Trending ▼   ResFinder  

galois theory

130 pages, 0 questions, 0 questions with responses, 0 total responses,    0    0
ranjitsran89
   		+Fave Message
ProfileTimelineUploads
 Home > ranjitsran89 >

Formatting page ...

An Introduction to Galois Theory Andrew Baker [14/11/2011] School of Mathematics & Statistics, University of Glasgow. E-mail address : a.baker@maths.gla.ac.uk URL: http://www.maths.gla.ac.uk/ ajb Q( 3 2, 3 ) Q( 3 2) N ggg n ggggg nnn ggggg nnnnn ggg 2 ggggg nnn 2 ggggg 3 Q( 2 ) pp pp pp 2 pp pp 2 pp3 pp Q( 3 2 3 ) pp I pp pp pp p H 3 yyy N ee ee yyy ee yyy ee yyy ee 3 yyy 3 ee yyy e yyy yyy eee yyy ee yyy ee e yyy 3 Q S kk kkkk 2 kkkk kk kkkk kkkk Q( 3 ) Q Gal(E/Q) S3 = ooo }} ooo }}} oo ooo3 }}} 3 oooo }} o 3 }} ooo } ooo }} oo } }} ooo }} ooo 2 {id, (1 2 3), (1 3 2)} w ww ww w ww 3 www w {id, (2 3)} {id, (1 2)} {id, (1 3)} ww ww ww ww 2 2 2 ww w {id} The Galois Correspondence for Q( 3 2, 3 )/Q Introduction: What is Galois Theory? Much of early algebra centred around the search for explicit formulae for roots of polynomial equations in one or more unknowns. The solution of linear and quadratic equations in a single unknown was well understood in antiquity, while formulae for the roots of general real cubics and quartics was solved by the 16th century. These solutions involved complex numbers rather than just real numbers. By the early 19th century no general solution of a general polynomial equation by radicals (i.e., by repeatedly taking n-th roots for various n) was found despite considerable e ort by many outstanding mathematicians. Eventually, the work of Abel and Galois led to a satisfactory framework for fully understanding this problem and the realization that the general polynomial equation of degree at least 5 could not always be solved by radicals. At a more profound level, the algebraic structure of Galois extensions is mirrored in the subgroups of their Galois groups, which allows the application of group theoretic ideas to the study of elds. This Galois Correspondence is a powerful idea which can be generalized to apply to such diverse topics as ring theory, algebraic number theory, algebraic geometry, di erential equations and algebraic topology. Because of this, Galois theory in its many manifestations is a central topic in modern mathematics. In this course we will focus on the following topics. The solution of polynomial equations over a eld, including relationships between roots, methods of solutions and location of roots. The structure of nite and algebraic extensions of elds and their automorphisms. We will study these in detail, building up a theory of algebraic extensions of elds and their automorphism groups and applying it to solve questions about roots of polynomial equations. The techniques we will meet can also be applied to study the following some of which may be met by people studying more advanced courses. Classic topics such as squaring the circle, duplication of the cube, constructible numbers and constructible polygons. Applications of Galois theoretic ideas in Number Theory, the study of di erential equations and Algebraic Geometry. There are many good introductory books on Galois Theory, some of which are listed in the Bibliography. In particular, [2, 3, 8] are all excellent sources and have many similarities to the present approach to the material. BY: A. J. Baker (2011) ii Contents Introduction: What is Galois Theory? ii Chapter 1. Integral domains, elds and polynomial rings Basic notions, convention, etc 1.1. Recollections on integral domains and elds 1.2. Polynomial rings 1.3. Identifying irreducible polynomials 1.4. Finding roots of complex polynomials of small degree 1.5. Automorphisms of rings and elds Exercises on Chapter 1 1 1 1 6 12 16 19 23 Chapter 2. Fields and their extensions 2.1. Fields and sub elds 2.2. Simple and nitely generated extensions Exercises on Chapter 2 27 27 29 33 Chapter 3. Algebraic extensions of elds 3.1. Algebraic extensions 3.2. Splitting elds and Kronecker s Theorem 3.3. Monomorphisms between extensions 3.4. Algebraic closures 3.5. Multiplicity of roots and separability 3.6. The Primitive Element Theorem 3.7. Normal extensions and splitting elds Exercises on Chapter 3 35 35 39 42 45 48 52 54 55 Chapter 4. Galois extensions and the Galois Correspondence 4.1. Galois extensions 4.2. Working with Galois groups 4.3. Subgroups of Galois groups and their xed elds 4.4. Sub elds of Galois extensions and relative Galois groups 4.5. The Galois Correspondence and the Main Theorem of Galois Theory 4.6. Galois extensions inside the complex numbers and complex conjugation 4.7. Galois groups of even and odd permutations 4.8. Kaplansky s Theorem Exercises on Chapter 4 57 57 58 60 61 62 64 65 68 71 Chapter 5. Galois extensions for elds of positive characteristic 5.1. Finite elds 73 73 iii 5.2. Galois groups of nite elds and Frobenius mappings 5.3. The trace and norm mappings Exercises on Chapter 5 77 79 80 Chapter 6. A Galois Miscellany 6.1. A proof of the Fundamental Theorem of Algebra 6.2. Cyclotomic extensions 6.3. Artin s Theorem on linear independence of characters 6.4. Simple radical extensions 6.5. Solvability and radical extensions 6.6. Symmetric functions Exercises on Chapter 6 83 83 84 88 90 92 96 97 Bibliography 101 Solutions Chapter Chapter Chapter Chapter Chapter Chapter 103 103 110 112 114 117 119 1 2 3 4 5 6 iv CHAPTER 1 Integral domains, elds and polynomial rings Basic notions, convention, etc In these notes, a ring will always be a unital ring, i.e., a ring with unity 1 = 0. Most of the rings encountered will also be commutative. An ideal I R will always mean a two-sided ideal. An ideal I R in a ring R is proper if I = R, or equivalently if I R. Under a ring homomorphism : R S , 1 R is sent to 1 S , i.e., (1) = 1. 1.1. Definition. Let : R S be a ring homomorphism. is a monomorphism if it is injective, i.e., if for r1 , r2 R, = (r1 ) = (r2 ) r1 = r2 , or equivalently if ker = {0}. is an epimorphism if it is surjective, i.e., if for every s S there is an r R with (r) = s. is an isomorphism if it is both a monomorphism and an epimorphism, i.e., if it is invertible (in which case its inverse is also an isomorphism). 1.1. Recollections on integral domains and elds The material in this section is standard and most of it should be familiar. Details may be found in [3, 5] or other books containing introductory ring theory. First we recall some important properties of elements in a ring. 1.2. Definition. Let R be a ring. An element u R is a unit if it is invertible, i.e., there is an element v R for which uv = 1 = vu. We usually write u 1 for this element v , which is necessarily unique and is called the (multiplicative ) inverse of u in R. We will denote the set of all invertible elements of R by R and note that it always forms a group under multiplication. 1.3. Definition. Let R be a commutative ring. Then a non-zero element z R is a zero-divisor if there is a non-zero element w R for which zw = wz = 0. A commutative ring R in which there are no zero-divisors is called an integral domain or an entire ring. This means that for u, v R, uv = 0 = u = 0 or v = 0. 1.4. Example. The following rings are integral domains. (i) The ring of integers, Z. 1 (ii) If p is a prime, the ring of integers modulo p, Fp = Z/p = Z/(p). (iii) The rings of rational numbers, Q, real numbers, R, and complex numbers, C. (iv) The polynomial ring R[X ], where R is an integral domain; in particular, the polynomial rings Z[X ], Q[X ], R[X ] and C[X ] are all integral domains. 1.5. Definition. Let I R be a proper ideal in a commutative ring R. I is a prime ideal if for u, v R, uv I = u I or v I. I is a maximal ideal R if whenever J R is a proper ideal and I J then J = I . I R is principal if I = (p) = {rp : r R} for some p R. Notice that if p, q R, then (q ) = (p) if and only if q = up for some unit u R. We also write p | x if x (p). p R is prime (or is a prime ) if (p) R is a prime ideal; this is equivalent to the requirement that whenever p | xy with x, y R then p | x or p | y . R is a principal ideal domain if it is an integral domain and every ideal I R is principal. A non-zero element p R is irreducible (or is an irreducible ) if for u, v R, p = uv = u or v is a unit. 1.6. Example. Every ideal I Z is principal, so I = (n) for some n Z which we can always take to be non-negative, i.e., n 0. Hence Z is a principal ideal domain. 1.7. Proposition. Let R be a commutative ring and I R an ideal. (i) The quotient ring R/I is an integral domain if and only if I is a prime ideal. (ii) The quotient ring R/I is a eld if and only if I is a maximal ideal. 1.8. Example. If n n is a prime. 0, the quotient ring Z/n = Z/(n) is an integral domain if and only if For any (not necessarily commutative) ring with unity there is an important ring homomorphism : Z R called the unit or characteristic homomorphism which is de ned by 1 + + 1 if n > 0, n (n) = n1 = (1 + + 1) if n < 0, n 0 if n = 0. Since 1 R is non-zero, ker Z is a proper ideal and using the Isomorphism Theorems we see that there is a quotient monomorphism : Z/ ker R which allows us to identify the quotient ring Z/ ker with the image Z R as a subring of R. By Example 1.6, there is a unique non-negative integer p 0 such that ker = (p); this p is called the characteristic of R and denoted char R. 1.9. Lemma. If R is an integral domain, its characteristic char R is a prime. 2 Proof. Consider p = char R. If p = 0 we are done. So suppose that p > 0. The quotient monomorphism : Z/ ker R identi es Z/ ker with the subring im = im of the integral domain R. But every subring of an integral domain is itself an integral domain, hence Z/ ker is an integral domain. Now by Proposition 1.7(i), ker = (p) is prime ideal and so by Example 1.8, p is a prime. 1.10. Remark. When discussing a ring with unit R, we can consider it as containing as a subring of the form Z/(char R) since the quotient homomorphism : Z/(char R) R gives an isomorphism Z/(char R) im , allowing us to identify these rings. In particular, every integral domain contains as a subring either Z = Z/(0) (if char R = 0) or Z/(p) if p = char R > 0 is a non-zero prime. This subring is sometimes called the characteristic subring of R. The rings Z and Z/n = Z/(n) for n > 0 are often called core rings. When considering integral domains, the rings Z and Fp = Z/p = Z/(p) for p > 0 a prime are called prime rings. Here is a useful and important fact about rings which contain a nite prime ring Fp . 1.11. Theorem (Idiot s Binomial Theorem). Let R be a commutative ring containing Fp for some prime p > 0. If u, v R, then (u + v )p = up + v p . Proof. We have p1 = 0 in R, hence pt = 0 for any t R. The Binomial Expansion yields () () ( ) p p 1 p p 2 2 p p p (1.1) (u + v ) = u + u v+ u v + + uv p 1 + v p . 1 2 p 1 Now suppose that 1 j p 1. Then we have () p p (p 1)! (p 1)! = =p . j j ! (p j )! j ! (p j )! There are no factors of p appearing in (p 1)!, j ! or (p j )!, so since this number is an integer it must be divisible by p, i.e., () p (1.2a) p| , j or equivalently (1.2b) () p 0 (mod p). j () p 1 = 0. j Combining the divisibility conditions of (1.2) with the expansion of (1.1), we obtain the required equation in R, (u + v )p = up + v p . Hence in R we have 1.12. Definition. A commutative ring k is a eld if every non-zero element u k is a unit. This is equivalent to requiring that k = k {0}. The familiar rings Q, R and C are all elds. 1.13. Example. If n 1, the quotient ring Z/n is a eld if and only if n is a prime. 1.14. Proposition. Every eld is an integral domain. 3 u 1 Proof. Let k be a eld. Suppose that u, v k and uv = 0. If u = 0, we can multiply by to obtain v = u 1 uv = 0, hence v = 0. So at least one of u, v must be 0. 1.15. Lemma. Let R be an integral domain. If p R is a non-zero prime then it is irreducible. Proof. Suppose that p = uv for some u, v R. Then p | u or p | v , and we might as well assume that u = tp for some t R. Then (1 tv )p = 0 and so tv = 1, showing that v is a unit with inverse t. Now let D be an integral domain. A natural question to ask is whether D is isomorphic to a subring of a eld. This is certainly true for the integers Z which are contained in the eld of rational numbers Q, and for a prime p > 0, the prime ring Fp is itself a eld. 1.16. Definition. The elds Q and Fp where p > 0 is a prime are the prime elds. Of course, we can view Z as a subring of any sub eld of the complex numbers so an answer to this question may not be unique! However, there is always a smallest such eld which is unique up to an isomorphism. 1.17. Theorem. Let D be an integral domain. (i) There is a eld of fractions of D, Fr(D), which contains D as a subring. (ii) If : D F is a ring monomorphism into a eld F , there is a unique homomorphism : Fr(D) F such that (t) = (t) for all t D Fr(D). D inc /F < ! Fr(D) Proof. (i) Consider the set P(D) = {(a, b) : a, b D, b = 0}. Now introduce an equivalence relation on P(D), namely (a , b ) (a, b) ab = a b. Of course, it is necessary to check that this relation is an equivalence relation; this is left as an exercise. We denote the equivalence class of (a, b) by [a, b] and the set of equivalence classes by Fr(D). We de ne addition and multiplication on Fr(D) by [a, b] + [c, d] = [ad + bc, bd], [a, b][c, d] = [ac, bd]. We need to verify that these operations are well de ned. For example, if [a , b ] = [a, b] and [c , d ] = [c, d], then (a d + b c )bd = a d bd + b c bd = ab d d + b bcd = (ad + bc)b d , 4 and so (a d + b c , b d ) (ad + bc, bd); hence addition is well de ned. A similar calculation shows that (a c , b d ) (ac, bd), so multiplication is also well de ned. It is now straightforward to show that Fr(D) is a commutative ring with zero 0 = [0, 1] and unit 1 = [1, 1]. In fact, as we will soon see, Fr(D) is a eld. Let [a, b] Fr(D). Then [a, b] = [0, 1] if and only if (0, 1) (a, b) which is equivalent to requiring that a = 0; notice that for any b = 0, [0, b] = [0, 1]. We also have [a, b] = [1, 1] if and only if a = b. Now let [a, b] Fr(D) be non-zero, i.e., a = 0. Then b = 0 and [a, b], [b, a] Fr(D) satisfy [a, b][b, a] = [ab, ba] = [1, 1] = 1, so [a, b] has [b, a] as an inverse. This shows that Fr(D) is a eld. We can view D as a subring of Fr(D) using the map j : D Fr(D); j (t) = [t, 1] which is a ring homomorphism; it is easy to check that it is a monomorphism. Therefore we may identify t D with j (t) = [t, 1] Fr(D) and D with the subring im j Fr(D). (ii) Consider the function : P(D) F ; (a, b) = (a) (b) 1 . If (a , b ) (a, b), then (a , b ) = (a ) (b ) 1 = (a ) (b) (b) 1 (b ) 1 = (a b) (b) 1 (b ) 1 = (ab ) (b ) 1 (b) 1 = (a) (b ) (b ) 1 (b) 1 = (a) (b) 1 = (a, b), so is constant on each equivalence class of . Hence we may de ne the function : Fr(D) F ; ([a, b]) = (a, b). It is now easy to verify that is a ring homomorphism which agrees with on D Fr(D). The next three corollaries are left as an exercise. 1.18. Corollary. If F is a eld then F = Fr(F ). 1.19. Corollary. If D is a subring of a eld F , then Fr(D) Fr(F ) = F and Fr(D) is the smallest sub eld of F containing D. 1.20. Corollary. Let D1 and D2 be integral domains and let : D1 D2 be a ring monomorphism. Then there is a unique induced ring homomorphism : Fr(D1 ) Fr(D2 ) which satis es (t) = (t) whenever t D1 Fr(D1 ). D1 inc Fr(D1 ) / D2 inc / Fr(D2 ) Moreover, this construction has the following properties. 5 If : D1 D2 and : D2 D3 are monomorphisms between integral domains then = ( ) as homomorphisms Fr(D1 ) Fr(D3 ). For any integral domain D, the identity homomorphism id : D D induces the identity homomorphism (id) = id : Fr(D) Fr(D). D1 / D2 inc Fr(D1 ) / D3 inc / Fr(D2 ) D inc / Fr(D3 ) id /D inc inc / Fr(D) Fr(D) id = id 1.21. Remarks. (a) When working with a eld of fractions we usually adopt the familiar notation a = a/b = [a, b] b for the equivalence class of (a, b). The rules for algebraic manipulation of such symbols are the usual ones for working with fractions, i.e., a1 a2 a1 b2 + a2 b1 + = , b1 b2 b1 b2 a1 a2 a1 a2 a1 a2 = = . b1 b2 b1 b2 b1 b2 The eld of fractions of an integral domain is sometimes called its eld of quotients, however as the word quotient is also associated with quotient rings we prefer to avoid using that terminology. (b) Corollary 1.20 is sometimes said to imply that the construction of Fr(D) is functorial in the integral domain D. 1.2. Polynomial rings Let R be a commutative ring. We will make frequent use of the ring R[X ] of polynomials over R in an indeterminate X . This consists of elements of form p(X ) = p0 + p1 X + + pm X m where m 0 and p0 , p1 , . . . , pm R; such p(X ) are called polynomials. Addition and multiplication in R[X ] are de ned by (p0 + p1 X + + pm X m ) + (q0 + q1 X + + qm X m ) = (p0 + q0 ) + (p1 + q1 )X + + (pm + qm )X m , and (p0 + p1 X + + pm X m )(q0 + q1 X + + qm X m ) = (p0 q0 ) + (p0 q1 + p1 q0 )X + + (p0 qm + p1 qm 1 + + pm 1 q1 + pm q0 )X 2m . Then R[X ] is a commutative ring with the constant polynomials 0 and 1 as its zero and unit. We identify r R with the obvious constant polynomial; this allows us to view R as a subring of R[X ] and the inclusion function inc : R R[X ] is a monomorphism. More generally, we inductively can de ne the ring of polynomials in n indeterminates X1 , . . . , Xn over R, R[X1 , . . . , Xn ] = R[X1 , . . . , Xn 1 ][Xn ] 6 for n 1. Again there is an inclusion monomorphism inc : R R[X1 , . . . , Xn ] which sends each element of R to itself considered as a constant polynomial. These polynomial rings have an important universal property. 