Q] = [Q ( ):Q( ) [Q( ):Q]

Now:{1, 21/3, 22/3} is a basis for Q( ) over Q

[Q( ):Q]=3 and [Q( ) : Q( ) ]=4, as

{1, 31/4,32/4,33/4} is a basis for Q( ) over Q( )

So,

[Q ( ):Q] =12

(*) To Show: Q(√2, ) = Q (√2+ )

Solution:

Claim 1: Q (√2+ ) Q (√2, )

Since √2, ∊Q(√2, )

=> √2+ ∊ Q(√2, )

=> Q (√2+ ) Q (√2, )

Claim 2: Q(√2,

24
It is sufficient to prove: √2-

√2 ∊ Q(√2+

Let √2+

=> = α-√2(cubic both sides)

=> 5=α3-2√2-3√2α2 +6α

=>√2= (α3+6α-5)/(2+3α2) ∊ Q(α)

=> √2=(α3+6α-5)(2+3α2)-1 ∊ Q(α)

Similarly, so,

√2 - ∊ Q(α)

PROPERTIES OF ALGEBRAIC EXTENSIONS

Theorem:

Algebraic over algebraic is algebraic

If K is an algebraic extension of E and E is an algebraic extension of F, then K is an

algebraic extension of F.

Proof:

Let a∊ K, as K is algebraic over F so there is b0, b1,. . . , bn ∊ E such that

b0+b1a+ . . . +bnan=0

25
Now

E is algebraic over F and b0, b1, . . . ,bn ∊ E so

b0, b1, . . . ,bn algebraic over F. So E |F is finite and E is isomorphic to

F (b0, b1, . . . ,bn)

Therefore F(b0, b1, . . .,bn) F is finite

=> [F(b0):F]= finite

[F(b1):F] = finite

- ---------
- ---------
- - - - - - - - -- -
b1 is algebraic over F
=> b1 is algebraic over F(b0)
Now, [F(b0b1): F(b0)]=finite
[F(b0b1): F]=[ [F(b0b1): F(b0)][F(b0):F]
(finite) (finite)

By separation we can write

[F (b0, b1, . . . ,bn):F]= finite

[M:F]= finite

a satisfies

=> b0+b1a+b2a2+ . . . + bnan = 0

=> a is algebraic over M

=> [M(a): M]= finite

26
Now,

[M(a): F] = [M(a):M][M:F]

= finite

=> M(a) is algebraic over F

=> ‘a’ is algebraic over F

=> k is algebraic over F.

Corollary:- Subfield of algebraic Elements

Let E be a field extension of field f.

Let S = {all elements of E which are algebraic over F}

Then ‘S’ is a sub field of E containing F.

Proof:

Let a1b(b≠0) ∊ E are algebraic over F.

To show: a+b, a-b, ab, a/b are algebraic over F.

It is sufficient to show.

[F(a,b):F]= finite, as a+b, a-b, ab a/b ∊F(a,b)

Now;

[F(a,b):F] = [F(a,b): F(b)][F(b):F]

Also a is algebraic over F, it is clearly algebraic over F(b) .

Thus, both [F(a,b): F(b)] and [F(b):F] are finite.

(*) For any extension E of a field F, the subfield of E of the elements that are algebraic
over F is called the algebraic closure of F in E. e.g. algebraic closure of Q in ₵.

27
CHAPTER- 3

An Introduction to Galois Theory.

28
[1]Galois Theory

In the previous discussion we proved the existence of a finite extension of a field F

which contains all the roots of a given polynomial f(x) whose coefficients are in F .The
main idea of Galois theory (named for Evariste Galois , 1811 - 1832) is to consider the
relation of the group of permutations of the roots of f(x) to the algebraic structure of its
splitting field.

Basic Definitions

Automorphism, Group Fixing F, Fixed Field of H

A ring isomorphism φ of field E with itself is called an automorphism of E. The

collection of automorphism of E is denoted by Aut (E)

 An automorphism φ ∊ Aut(E) is said to fix an elements α∊E if φα=α

If F is subset of E, then automorphism φ is said to fix F if φa=a ∀a ∊F

Any field has at least one automorphism, the identity map (trivial automorphism).

Definition: Let E be an extension field of the field F.