1.22. Theorem (Homomorphism Extension Property). Let : R S be a ring homomorphism. (i) For each s S there is a unique ring homomorphism s : R[X ] S for which s (r) = (r) for all r R, s (X ) = s. R inc /S = ! s R[X ] (ii) For n 1 and s1 , . . . , sn S , there is a unique ring homomorphism s1 ,...,sn : R[X1 , . . . , Xn ] S for which s1 ,...,sn (r) = (r) for all r R, s1 ,...,sn (Xi ) = si for i = 1, . . . , n. R inc 8/ S ! s1 ,...,sn R[X1 , . . . , Xn ] Proof. (Sketch) (i) For a polynomial p(X ) = p0 + p1 X + + pm X m R[X ], we de ne (1.3) s (p(X )) = p0 + p1 s + + pm sm S. It is then straightforward to check that s is a ring homomorphism with the stated properties and moreover is the unique such homomorphism. (ii) is proved by induction on n using (i). We will refer to s1 ,...,sn as the extension of by evaluation at s1 , . . . , sn . It is standard to write p(s1 , . . . , sn ) = s1 ,...,sn (p(X1 , . . . , Xn )). An important special case occurs when we start with the identity homomorphism id : R R and r1 , . . . , rn R; then we have the homomorphism r1 ,...,rn = idr1 ,...,rn : R[X1 , . . . , Xn ] R. Slightly more generally we may take the inclusion of a subring inc : R S and s1 , . . . , sn S ; then s1 ,...,sn = incs1 ,...,sn : R[X1 , . . . , Xn ] S is called evaluation at s1 , . . . , sn and we denote its image by R[s1 , . . . , sn ] = s1 ,...,sn R[X1 , . . . , Xn ] S. 7 Then R[s1 , . . . , sn ] is a subring of S , called the subring generated by s1 , . . . , sn over R. Here is an example illustrating how we will use such evaluation homomorphisms. 1.23. Example. Consider the inclusion homomorphism inc : Q C. We have the evaluation at i homomorphism i , for which i (X ) = i. We easily see that i Q[X ] C is a subring Q[i] C consisting of the complex numbers of form a + bi with a, b Q. Notice that if we had used i instead of i, evaluation at i, i , we would also have i Q[X ] = Q[i]. These evaluation homomorphisms are related by complex conjugation since i (p(X )) = i (p(X )), which is equivalent to the functional equation i = ( ) i . Notice also that in these examples we have ker i = ker i = (X 2 + 1) Q[X ], hence we also have Q[i] Q[X ]/(X 2 + 1). = In fact (X 2 + 1) is actually a maximal ideal and so Q[i] C is a sub eld; later we will write Q(i) for this sub eld. 1.24. Proposition. Let R be an integral domain. (i) The ring R[X ] of polynomials in an indeterminate X over R is an integral domain. (ii) The ring R[X1 , . . . , Xn ] of polynomials in the indeterminates X1 , . . . , Xn over R is an integral domain. 1.25. Corollary. Let k be a eld and n 1. Then the polynomial ring k[X1 , . . . , Xn ] in the indeterminates X1 , . . . , Xn is an integral domain. As we will make considerable use of such rings we describe in detail some of their important properties. First we recall long division in a polynomial ring k[X ] over a eld k; full details can be found in a basic course on commutative rings or any introductory book on this subject. 1.26. Theorem (Long Division). Let k be a eld. Let f (X ), d(X ) k[X ] and assume that d(X ) = 0 so that deg d(X ) > 0. Then there are unique polynomials q (X ), r(X ) k[X ] for which f (X ) = q (X )d(X ) + r(X ) and either deg r(X ) < deg d(X ) or r(X ) = 0. In the situation discussed in this result, the following names are often used. We refer to the process of nding q (X ) and r(X ) as long division of f (X ) by d(X ). Also, f (X ) = the dividend , d(X ) = the divisor , q (X ) = the quotient , r(X ) = the remainder . 1.27. Example. For k = Q, nd the quotient and remainder when f (X ) = 6X 4 6X 3 + 3X 2 3X + 1 is divided by d(X ) = 2X 2 + 1. 8 Solution. In the usual notation we have the following calculation. 3X 2 3X 2X 2 + 1 | 6X 4 6X 3 + 3X 2 3X + 1 6X 4 + 0X 3 + 3X 2 + 0X + 0 6X 3 + 0X 2 3X + 1 6X 3 + 0X 2 3X + 0 1 Hence 6X 4 6X 3 + 3X 2 3X + 1 = (3X 2 3X )(2X 2 + 1) + 1, giving q (X ) = 3X 2 3X and r(X ) = 1. 1.28. Example. For k = F5 , nd the quotient and remainder when f (X ) = 10X 5 + 6X 4 6X 3 + 3X 2 3X + 1 is divided by d(X ) = 2X 2 + 1. Solution. First notice that working modulo 5 we have f (X ) = 10X 5 + 6X 4 6X 3 + 3X 2 3X + 1 X 4 + 4X 3 + 3X 2 + 2X + 1 (mod 5). Notice also following multiplicative inverses in F5 : 2 1 3 (mod 5), 3 1 2 (mod 5), 4 1 4 (mod 5). We have the following calculation. 3X 2 + 2X 2X 2 + 1| 6X 4 + 4X 3 + 3X 2 + 2X + 1 6X 4 + 0X 3 + 3X 2 + 0X + 0 4X 3 + 0X 2 + 2X + 1 4X 3 + 0X 2 + 2X + 0 1 Hence 6X 4 6X 3 + 3X 2 3X + 1 (3X 2 + 2X )(2X 2 + 1) + 1 (mod 5), giving q (X ) = 3X 2 + 2X and r(X ) = 1. An important consequence of Theorem 1.26 is the following which makes use of the Euclidean Algorithm. 1.29. Corollary. Let k be a eld and X an indeterminate. Let f (X ), g (X ) k[X ] be non-zero. Then there are a(X ), b(X ) k[X ] such that a(X )f (X ) + b(X )g (X ) = gcd(f (X ), g (X )). Here the greatest common divisor gcd(f (X ), g (X )) of f (X ), g (X ) is the monic polynomial of greatest degree which divides both of f (X ), g (X ). 9 1.30. Proposition. Let k be a eld and X an indeterminate. Then a non-constant polynomial p(X ) k[X ] is irreducible if and only if it is a prime. Proof. By Lemma 1.15 we already know that p(X ) is irreducible if it is prime. So suppose that p(X ) is irreducible and that p(X ) | u(X )v (X ) for u(X ), v (X ) k[X ]. Then by Corollary 1.29, there are a(X ), b(X ) k[X ] such that a(X )p(X ) + b(X )u(X ) = gcd(p(X ), u(X )). But since p(X ) is irreducible, gcd(p(X ), u(X )) = p(X ) or gcd(p(X ), u(X )) = 1. In the latter case, a(X )p(X ) + b(X )u(X ) = 1, and multiplying through by v (X ) gives a(X )p(X )v (X ) + b(X )u(X )v (X ) = v (X ) and so p(X ) | v (X ). This shows that p(X ) | u(X ) or p(X ) | v (X ), and so p(X ) is prime. 1.31. Theorem. Let k be a eld and X an indeterminate. (i) Every ideal I k[X ] is principal, i.e., I = (h(X )) for some h(X ) k[X ]. (ii) The ideal (p(X )) k[X ] is prime if and only if p(X ) = 0 or p(X ) is irreducible in k[X ]. (iii) The quotient ring k[X ]/(p(X )) is an integral domain if and only if p(X ) = 0 or p(X ) is irreducible in k[X ]. (iv) The quotient ring k[X ]/(p(X )) is a eld if and only if p(X ) is an irreducible in k[X ]. Proof. (i) Let I k[X ] and assume that I = (0). Then there must be at least one element of I with positive degree and so we can choose h(X ) I of minimal degree, say d = deg h(X ). Now let p(X ) I . By Long Division, there are q (X ), r(X ) k[X ] such that p(X ) = q (X )h(X ) + r(X ) and deg r(X ) < d or r(X ) = 0. Since p(X ) and h(X ) are in the ideal I , we also have r(X ) = p(X ) q (X )h(X ) I. If r(X ) = 0, this would contradict the minimality of d, so we must have r(X ) = 0, showing that p(X ) = q (X )h(X ). Thus I (p(X )) I and therefore I = (p(X )). (ii) This follows from Proposition 1.30. (iii) This follows from Proposition 1.7(i). (iv) Since k[X ] is an integral domain and not a eld, it follows that if k[X ]/(p(X )) is a eld then because it is an integral domain, p(X ) is an irreducible by (iii). Suppose that p(X ) is irreducible (and hence is non-zero). Then for any q (X ) k[X ] with q (X ) (p(X )), by Corollary 1.29 we can nd suitable a(X ), b(X ) k[X ] for which / a(X )p(X ) + b(X )q (X ) = gcd(p(X ), q (X )). But gcd(p(X ), q (X )) = 1 since p(X ) is irreducible, so a(X )p(X ) + b(X )q (X ) = 1. This shows that in the quotient ring k[X ]/(p(X )) the residue class of q (X ) has the residue class of b(X ) as its inverse. 10 1.32. Remark. In connection with Theorem 1.31(i), notice that if p(X ) k[X ], then provided d = deg p(X ) > 0, we have for some pd = 0, p(X ) = p0 + p1 X + + pd X d = pd q (X ), where q (X ) = p 1 p0 + p 1 p1 X + + p 1 pd 1 X d 1 + X d . d d d This easily implies that as ideals of k[X ], (p(X )) = (q (X )). So we can always nd a monic polynomial as the generator of a given ideal, and this monic polynomial is unique. 1.33. Proposition (Unique Factorization Property). Every non-constant polynomial f (x) k[X ] has a factorization f (x) = cp1 (X ) pk (X ), where c k, and p1 (X ), . . . , pk (X ) k[X ] are irreducible monic polynomials. Moreover, c is unique and the sequence of polynomials p1 (X ), . . . , pk (X ) is unique apart from the order of the terms. Proof. (Sketch) Existence is proved by induction on the degree of f (X ) and begins with the obvious case deg f (X ) = 1. If deg f (X ) > 1, then either f (X ) is already irreducible, or f (X ) = f1 (X )f2 (X ) with both factors of positive degree, and therefore deg fj (X ) < deg f (X ). This gives the inductive step. To prove uniqueness, suppose that p1 (X ) pk (X ) = q1 (X ) q (X ) where pi (X ), qj (X ) k[X ] are irreducible monic polynomials. Then by Proposition 1.30, each pi (X ) is prime, hence divides one of the qj (X ), hence must equal it. By reordering we can assume that pi (X ) = qi (X ) and k . After cancelling common factors we obtain qk+1 (X ) q (X ) = 1, and so we see that k = . 1.34. Corollary. Suppose that f (X ) k[X ] factors into linear factors f (X ) = c(X u1 ) (X ud ), where u1 , . . . , ud k. Then the sequence of roots u1 , . . . , ud is unique apart from the order. In particular, if v1 , . . . , vr are the distinct roots, then f (X ) = c(X v1 )m1 (X vr )mr , where mi > 0 and this factorization is unique apart from the order of the pairs (vi , mi ). 1.35. Corollary. The number of distinct roots of a non-constant polynomial f (X ) k[X ] is at most deg f (X ). 1.36. Definition. If k is a eld and X an indeterminate, then the eld of fractions of k[X ] is the eld of rational functions, k(X ). The elements of k(X ) are fractions of the form a0 + a1 X + + am X m b0 + b1 X + + bn X n with ai , bj k and b0 + b1 X + + bn X n = 0. 11 1.3. Identifying irreducible polynomials When k is a eld, we will need some e ective methods for deciding when a polynomial in k[X ] is irreducible. Let us consider factorisation of polynomials over Q. If f (X ) Z[X ] then we can also consider f (X ) as an element of Q[X ]. If R = Z or Q, we say that f (X ) has a proper factorisation over R if f (X ) = g (X )h(X ) for some g (X ), h(X ) R[X ] with deg g (X ) > 0 and deg h(X ) > 0. 1.37. Proposition (Gauss s Lemma). Let f (X ) Z[X ]. Then f (X ) has a proper factorisation over Z if and only it has a proper factorisation over Q. So to nd factors of f (X ) it is su cient to look for factors in Z[X ]. Our next result is a special case of the Eisenstein Irreducibility Test. The version here is slightly more general than the more usual one which corresponds to taking s = 0. 1.38. Proposition (Eisenstein Test). Let f (X ) Z[X ] and s Z. Choose ai Z so that f (X ) = a0 + a1 (X s) + + ad 1 (X s)d 1 + ad (X s)d , where d = deg f (X ). Suppose that p > 0 is a prime for which the following three conditions hold: ak 0 (mod p) for k = 0, . . . , d 1; a0 0 (mod p2 ); ad 0 (mod p). Then f (X ) is irreducible in Q[X ] and hence also in Z[X ]. 1.39. Example. Let p 2 be a prime. Then the polynomial p (X ) = 1 + X + + X p 1 Z[X ] is irreducible in Q[X ] and hence also in Z[X ]. Proof. Working in Z[X ], p (X )(X 1) = (1 + X + + X p 1 )(X 1) = Xp 1 = (1 + (X 1))p 1 p (p) = (X 1)k k k=1 (X 1)p since by (1.2a), p divides (mod p), () p p! = k ! (p k )! k when k = 1, . . . , p 1. Hence p (X ) (X 1)p 1 Also, (mod p) () p = p 0 (mod p2 ), 1 12 giving p (X ) = (X 1)p 1 + cp 2 (X 1)p 2 + + c1 (X 1) + c0 (1.4) with cr 0 (mod p) and c0 = p. So the Eisenstein Test can be applied here with s = 1 to show that p (X ) is irreducible in Z[X ]. 1.40. Example. As examples we have the irreducible polynomials 2 (X ) = 1 + X, 3 (X ) = 1 + X + X 2 , 5 (X ) = 1 + X + X 2 + X 3 + X 4 , 7 (X ) = 1 + X + X 2 + X 3 + X 4 + X 5 + X 6 , 11 (X ) = 1 + X + X 2 + X 3 + X 4 + X 5 + X 6 + X 7 + X 8 + X 9 + X 10 . These are examples of the cyclotomic polynomials n (X ) Z[X ] which are de ned for all n 1 by (1.5a) Xn 1 = d (X ), d|n where the product is taken over all the positive divisors of n. For example, X 2 1 = (X 1)(X + 1) = 1 (X ) 2 (X ), X 3 1 = (X 1)(X 2 + X + 1) = 1 (X ) 3 (X ), X 4 1 = (X 1)(X + 1)(X 2 + 1) = 1 (X ) 2 (X ) 4 (X ), X 5 1 = (X 1)(X 4 + X 3 + X 2 + X + 1) = 1 (X ) 5 (X ), X 6 1 = (X 1)(X + 1)(X 2 + X + 1)(X 2 X + 1) = 1 (X ) 2 (X ) 3 (X ) 6 (X ), X 12 1 = (X 1)(X + 1)(X 2 + X + 1)(X 2 + 1)(X 2 X + 1)(X 4 X 2 + 1) = 1 (X ) 2 (X ) 3 (X ) 4 (X ) 6 (X ) 12 (X ). Cyclotomic polynomials can be computed recursively using Equation (1.5a). If we know k (X ) for k < n, then (1.5b) Xn 1 n (X ) = . d (X ) d|n d<n The degree of n (X ) involves a function of n probably familiar from elementary Number Theory. 1.41. Definition. The Euler function : N N is de ned by (n) = number of k = 1, . . . , n for which gcd(n, k ) = 1 = |(Z/n) | = number of units in Z/n = number of generators of the cyclic group Z/n. In particular, if p 2 is a prime then (p) = p 1. Of course, (1) = 1. 13 It can be shown that for each natural number n, (1.6) (d) = n. d|n Notice that we can inductively determine (n) using this equation. For example, if p and q are distinct primes, then (pq ) = pq ( (p) + (q ) + (1)) = pq (p 1) (q 1) 1 = (p 1)(q 1). It is also true that whenever m, n are coprime, i.e., when gcd(m, n) = 1, (1.7) (mn) = (m) (n). Thus if n = pr1 prs where p1 < p2 < < ps are the prime factors of n and rj > 0, then s 1 (1.8) (n) = (pr1 ) (prs ). s 1 Furthermore, if p is a prime and r > 0, then (pr ) = (p 1)pr 1 . (1.9) Notice that as a result, (n) is even when n > 2. 1.42. Remark. For those who know about the M bius function (which takes values 0, 1) o and M bius inversion, the latter can be used to solve Equation (1.6) for , giving o n (1.10) (n) = (d) . d d|n Similarly, the formulae of (1.5) lead to (1.11) n ( X ) = (X n/d 1) (d) . d|n So for example, if p, q are distinct primes, then using standard properties of , pq (X ) = (X pq 1) (1) (X pq/p 1) (p) (X pq/q 1) (q) (X pq/pq 1) (pq) = (X pq 1)(X q 1) 1 (X p 1) 1 (X 1) = (X pq 1)(X 1) . (X q 1)(X p 1) Recall that an element of a eld K is a primitive n-th root of unity if min{k : 1 k and k = 1} = n. We think of n = e2 i/n as the standard complex primitive n-th root of unity. Then every k complex n-th root of unity has the form n = e2 ik/n for k = 0, 1, . . . , n 1. 1.43. Theorem. For each n 1, the cyclotomic polynomial n (X ) is irreducible in Q[X ] and hence in Z[X ]. The complex roots of n (X ) are the primitive n-th roots of unity, k n = e2 ik/n (0 n 1, gcd(k, n) = 1). k and the number of these is deg n (X ) = (n). Hence, t n (X ) = (X n ). t=1,...,n 1 gcd(t,n)=1 Proof. We will give a reformulation and proof of this in Theorem 6.2. 14 1.44. Example. For n = 6 we have 6 = e 2 i/6 =e i/3 3 1 =+ i. 2 2 Then (6) = 2 and 5 6 (X ) = X 2 X + 1 = (X 6 )(X 6 ). It is also worth recording a related general result on cyclic groups. 1.45. Proposition. Let n 1 and C = g be a cyclic group of order n and a generator g . Then an element g r C is a generator if and only if gcd(r, n) = 1; the number of such elements of C is (n). This leads to a useful group theoretic result. 1.46. Lemma. Let G be a nite group satisfying the following condition: For each n 1, there are at most n solutions of xn = in G. Then G is cyclic and in particular is abelian. Proof. Let G (d) denote the number of elements in G of order d. By Lagrange s Theorem, G (d) = 0 unless d divides |G|. Since G= {g G : |g | = d}, d||G| we have |G| = G (d). d||G| Recall the Euler -function satis es Equation (1.6), hence |G| = (d). d||G| Combining these we obtain (1.12) G (d) = d||G| (d). d||G| Let d be a divisor of |G|. By Proposition 1.45, for each element g G of order d, the cyclic subgroup g G has (d) generators, each of order d. As there are at most d such elements g in G, this gives G (d) (d). So G (d) (d). d||G| d||G| Now if G (d) < (d) for some d, we would have a strict inequality in place of Equation (1.12). Hence G (d) = (d) for all d. In particular, there are (|G|) elements of order |G|, hence there must be an element of order |G|, so G is cyclic. The above results for polynomials over Q and Z have analogues over the eld of fractions k(T ) and polynomial ring k[T ], where k is a eld. A polynomial f (X ) k[T ][X ] is an element of k(T )[X ]. If R = k[T ] or k(T ), we say that f (X ) has a proper factorisation over R if f (X ) = g (X )h(X ) for some g (X ), h(X ) R[X ] with deg g (X ) > 0 and deg h(X ) > 0. 15 1.47. Proposition (Gauss s Lemma). Let f (X ) k[T ][X ]. Then f (X ) has a proper factorisation over k[T ] if and only it has a proper factorisation over k(T ). Here is another version of the Eisenstein Test ; again we state a version which is slightly more general than the usual one which corresponds to the case where s = 0. 1.48. Proposition (Eisenstein Test). Let f (X ) k[T ][X ] and s k[T ]. Choose ai k[T ] so that f (X ) = a0 + a1 (X s) + + ad 1 (X s)d 1 + ad (X s)d , where d = deg f (X ). Suppose that p(T ) k[T ] is an irreducible for which the following three conditions hold: ak 0 (mod p(T )) for k = 0, . . . , d 1; a0 0 (mod p(T )2 ); ad 0 (mod p(T )). Then f (X ) is irreducible in k(T )[X ] and hence also in k[T ][X ]. 1.49. Example. Let k be a eld. Then the polynomial X n T is irreducible in k(T )[X ]. 1.4. Finding roots of complex polynomials of small degree In this section we work within the complex numbers and take k C. In practice we will usually have k = R or k = C. For monic linear (degree 1) or quadratic (degree 2) polynomials, methods of nding roots are very familiar. Let us consider the cases of cubic (degree 3) and quartic (degree 4) polynomials. Cubic polynomials: Cardan s method. The following 16th century method of nding roots of cubics is due to Jer me Cardan who seems to have obtained some preliminary versions o from Niccol` Tartaglia by somewhat disreputable means! For historical details see [2, 3]. o A monic cubic f (X ) = X 3 + a2 X 2 + a1 X + a0 C[X ] can be transformed into one with no quadratic term by a change of variables X X a2 /3 giving ) ( ) ( a1 a2 2a3 12 3 2 g (X ) = f (X a2 /3) = X a1 a2 X + a0 + C[X ]. 3 3 27 Clearly nding the roots of f (X ) is equivalent to nding those of g (X ), so we may as well assume that we want to nd the complex roots of f (X ) = X 3 + pX + q C[X ]. Suppose that x C is a root of f (X ), i.e., x3 + px + q = 0. (1.13) If we introduce u C for which x=u then p , 3u ( ( p) p )3 +p u +q =0 u 3u 3u 16 and so u3 p3 + q = 0, 27u3 hence u6 + qu3 Solving for u3 we obtain q1 u = 22 3 where equation q2 + p3 = 0. 27 q2 + 4p3 , 27 4 p3 denotes one of the complex square roots of the discriminant of the quadratic 27 p3 = 0. 27 Now if we take u to be a cube root of one of the complex numbers q1 4 p3 q2 + 22 27 we obtain the desired root of f (X ) as x = u p/3u. Notice that we have a choice of 2 values for u3 and for each of these a choice of 3 values for u, di ering by factors of the form r for r = 0, 1, 2 where = e2 i/3 is a primitive cube root of 1. However, since ( ) 4p3 3 4p q q 2 + q q 2 + 27 1 27 = 27 =2 , 2 + 4p3 /27) 3 q (q 4p 4 p3 q + q 2 + 27 it is easy to verify that there are in fact only 3 choices of the root x which we can write symbolically as 3 3 3 q1 4p q1 4p3 2+ x= + (1.14) q + q2 + 22 27 22 27 U 2 + qU or more precisely as (1.15) x= 3 q1 + 22 q2 + 4 p3 p . 27 3 q1 4p 3 3 + q2 + 22 27 1.50. Example. Find the complex roots of the polynomial f (X ) = X 3 + 3X 10 R[X ]. Solution. Applying the method above, we reduce to the quadratic equation U 2 10U 1 = 0 whose roots are 5 26 R. Notice that 5 + 26 > 0 and 5 26 < 0; we also have 1 . 5 26 = 5 + 26 Now 5 + 26 has the complex cube roots 3 3 3 5 + 26, 5 + 26 , 5 + 26 2 . 17 Here we have x = u 1/u, so the 3 complex roots of f (X ) are ) ( 1 3 5 + 26 r (r = 0, 1, 2). 3 5 + 26 Notice that one of these is real, namely 3 5+ 1 26 = 3 5 + 26 ( 3 )2 5 + 26 1 . 3 5 + 26 Quartic polynomials: Ferrari s method. The following method of nding roots of quartics was publicised by Cardan who attributed it to his student Lodovicio Ferrari. A general monic quartic polynomial f (X ) = X 4 + a3 X 3 + a2 X 2 + a1 X + a0 C[X ] can be transformed into one with no cubic term by a change of variables X X a2 /3 giving g (X ) = f (X a3 /4) = ( ) ( ) ( ) 32 13 1 1 341 4 2 2 Y + a2 a3 Y + a a2 a3 + a1 Y a2 a3 a + a1 a3 + a0 . 8 83 2 16 256 3 4 Clearly nding the roots of f (X ) is equivalent to nding those of g (X ), so we may as well assume that we want to nd the complex roots of f (X ) = X 4 + pX 2 + qX + r C[X ]. Suppose that x is a root and introduce numbers y, z such that z = x2 + y (we will x the values of these later). Then z 2 = x4 + 2x2 y + y 2 = px2 qx r + 2x2 y + y 2 = (2y p)x2 qx + y 2 r. Now choose y to make the last quadratic expression in x a square, (2y p)x2 qx + (y 2 r) = (Ax + B )2 . (1.16) This can be done by requiring the vanishing of the discriminant q 2 4(2y p)(y 2 r) = 0. (1.17) Notice that if y = p/2 then we would require q = 0 and then f (X ) = X 4 + pX 2 + r = (X 2 )2 + p(X 2 ) + r = 0 can be solved by solving Z 2 + pZ + r = 0. Since Equation (1.17) is a cubic in y , we can use the method of solution of cubics to nd a root y = t say. Then for Equation (1.16) we have (x2 + t)2 = (Ax + B )2 , whence x2 = t (Ax + B ). 18 Thus taking the two square roots of the right hand side we obtain 4 values for x, which we write symbolically as x = t (Ax + B ). 1.51. Remark. In the case of cubic and quartic polynomials over C we can obtain all the roots by repeatedly taking square or cube roots (or radicals ). Consequently such polynomials are said to be solvable by radicals. Later we will see that this is not true in general for polynomials of degree at least 5; this is one of the great early successes of this theory. 1.5. Automorphisms of rings and elds 1.52. Definition. Let R be a ring and R0 R a subring. An automorphism of R is a ring isomorphism : R R. The set of all such automorphisms is denoted Aut(R). An automorphism of R over R0 is a ring isomorphism : R R for which (r) = r whenever r R0 . The set of all automorphisms of R over R0 is denoted AutR0 (R). 1.53. Proposition. For a ring R with a subring R0 R, Aut(R) and AutR0 (R) form groups under composition of functions. Proof. The composition of two automorphisms , : R R is also an automorphism of R as is the inverse of . The identity function id = idR : R R is an automorphism. Hence Aut(R) forms a group under composition. The argument for AutR0 (R) is similar. 1.54. Proposition. Let R be one of the core rings Z or Z/n with n > 1. Then (i) The only automorphism of R is the identity, i.e., Aut(R) = {id}. (ii) If S is a ring containing a core ring R and Aut(S ), then restricts to the identity on R, i.e., (r) = r for all r R. Hence, Aut(S ) = AutR (S ). Proof. (i) For such a core ring R, every element has the form k 1 for some k Z. For an automorphism of R, (1) + + (1) if k > 0, k (k 1) = ( (1) + + (1)) if k < 0, k (0) if k = 0 1 + + 1 if k > 0, k = (1 + + 1) if k < 0, k 0 if k = 0 =k 1. Thus = id. (ii) For Aut(S ), (1) = 1 and a similar argument to that for (i) shows that (r) = r for all r R. 19 1.55. Proposition. Let D be an integral domain and : D D be an automorphism. Then the induced homomorphism gives an automorphism : Fr(D) Fr(D). Proof. Given , the induced homomorphism : Fr(D) Fr(D) exists and we need to show it has an inverse. The inverse automorphism 1 : D D also gives rise to an induced homomorphism ( 1 ) : Fr(D) Fr(D). Since 1 = id = 1 , we can apply Corollary 1.20 to show that ( 1 ) ( ) = id = ( ) ( 1 ) . Hence ( ) is invertible with inverse ( 1 ) . 1.56. Corollary. There is a monomorphism of groups ( ) : Aut(D) Aut(Fr(D)); . 1.57. Example. The eld of fractions of the ring of integers Z is the eld of rationals Q. The homomorphism ( ) : Aut(Z) Aut(Q); is an isomorphism and hence Aut(Q) = {id}. Combining this example with Proposition 1.54(ii) we obtain another useful result. 1.58. Proposition. Let k be one of the prime elds Q or Fp with p > 0 prime. If R is a ring containing k as a subring, then every automorphism of R restricts to the identity on k, i.e., Aut(R) = Autk (R). Recalling De nition 1.36, we have an example which shows that the monomorphism of Corollary 1.56 need not be an epimorphism. Here we take D = Q[X ] and Fr(Q[X ]) = Q(X ). 1.59. Example. The homomorphism ( ) : Aut(Q[X ]) Aut(Q(X )); is a monomorphism but it is not an epimorphism since there is an automorphism : Q(X ) Q(X ); (f (X )) = f (1/X ) which sends X Q[X ] Q(X ) to 1/X Q[X ] and so does not restrict to an automorphism of / Q[X ]. Let k be a eld. The group of invertible 2 2 matrices over k is the 2 2 general linear group over k, {[ ] } a11 a12 GL2 (k) = : aij k, a11 a22 a12 a21 = 0 a21 a22 The scalar matrices form a normal subgroup Scal2 (k) = {diag(t, t) : t k, t = 0} GL2 (k). The quotient group is called the 2 2 projective general linear group over k, PGL2 (k) = GL2 (k)/ Scal2 (k). 20 Notice that GL2 (k) has another interesting subgroup called the a ne subgroup, {[ ] } ab A 1 (k) = : a, b k, a = 0 GL2 (k). 01 1.60. Example. Let k be a eld and X an indeterminate. Then Autk (k[X ]) and hence Autk (k(X )), contains a subgroup isomorphic to A 1 (k). In fact, Autk (k[X ]) A 1 (k). = Proof. We begin by showing that to each a ne matrix [ ] ab A= A 1 (k) 01 there is an associated automorphism A : k[X ] k[X ]. For this we use the element aX + b k[X ] together with the extension result of Theorem 1.22(i) to obtain a homomorphism A : k[X ] k[X ] with A (X ) = aX + b. Using the inverse matrix [ ] a 1 a 1 b 1 A= 0 1 we similarly obtain a homomorphism A 1 : k[X ] k[X ] for which A 1 (X ) = a 1 X a 1 b. Using the same line of argument as in the proof of Proposition 1.55 (or doing a direct calculation) we see that A 1 is the inverse of A an so A Autk (k[X ]). It is straightforward to check that for A1 , A2 A 1 (k), A2 A1 = A1 A2 , (note the order!) hence there is a homomorphism of groups A 1 (k) Autk (k[X ]); A A 1 , which is easily seen to be a monomorphism. Composing with ( ) we see that there is a monomorphism A 1 (k) Autk (k(X )). In fact, this is also an epimorphism and we leave the proof of this as an exercise. 1.61. Example. Let k be a eld and X an indeterminate. Then (i) Autk (k(X )) contains a subgroup isomorphic to PGL2 (k). (ii) In fact, Autk (k(X )) PGL2 (k). = Proof. (i) We begin by showing that to each invertible matrix [ ] a11 a12 A= GL2 (k) a21 a22 there is an associated automorphism A : k(X ) k(X ). We begin by choosing the element (a11 X + a12 )/(a21 X + a22 ) k(X ) and then using Theorem 1.22(i) to obtain a homomorphism k[X ] k(X ) that sends X to (a11 X +a12 )/(a21 X +a22 ). By applying ( ) to this we obtain a homomorphism (known as a fractional linear transformation ) A : k(X ) k(X ) for which A (X ) = a11 X + a12 . a21 X + a22 21 Again we nd that A2 A1 = A1 A2 . 1 There is an associated homomorphism of groups GL2 (k) Autk (k(X )) sending A to A . However, this is not an injection in general since for each scalar matrix diag(t, t), diag(t,t) (X ) = tX = X, t showing that diag(t,t) is the identity function. In fact it is easy to see that Scal2 (k) GL2 (k) is the kernel of this homomorphism. Therefore passing to the quotient PGL2 (k) = GL2 (k)/ Scal2 (k) we obtain a monomorphism PGL2 (k) Autk (k(X )). There is one case where Scal2 (k) is the trivial group, namely k = F2 . (ii) To show that every automorphism of k(X ) is a fractional linear transformation is less elementary. We give a sketch proof for the case of k = C; actually this argument can be modi ed to work for any algebraically closed eld, but an easy argument then shows the general case. Let AutC (C(X )). There is an associated rational (hence meromorphic) function f given by z f (z ), where (X ) = f (X ), de ned on C with the poles of f deleted. If we write f (X ) = p(X ) q (X ) where p(X ), q (X ) C[X ] have no common factors of positive degree, then the order of f (X ) is ord f = max{deg p(X ), deg q (X )}. Now let c C. Then the number of solutions counted with algebraic multiplicity of the equation f (z ) = c turns out to be ord f . Also, if deg p(X ) deg q (X ) then the number of poles of f counted with algebraic multiplicity is also ord f . Finally, if deg p(X ) > deg q (X ) then we can write p0 (X ) , f (X ) = p1 (X ) + q (X ) where p0 (X ), p1 (X ) C[X ] and deg p0 (X ) < deg q (X ). Then the number of poles of f counted with algebraic multiplicity is p0 deg p1 (X ) + ord . q Now it is easy to see that since is invertible so is the function f . But this can only happen if the function f is injective which means that all of these numbers must be 1, hence ord f = 1. Thus aX + b f (X ) = = constant cX + d [ ] ab and the matrix must be invertible. cd Clearly not every fractional linear transformation A : k(X ) k(X ) maps polynomials to polynomials so ( ) : Autk (k[X ]) Autk (k(X )) is not an epimorphism. Now we turn to a more familiar eld R, the real numbers. 1.62. Proposition. The only automorphism of the eld R is the identity function, hence Aut(R) = {id}. 22 Proof. First we note that Q R is a subring and if Aut(R) then (q ) = q for q Q by Example 1.57. We recall from Analysis that the rational numbers are dense in the real numbers in the sense that each r R can be expressed as a limit r = limn qn , where qn Q. Then for a continuous function f : R R, its value at r depends on its values on Q since f (r) = f ( lim qn ) = lim f (qn ). n n We will show that an automorphism Aut(R) is continuous. First recall that for x, y R, x<y 0<y x y x = t2 for some non-zero t R. Now for Aut(R) and s R, we have (s2 ) = (s)2 . Hence, x < y = (y ) (x) = (t)2 for some non-zero t R = (x) < (y ). So preserves order and xes rational numbers. Now let x R and > 0. Then we can choose a rational number q such that 0 < q Taking = q we nd that for y R with |y x| < (i.e., < y x < ) we have . = ( ) < (y ) (x) < ( ) = , hence | (y ) (x)| < . This shows that is continuous at x. Thus every automorphism of R is continuous function which xes all the rational numbers, hence it must be the identity function. 1.63. Remark. If we try to determine Aut(C) the answer turns out to be much more complicated. It is easy to see that complex conjugation ( ) : C C is an automorphism of C and xes every real number, i.e., ( ) AutR (C); in fact, AutR (C) = {id, ( )}. However, it is not true that every Aut(C) xes every real number! The automorphism group Aut(C) is actually enormous but it is hard to nd an explicit element other than id and ( ). Note that given an automorphism Aut(C), the composition ( ) 1 is also self inverse, so there are many elements of order 2 in the group Aut(C). Exercises on Chapter 1 1.1. Let R be a ring. Show that {n Z : n > 0 and n1 = 0} = {n Z : n > 0 and nr = 0 for all r R}. Deduce that if char R > 0 then these sets are non-empty and char R = min{n Z : n > 0 and nr = 0 for all r R}. 1.2. Let R be an integral domain. (a) Show that every subring S R is also an integral domain. What is the relationship between char S and char R ? 23 (b) If R is a eld, give an example to show that a subring of R need not be a eld. 1.3. For each of the following rings R, nd the characteristic char R and the characteristic subring of R. Determine which of these rings is an integral domain. In (b) and (c), A is an arbitrary commutative ring. (a) Any subring R C. (b) The polynomial ring R = A[X ]. (c) The ring of n n matrices over A, a11 . . . a1n . .. . . R = Matn (A) = . : aij A . . . . a . . . ann n1 1.4. If R is a commutative ring with unit containing the prime eld Fp for some prime p > 0, show that the function : R R given by (t) = tp , de nes a ring homomorphism. Give examples to show that need not be surjective or injective. 1.5. Let R and S be rings with unity and Q S a prime ideal. (a) If : R S is a ring homomorphism, show that 1 Q = {r R : (r) Q} R is a prime ideal of R. (b) If R S is a subring, show that Q R is a prime ideal of R. (c) If the word prime is replaced by maximal throughout, are the results in parts (a) and (b) still true? [Hint: look for a counterexample.] (d) If R S is a subring and P R is a maximal ideal, suppose that Q S is a prime ideal for which P Q. Show that Q R = P . 1.6. Let k be a eld, R be a ring with unit and let : k R be a ring homomorphism. Show that is a monomorphism. 1.7. Consider the sets Z(i) = {u + vi : u, v Z} C, Q(i) = {u + vi : u, v Q} C. (a) Show that Z(i) and Q(i) are subrings of C. Also show that Z(i) is an integral domain, Q(i) is a eld and Z(i) is a subring of Q(i). (b) Show that the inclusion homomorphism inc : Z(i) Q(i) extends to a monomorphism inc : Fr(Z(i)) Q(i). (c) Show that inc is an isomorphism, so Fr(Z(i)) = Q(i). 1.8. Let R be a commutative ring. (a) If a, b R, show that there is a unique ring homomorphism a,b : R[X ] R[X ] for which a,b (r) = r if r R and a,b (X ) = aX + b. If c, d R, determine a,b c,d . If a is a unit, show that a,b is an isomorphism and nd its inverse. (b) Now suppose that R = k is a eld and a, b k with a = 0. Prove the following. (i) If f (X ) k[X ], the deg a,b (f (X )) = deg f (X ). (ii) If p(X ) k[X ] is a prime then so is a,b (p(X )). 24 (iii) If p(X ) k[X ] is an irreducible then so is a,b (p(X )). 1.9. Let k be a eld and k[[X ]] be the set consisting of all power series ak X k = a0 + a1 X + + ak X k + , k=0 with ak k. (a) Show that this can be made into an integral domain containing k[X ] as a subring by de ning addition and multiplication in the obvious way. (b) Show that ak X k k[[X ]] is a unit if and only if a0 = 0. k=0 (c) Show that Fr(k[[X ]]) consists of all nite-tailed Laurent series ak X k = a X + a +1 X +1 + + ak X k + k = for some Z and ak k. 1.10. Taking k = Q, nd the quotient and remainder when performing long division of f (X ) = 6X 4 6X 3 + 3X 2 3X 2 by d(X ) = 2X 3 + X + 3. 1.11. Taking k = F3 , nd the quotient and remainder when performing long division of f (X ) = 2X 3 + 2X 2 + X + 1 by d(X ) = 2X 3 + 2X . 