Let Aut(E/F) be the group of automorphisms of E which fix F

= set of F automorphism of E

= Galois group of E/F

= Gal (E/F)

i ,e. Gal (E/F) = { φ| φ: E -> E S.T. φ (α)=α ∀ α ∊ F}

If H is a subgroup of Gal (E/F), the set

EH = {x ∊ E | φ(x) = x for all φ ∊ H} is called fixed field of H

29
Example

1. Consider the extension Q(√2) of Q

Since Q (√2) = { a+ b √2| a ,b ∊ Q}

φ: Q (√2) -> Q(√2) such that φ(a)=a ∀a ∊ Q

φ (a+ b √2) = φ (a)+ φ(b √2)

= a+ φ (b) φ (√2)

= a+ b φ (√2)

An automorphism φ of Q (√2) is determine if φ (√2) is to be calculate

2= φ( 2)= φ(√2√2)= φ(√2)φ(√2)= (φ(√2))2

 φ(√2)= ±√2

this proves that the group Gal(Q(√2)|Q) has two elements, the identity mapping and
mapping that sends a+ b√2 to a- b√2 the fixed field of Gal(Q(√2)|Q)is just Q as
everything is fixed by identity automorphism

i.e. a+ b √2 -> a+ b √2

a+ b √2 -> a- b √2

Which is equivalent to

a +b √2= a-b √2

 b=0

2. Consider the extension Q (3√2) of Q similarly as above example, an automorphism φ

of Q (3√2) is completely determines by φ (3√2).

φ (3√2) is a cube root of 2.

Therefore φ (3√2) = 3√2, 3√2ω, or 3√2ω2 where ω3=1, ω ≠ 1

30
Because φ (3√2) is real (subset of real number), the only possibility is Q (3√2) = 3√2,

Hence

a+b3√2 -> a+b3√2

Gal (Q (3√2) |Q) has one elements and fixed field of Gal (Q (3√2) |Q) is Q (3√2)

3. Consider the extension Q (√3, √5) of Q.

As we known that Q (√3, √5) = {a + b √3 + c √5 + d √3 √5|a, b, c, d ∊ Q}

Any automorphism φ of Q (√3, √5) is completely determined by the two values φ (√3)
and φ.(√5). This time there are four automorphism:

ϵ α β αβ
√3 -> √3 √3 -> -√3 √3 -> √3 √3 -> -√3
√5 -> √5 √5 -> √5 √5 ->- √5 √5 ->- √5

Gal (Q (√3, √5) |Q) = {ϵ, α, β, αβ} isomorphic to Z2 ⊕Z2

Let H= {Id, α}: α: a + b √3 -> a - b √3

a + b √5 -> a + b √5

Fixed field of {ϵ, α} is Q (√5)

Fixed field of {ϵ, β} is Q (√3)

Fixed field of {ϵ, α β} is Q (√3√5)

The lattice of subgroup of Gal (Q (√3, √5) |Q) and the lattice of subfield of Q (√3, √5) are
shown below:

31
 {ϵ, α, β, αβ} Q(√3,√5)

{ϵ, α} Q (√3)
{ϵ, β} Q (√5) Q (√3√5)
{ϵ, αβ}

{ϵ} Q

[1]Hence we see that there is an intimate connection between the lattice of subfield
between E and F and the lattice of subgroup of Gal (E/F).

Let A -> be the lattice of subfield of E containing F

B -> be the lattice of subgroup of Gal (E/F).

Then for each K in A , the group Gal (E/K) is in B and for each H in B, the field E H is in
A.

Define a mapping

g: A -> B by

g (K)=Gal(E/K)and

f: B -> A by

f (H)=EH

If K and L belongs to A and K ⊆L then g(K) ⊇ g(L) similarly if G and H belongs to B and
G⊆H then f(G) ⊇ f(H). Thus f and g are inclusion reversing mapping between A and B.

32
However, when E is suitably chosen extension of F, the fundamental theorem of Galois
Theory says that f and g are inverse of each other so that the inclusions are equalities.
In particular, f and g ate inclusion reversing isomorphism between the lattices A and B

[4]Galois Extension:
A finite extension E/F is said to be Galois extension if

|Gal (E/F)|= [E: F]

Example: In above example --- (1)

Gal (Q (√2) |Q) is Galois extension, whereas Gal (Q (3√2) |Q) in example --- (2) is not
Galois extension.

Normal Extension:
let E be a finite extension of field F. The E is said to be normal extension of F if the
fixed field of G (E/F) is F itself.