1.12. Let p > 0 be a prime. Suppose that f (X ) = a0 + a1 X + + an X n Z[X ] with p an and that f (X ) Fp [X ] denotes the polynomial obtained by reducing the coe cients of f (X ) modulo p. If f (X ) is irreducible, show that f (X ) is irreducible. Which of the following polynomials in Z[X ] is irreducible? X 3 X + 1, X 3 + 2X + 1, X 3 + X 1, X 5 X + 1, X 5 + X 1, 5X 3 10X + X 2 2. 1.13. Find generators for each of the following ideals: I2 = {f (X ) Q[X ] : f ( 2 i) = 0} Q[X ], I4 = {f (X ) R[X ] : f ( 2) = 0} R[X ], I1 = {f (X ) Q[X ] : f (i) = 0} Q[X ], I3 = {f (X ) Q[X ] : f ( 2) = 0} Q[X ], I5 = {f (X ) R[X ] : f ( 2 i) = 0} R[X ], I6 = {f (X ) R[X ] : f ( 3 ) = 0} R[X ]. 1.14. Consider the inclusion inc : Q C and its extension to 2 : Q[X ] C. Determine the image 2 Q[X ] C. What is 2 Q[X ] C? Find ker 2 Q[X ] and ker 2 Q[X ]; are these maximal ideals? 1.15. Let = ( 1 + 3i)/2 C. Consider the inclusion inc : Q C and its extension to : Q[X ] C. Determine the image Q[X ] C. Determine ker Q[X ] and decide whether it is maximal. Find another evaluation homomorphism with the same kernel and image. 1.16. Consider the inclusion inc : Q C and its extension to : Q[X ] C where is one of the 4 complex roots of the polynomial f (X ) = X 4 2 Q[X ]. Determine the image Q[X ] C and the ideal ker Q[X ]; is the latter ideal maximal? What happens if is replaced by one of the other roots of f (X )? Repeat this problem starting with the inclusion of the real numbers into the complex numbers inc : R C and : R[X ] C. 25 1.17. Use Cardan s method to nd the complex roots of the polynomial f (X ) = X 3 9X 2 + 21X 5. 1.18. Consider the real numbers 3 3 = 10 + 108 + 10 108, 3 = 2 1+ 3 7 + 3 3 2 1 3 7 . 3 Find rational cubic polynomials f (X ) and g (X ) for which f ( ) = 0 = g ( ). Hence determine these real numbers. 1.19. Prove the nal part of Example 1.60 by showing that there is an isomorphism of groups A 1 (k) Autk (k[X ]). = 1.20. Let k be any eld. Consider the 6 automorphisms j : k(X ) k(X ) (j = 1, . . . , 6) de ned by 1 (f (X )) = f (X ), 2 (f (X )) = f (1 X ), 3 (f (X )) = f (1/X ), 4 (f (X )) = f ((X 1)/X ), 5 (f (X )) = f (1/(1 X )), 6 (f (X )) = f (X/(X 1)). Show that the set consisting of these elements is a subgroup k the symmetric group S3 . When k = F2 , show that k GL2 (k). = Autk (k(X )) isomorphic to 1.21. Determine the cyclotomic polynomial 20 (X ). 1.22. Let p > 0 be a prime. (a) Show that for k 1, the cyclotomic polynomial pk (X ) satis es k 1 pk (X ) = p (X p ) and has as its complex roots the primitive pk -th roots of 1. (b) Show that pk (X ) Q[X ] is irreducible. (c) Generalize part (a) to show that if n = pr1 prk is the prime power factorization of n 1 k with the pi being distinct primes and ri > 0, then r 1 r1 1 pkk n (X ) = p1 pk (X p1 1.23. For n ). 2, show that X (n) n (X 1 ) = n (X ). 1.24. Show that for n 1, n + n 1 = 2 cos(2 /n). 2 Find expressions for 5 + 5 1 and 5 + 5 2 in terms of cos(2 /5). Hence nd a rational polynomial which has cos(2 /5) as a root. 1.25. Let p > 0 be a prime and K be a eld with char K = p. (a) Show that if K is a p-th root of 1 then = 1. Deduce that if m, n > 0 and p n, then every npm -th root of 1 in K is an n-th root of 1. (b) If a K , show that the polynomial X p a K [X ] has either no roots or exactly one root in K . 26 CHAPTER 2 Fields and their extensions 2.1. Fields and sub elds 2.1. Definition. Let K and L be elds and suppose that K L is a subring. Then we say that K is a sub eld of L; L is also said to be an extension ( eld ) of K . We write K L or L/K to indicate this, and write K < L if K is a proper sub eld of L, i.e., if K = L. An important fact about an extension of elds L/K is that L is a K -vector space whose addition is the addition in the eld L while scalar multiplication is de ned by u x = ux (u K, x L). 2.2. Definition. We will call dimK L the degree or index of the extension L/K and use the notation [L : K ] = dimK L. An extension of elds L/K is nite (dimensional ) if [L : K ] < , otherwise it is in nite (dimensional ). 2.3. Example. Show that the extension C/R is nite, while R/Q and C/Q are both in nite. Solution. We have C = {x + yi : x, y R}, so 1, i span C as a vector space over R. Since i R, these elements are also linearly independent / over R and therefore they form a basis, whence [C : R] = 2. The in niteness of R/Q and C/Q are consequences of the fact that any nite dimensional vector space over Q is countable, however R and C are uncountable. A basis for the Q-vector space R is known as a Hamel basis. 2.4. Example. Consider the extension Q( 2)/Q where Q( 2) = {x + y 2 : x, y Q}. Show that [Q( 2) : Q] = 2. Solution. The elements 1, 2 clearly span the Q-vector space Q( 2). Now recall that 2 Q. If the elements 1, 2 were linearly dependent we would have u + v 2 = 0 for some / u, v Q not both zero; in fact it is easy to see that we would then also have u, v both non-zero. Thus we would have u 2 = Q, v which we know to be false. Hence 1, 2 are linearly independent and so form a basis for Q( 2) over Q and [Q( 2) : Q] = 2. If we have two extensions L/K and M/L then it is a straightforward to verify that K and so we have another extension M/K . M 2.5. Definition. Given two extensions L/K and M/L, we say that L/K is a subextension of M/K and sometimes write L/K M /K . 27 2.6. Theorem. Let L/K be a subextension of M/K . (i) If one or both of the dimensions [L : K ] or [M : L] is in nite then so is [M : K ]. (ii) If the dimensions [L : K ] and [M : L] are both nite then so is [M : K ] and [M : K ] = [M : L] [L : K ]. Proof. (i) If [M : K ] is nite, choose a basis m1 , . . . , mr of M over K . Now any element u M can be expressed as u = t1 m1 + + tr mr , where t1 , . . . , tr K ; but since K L, this means that m1 , . . . , mr spans M over L and so [M : L] < . Also L is a K -vector subspace of the nite dimensional K -vector space M , hence [L : K ] < . (ii) Setting r = [L : K ] and s = [M : L], choose a basis 1 , . . . , r of L over K and a basis m1 , . . . , ms of M over L. Now let v M . Then there are elements y1 , . . . , ys L for which v = y1 m1 + + ys ms . But each yj can be expressed in the form yj = x1j 1 + + xrj r for suitable xij K . Hence, v= (r s j =1 ) mj = xij i s r xij ( i mj ), j =1 i=1 i=1 where each coe cient xij is in K . Thus the elements i mj (i = 1, . . . , r, j = 1, . . . , s) span the K -vector space M . Now suppose that for some tij K we have r s tij ( i mj ) = 0. j =1 i=1 On collecting terms we obtain where each coe cient means that for each j , r (r s j =1 i=1 tij i ) tij i mj = 0 , i=1 is in L. By the linear independence of the mj over L, this r tij i = 0. i=1 By the linear independence of the i over K , each tij = 0. Hence the i mj form a basis of M over K and so [M : K ] = rs = [M : L] [L : K ]. 28 We will often indicate subextensions in diagrammatic form where larger elds always go above smaller ones and the information on the lines indicates dimensions MF [M :L] A 6 L1 [L:K ] ! & [M :K ]=[M :L] [L:K ] K We often suppress composite lines such as the dashed one. Such towers of extensions are our main objects of study. We can build up sequences of extensions and form towers of arbitrary length. Thus, if L1 /K, L2 /L1 , . . . , Lk /Lk 1 is a such a sequence of extensions, there is a diagram Lk Lk 1 L1 K 2.2. Simple and nitely generated extensions F . Given elements u1 , . . . , ur F we set K (u1 , . . . , ur ) = L 2.7. Definition. Let F be a eld and K KLF u1 ,...,ur L which is the smallest sub eld in F that contains K and the elements u1 , . . . , ur . The extension K (u1 , . . . , ur )/K is said to be generated by the elements u1 , . . . , ur ; we also say that K (u1 , . . . , ur )/K is a nitely generated extension of K . An extension of the form K (u)/K is called a simple extension of K with generator u. We can extend this to the case of an in nite sequence u1 , . . . , ur , . . . in F and denote by K (u1 , . . . , ur , . . .) F the smallest extension eld of K containing all the elements ur . It can be shown that (2.1) K (u1 , . . . , ur ) = } { f (u1 , . . . , ur ) F : f (X1 , . . . , Xr ), g (X1 , . . . , Xr ) K [X1 , . . . , Xr ], g (u1 , . . . , ur ) = 0 . g (u1 , . . . , ur ) Reordering the ui does not change K (u1 , . . . , un ). 29 2.8. Proposition. Let K (u)/K and K (u, v )/K (u) be simple extensions. Then K (u, v ) = K (u)(v ) = K (v )(u). More generally, K (u1 , . . . , un ) = K (u1 , . . . , un 1 )(un ) and this is independent of the order of the sequence u1 , . . . , un . 2.9. Theorem. For a simple extension K (u)/K , exactly one of the following conditions holds. (i) The evaluation at u homomorphism u : K [X ] K (u) is a monomorphism and on passing to the fraction eld gives an isomorphism ( u ) : K (X ) K (u). In this case, K (u)/K is in nite and u is said to be transcendental over K . (ii) The evaluation at u homomorphism u : K [X ] K (u) has a non-trivial kernel ker u = (p(X )) where p(X ) K [X ] is an irreducible monic polynomial of positive degree and the quotient homomorphism u : K [X ]/(p(X )) K (u) is an isomorphism. In this case K (u)/K is nite with [K (u) : K ] = deg p(X ) and u is said to be algebraic over K . Proof. (i) If ker u = (0), all that needs checking is that ( u ) is an epimorphism; but as u is in the image of ( u ) this is obvious. (ii) When ker u = (0), Theorem 1.31(iv) implies that the image of u is a sub eld of K (u) and since it contains u it must equal K (u). Hence u is an isomorphism. Using Long Division, we nd that every element of K [X ]/(p(X )) can be uniquely expressed as a coset of the form f (X ) + (p(X )), where deg f (X ) < deg p(X ). Hence every element of K [X ]/(p(X )) can be uniquely expressed as a linear combination over K of the d cosets 1 + (p(X )), X + (p(X )), X 2 + (p(X )), . . . , X d 1 + (p(X )), where d = deg p(X ). Via the isomorphism u under which u (X k + (p(X ))) = uk , we see that the elements 1, u, . . . , ud 1 form a basis for K (u) over K . 2.10. Example. For the extension Q( 2, 3)/Q we have [Q( 2, 3) : Q] = 4. Proof. By Example 2.4 we know that [Q( 2) : Q] = 2. We have the following tower of extensions. Q( 2, 3) [Q( 2, 3):Q( 2)] Q( 2) [Q( 2, 3):Q]=2[Q( 2, 3):Q( 2)] 2 Q We will show that [Q( 2, 3) : Q( 2)] = 2. 30 Notice that if u Q( 2, 3) = Q( 2)( 3) then u = a + b 3 for some a, b Q( 2), so 1, 3 span Q( 2, 3) over Q( 2). But if these are linearly dependent then 3 Q( 2). Writing 3=v+w 2 with v, w Q, we nd that v 2 + 2w2 + 2vw 2 = 3 Q, and hence 2vw 2 Q. The possibilities v = 0 or w = 0 are easily ruled out, while v, w = 0 would implies that 2 Q which is false. So 1, 3 are linearly independent over Q( 2) and therefore form a basis of Q( 2, 3). This shows that [Q( 2, 3) : Q( 2)] = 2 and so [Q( 2, 3) : Q] = 4. 2.11. Remark. There are some other sub elds of Q( 2, 3) which are conveniently displayed in the following diagram. Q( 2, 3) r rrr rrr rrr 2 vvv vvv vvv 2 vvv 2 2 Q( 2) vvv vvv2 vvv v Q( 3) Q Q( 6) r rrr rrr rr 2 rrr One idea in the veri cation of Example 2.10 can be extended to provide a useful general result whose proof is left as an exercise. 2.12. Proposition. Let p1 , . . . , pn be a sequence of distinct primes pi > 0. Then pn Q( p1 , . . . , pn 1 ). / Hence [Q( p1 , . . . , pn ) : Q( p1 , . . . , pn 1 )] = 2 and [Q( p1 , . . . , pn ) : Q] = 2n . 2.13. Example. For the extension Q( 2, i)/Q we have [Q( 2, i) : Q] = 4. / Proof. We know that [Q( 2) : Q] = 2. Also, i Q( 2) since i is not real and Q( 2) R. Since i2 + 1 = 0, we have Q( 2, i) = Q( 2)(i) and [Q( 2, i) : Q( 2)] = 2. Using the formula [Q( 2, i) : Q] = [Q( 2, i) : Q( 2)] [Q( 2) : Q], we obtain [Q( 2, i) : Q] = 4. 31 This example also has several other sub elds, with only Q( 2) = Q( 2, i) R being a sub eld of R. C 2 R Q( 2, i) tt tt tt t tt 2 Q( 2) 2 Q(i) uu uu uu u 2 uuu u 2.14. Example. For n n-th root of 2. uu uu 2 uu uu u 2 Q Q ( 2 i) rr rr rr r rr 2 rr 1, let En = Q(21/n ) R, where 21/n R denotes the positive real (i) Show that [En : Q] = n. (ii) If m 1 with m | n, show that Em En and determine [En : Em ]. (iii) If m, n are coprime, show that Emn = Q(21/m , 21/n ). Solution. (i) Consider the evaluation homomorphism 21/n : Q[X ] En . Applying the Eisenstein Test 1.38 using the prime 2 to the polynomial X n 2 Z[X ], we nd that ker 21/n = (X n 2) Q[X ], and the induced homomorphism 21/n : Q[X ]/(X n 2) En is an isomorphism. Hence [En : Q] = n. (ii) Since n/m is an integer, 21/m = (21/n )n/m En , so Em = Q(21/m ) En . By Theorem 2.6 we have n = [En : Q] = [En : Em ] [Em : Q] = m[En : Em ], whence [En : Em ] = n/m. (iii) By (ii) we have Em Emn and En Emn , hence Q(21/m , 21/n ) there are integers r, s for which rm + sn = 1 and so Emn . As gcd(m, n) = 1, 1 rm + sn r s = =+. mn mn nm This shows that 21/mn = (21/n )r (21/m )s Q(21/m , 21/n ), whence Emn Q(21/m , 21/n ). Combining these inclusions we obtain Emn = Q(21/m , 21/n ). 32 Exercises on Chapter 2 2.1. Let p N be an prime. Show that the extension Q( p)/Q has [Q( p) : Q] = 2. 2.2. Let p, q > 0 be distinct primes. Show that [Q( p, q ) : Q( p)] = 2. 2.3. Prove Proposition 2.12 by induction on n. 2.4. Let K a eld with char K = 2 and suppose that L/K is an extension. If a, b K are distinct, suppose that u, v L satisfy u2 = a and v 2 = b. Show that K (u, v ) = K (u + v ). [Hint: rst show that u v = 0 and deduce that u v K (u + v ); then show that u, v K (u + v ).] 2.5. Show that [Q(i) : Q] = 2. 2.6. Show that [Q( 3, i) : Q] = 4. Find the three sub elds L Q( 3, i) with [L : Q] = 2 and display their relationship in a diagram, indicating which ones are sub elds of R. 2.7. Let 5 = e2 i/5 C. (a) Explain why [Q( 5 ) : Q] = 4. (b) Show that cos(2 /5), sin(2 /5) i Q( 5 ). (c) Show that for t R, cos 5t = 16 cos5 t 20 cos3 t + 5 cos t. (d) Show that the numbers cos(2k /5) with k = 0, 1, 2, 3, 4 are roots of the polynomial f (X ) = 16X 5 20X 3 + 5X 1 = (X 1)(4X 2 + 2X 1)2 and deduce that [Q(cos(2 /5)) : Q] = 2. (e) Display the relationship between the elds Q, Q(cos(2 /5)), and Q( 5 ) in a suitable diagram. 2.8. This question is for those who like lots of calculation or using Maple. Let 7 = e2 i/7 C. (a) Explain why [Q( 7 ) : Q] = 6. (b) Show that cos(2 /7), sin(2 /7) i Q( 7 ). (c) Show cos 7t = 64 cos7 t 112 cos5 t + 56 cos3 t 7 cos t. Show that the numbers cos(2k /7) with k = 0, 1, . . . , 6 are roots of the polynomial f (X ) = 64X 7 112X 5 + 56X 3 7X 1 = (X 1)(8X 3 + 4X 2 4X 1)2 and deduce that [Q(cos(2 /7)) : Q] = 3. (d) Show that sin(2 /7) i is a root of g (X ) = 64X 7 + 112X 5 + 56X 3 + 7X = X (64X 6 + 112X 4 + 56X 2 + 7) and that 64X 6 + 112X 4 + 56X 2 + 7 Q[X ] is irreducible. What is [Q(sin(2 /7) i) : Q]? (e) Display the relationship between the elds Q, Q(cos(2 /7)), Q(sin(2 /7) i) and Q( 7 ) in a diagram. (f) Is i Q( 7 )? 2.9. In this question we continue to consider the situation described in Example 2.14. 33 (a) Show that AutQ (En ) = {id} if n is odd, {id, n } Z/2 if n is even, = where n composition order 2. has (b) Let E = En R. Show that AutQ (E ) = {id}. n1 (c) Display the 6 sub elds of E12 in a diagram. (d) Which of the sub elds in part (c) contain the element 21/2 + 21/3 ? 34 CHAPTER 3 Algebraic extensions of elds 3.1. Algebraic extensions Let L/K be an extension of elds. From Theorem 2.9(ii), recall the following notion. 3.1. Definition. An element t L is algebraic over K if there is a non-zero polynomial p(X ) K [X ] for which p(t) = 0. Notice in particular that for an element t K , the polynomial p(X ) = X t K [X ] satis es p(t) = 0, so t is algebraic over K . Theorem 2.9 allows us to characterize algebraic elements in other ways. 3.2. Proposition. Let t L. Then the following conditions are equivalent. (i) t is algebraic over K . (ii) The evaluation homomorphism t : K [X ] L has non-trivial kernel. (iii) The extension K (t)/K is nite dimensional. 3.3. Definition. If t L is algebraic over K then by Proposition 3.2, ker t = (minpolyK,t (X )) = (0), where minpolyK,t (X ) K [X ] is an irreducible monic polynomial called the minimal polynomial of t over K . The degree of minpolyK,t (X ) is called the degree of t over K and is denoted degK t. 3.4. Proposition. If t L is algebraic over K then [K (t) : K ] = deg minpolyK,t (X ) = degK t. Proof. This follows from Theorem 2.9(ii). 3.5. Remark. Suppose that t L is algebraic over K and that p(X ) ker t with deg p(X ) = deg minpolyK,t (X ). Then minpolyK,t (X ) | p(X ) and so p(X ) = u minpolyK,t (X ) for some u K . In particular, when p(X ) is monic, p(X ) = minpolyK,t (X ). We will often use this without further comment. 3.6. Example. Consider C/Q. The minimal polynomial of minpolyQ, 2 (X ) = X 2 2. 35 2 C over Q is Proof. Clearly X 2 2 ker 2 since ( 2)2 2 = 0. By Example 2.4, deg minpolyQ, 2 (X ) = [Q( 2) : Q] = 2, hence minpolyQ, 2 (X ) = X 2 2. 3.7. Example. Consider C/Q. The minimal polynomial of i C over Q is X 2 + 1. Proof. Clearly X 2 + 1 ker i since i2 + 1 = 0. As [Q(i) : Q] = 2, we have minpolyQ,i (X ) = X 2 + 1. 3.8. Example. Consider C/Q. Find the minimal polynomial of the primitive 6-th root of unity, 6 C over Q. Solution. Recall from Example 1.44 that 6 is a root of the irreducible cyclotomic polynomial 6 (X ) = X 2 X + 1. Then 6 (X ) ker 6 so minpolyQ, 6 (X ) | 6 (X ). Since 6 (X ) is irreducible and monic, we must have minpolyQ, 6 (X ) = 6 (X ) and so degQ 6 = 2. 3.9. Example. Consider C/Q. Find the minimal polynomial of 2+ 3 over Q. Solution. Notice that ( 3 2)( 3 + 2) 1 Q( 2 + 3). 3 2= = ( 3 + 2) 2+ 3 So we have hence Q( 2, 3) ) 1( 2= ( 2 + 3) ( 3 2) Q( 2 + 3), 2 ) 1( 3= ( 2 + 3) + ( 3 2) Q( 2 + 3), 2 Q( 2 + 3). Since Q( 2 + 3) Q( 2, 3) we must have Q( 2 + 3) = Q( 2, 3). Referring to Example 2.10 we see that degQ ( 2 + 3) = 4. Let us nd a non-zero polynomial in ker 2+ 3 Q[X ]. Referring to Example 2.10 or Proposition 2.12 we see that 2 + 3 Q( 2), hence / degQ( 2) ( 2 + 3) = 2. One polynomial in ker 2+ 3 Q( 2)[X ] is (X ( 2 + 3))(X ( 2 3)) = X 2 2 2X 1. Since this is monic and of degree 2, minpolyQ( 2), 2+ 3 (X ) = X 2 2 2X 1. 36 Similarly, minpolyQ( 2), 2+ 3 (X ) = X 2 + 2 2X 1. Consider p(X ) = minpolyQ( 2), 2+ 3 (X ) minpolyQ( 2), 2+ 3 (X ) = (X 2 2 2X 1)(X 2 + 2 2X 1) = X 4 10X 2 + 1. Then p( 2 + 3) = 0 so p(X ) ker t . Since deg p(X ) = 4 and p(X ) is monic, we have minpolyQ, 2+ 3 (X ) = X 4 10X 2 + 1. 3.10. Definition. Let L/K be a nite extension. An element u L for which L = K (u) is called a primitive element for the extension L/K . If L/K such a primitive element exists, then L/K is called a simple extension. Later we will see that when char K = 0 every nite extension L/K has a primitive element, hence every such extension is simple. 3.11. Lemma. Let L/K be a nite extension and u L. Then u is a primitive element for L/K if and only if degK u = [L : K ]. Proof. K (u) L is a nite dimensional K -vector subspace. Then K (u) = L if and only dimK K (u) = dimK L. Since degK u = dimK K (u) and [L : K ] = dimK L the result follows. Sometimes the minimal polynomial of an element in an extension is introduced in a di erent but equivalent way. 3.12. Proposition. Let t L be algebraic over K . Then I(t) = {f (X ) K [X ] : f (t) = 0} K [X ] is an ideal which is principal and has an irreducible monic generator q (X ) K [X ]. In fact, q (X ) = minpolyK,t (X ). Proof. It is easy to see that I(t) K [X ] and therefore I(t) = (q (X )) for some monic generator q (X ). To see that q (X ) is irreducible, suppose that q (X ) = q1 (X )q2 (X ) with deg qi (X ) < deg q (X ). Now as q1 (t)q2 (t) = 0, we must have q1 (t) = 0 or q2 (t) = 0, hence q1 (X ) I(t) or q2 (X ) I(t). These possibilities give q (X ) | q1 (X ) or q (X ) | q2 (X ) and so deg q (X ) deg q1 (X ) or deg q (X ) deg q2 (X ), contradicting the above assumption that deg qi (X ) < deg q (X ). The irreducible monic polynomial minpolyK,t (X ) is in I(t) so q (X ) | minpolyK,t (X ) and therefore q (X ) = minpolyK,t (X ). The next Lemma will often be useful. 3.13. Lemma. Let L/K be an extension and suppose that u1 , . . . , un L are algebraic. Then K (u1 , . . . , un )/K is a nite extension. Proof. Use induction on n together with Proposition 2.8 and Theorem 2.6(ii). We now come to an important notion for extensions. 37 3.14. Definition. The extension L/K is algebraic or L is algebraic over K if every element t L is algebraic over K . 3.15. Proposition. Let L/K be a nite extension. Then L/K is algebraic. Proof. Let t L. Since the K -vector space L is nite dimensional, when viewed as elements of this vector space, the powers 1, t, . . . , tn , . . . must be linearly dependent over K . Hence for suitable coe cients cj K not all zero and some m 1 we have c0 + c1 t + + cm tm = 0. But this means that t is algebraic over K . 3.16. Proposition. Let M/L and L/K be algebraic extensions. Then the extension M/K is algebraic. Proof. Let u M . Then u is algebraic over L, so there is a polynomial p(X ) = p0 + p1 X + + pm X m L[X ] of positive degree with p(u) = 0. By Lemma 3.13, the extension K (p0 , . . . , pm )/K is nite and so is K (p0 , . . . , pm , u)/K (p0 , . . . , pm ). By Theorem 2.6(ii), K (p0 , . . . , pm , u)/K is nite, so by Proposition 3.15, u is algebraic over K . 3.17. Definition. For an extension L/K , let Lalg = {t L : t is algebraic over K } L. 3.18. Proposition. For an extension L/K , Lalg is a sub eld containing K and Lalg /K is algebraic. Proof. Clearly K Lalg . We must show that Lalg L. Let u, v Lalg . Then by Lemma 3.13, K (u, v )/K is a nite dimensional extension, hence every element of K (u, v ) is algebraic over K . In particular, u + v and uv are in K (u, v ) and if u = 0, u 1 is also in K (u, v ). Therefore u + v , uv and u 1 are all algebraic over K . 3.19. Example. In the extension C/Q we can consider Calg C which is called the sub eld of algebraic numbers. Similarly, in the extension R/Q the sub eld Ralg = Calg R C consists of all the real algebraic numbers. Elements of C Calg are called transcendental complex numbers; examples are e and . The sets Calg and Ralg are both countable, whereas C and R are uncountable, so there are in fact many more transcendental numbers but it can be hard to determine whether a given number is transcendental or not. A more usual notation for Calg is Q since this is the algebraic closure of Q which will be discussed later. When dealing with algebraic extensions of Q we will usually work with sub elds of Q = Calg . We end this section with a technical result. 3.20. Proposition. Let K (u)/K be a nite simple extension. Then there are only nitely many subextensions F/K K (u)/K . 38 Proof. Consider the minimal polynomial minpolyK,u (X ) K [X ]. Now for any subextension F/K K (u)/K we can also consider minpolyF,u (X ) = c0 + c1 X + + ck 1 X k 1 + X k F [X ], which divides minpolyK,u (X ) in F [X ]. The Unique Factorization Property 1.33 implies that minpolyK,u (X ) has only nitely many monic divisors in K (u)[X ], so there are only a nite number of possibilities for minpolyF,u (X ). Now consider F0 = K (c0 , c1 , . . . , ck 1 ), the extension eld of K generated by the coe cients of minpolyF,u (X ). Then F0 F and so minpolyF,u (X ) F0 [X ] is irreducible since it is irreducible in F [X ]; hence minpolyF,u (X ) = minpolyF0 ,u (X ). We have [K (u) : F ] = deg minpolyF,u (X ) = deg minpolyF0 ,u (X ) = [K (u) : F0 ], hence F = F0 . This shows that there are only nitely many subextensions F/K has the form K (a0 , a1 , . . . , a 1 ), where K (u)/K , each of which a0 + a1 X + + a 1 X 1 + X K (u)[X ] is a factor of minpolyK,u (X ) in K (u)[X ]. 3.2. Splitting elds and Kronecker s Theorem We can now answer a basic question. Let K be a eld and p(X ) K [X ] be a polynomial of positive degree. 3.21. Question. Is there an extension eld L/K for which p(X ) has a root in L? A stronger version of this question is the following. 3.22. Question. Is there an extension eld E/K for which p(X ) factorizes into linear factors in E [X ]? 3.23. Definition. p(X ) K [X ] splits in E/K or over E if it factorizes into linear factors in E [X ]. Of course, if we have such a eld E then the distinct roots u1 , . . . , uk of p(X ) in E generate a sub eld K (u1 , . . . , uk ) E which is the smallest sub eld of E that answers Question 3.22. 3.24. Definition. Such a minimal extension of K is called a splitting eld of p(X ) over K and we will sometimes denote it by K (p(X )) or Kp . We already know how to answer Question 3.21. 3.25. Theorem (Kronecker s Theorem: rst version). Let K be a eld and p(X ) K [X ] be a polynomial of positive degree. Then there is a nite extension L/K for which p(X ) has a root in L. Proof. We begin by factorizing p(X ) K [X ] into irreducible monic factors qj (X ) together with a constant factor c: p(X ) = cq1 (X ) qr (X ). 39 Now for any j we can form the quotient eld K [x]/(qj (X )) which is a nite dimensional (simple) extension of K and in which the coset X + (qj (X )) satis es the equation qj (X + (qj (X ))) = 0 + (qj (X )). Hence p(X ) has a root in K [x]/(qj (X )). Of course, this construction is only interesting if qj (X ) to has degree bigger than 1 since a linear polynomial already has a root in K . To answer Question 3.22 we iterate this construction. Namely, having found one root u1 in an extension L1 /K we discard the linear factor X u1 and consider the polynomial p1 (X ) = p(X ) L1 [X ]. X u1 We can repeat the argument to form a nite extension of L1 (and hence of K ) containing a root of p1 (X ) and so on. At each stage we either already have another root in L1 or we need to enlarge the eld to obtain one. 3.26. Theorem (Kronecker s Theorem: second version). Let K be a eld and p(X ) K [X ] be a polynomial of positive degree. Then there is a nite extension E/K which is a splitting eld of p(X ) over K . In practise we often have extension elds lying around in nature containing roots and we can work inside of these. When working over Q (or any other sub eld of C) we can always nd roots in C by the Fundamental Theorem of Algebra. We then refer to a sub eld of C which is a splitting eld as the splitting sub eld. 3.27. Example. Find a splitting eld E/Q for p(X ) = X 4 4 over Q and determine [E : Q]. Solution. Notice that p(X ) = (X 2 2)(X 2 + 2), so rst we adjoin the roots 2 of (X 2 2) to form Q( 2, 2) = Q( 2) which gives an extension Q( 2)/Q of degree 2. Next consider the polynomial X 2 + 2 Q( 2)[X ]. The complex roots of X 2 + 2 are 2i and these are not real, so this polynomial is irreducible in Q( 2)[X ]. Hence we need to consider Q( 2, 2i) = Q( 2, i) and the extension Q( 2, i)/Q( 2) which has degree 2. C Q ( 2 , i) adjoin roots of X 2 + 2 2 Q( 2) adjoin roots of X 2 2 2 Q Thus the splitting sub eld of p(X ) over Q in C is Q( 2, i) and [Q( 2, i) : Q] = 4. 40 Of course we could have started by rst adjoining roots of X 2 + 2 and then adjoining roots of X 2 2, thus giving the tower C Q ( 2 , i) adjoin roots of X 2 2 2 Q( 2i) adjoin roots of X 2 + 2 2 Q An important point is that if a splitting eld exists inside of a given extension eld F/K , it is unique as a sub eld of F . 3.28. Proposition. Let F/K be an extension eld and p(X ) K [X ]. If E1 , E2 splitting sub elds for p(X ) over K then E1 = E2 . F are Proof. Let u1 , . . . , uk F be the distinct roots of p(X ) in F . By de nition, K (u1 , . . . , uk ) is the smallest sub eld containing K and all the uj . But K (u1 , . . . , uk ) must be contained in any splitting sub eld, so E1 = K (u1 , . . . , uk ) = E2 . Since we will frequently encounter quadratic polynomials we record a useful result on roots of such polynomials. Recall that p(X ) = aX 2 + bX + c K [X ] is quadratic if a = 0 and its discriminant is = b2 4ac K. The proof of the next result is the standard one which works provided 2 has an inverse in K , i.e., when char K = 2. 3.29. Proposition. Let K be a eld of characteristic di erent from 2. Then the quadratic polynomial p(X ) = aX 2 + bX + c K [X ] has no roots in K if is not a square in K ; one root b/(2a) = (2a) 1 b if = 0; two distinct roots b + b = (2a) 1 ( b + ), = (2a) 1 ( b ), 2a 2a if = 2 for some non-zero K . In particular, the splitting eld of p(X ) over K is K if is a square in K and K ( ) otherwise, where is one of the two square roots of in some extension of K such as the algebraic closure K which we will introduce in Section 3.4. 3.30. Example. Find a splitting eld E/Q for p(X ) = X 3 2 over Q and determine [E : Q]. Solution. By the Eisenstein Test 1.38, p(X ) is irreducible over Q. One root of p(X ) is 3 2 R so we adjoin this to Q to form an extension Q( 3 2)/Q of degree 3. Now 3 3 3 p(X ) = (X 2)(X 2 + 2X + ( 2)2 ) 41 2 and the second factor has the non-real complex roots 3 2 3 , 3 2 3 lying in the extension Q( 3 2, 3 )/Q( 3 2) of degree 2. So the splitting sub eld of X 3 2 in C over Q is Q( 3 2, 3 ) with [Q( 3 2, 3 ) : Q] = 6. 2 An alternative strategy would have been to adjoin one of the other roots 3 2 3 or 3 2 3 rst. We could also have begun by adjoining 3 to form the extension Q( 3 )/Q, but none of the roots of p(X ) lie in this eld so the extension Q( 3 2, 3 )/Q( 3 ) of degree 3 is obtained by adjoining one and hence all of the roots. Figure 3.1 shows all the sub elds of the extension Q( 3 2, 3 )/Q. C 2 R Q( 3 2, 3 ) VV ii r iiii rr V 2 iiii rrrr i 2 VVV iiii rr V ii r 2 VV 3 iiii Q( 3 2 3 ) VVV Q( 3 2) Q( 3 2 3 ) `` vvv VV `` vvv VV `` vvv VV `` vvv `` 3 vv3 vvv ` 3 Q( 3 ) vvv ``` ss vvv `` 2 ss s vvv `` sss ` vvv ss s 2 Q Figure 3.1. The sub elds of Q( 3 2, 3 )/Q 3.3. Monomorphisms between extensions 3.31. Definition. For extensions F/K and L/K , let MonoK (L, F ) denote the set of all monomorphisms L F which x the elements of K . 3.32. Remark. We always have AutK (F ) MonoK (F, F ) and MonoK (F, F ) is closed under composition but is not always a group since elements are not necessarily invertible. If F/K is nite, then we do have MonoK (F, F ) = AutK (F ) since every injective K -linear transformation is surjective and so invertible. We will also use the following notation. 3.33. Definition. Let F/K be an extension and p(X ) K [X ]. Set Roots(p, F ) = {u F : p(u) = 0}, the set of roots of p(X ) in F . This is always a nite set which may of course be empty, which happens precisely when p(X ) has no root in F . 42 Suppose that p(X ) K [X ] is an irreducible polynomial which we might as well assume is monic, and let F/K be an extension. Then if t F is a root of p(X ), the evaluation homomorphism t : K [X ] F factors through the quotient monomorphism t : K [X ]/(p(X )) F whose image is K (t) F . Of course, there is one such monomorphism for each root of p(X ) in F . If we x one such root t0 and identify K [X ]/(p(X )) with K (t0 ) via t0 , then each root of p(X ) in F gives rise to a monomorphism t = t 1 : K (t0 ) F for which t (t0 ) = t. t0 t = t 1 t 0 K (t0 ) o t0 = K [X ]/(p(X )) t /* F Notice that if : K [X ]/(p(X )) F is any homomorphism extending the identity function on K , then the coset X + (p(X )) must be sent by to a root of p(X ) in F , hence every such homomorphism arises this way. This discussion is summarized in the following result. 3.34. Proposition. Let F/K be a eld extension. Let p(X ) K [X ] be an irreducible polynomial with t0 F be a root of p(X ). Then there is a bijection Roots(p, F ) MonoK (K (t0 ), F ) given by t t , where t : K (t0 ) F has the e ect t (t0 ) = t. 3.35. Example. Show that MonoQ (Q( 2), C) has two elements. Solution. We have Q( 2) Q[X ]/(X 2 2) where X 2 2 is irreducible over Q. Hence = the Q-monomorphisms we want send 2 to 2 which are the complex roots of X 2 2. In fact both possibilities occur, giving monomorphisms id, : Q( 2) C, where (a + b 2) = a b 2. We can replace C by Q( 2) to obtain MonoQ (Q( 2), C) = MonoQ (Q( 2), Q( 2)) = AutQ (Q( 2)). We will see that this is not always true. 3.36. Example. Show that MonoQ (Q( 3 2), C) has 3 elements but MonoQ (Q( 3 2), Q( 3 2)) contains only the identity function. Solution. Here minpolyQ, 2 (X ) = X 3 2 and there are 3 complex roots 3 2, 3 2 3 , 3 2 3 2 3 . As two of these roots are not real, MonoQ (Q( 3 2), Q( 3 2)) contains only the identity since Q( 3 2) R. 2 Each of the above roots corresponds to one of the sub elds Q( 3 2), Q( 3 2 3 ) or Q( 3 2 3 ) of C and there are 3 monomorphisms 0 , 1 , 2 : Q( 3 2) C given by 3 3 3 3 0 (a + b 2 + c( 2)2 ) = a + b 2 + c( 2)2 , 3 3 3 3 2 1 (a + b 2 + c( 2)2 ) = a + b 2 3 + c( 2)2 3 , 2 3 3 3 3 2 (a + b 2 + c( 2)2 ) = a + b 2 3 + c( 2)2 3 . These mappings have images 3 3 0 Q( 2) = Q( 2), 3 3 1 Q( 2) = Q( 2 3 ), 43 2 3 3 2 Q( 2) = Q( 2 3 ). 3.37. Proposition. Let F/K and L/K be extensions. (i) For p(X ) K [X ], each monomorphism MonoK (L, F ) restricts to a function p : Roots(p, L) Roots(p, F ) which is an injection. (ii) If MonoK (L, L), then p : Roots(p, L) Roots(p, L) is a bijection. Proof. (i) For u Roots(p, L) we have p( (u)) = (p(u)) = (0) = 0, so maps Roots(p, L) into Roots(p, F ). Since is an injection its restriction to Roots(p, L) L is also an injection. (ii) From (i), p : Roots(p, L) Roots(p, L) is an injective function from a nite set to itself, hence it is also surjective by the Pigeon Hole Principle. Thus p : Roots(p, L) Roots(p, L) is a bijection. Part (ii) says that any automorphism of L/K permutes the set of roots in L of a polynomial p(X ) K [X ]. This gives us a strong hold on the possible automorphisms. In the case of nite, or more generally algebraic, extensions it is the key to understanding the automorphism group and this is a fundamental insight of Galois Theory. 3.38. Example. Determine MonoQ (Q( 3 2, 3 ), C). Solution. We have already met the extension Q( 3 2, 3 )/Q in Example 3.30 and we will make use of information from there. We build up the list of monomorphisms in stages. First consider monomorphisms that x 3 2 and hence x the sub eld Q( 3 2). These form the subset 3 3 MonoQ( 2) (Q( 2, 3 ), C) MonoQ (Q( 2, 3 ), C). 3 We know that Q( 3 2, 3 ) = Q( 3 2)( 3 ) and that 3 is a root of the irreducible cyclotomic polynomial 3 (X ) = X 2 + X + 1 Q( 3 2)[X ]. So there are two monomorphisms id, 0 xing Q( 3 2), where 0 has the e ect ( ) 3 2 3 2 0 : . 2 3 3 Next we consider monomorphisms that send 3 2 to 3 2 3 . This time we have 2 distinct ways to extend to elements of MonoQ (Q( 3 2, 3 ), Q( 3 2, 3 )) since again we can send 3 to either 3 2 or 3 . The possibilities are ) ( ) ( 3 3 2 3 2 3 2 3 2 3 , 1 : . 