. Moreover –

These statements are equivalent

(i) E is normal extension of F

(ii) F is the fixed field of Gal(E/F)
(iii) [E:F]= |Gal (E/F)|

Results

(i) Let E be a normal Extension of F and H be a subgroup of Gal (E/F); let

EH= {a ∊ E: σ (a) =a ∀ σ ∊ H} be the fixed field of H. Then
a. [E:EH]=|H|
b. H=Gal(E:EH)
(ii) E is a normal extension of F  E is the splitting field of some polynomial over F

33
Conjugate element:

let E be a finite extension of a field F, then two elements α and β of a field E are said to
be conjugate over F if they have the same minimal polynomial over F.

[1]FUNDAMENTAL THEOREM OF GALOIS THEORY

Statement:

Let F be a field of characteristic 0 or a finite field .If E is the splitting field over
F for some polynomial in F[x], then the mapping from the set of subfields of E containing
F to the set of subgroup of Gal(E/F) given by K→ Gal(E/K) is a one- to -one
correspondence. Furthermore for any subfield K of E containing F.

1. [E: K] = |Gal (E/K)| and [K: F] =| Gal (E/F)|/ |Gal (E/K)|.

[The index of Gal (E/K) in Gal (E/F) equals the degree of K over F]

2. If K is the splitting field of some polynomial in F[x], then Gal (E/K) is normal subgroup
of Gal (E/F) and Gal (K/F) is isomorphic to Gal (E/F)/ Gal (E/K).

3. K= Gal (E/K). [The fixed field of Gal (E/K) is K]

4. If H is a subgroup of Gal (E/F) ,then H= Gal(E/ )[The automorphism group of E

fixing is H]

34
Proof:

First we prove that there exists a one-one correspondence between the set of
subfields of E containing F to the set of subgroup of Gal (E/F) given by

K→ Gal (E/K).

Let K be any subfield of E containing F and Gal (E/K) be group of all

K-automorphism of E.

Since F K E, so that Gal (E/K) Gal (E/F). Also Gal (E/K) and Gal (E/F) are the
subgroups of the group of all automorphism of E; therefore Gal (E/K) is a subgroup of
Gal (E/F).

Thus for each subfield K of E containing, we can find a subgroup Gal (E/K) of Gal (E/F).

Consider a mapping of the set of all subfields of E containing F into the set of all
subgroups of Gal (E/F), defined by

Ψ (K) = Gal (E/K) for all subfield K of E containing F.

To prove ψ is one-one

Let and be any two subfields of E containing F and suppose that

Ψ( Ψ( )

⟹ Gal (E/ ) = Gal (E/ )

The fixed field of Gal (E/ the fixed field of Gal (E/ )

= ( E is a splitting over F

 E is a normal extension of F)

35
Now to prove ψ is onto:

Let H be an arbitrary subgroup of Gal (E/F), then the fixed field of H denoted by
is

= {aєE : (a)=a ∀ ∊ H}

Then,

H= Gal (E/ ).

This shows that each subgroup of Gal (E/F) is of the form Gal (E/ ) such that

F E[ is fixed field] and corresponding to this subgroup Gal (E/ ) there exist
a subfield of E containing F such that ψ ( )= Gal (E/ ).

i) E is a normal extension of F and K is a subfield of E containing F such that

F K E then E is normal extension of K therefore we have

[E: F]= | Gal (E/F)| and

[E: K]= | Gal (E/K)|.

Moreover

[E: F]= [E: K] [K: F]

⟹| Gal (E/F)|= | Gal (E/K)| [K: F].

⟹ [K: F] = | Gal (E/F)|/ | Gal (E/K)|.

36
2. K is the splitting field of some polynomial in F[x], means K is the normal extension of
F, to show Gal (E/K) is a normal subgroup of Gal (E/F).

For any ∊ Gal (E/F) and ψ ∊ Gal (E/K).

To show: ψ ∊ Gal (E/K)

Let be any arbitrary element of K .Since K is a normal extension of F, so that the

splitting field of the minimal polynomial of over F is contained in K and every

conjugate of is therefore in K.

Since ( ) is conjugate of for any ∊ Gal (E/F),

then ( ) ∊ K . Thus for any automorphism ψ ∊ Gal (E/K), ψ ( ( )) = ( )

Now

( ψ )( ) = [ψ ( ( ))]

= ( ( ))

⟹ ψ ∊ Gal (E/K) ∀ ∊ Gal (E/F)

Ψ ∊ Gal (E/K).