1 : 2 3 3 3 3 2 Finally we consider monomorphisms that send 3 2 to 3 2 3 . There are again two possibilities ( ( 2) 2) 3 3 2 3 2 3 2 3 2 3 , 2 : . 2 : 2 3 3 3 3 These are all 6 of the required monomorphisms. It is also the case here that 3 3 3 3 MonoQ (Q( 2, 3 ), C) = MonoQ (Q( 2, 3 ), Q( 2, 3 )) = AutQ (Q( 2, 3 )), so these form a group. It is a nice exercise to show that AutQ (Q( 3 2, 3 )) S3 , the symmetric = group on 3 objects. It is also worth remarking that | AutQ (Q( 3 2, 3 ))| = [Q( 3 2, 3 ) : Q]. 44 We end this section with another useful result. 3.39. Proposition. Let L/K be an extension and MonoK (L, L). Then restricts to an automorphism alg : Lalg Lalg . Proof. Suppose that u Lalg , say p(u) = 0 for some p(X ) K [X ] of positive degree. Then p( (u)) = (p(u)) = (0) = 0, so maps Lalg L into itself and therefore gives rise to a restriction alg : Lalg Lalg which is also a monomorphism. We must show that alg is a bijection by showing it is surjective. Let v Lalg and suppose that q (v ) = 0 for some q (X ) K [X ] of positive degree. Now Roots(q, L) = since it contains v , and it is also nite. Then q : Roots(q, L) Roots(q, L) is a bijection by Proposition 3.37(ii), hence v = q (w) = (w) for some w Roots(q, L) Lalg . This shows that v im and so alg is surjective. 3.4. Algebraic closures An important property of the complex numbers is that C is algebraically closed. 3.40. Theorem (Fundamental Theorem of Algebra for C). Every non-constant polynomial p(X ) C[X ] has a root in C. 3.41. Corollary. Every non-constant polynomial p(X ) C[X ] has a factorization p(X ) = c(X u1 ) (X ud ), where c, u1 , . . . , ud C and this is unique apart from the order of the roots uj . It is natural to pose the following question. 3.42. Question. Let K be a eld. Is there an algebraically closed eld F containing K ? By taking F alg we might as well ask that such a eld be algebraic over K . 3.43. Definition. Let K be a eld. An extension F/K is called an algebraic closure of K if F is algebraic over K and algebraically closed. 3.44. Theorem. Let K be a eld. (i) There is an algebraic closure of K . (ii) Let F1 and F2 be algebraic closures of K . Then there is an isomorphism : F1 F2 which xes the elements of K . K } eee ee }} ee }} e }} ~} / F2 F1 Hence algebraic closures are essentially unique. Proof. See [3] for a proof using Zorn s Lemma (see Axiom 3.48) which is logically equivalent to the Axiom of Choice. 45 Because of the uniqueness we usually x some choice of algebraic closure of K and write K or K alg cl , referring to it as the algebraic closure of K . We are already familiar with the example C = C. There are some immediate consequences of Theorem 3.44. We will temporarily write . E1 = E2 to indicate that for extensions E1 /K and E2 /K there is an isomorphism E1 E2 xing the elements of K . 3.45. Proposition. Let K be a eld. . (i) If L/K is an algebraic extension, then L = K . . (ii) If L/K is an extension, then so is L/K and (L)alg = K . Proof. (i) By Proposition 3.16, every element of L is algebraic over K . Since L is algebraically closed it is an algebraic closure of K . (ii) Every non-constant polynomial in (L)alg [X ] has a root in L; indeed, by Proposition 3.16, all of its roots are in fact algebraic over K since (L)alg is algebraic over K . Hence these roots lie in (L)alg , which shows that it is algebraically closed. For example, we have Q = Calg and R = C. There is a stronger result than Theorem 3.44(ii), the Monomorphism Extension Theorem, which we will nd useful. Again the proof uses Zorn s Lemma which we state below. First we need some de nitions. 3.46. Definition. A partially ordered set (X, ) consists of a set X and a binary relation such that whenever x, y, z X , x x; if x y and y if x y and y z then x z ; x then x = y . (X, ) is totally ordered if for every pair x, y X , at least one of x y or y x is true. 3.47. Definition. Let (X, ) be a partially ordered set and Y X . y X is an upper bound for Y if for every y Y , y An element x X is a maximal element of X if x y = y. y = x. 3.48. Axiom (Zorn s Lemma). Let (X, ) be a partially ordered set in which every totally ordered subset has an upper bound. Then X has a maximal element. 3.49. Theorem (Monomorphism Extension Theorem). Let M/K be an algebraic extension and L/K M /K . Suppose that 0 : L K is a monomorphism xing the elements of K . 46 Then there is an extension of 0 to a monomorphism : M K . 8K @ M 0 L K = /K Proof. We consider the set X consisting of all pairs (F, ), where F/L M /L and : F K extends 0 . We order X using the relation for which (F1 , 1 ) (F2 , 2 ) whenever F1 F2 and 2 extends 1 . Then (X, ) is a partially ordered set. Suppose that Y X is a totally ordered subset. Let F= F. (F, ) Y Then F /L M /L. Also there is a function : F K de ned by (u) = (u) whenever u F for (F, ) Y . It is straightforward to check that if u F for (F , ) Y then (u) = (u), so is well-de ned. Then for every (F, ) Y we have (F, ) (F , ), so (F , ) is an upper bound for Y . By Zorn s Lemma there must be a maximal element of X , (M0 , 0 ). Suppose that M0 = M , so there is an element u M for which u M0 . Since M is algebraic / over K it is also algebraic over M0 , hence u is algebraic over M0 . If minpolyM0 ,u (X ) = a0 + + an 1 X n 1 + X n , then the polynomial f (X ) = 0 (a0 ) + + 0 (an 1 )X n 1 + X n ( 0 M0 )[X ] is also irreducible and so it has a root v in K (which is also an algebraic closure of 0 M0 K ). The Homomorphism Extension Property 1.22 of the polynomial ring M0 [X ] applied to the monomorphism 0 : M0 K yields a homomorphism 0 : M0 [X ] K extending 0 and for which 0 (u) = v . This factors through the quotient ring M0 [X ]/(minpolyM0 ,u (X )) to give a monomorphism 0 : M0 (u) K extending 0 . But then (M0 , 0 ) (M0 (u), 0 ) and (M0 , 0 ) = (M0 (u), 0 ), contradicting the maximality of (M0 , 0 ). Hence M0 = M and so we can take = 0 . 3.50. Example. Let u K and suppose that p(X ) = minpolyK,u (X ) K [X ]. Then for any other root of p(X ), v K say, there is a monomorphism v : K (u) K with v (u) = v . This extends to a monomorphism : K K . 3.51. Definition. Let u, v K . Then v is conjugate to u over K or is a conjugate of u over K if there is a monomorphism : K K xing K for which v = (u). 47 3.52. Lemma. If u, v K , then v is conjugate to u over K if and only if minpolyK,u (v ) = 0. Proof. Suppose that v = (u) for some MonoK (K, K ). If minpolyK,u (X ) = a0 + a1 X + + ad 1 X d 1 + X d , then a0 + a1 u + + ad 1 ud 1 + ud = 0 and so a0 + a1 v + + ad 1 v d 1 + v d = (a0 + a1 u + + ad 1 ud 1 + ud ) = 0. The converse follows from Example 3.50. 3.5. Multiplicity of roots and separability Let K be a eld. Suppose that f (X ) K [X ] and u K is a root of f (X ), i.e., f (u) = 0. Then we can factor f (X ) as f (X ) = (X u)f1 (X ) for some f1 (X ) K [X ]. 3.53. Definition. If f1 (u) = 0 then u is a multiple or repeated root of f (X ). If f1 (u) = 0 then u is a simple root of f (X ). We need to understand more clearly when an irreducible polynomial has a multiple root since this turns out to be important in what follows. Consider the formal derivative on K [X ], i.e., the function : K [X ] K [X ] given by (f (X )) = f (X ) = a1 + 2a2 X + + dad X d 1 , where f (X ) = a0 + a1 X + a2 X 2 + + ad X d with aj K . 3.54. Proposition. The formal derivative : K [X ] K [X ] has the following properties. (i) is K -linear. (ii) is a derivation, i.e., for f (X ), g (X ) K [X ], (f (X )g (X )) = (f (X ))g (X ) + f (X ) (g (X )). (iii) If char K = 0, then ker = K and is surjective. (iv) If char K = p > 0, then ker = {h(X p ) : h(X ) K [X ]} and im is spanned by the monomials X k with p (k + 1). Proof. (i) This is routine. (ii) By K -linearity, it su ces to verify this for the case where f (X ) = X r and g (X ) = X s with r, s 0. But then (X r+s ) = (r + s)X r+s 1 = rX r 1 X s + sX r X s 1 = (X r )X s + X r (X s ). (iii) If f (X ) = a0 + a1 X + a2 X 2 + + ad X d then (f (X )) = 0 a1 = 2a2 = = dad = 0. So (f (X )) = 0 if and only if f (X ) = a0 K . It is also clear that every polynomial g (X ) K [X ] has the form g (X ) = (f (X ) where f (X ) is an anti-derivative of g (X ). 48 (iv) For a monomial X m , (X m ) = mX m 1 and this is zero if and only if p | m. Using this we see that (a0 + a1 X + a2 X 2 + + ad X d ) = 0 am = 0 whenever p m. Also, im is spanned by the monomials X k for which (X k+1 ) = 0, which are the ones with p (k + 1). We now apply the formal derivative to detect multiple roots. 3.55. Proposition. Let f (X ) K [X ] have a root u L for some extension L/K . Then u is a multiple root of f (X ) if and only if f (X ) and f (X ) have a common factor of positive degree in K [X ] which vanishes at u. Proof. Working in L[X ], let f (X ) = (X u)f1 (X ). Then f (X ) = f1 (X ) + (X u)f1 (X ), so f (u) = f1 (u). Hence u is a multiple root if and only if f (X ) and f (X ) have a common factor in L[X ] (and hence in K [X ] by Proposition 3.12) and which vanishes at u. 3.56. Corollary. If f (X ) is irreducible in K [X ] then a root u is a multiple root if and only if f (X ) = 0. In particular, this can only happen if char K > 0. 3.57. Corollary. If char K = 0 and f (X ) is irreducible in K [X ], then every root of f (X ) is simple. 3.58. Example. For n 1, show that each of the roots of f (X ) = X n 1 in C is simple. Solution. We have f (X ) = (X n 1) = nX n 1 , so for any root of f (X ), f ( ) = n n 1 = 0. 3.59. Example. Show that 2i is a multiple root of f (X ) = X 4 + 8X 2 + 16. Solution. We have f (X ) = 4X 3 +16X . Using Long Division and the Euclidean Algorithm we nd that gcd(f (X ), f (X )) = X 2 + 4, where 2i is also a root of X 2 + 4. Hence 2i is a multiple root of f (X ). In fact, X 4 + 8X 2 + 16 = (X 2 + 4)2 , so this is obvious. 3.60. Example. Let p > 0 be a prime and suppose that L/Fp is an extension. Show that each of the roots of f (X ) = X p 1 in L is multiple. Solution. We have f (X ) = (X p 1) = pX p 1 = 0, so if is any root of f (X ) then f ( ) = 0. Later we will see that 1 is the only root of X p 1. 3.61. Definition. An irreducible polynomial p(X ) K [X ] is separable over K if every root of p(X ) in an extension L/K is simple. By Corollary 3.56, this is equivalent to requiring that p (X ) = 0. If u L is a multiple root of p(X ), then the multiplicity of u in p(X ) is the maximum m such that p(X ) = (X u)m q (X ) for some q (X ) L[X ]. 3.62. Proposition. Let K be a eld and let K be an algebraic closure. If the irreducible polynomial p(X ) K [X ] has distinct roots u1 , . . . , uk K , then the multiplicities of the uj are equal. Hence in K [X ], p(X ) = c(X u1 )m (X uk )m , where c K and m 1. 49 Proof. Let u K be a root of p(X ) and suppose that it has multiplicity m, so we can write p(X ) = (X u)m p1 (X ) where p1 (X ) K (u)[X ] and p1 (u) = 0. Now let v K be any other root of p(X ). By Proposition 3.34, there is a monomorphism v : K (u) K for which v (u) = v . When p(X ) is viewed as an element of K (u)[X ], the coe cients of p(X ) are xed by v . Then v ((X u)m p1 (X )) = (X u)m p1 (X ), and so (X v )m p1 (X ) = (X u)m p1 (X ), where p1 (X ) K [X ] is obtained applying v to the coe cients of p1 (X ). Now by Corollary 1.34, (X v )m must divide p1 (X ) in K [X ], and therefore the multiplicity of v must be at least m. Interchanging the r les of u and v we nd that the multiplicities of u and v are in fact equal. o 3.63. Corollary. Let K be a eld and let K be an algebraic closure. If the irreducible polynomial p(X ) K [X ] has distinct roots u1 , . . . , uk K which are all simple then in K [X ], p(X ) = c(X u1 ) (X uk ), where c K and k = deg p(X ). 3.64. Corollary. Let K be a eld and let u K . Then the number of distinct conjugates of u is deg minpolyK,u (X ) , m where m is the multiplicity of u in minpolyK,u (X ). 3.65. Definition. An algebraic element u L in an extension L/K is separable if its minimal polynomial minpolyK,u (X ) K [X ] is separable. 3.66. Definition. An algebraic extension L/K is called separable if every element of L is separable over K . 3.67. Example. An algebraic extension L/K of a eld of characteristic 0 is separable by Corollary 3.57. 3.68. Definition. Let L/K be a nite extension. The separable degree of L over K is (L : K ) = | MonoK (L, K )|. 3.69. Lemma. For a nite simple extension K (u)/K , (K (u) : K ) = | Roots(minpolyK,u , K )|. If K (u)/K is separable, then [K (u) : K ] = (K (u) : K ). Proof. This follows from Proposition 3.34 applied to the case L = K . Any nite extension L/K can be built up from a succession of simple extensions (3.1) K (u1 )/K, K (u1 , u2 )/K (u1 ), , L = K (u1 , . . . , uk )/K (u1 , . . . , uk 1 ). So we can use the following to compute (L : K ) = (K (u1 , . . . , uk ) : K ). 50 3.70. Proposition. Let L/K and M/L be nite extensions. Then (M : K ) = (M : L)(L : K ). Proof. For MonoK (M, K ) let L MonoK (L, K ) be its restriction to L. By the Monomorphism Extension Theorem 3.49, each element of MonoK (L, K ) extends to a monomorphism M K , so every element MonoK (L, K ) has the form = L for some MonoK (M, K ). Since (L : K ) = | MonoK (L, K )|, we need to show that the number of such is always (M : L) = | MonoL (M, K )|. So given MonoK (L, K ), choose any extension to a monomorphism : K K ; by Proposition 3.39, is an automorphism. Of course, restricting to M K we obtain a monomorphism M K . Now for any extension : M K of we can form the composition 1 : M K ; notice that if u L, then 1 (u) = 1 ( (u)) = u, hence 1 MonoL (M, K ). Conversely, each MonoL (M, K ) gives rise to a monomorphism : M K which extends . In e ect, this shows that there is a bijection { } extensions of to monomorphism a M K MonoL (M, K ), so (M : L) = | MonoL (M, K )| agrees with the number of extensions of to a monomorphism M K . Therefore we have the desired formula (M : K ) = (M : L)(L : K ). 3.71. Corollary. Let L/K be a nite extension. Then (L : K ) | [L : K ]. Proof. If L/K is a simple extension then by Propositions 3.62 and 3.34 we know that this is true. The general result follows by building up L/K as a sequence of simple extensions as in (3.1) and then using Theorem 2.6(ii) which gives [L : K ] = [K (u1 ) : K ] [K (u1 , u2 ) : K (u1 )] [K (u1 , . . . , uk ) : K (u1 , . . . , uk 1 )]. For each k , (K (u1 , . . . , uk ) : K (u1 , . . . , uk 1 )) divides [K (u1 , . . . , uk ) : K (u1 , . . . , uk 1 )], so the desired result follows. 3.72. Proposition. Let L/K be a nite extension. Then L/K is separable if and only if (L : K ) = [L : K ]. Proof. Suppose that L/K is separable. If K E L, then for any u L, u is algebraic over E , and in the polynomial ring E [X ] we have minpolyE,u (X ) | minpolyK,u (X ). As minpolyK,u (X ) is separable, so is minpolyE,u (X ), and therefore L/E is separable. Clearly E/K is also separable. We have (L : K ) = (L : E ) (E : K ) and [L : K ] = [L : E ] [E : K ], so to verify that (L : K ) = [L : K ] it su ces to show that (L : E ) = [L : E ] and (E : K ) = [E : K ]. Expressing L/K in terms of a sequence of simple extensions as in (3.1), we have (L : K ) = (K (u1 ) : K ) (L : K (u1 , . . . , uk 1 )), [L : K ] = [K (u1 ) : K ] [L : K (u1 , . . . , uk 1 )]. Now we can apply Lemma 3.69 to each of these intermediate separable simple extensions to obtain (L : K ) = [L : K ]. For the converse, suppose that (L : K ) = [L : K ]. We must show that for each u L, u is separable. For the extensions K (u)/K and L/K (u) we have (L : K ) = (L : K (u)) (K (u) : K ) 51 and [L : K ] = [L : K (u)] [K (u) : K ]. By Corollary 3.71, there are some positive integers r, s for which [L : K (u)] = r(L : K (u)) and [K (u) : K ] = s(K (u) : K ). Hence (L : K (u))(K (u) : K ) = rs(L : K (u))(K (u) : K ), which can only happen if r = s = 1. Thus (K (u) : K ) = [K (u) : K ] and so u is separable. 3.73. Proposition. Let L/K and M/L be nite extensions. Then M/K is separable if and only if L/K and M/L are separable. Proof. If M/K is separable then [M : K ] = (M : K ) and so by Proposition 3.70, [M : L][L : K ] = (M : L)(L : K ). This can only happen if [M : L] = (M : L) and [L : K ] = (L : K ), since (M : L) [M : L] and (L : K ) [L : K ]. By Proposition 3.72 this implies that L/K and M/L are separable. Conversely, if L/K and M/L are separable then [M : L] = (M : L) and [L : K ] = (L : K ), hence [M : K ] = [M : L][L : K ] = (M : L)(L : K ) = (M : K ). Therefore M/K is separable. 3.6. The Primitive Element Theorem Recall from De nition 3.10 that a nite extension L/K is simple if there is an element u L for which L = K (u), and such an element is called a primitive element. 3.74. Theorem (Primitive Element Theorem). Let L/K be a nite separable extension. Then L has a primitive element, hence L/K is a simple extension. Proof. The case where K is a nite eld will be dealt with in Proposition 5.16. So we will assume that K is in nite. Since L is built up from a sequence of simple extensions it su ces to consider the case L = K (u, v ). Let p(X ), q (X ) K [X ] be the minimal polynomials of u and v over K . Suppose that the distinct roots of p(X ) in K are u = u1 , . . . , ur , while the distinct roots of q (X ) are v = v1 , . . . , vs . By the separability assumption, r = deg p(X ) and s = deg q (X ). Since K is in nite, we can choose an element t K for which t = u ui vj v whenever j = 1. Then taking w = u + tv L, we nd that w = ui + tvj whenever j = 1. De ne the polynomial (of degree r) h(X ) = p(w tX ) K (w)[X ] L[X ]. Then h(v ) = p(u) = 0, but h(vj ) = p(ui ) = 0 for any j = 1 by construction of t, so none of the other vj is a zero of h(X ). Now since the polynomials h(X ), q (X ) K (w)[X ] have exactly one common root in K , namely v , by separability their greatest common divisor in K (w)[X ] is a linear polynomial which must be X v , hence v K (w) and so u = w tv K (w). This shows that K (u, v ) K (w) and therefore K (w) = K (u, v ). 52 3.75. Corollary. Let L/K be a nite extension of a eld of characteristic 0. Then L has a primitive element. Proof. Since Q K , K is in nite and by Example 3.67 L/K is separable. To nd a primitive element we can always use the method suggested by the proof of Theorem 3.74, however a try it and see approach will often be su cient. 3.76. Example. Find a primitive element for the extension Q( 3, i)/Q. Solution. Consider 3 + i. Then working over the sub eld Q( 3) Q( 3, i) we nd that i Q( 3) R and / (X ( 3 + i))(X ( 3 i)) = X 2 2 3X + 4 Q( 3)[X ], hence Now taking X 2 2 3X + 4 = minpolyQ( 3), 3+i (X ). (X 2 2 3X + 4)(X 2 + 2 3X + 4) = X 4 4X 2 + 16 Q[X ], we see that minpolyQ, 3+i (X ) | (X 4 4X 2 + 16) in Q[X ]. Notice that ( 3 i) 1 ( 3 i) 1 ( 3 + i) = = ( 3 i) Q( 3 + i), = 3+1 4 ( 3 + i)( 3 i) since ( 3 + i) 1 Q( 3 + i). Hence 1 1 3 = (( 3 + i) + ( 3 i)), i = (( 3 + i) ( 3 i)), 2 2 are both in Q( 3 + i), showing that Q( 3, i) Q( 3 + i) and so Q( 3, i) = Q( 3 + i). Thus we must have deg minpolyQ, 3+i (X ) = 4, and so minpolyQ, 3+i (X ) = X 4 4X 2 + 16. There is a general phenomenon illustrated by Example 3.76. 3.77. Proposition. Let u K be separable over K . Then minpolyK,u (X ) = (X 1 (u)) (X d (u)), where 1 , . . . , d are the elements of MonoK (K (u), K ). In particular, the polynomial (X 1 (u)) (X d (u)) K [X ] is in K [X ] and is irreducible therein. Proof. Since K (u) is separable then by Lemma 3.52, d = deg minpolyK,u (X ) = [K (u) : K ] = (K (u) : K ). In Example 3.76 we have [Q( 3, i) : Q] = [Q( 3, i) : Q( 3)][Q( 3) : Q] = 2 2 = 4. There are four monomorphisms k : Q( 3, i) Q( 3, i) given by ( ( ( ) ) ) 3 3 3 3 3 3 , 3 = , 4 = . 1 = id, 2 = i i i i i i 53 Then 2 ( 3 + i) = ( 3 i), 3 ( 3 + i) = ( 3 + i), 4 ( 3 + i) = ( 3 i) , so 3 i)(X 3 + i)(X + 3 i)(X + 3 + i) = X 4 4X 2 + 16 Q[X ]. Hence this polynomial is irreducible. So we have [Q( 3 + i) : Q] = 4 and Q( 3 + i) = Q( 3, i). (X 3.7. Normal extensions and splitting elds Let K be an algebraic closure for the eld K and let E/K K /K be a nite extension. If MonoK (E, K ), then by Remark 3.32, E = E if and only if E E . 3.78. Definition. E/K is normal if E = E for every MonoK (E, K ). 3.79. Remark. If E/K is a normal extension then whenever an irreducible polynomial p(X ) K [X ] has a root in E , it splits in E since by Lemma 3.52 each pair of roots of p(X ) is conjugate over K and one can be mapped to the other by a monomorphism K K which must map E into itself. 3.80. Theorem. A nite extension E/K is normal if and only if it is a splitting eld over K for some polynomial f (X ) K [X ]. Proof. Suppose that E/K is normal. Then there is a sequence of extensions K K (u1 ) K (u1 , u2 ) K (u1 , . . . , un ) = E Construct a polynomial by taking f (X ) = minpolyK,u1 (X ) minpolyK,u2 (X ) minpolyK,un (X ). Then by Remark 3.79, f (X ) splits in E . Also, E is generated by some of the roots of f (X ). Hence E is a splitting eld for f (X ) over K . Now suppose that E is a splitting eld for g (X ) K [X ], so that E = K (v1 , . . . , vk ), where v1 , . . . , vk are the distinct roots of g (X ) in E . Now any monomorphism MonoK (E, K ) must map these roots to (v1 ), . . . , (vk ) which are also roots of g (X ) and therefore lie in E (see Proposition 3.34). Since permutes the roots vj , we have E = K (v1 , . . . , vk ) = K ( (v1 ), . . . , (vk )) = K (v1 , . . . , vk ) = E. 3.81. Corollary. Let E/L and L/K be nite extensions. If E/K is normal then E/L is normal. Proof. If E is the splitting eld of a polynomial f (X ) K [X ] over K , then E is the splitting eld of f (X ) over L. These result makes it easy to recognize a normal extension since it is su cient to describe it as a splitting eld for some polynomial over K . In Chapter 4 we will see that separable normal extensions play a central r le in Galois Theory, indeed these are known as Galois extensions. o 54 Exercises on Chapter 3 3.1. Prove Proposition 3.2. 3.2. Finding splitting sub elds E polynomials. p1 (X ) = X 4 X 2 +1, C over Q and determine [E : Q] for each of the following p2 (X ) = X 6 2, p3 (X ) = X 4 +2, p4 (X ) = X 4 +5X 3 +10X 2 +10X +5. [Hint: for p4 (X ), consider p4 (Y 1) Q[Y ].] 3.3. Prove that AutQ (Q( 3 2, 3 )) S3 , the symmetric group on 3 elements, as claimed in the = solution of Example 3.38. [Hint: work out the e ect of each automorphism on the three roots of the polynomial X 3 2.] 3.4. Let k be a eld of characteristic char k = p > 0 and k(T ) be the eld of rational functions in T over k. Show that the polynomial g (X ) = X p T k(T )[X ] is irreducible and has a multiple root in k(T ). How does g (X ) factor in k(T )[X ]? 3.5. Find primitive elements for the extensions Q( 5, 10)/Q, Q( 2, i)/Q, Q( 3, i)/Q, Q( 4 3, i)/Q, in each case nding it minimal polynomial over Q. [Hint: look for elements of high degree over Q, or use the method of proof of Theorem 3.74.] 3.6. Prove the following converse of Proposition 3.20: Let L/K be a nite extension. If there are only nitely many subextensions F/K then L/K is simple, i.e., L = K (w) for some w L. L/K , [Hint: First deal with the case where L = K (u, v ), then use induction on n to prove the general case L = K (u1 , . . . , un ).] 3.7. Let K be a eld. Show that every quadratic (i.e., of degree 2) extension E/K is normal. Is such an extension always separable? 3.8. Let f (X ) Q[X ] be an irreducible polynomial of odd degree greater than 1 and having only one real root u R. Show that Q(u)/Q is not a normal extension. 55 CHAPTER 4 Galois extensions and the Galois Correspondence In this Chapter we will study the structure of Galois extensions and their associated Galois groups, in particular we will explain how these are related through the Galois Correspondence. Throughout the chapter, let K be a eld. 4.1. Galois extensions 4.1. Definition. A nite extension E/K is a ( nite ) Galois extension if it is both normal and separable. From Section 3.5 we know that for such a Galois extension E/K , [E : K ] = (E : K ) and also every monomorphism MonoK (E, K ) maps E into itself, hence restricts to an automorphism of E which will be denoted |E . >K ~~ ~ ~~ ~~ ~ = E / E |E K = /K Also, by the Monomorphism Extension Theorem 3.49, every automorphism AutK (E ) extends to a monomorphism E K xing elements of K . So there is a bijection MonoK (E, K ) AutK (E ) and we have (4.1) | AutK (E )| = (E : K ) = [E : K ]. 4.2. Definition. For a nite Galois extension E/K , the group Gal(E/K ) = AutK (E ) is called the Galois group of the extension or the Galois group of E over K . The elements of Gal(E/K ) are called (Galois ) automorphisms of E/K . Notice that Equation (4.1) implies (4.2) | Gal(E/K )| = (E : K ) = [E : K ]. We can also reformulate the notion of conjugacy introduced in De nition 3.51. 4.3. Definition. Let E/K a nite Galois extension and u, v E . Then v is conjugate to u if there is a Gal(E/K ) for which v = (u); we also say that v is a conjugate of u. 57 It is easy to see that for u, v K , there is a nite Galois extension E/K in which v is a conjugate of u if and only v is a conjugate of u over K in the old sense. Here is a slightly di erent way to understand this. First notice that every element AutK (K, K ) restricts to a monomorphism E K whose image is contained in E , hence gives rise to an automorphism E : E E . Similarly, if F/K is any nite normal extension with E F , every automorphism F : F F restricts to an automorphism E : E E . The proof of the next result is left as an exercise. 4.4. Proposition. If E/K is a nite Galois extension, then the function AutK (K, K ) AutK (E, E ); E is a surjective group homomorphism. If F/K K /K is any nite normal extension with E then there is a surjective group homomorphism AutK (F, F ) AutK (E, E ); F F E . Furthermore, for AutK (K, K ) we have ( F )F = E . E 4.2. Working with Galois groups Let E/K be a nite Galois extension. Then we know that E is a splitting eld for some polynomial over K since E/K is normal. We also know that E is a simple extension of K since E/K is separable. Hence E is a splitting eld for the minimal polynomial of any primitive element for E/K ; this minimal polynomial has degree [E : K ]. It is often convenient to use these facts to interpret elements of the Galois group as permutations of the roots of some polynomial which splits over E . 4.5. Example. Describe the Galois group Gal(Q( 2, 3)/Q) as a subgroup of the group of permutations of the roots of (X 2 2)(X 2 3) Q[X ]. Solution. We have [Q( 2, 3) : Q] = [Q( 2, 3) : Q( 2)] [Q( 2) : Q] = 4, and the following non-trivial elements of the Galois group together with the element identity 1 = id: 2 2 2 2 2 2 2 2 2 2 2 2 , 3 = , 4 = . 2 = 3 3 3 3 3 3 3 3 3 3 3 3 Writing the roots in the list 2, 2, 3, 3 and numbering them from 1 to 4, these automorphisms correspond to the following permutations in S4 expressed in cycle notation: 2 (1 2), 3 (3 4), 4 (1 2)(3 4). 4.6. Example. Using a primitive element u for the extension, describe the Galois group Gal(Q( 2, 3)/Q) as a subgroup of the group of permutations of the roots of minpolyQ,u (X ) Q[X ]. 58 Solution. We have Q( 2, 3) = Q( 2 + 3) and the conjugates of u = 2 + 3 are 2 3. Listing these as 2 + 3, 2 3, 2 + 3, 2 3, and after numbering them accordingly, we nd the correspondences 2 (1 3)(2 4), 3 (1 2)(3 4), 4 (1 4)(2 3). Next we summarize the properties of Galois groups that can be deduced from what we have established so far. Recall that for an extension F/K and a polynomial f (X ) K [X ], Roots(f, F ) denotes the set of roots of f (X ) in F . 4.7. Recollection. Recall that an action of a group G on a set X is transitive if for every pair of elements x, y X , there is an element g G such that y = gx (so there is only one orbit); the action is faithful or e ective if for every non-identity element h G, there is an element z X such that hz = z . 4.8. Theorem. Let E/K be a nite Galois extension. Suppose that E is the splitting eld of a separable irreducible polynomial f (X ) K [X ] of degree n. Then the following are true. (i) Gal(E/K ) acts transitively and faithfully on Roots(f, E ). (ii) Gal(E/K ) can be identi ed with a subgroup of the group of permutations of Roots(f, E ). If we order the roots u1 , . . . , un then Gal(E/K ) can be identi ed with a subgroup of Sn . (iii) | Gal(E/K )| divides n! and is divisible by n. As we have seen in Examples 4.5 and 4.6, in practise it is often easier to use a not necessarily irreducible polynomial to determine and work with a Galois group. 4.9. Example. The Galois extension Q( 8 )/Q has degree [Q( 8 ) : Q] = 4 and it has the following automorphisms apart from the identity: 3 : 8 8 , 5 : 8 8 , 7 : 8 8 . If we list the roots of the minimal polynomial minpolyQ, (X ) = 8 (X ) = X 4 + 1 357 in the order 8 , 8 , 8 , 8 , we nd that these automorphisms correspond to the following permutations in S4 : (1 2)(3 4), (1 3)(2 4), (1 4)(2 3). So the Galois group Gal(Q( 8 )/Q) corresponds to {id, (1 2)(3 4), (1 3)(2 4), (1 4)(2 3)} S4 . Noticing that 1 1 8 = + i, 2 2 we easily nd that 2, i Q( 8 ); hence Q( 2, i) Q( 8 ). Since [Q( 2, i) : Q] = 4, we have Q( 2, i) = Q( 8 ). Notice that Q( 2, i) is the splitting eld of f (X ) = (X 2 2)(X 2 + 1) over 59 Q. Now list the roots of f (X ) in the order 2, 2, i, i, and observe that 2 2 2 2 2 2 2 2 : : (1 2)(3 4), (1 2), i i i i i i i i 2 2 2 2 : (3 4). i i i i In this description, the Galois group Gal(Q( 8 )/Q) = Gal(Q( 2, i)/Q) corresponds to the subgroup {id, (1 2), (3 4), (1 2)(3 4)} S4 . While it can be hard to determine Galois groups in general, special arguments can sometimes be exploited. 4.10. Example. Suppose that f (X ) = X 3 + aX 2 + bX + c Q[X ] is an irreducible cubic and that f (X ) has only one real root. Then Gal(Q(f (X ))/Q) S3 . = Proof. Let u1 R be the real root of f (X ) and let u2 , u3 be the remaining complex roots. Then Q(f (X )) = Q(u1 , u2 , u3 ) and in fact [Q(f (X )) : Q] = 6 since [Q(f (X )) : Q] | 6 and u2 Q(u1 ) / R. Hence Gal(Q(f (X ))/Q) is isomorphic to a subgroup of S3 and so Gal(Q(f (X ))/Q) S3 since the orders agree. We also have Q(f (X )) R = Q(u1 ). = The Galois group Gal(Q(f (X ))/Q) contains an element of order 3 which corresponds to a 3-cycle when viewed as a permutation of the roots u1 , u2 , u3 ; we can assume that this is (1 2 3). It also contains an element of order 2 obtained by restricting complex conjugation to Q(f (X )); this xes u1 and interchanges u2 , u3 , so it corresponds to the transposition (2 3). 4.11. Remark. Such examples occur when the cubic polynomial f (X ) has local maximum and minimum at real values c+ and c with f (c+ ), f (c ) > 0 or f (c+ ), f (c ) < 0. This happens for example with f (X ) = X 3 3X + 3 which has local extrema at 1 and f (1) = 1, f ( 1) = 5. Given a Galois extension E/K , we will next study subextensions L/K E /K and subgroups Gal(E/K ), focusing on the relationship between objects of these types. 4.3. Subgroups of Galois groups and their xed elds Let E/K a Galois extension and suppose that Gal(E/K ). Consider the subset of elements of E xed by , E = {u E : , (u) = u}. 4.12. Lemma. E E is a sub eld of E containing K . Proof. For u, v E and , (u + v ) = (u) + (v ) = u + v, Also, if u = 0, (uv ) = (u) (v ) = uv. (u 1 ) = (u) 1 = u 1 . 60 Finally, if t K then (t) = t, so K 4.13. Definition. E E . E is the xed sub eld of . By Proposition 3.73, the extensions E/E and E /K are separable. E/E is also normal, so this is a Galois extension; we will identify its Galois group. Notice that [E : E ] = (E : E ) = | Gal(E/E )|. Now each element of Gal(E/E ) is also an element of Gal(E/K ) and Gal(E/E ) Gal(E/K ). Notice that by de nition Gal(E/E ), so Lagrange s Theorem implies that | | divides | Gal(E/E )|. In fact we have 4.14. Proposition. For Gal(E/K ), we have Gal(E/E ) = and the equations [E : E ] = | Gal(E/E )| = | |, [E : K ] = | Gal(E/K )| . | | Proof. We know that E/E is separable, so by the Primitive Element Theorem 3.74 it is simple, say E = E (u). Now let the distinct elements of be 1 = id, 2 , . . . , h , where h = | |. Consider the polynomial of degree h f (X ) = (X u)(X 2 (u)) (X h (u)) E [X ]. Notice that f (X ) is unchanged by applying any k to its coe cients since the roots j (u) are permuted by k . Hence, f (X ) E [X ]. This shows that [E : E ] = [E (u) : E ] Since h = | |. Gal(E/E ), we also have h = | | | Gal(E/E )| = [E : E ]. Combining these two inequalities we obtain [E : E ] = | Gal(E/E )| = | | = h and therefore = Gal(E/E ). 4.4. Sub elds of Galois extensions and relative Galois groups Let E/K a Galois extension and suppose that L/K E /K (i.e., K L E ). Then E/L is also a Galois extension whose Galois group Gal(E/L) is sometimes called the relative Galois group of the pair of extensions E/K and L/K . The following is immediate. 4.15. Lemma. The relative Galois group of the pair of extensions L/K E /K is a subgroup of Gal(E/K ), i.e., Gal(E/L) Gal(E/K ), and its order is | Gal(E/L)| = [E : L]. 4.16. Proposition. Let L/K E /K . Then L = E Gal(E/L) . Proof. Clearly L E Gal(E/L) . Suppose that u E L. By Theorem 4.8(i), there is an automorphism Gal(E/L) such that (u) = u, hence u E Gal(E/L) . This shows that / Gal(E/L) Gal(E/L) = L. E L and therefore E We need to understand when Gal(E/L) Gal(E/K ) is actually a normal subgroup. The next result explains the connection between the two uses of the word normal which both ultimately derive from their use in Galois theory. 61 4.17. Proposition. Let E/K be a nite Galois extension and L/K E /K . (i) The relative Galois group Gal(E/L) of the pair of extensions L/K E /K is a normal subgroup of Gal(E/K ) if and only if L/K is a normal extension. (ii) If L/K is normal and hence a Galois extension, then there is a group isomorphism = Gal(E/K )/ Gal(E/L) Gal(L/K ); Gal(E/L) |L . Proof. (i) Suppose that Gal(E/L) Gal(E/K ), i.e., for all Gal(E/L) and Gal(E/K ), we have 1 Gal(E/L). Now if u L, then for any Gal(E/K ) and Gal(E/L), (u) E satis es (u) = ( 1 (u)) = (u), since 1 Gal(E/L); hence (u) E Gal(E/L) = L. By the Monomorphism Extension Theorem 3.49, every monomorphism L K xing K extends to a monomorphism E K which must have image E , so the above argument shows that L/K is normal. Conversely, if L/K is normal, then for every Gal(E/K ) and v L, (v ) L, so for every Gal(E/L), ( (v )) = (v ) and therefore 1 (v ) = v. This shows that 1 Gal(E/L). Hence for every Gal(E/K ), Gal(E/L) 1 = Gal(E/L), which shows that Gal(E/L) Gal(E/K ). (ii) If Gal(E/K ), then L = L since L/K is normal. Hence we can restrict to an automorphism of L, |L : L L; |L (u) = (u). Then |L is the identity function on L if and only if Gal(E/L). It is easy to see that the function Gal(E/K ) Gal(L/K ); |L is a group homomorphism whose kernel is Gal(E/L). Thus we obtain an injective homomorphism Gal(E/K )/ Gal(E/L) Gal(L/K ) for which [E : K ] = [L : K ] = | Gal(L/K )|. [E : L] Hence this homomorphism is an isomorphism. | Gal(E/K )/ Gal(E/L)| = 4.5. The Galois Correspondence and the Main Theorem of Galois Theory We are now almost ready to state our central result which describes the Galois Correspondence associated with a nite Galois extension. We will use the following notation. For a nite Galois extension E/K , let S(E/K ) = the set of all subgroups of Gal(E/K ); F(E/K ) = the set of all subextensions L/K of E/K. 62 Each of these sets is ordered by inclusion. Since every subgroup of a nite group is a nite subset of a nite set, S(E/K ) is also a nite set. De ne two functions by E/K : F(E/K ) S(E/K ); E/K (L) = Gal(E/L), E/K : S(E/K ) F(E/K ); E/K ( ) = E . 