Let K is the normal extension of F. Let be any element of G (E|F).Define a mapping

’of K into E be splitting ’ ( ) = ( )∀ ∊ K. Since is an F- automorphism of E and

K is a normal extension of F, so that K = F ( ), therefore ’ is F –automorphism of K

37
i.e., ’ ∊ G (K|F)

Thus (K) = ’ (K) =K.

Now consider a mapping

: G (E|F) → G (K|F) by setting

( ) = ’ ∀ є G (E|F)

This mapping is a group homomorphism; for if and are any two elements of

G(E|F) and ∊ K, then

( ( ))( ) = (( )’) ( ) [By above define]

=( )( )

= ( ( ))

and ( ( ) ( )( ) = ( )( ( )( ))

=( ( ) )( ( ))

=( ( ) )( ( ))

= ’ ( ))

= ( ( ))

So, )= )( )

38
Consider any ψ ∊ Gal (K/F), then

Ψ ( ) is conjugate of over F, so there exists an F-automorphism σ of E

Such that σ ( ) = Ψ ( ).

Also σ and Ψ are both identity of F and K and K =F( ),

So that σ (a) = Ψ (a) a ∊ F ( ) = K.

ψ = σ’= ( )

Hence f is onto

Furthermore,

Kernel = { ∊ G (E|F) :( ) = , the identity of Gal (K/F)}

= { ∊ G (E|F): ’= }

= { ∊ G (E|F): ’ ( ) = ( ) = єK}

= { ∊ G (E|F) : ( )= єK}

= Gal (E/K)

Then by Fundamental theorem of homomorphism of groups

Gal (K/F) is isomorphic to Gal (E/F)/ Gal (E/K).

3. E is a normal extension of K (from above discussion)

So, by definition of normal extension the fixed field of G (E|K) is K.

39
4∵H is a subgroup of Gal (E/F) so that H Gal (E/F)

Also is the fixed field of H, then we have

= {a ∊ E | (a) =a ∊H}

∵ is subfield of E.

Since E is a normal extension of F, so that E is a finite extension of F

Gal (E/ ) |≤ [E: ] …………………(i)

Now Gal (E/ ) = { ∊A (K); (a) =a a∊ } ………………………(ii)

Let be any automorphism in H, then

∊H⟹ (b) =b b∊ [using ii]

⟹ ∊ Gal (E/ ) [using ii]

Therefore H Gal (E/ )

Therefore |H|≤| Gal (E/ )| …………………………. (iii)

From (i) and (iii)

|H|≤| Gal (E/ ) |≤ [E: ]

and |H|≤| Gal (E/ ) |≤ |H|

⟹|H|=| Gal (E/ )|

Also H is a subgroup of Gal (E/ )

Hence H=Gal (E/ ).

40
Summary

In the first chapter of project –I have discussed some fundamental concepts

which are related and useful in the Field Theory and Galois Theory like –Euclidean
domains, Principle ideal domain and Unique Factorization Domain (which has an
important role basically in the fundamental theorem of Field Theory) and relation among
them.

In the main discussion, i.e. from the 2nd chapter it is starting from definition,
different examples, and related theorem of Extentions Field, Splitting Fields, Existence
of Splitting Fields, Algebraic Extentions, Characteristics of Extentions, Degree of
extension and the properties of Algebraic Extentions.

In the last chapter (Galois Theory) – I discussed mainly Fundamental Theorem of

Galois Theory that’s show the intimate connection between the lattices of Subfield
between E (Extentions Field) and F (Field) and the lattice of subgroups of Gal (E/F).

41
References

[1] Joseph A. Gallian, Contemporary Abstract Algebra, New Delhi Narosa Publishing
House , 1999.

[2] David S. Dummit & Richard M.Foote , Abstract Algebra ,John Wiley & Sons,Inc.
New York.

[3] P.B.Bhattacharya, S.K. Jain,S.R.Nagpal- Basic Abstract Algebra Cambridge

University Press ,1995.

[4] John B. Fraleigh, A First Course In Abstract Algebra, Pearson Education (Singapur)
Pte.Ltd. , Indian Branch, 482 F.I.E. Patparganj,New Delhi.

[5] Bhupendra Singh, Advance Abstract Algebra Pragati Prakashan, Meerut 1999

42