4.18. Theorem (Main Theorem of Galois Theory). Let E/K be a nite Galois extension. Then the functions E/K and E/K are mutually inverse bijections which are order-reversing. E/K F(E/K ) o / S(E/K ) E/K Under this correspondence, normal subextensions of E/K correspond to normal subgroups of Gal(E/K ) and vice versa. Proof. We know from Proposition 4.16 that for an extension L/K in F(E/K ), E/K ( E/K (L)) = E/K (Gal(E/L)) = E Gal(E/L) = L. Also, by Proposition 4.14 for H S(E/K ) we have E/K ( E/K ( )) = E/K (E ) = Gal(E/E ) = . This shows that E/K and E/K are mutually inverse and so are inverse bijections. Let L1 /K, L2 /K F(E/K ) satisfy L1 /K L2 /K . Then Gal(E/L2 ) Gal(E/L1 ) since L1 L2 and so if Gal(E/L2 ) then xes every element of L1 . Hence E/K (L2 ) E/K (L1 ) and so E/K reverses order. Similarly, if 1 , 2 S(E/K ) and 1 2 , then E 2 E 1 since if w E 2 then it is xed by every element of 1 (as 1 is a subset of 2 ). Hence E/K reverses order. There is an immediate consequence of the Main Theorem 4.18 which is closely related to Proposition 3.20. 4.19. Corollary. Let E/K be a nite Galois extension. Then there are only nitely many subextensions L/K E /K . Proof. Since the set S(E/K ) is nite, so is F(E/K ). When dealing with a nite Galois extension E/K , we indicate the subextensions in a diagram with a line going upwards indicating an inclusion. We can also do this with the subgroups of the Galois group Gal(E/K ) with labels indicating the index of the subgroups. In e ect, the Galois Correspondence inverts these diagrams. 4.20. Example. Figure 4.1 shows the Galois Correspondence for the extension of Example 3.30. As noted at the end of Example 3.38, the Galois group here is Gal(Q( 3 2, 3 )/Q) S3 . It = is useful to make this isomorphism explicit. First take the 3 roots of the polynomial X 3 2 for 2 which E is the splitting eld over Q; these are 3 2, 3 2 3 , 3 2 3 which we number in the order they are listed. Then the monomorphisms id, 0 , 1 , 1 , 2 , 2 extend to automorphisms of E , each of which permutes these 3 roots in the following ways given by cycle notation: 0 = (2 3), 1 = (1 2 3), 1 = (1 2), 63 2 = (1 3 2), 2 = (1 3). E = Q( 3 2, 3 ) I ii ee eeeeee kkkk ii eeeeee kkkkk ii ee 2 ii eeeeee 2 kkk 2 eee kkk ii eeeeee eee 2 3 ii Q( 3 2) Q( 3 2 3 ) Q( 3 2 3 ) ii ii N N qq I qq ii qq ii qq ii qq 3 qq3 qq 3 Q( 3 ) qq Q lll qq 2lllll qq qq lll qq lll lll l Q S E/K Gal(E/Q) k 2 kkk ww kkk wwww k kkk w kkk 3 www kk 3k ww kkk Gal(E/Q( 3 )) 3 ww kkk y ww kkk ww kkk yy w kk yy ww kkk yy kkk ww yy 3 3 3 2 )) 3 yy Gal(E/Q( 2)) Gal(E/Q( 2 3 )) Gal(E/Q( 2 3 yy y yy yy 2 yy 2 2 yy {id} Figure 4.1. The Galois Correspondence for E = Q( 3 2, 3 )/Q We nd that 3 Gal(E/Q( 3 )) = {id, 1 , 2 } {id, (1 2 3), (1 3 2)}, Gal(E/Q( 2)) = {id, 0 } {id, (2 3)}, = = 2 3 3 Gal(E/Q( 2 3 )) = {id, 2 } {id, (1 3)}, Gal(E/Q( 2 3 )) = {id, 1 } {id, (1 2)}. = = Notice that {id, (1 2 3), (1 3 2)} S3 and so Q( 3 )/Q is a normal extension. Of course Q( 3 ) is the splitting eld of X 3 1 over Q. 4.6. Galois extensions inside the complex numbers and complex conjugation When working with Galois extensions contained in the complex numbers it is often useful to make use of complex conjugation as an element of a Galois group. Let E/Q be a nite Galois extension with E/Q C/Q. Setting ER = R E , we have Q ER E . 4.21. Proposition. Complex conjugation ( ) : C C restricts to an automorphism of E over Q, ( )E/Q : E E . Furthermore, (i) ( )E/Q agrees with the identity function if and only if ER = E . (ii) If ER = E , then ( )E/Q = {id, ( )E/Q } Z/2, = ( )E/Q hence, E = E and [E : E ] = 2. R R 64 Proof. Let u E . As E/Q is normal, minpolyQ,u (X ) Q[X ] splits over E , so all of its complex roots lie in E . But ( ) permutes the roots of this minimal polynomial. Therefore ( ) maps E into itself. (i) For z C, z = z if and only if z R. (ii) Here | ( )E/Q | = 2, and E ( )E/Q = {u E : u = u} = ER . C 2 R E nnn nnn nn nnn nnn 2 n ER ee ee ee ee Q We will usually write ( ) rather than ( )E/Q when no confusion seems likely to result. 4.22. Example. Consider the cyclotomic extension Q( 8 )/Q where 1 1 8 = e i/4 = + i. 2 2 From Example 4.9 we know that Q( 8 ) = Q( 2, i), and we easily see that [Q( 8 ) : Q] = 4, Q( 8 )R = Q( 2). 4.7. Galois groups of even and odd permutations We have seen that for a monic separable polynomial f (X ) K [X ] of degree n, the Galois group of its splitting eld E over K can naturally be thought of as a subgroup of the symmetric group Sn , where we view the latter as permuting the roots of f (X ). It is reasonable to ask when Gal(E/K ) An rather than just Gal(E/K ) Sn . We rst recall an interpretation of the sign of a permutation Sn , sgn = 1. For each pair i, j with 1 i<j n, exactly one of the inequalities (i) < (j ) or (j ) < (i) must hold and the ratio ( (j ) (i))/(j i) is either positive or negative. It is easily veri ed that the right-hand side of the following equation must have value 1 and so (4.3) sgn = 1 i<j n (j ) (i) . j i Note that this is sometimes used as the de nition of sgn . 65 Suppose that f (X ) factorizes over E as f (X ) = (X u1 ) (X un ) = n (X ui ). i=1 Here u1 , . . . , un E are the roots of f (X ); as we have assumed that f (X ) is separable, the ui are distinct. 4.23. Definition. The discriminant of f (X ) is Discr(f (X )) = (uj ui )2 E. 1 i<j n Notice that Discr(f (X )) = 0 since ui = uj if i = j . 4.24. Remark. There is an explicit formula for computing Discr(f (X )) is terms of its coe cients. For polynomials p(X ) = a0 + a1 X + + am X m , q (X ) = b0 + b1 X + + bn X n , their resultant is the (m + n) (m + n) determinant (with n rows of ai s and m rows of bi s) a0 0 .. . (4.4) Res(p(X ), q (X )) = a1 a0 .. . . . . . . . . am 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . a1 . . . . . . . am 0 . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. .. .. .. . . . . . . . . . . . 0 0 .. . 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 a 0 a 1 . . . . . . . am . b0 b1 . . . . . . . bn 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 0 b0 b1 . . . . . . . bn 0 . . . . . . . . . . . . . . . . . . . . . . . . . . 0 .. .. .. .. .. .. .. .. .. .. .. .. .. .. . . . . . . . . . . . . . . 0 .......................... 0 b0 b1 ............ bn Then if f (X ) is monic with d = deg f (X ), Discr(f (X )) = ( 1)d(d 1)/2 Res(f (X ), f (X )). (4.5) So for example, Discr(X 3 + pX + q ) = ( 1)3 Res(X 3 + pX + q, 3X 2 + p) q 0 = ( 1) p 0 0 p q 0 p 0 0 p 3 0 p 1 0 0 3 0 0 1 0 = 4p3 27q 2 . 0 3 Here are some low degree examples of discriminants obtained with the aid of Maple. n = 2: n = 3: n = 4: Discr(a0 + a1 X + X 2 ) = 4a0 + a2 . 1 Discr(a0 + a1 X + a2 X 2 + X 3 ) = 27a2 + 18a0 a1 a2 + a2 a2 4a3 a0 4a3 . 0 12 2 1 Discr(a0 + a1 X + a2 X 2 + a3 X 3 + X 4 ) = 18a3 a3 a2 6a2 a2 a0 192a3 a1 a2 27a4 1 31 0 1 + 144a2 a2 a2 + 144a0 a2 a2 + 256a3 4a3 a3 128a2 a2 + 16a4 a0 4a3 a2 30 1 0 31 20 2 21 + 18a3 a1 a2 a0 80a3 a1 a2 a0 27a4 a2 + a2 a2 a2 4a3 a2 a0 . 3 2 30 231 23 66 n = 5: Discr(a0 + a1 X + a2 X 2 + a3 X 3 + a4 X 4 + X 5 ) = 2250a4 a2 a3 36a0 a3 a3 128a2 a4 30 41 31 + 2000a2 a3 a2 900a1 a3 a2 2500a3 a4 a1 50a2 a2 a2 900a4 a3 a2 27a4 a4 3750a3 a2 a3 0 1 30 0 041 20 41 0 + 356a3 2 a2 2 a4 a1 a0 + 560a3 a2 a2 a2 2050a3 a2 a2 a4 a1 80a2 a2 a4 a1 3 630a3 3 a2 a4 a0 2 240 0 3 + 825a2 a2 a2 + 16a3 a3 a0 + 2000a2 a2 a3 6a2 a2 a3 128a2 a4 a2 + 16a4 a3 a0 4a3 a3 a2 320 32 40 241 240 24 241 + 108a5 a2 + 108a5 a0 746a3 a2 a0 a4 2 a1 2 27a4 a2 + 256a5 a3 4a3 a2 a2 + 144a3 a2 a3 30 2 21 40 321 21 + 144a2 a4 a3 + 3125a4 + 256a5 72a4 a2 a1 a0 + 18a3 a2 a3 a3 + 560a2 a2 a0 a2 + 16a4 a3 41 0 1 3 41 3 1 31 + 18a3 a2 3 a4 a1 2 72a3 a2 4 a4 a0 + 144a3 2 a2 a4 3 a0 2 192a4 4 a1 a3 a0 2 630a3 a2 3 a1 a0 + 24a2 3 a4 2 a1 a0 + a3 2 a2 2 a4 2 a1 2 6a4 3 a1 2 a3 2 a0 80a3 a2 2 a4 3 a1 a0 4a3 2 a2 3 a4 2 a0 + 2250a1 a2 a2 1600a3 a3 a3 192a4 a4 a2 1600a0 a3 a2 4a3 a3 a2 27a4 a2 a2 20 40 1 1 314 340 + 1020a4 2 a3 2 a0 2 a1 + 18a3 3 a2 a4 2 a0 a1 + 160a2 a4 3 a0 2 a1 + 144a2 a4 4 a0 a1 2 + 24a4 a1 2 a3 3 a0 + 1020a0 a4 a2 2 a1 2 + 160a0 a4 a1 3 a3 . So for example, Discr(X 5 + a4 X 4 + a0 ) = a3 (3125a0 + 256a5 ), 0 4 Discr(X 5 + a1 X + a0 ) = 256a5 + 3125a4 . 1 0 4.25. Proposition. For every Gal(E/K ), (Discr(f (X ))) = Discr(f (X )). Hence Discr(f (X )) E Gal(E/K ) = K . Proof. For Gal(E/K ) (Discr(f (X ))) = Sn , we have (u (j ) u (i) )2 = 1 i<j n 2 (u (j ) u (i) ) . 1 i<j n Now for each pair i, j with i < j , (uj ui ) = u (j ) u (i) , and by Equation (4.3) (4.6) (u (j ) u (i) ) = sgn 1 i<j n (uj ui ) = ( 1) 1 i<j n (uj ui ). 1 i<j n Hence (Discr(f (X ))) = Discr(f (X )). Since E Gal(E/K ) = K , we have Discr(f (X )) K . Now let (f (X )) = (uj ui ) E. 1 i<j n Then = Discr(f (X )), so the square roots of Discr(f (X )) are (f (X )). Now consider the e ect of Gal(E/K ) on (f (X )) E . By Equation (4.6), (f (X ))2 ( (f (X ))) = sgn (f (X )) = (f (X )). If (f (X )) K , this means that sgn = 1. On the other hand, if (f (X )) K then / K ( (f (X ))) = E Gal(E/K ) An . Of course | Gal(E/K )/ Gal(E/K ) An | = 2. 67 4.26. Proposition. The Galois group Gal(E/K ) Discr(f (X )) is a square in K . Sn is contained in An if and only if 4.27. Example. For the polynomials of Examples 6.40 and 6.42 we obtain Discr(X 5 35X 4 + 7) = 4611833296875 = 33 56 74 29 157, (X 5 35X 4 + 7) = 53 3 72 3 29 157 i = 18375 13659 i Q; / Discr(X 5 + 20X + 16) = 1024000000 = 216 56 , (X 5 + 20X + 16) = 28 53 Q. 4.8. Kaplansky s Theorem In this section we give a detailed account of the Galois theory of irreducible rational polynomials f (X ) = X 4 + aX 2 + b Q[X ]. The following result describes the Galois groups that occur and the proof introduces some useful computational techniques. 4.28. Theorem (Kaplansky s Theorem). Let f (X ) = X 4 + aX 2 + b Q[X ] be irreducible. (i) If b is a square in Q then Gal(Q(f (X ))/Q) Z/2 Z/2. = (ii) If b(a2 4b) is a square in Q then Gal(Q(f (X ))/Q) Z/4. = (iii) If neither b nor b(a2 4b) is a square in Q then Gal(Q(f (X ))/Q) = D8 . Proof. Let g (X ) = X 2 + aX + b Q[X ]. Notice that g (X ) must be irreducible since otherwise f (X ) would factorize, hence (a2 4b) is not a square in Q. Setting d = (a2 4b) Q and taking to be a square root of d (so Q), we nd that the roots of g (X ) are ( a )/2 Q. / / Then the roots of f (X ) are u, v , where ( a + ) ( a ) , v2 = , 2 2 so the splitting eld of f (X ) over Q is E = Q(u, v ) which contains the quadratic extension Q( )/Q. Since deg f (X ) = 4, we also have 4 | [E : Q]. In fact, since E is obtained by at most 3 successive quadratic extensions we also have [E : Q] | 8. (i) We have a2 d 4b (uv )2 = u2 v 2 = = = b, 4 4 hence uv is a square root of b which is in Q. Setting c = uv Q, we nd that v = c/u Q(u). This shows that E = Q(u) and we have the following Galois tower. u2 = E = Q(u) 2 Q( ) 2 Q In particular [E : Q] = 4 = | Gal(E/Q)|. Notice that for the Galois extension Q( )/Q there must be a normal subgroup N Gal(E/Q) with Q( ) = E N , Gal(Q( )/Q) = Gal(E/Q)/N. 68 Hence there is an element Gal(E/Q) for which ( ) = . This element must also have the e ects (u) = v and (v ) = u. Given u we might as well choose v so that (u) = v . There is also an element N for which (u) = u and we also have (v ) = v . Notice that if (v ) = u then easy calculation shows that (v ) = (v ) = u, ( ) = ( ) = , hence we might as assume that (v ) = u since if necessary we can replace our original choice by . We now have c c (u) = , (u) = u, (u) = (u) = . u u These satisfy 2 = 2 = ( )2 = id = the identity, = . This shows that Gal(Q(f (X ))/Q) = Gal(E/Q) = {id, , , } Z/2 Z/2 = the Klein 4-group. = (ii) If bd is a square in Q, then (uv )2 = u2 v 2 d = bd, which is a square in Q, so we can write uv = c Q or equivalently v = c/(u ) Q(u) since Q( ) Q(u). This shows that E = Q(u, v ) = Q(u) and again we have a Galois tower E = Q(u) 2 Q( ) 2 Q with [E : Q] = 4 = | Gal(E/Q)|. Since Q( )/Q is Galois there is an element Gal(E/Q) with ( ) = and this has the e ect (u) = v ; given u we might as well choose v so that (u) = v . Notice that c c (v ) = = = u, (u ) v so 2 (u) = u. This shows that Gal(Q(f (X ))/Q) = Gal(E/Q) = {id, , 2 3 } Z/4 = a cyclic group of order 4. = (iii) Suppose that d, b and bd are not squares in Q. By an easy calculation we nd that (uv )2 = b, so uv E is a square root of b in E . Suppose that uv Q( ); then uv = p + q for some p, q Q. By squaring we obtain b = (p2 + q 2 d) + 2pq , and so pq = 0. We cannot have q = 0 since this would imply that b was a square in Q; if p = 0 then b = q 2 d and so bd = (qd)2 , implying that bd was a square in Q. Thus we have Q(uv ) Q( ) = Q. A similar discussion shows that Q(uv ) Q( ) = Q = Q(uv ) Q(uv ). 69 So we have a Galois tower which includes the following sub elds. E = Q(u, v ) Q(uv, ) Q( ) r rrr rrr rrr www www www w Q(uv ) vvv vvv vvv vvv 2 Q(uv ) ppp ppp 2 ppp 2 ppp Q Choose Gal(E/Q(uv )) Gal(E/Q) so that ( ) = . By renaming v to v if necessary, we may assume that v = (u) and so u = (v ). Notice that 2 = id. Choose Gal(E/Q( )) Gal(E/Q) with (uv ) = uv . We must have either (u) = u or (v ) = v , so by interchanging if necessary we can assume that (u) = u and (v ) = v . Notice that 2 = id. Choose Gal(E/Q( , uv )) Gal(E/Q) so that (u) = u. Then we must have (v ) = v since (uv ) = uv . Notice that 2 = id. Setting = we nd (u) = v and (v ) = u. Then 2 = and has order 4. Also, = = 1 . The eight elements id, , , 1 , , , , 1 form a group isomorphic to the dihedral group of order 8, D8 . Therefore we have Gal(Q(f (X ))/Q) = Gal(E/Q) D8 , = and [E : Q] = 8. The corresponding Galois tower is E = Q(u, v ) 2 Q(uv, ) Q( ) r 2 rrr rr r rrr www www 2 www w 2 Q(uv ) vvv vvv vvv vvv 2 Q(uv ) p ppp 2 ppp pp 2 ppp Q 70 4.29. Example. We have the following Galois groups: Gal(Q(X 4 + 1)/Q) Z/2 Z/2; = Gal(Q(X 4 + 4X 2 + 2)/Q) Z/4; = Gal(Q(X 4 + 2X 2 + 2)/Q) D8 . = Exercises on Chapter 4 4.1. If f (X ) K [X ] is a separable polynomial, prove that the splitting eld of f (X ) over K is a nite Galois extension of K . 4.2. Let K be a eld for which char K = 2, 3 and suppose that f (X ) K [x] is a cubic polynomial. (a) Show that there u, v K with u = 0 such that f (uX + v ) = X 3 + aX + b for some a, b K . If f (X ) is monic, deduce that a, b K ; under what conditions is this always true? (b) If g (X ) = X 3 + aX + b K [x] is irreducible and E = K (g (X )) is its splitting eld over K , explain why Gal(E/K ) is isomorphic to one of the groups S3 or A3 . (c) Continuing with the notation and assumptions of (b), suppose that w1 , w2 , w3 are the distinct roots of g (X ) in E and let = (w1 w2 )2 (w2 w3 )2 (w1 w3 )2 E. Show that = 4b3 27a2 , and hence K . If = (w1 w2 )(w3 w3 )(w1 w3 ), show that A if K, 3 Gal(E/K ) = S3 if K. / [Hint: Consider K ( ) in Gal(E/K ) S3 .] E and the e ect on the element of even and odd permutations 4.3. Show that f (X ) = X 3 3X +1 Q[X ] is irreducible over Q, and show that its discriminant is a square in Q. Prove that the Galois group of f (X ) over Q is cyclic. 4.4. This is a revision exercise on nite groups of small order. (a) Show that every non-abelian nite group has order at least 6. (b) Let D8 be the dihedral group with the eight elements , , 2 , 3 , , , 2 , 3 satisfying 4 = , = 1 = 3 . 2 = , Find all the normal subgroups of D8 . 71 4.5. Use Kaplansky s Theorem 4.28 to nd the Galois group of the splitting eld E of the polynomial X 4 + 3 Q[X ] over Q. Determine all the subextensions F E for which F/Q is Galois. 4.6. Find the Galois groups for each of the following extensions: Q(X 3 10)/Q; Q( 2)(X 3 10)/Q( 2); Q( 3 i)(X 3 10)/Q( 3 i); Q( 23 i)(X 3 X 1)/Q( 23 i); K (X 3 X 1)/K for K = Q, Q( 5), Q( 5 i), Q(i). 4.7. Let p > 0 be a prime. Let K be a eld with char K = p. Suppose that 0 = a K and f (X ) = X p a K [X ]. Let L/K where L is a splitting eld for f (X ) over K . (a) Show that f (X ) has p distinct roots in L. If u L is one such root, describe the remaining roots and show that L contains p distinct p-th roots of 1. (b) Suppose that K contains p distinct p-th roots of 1. Show that either f (X ) is irreducible over K or it factors into p distinct linear factors over K . (c) Suppose that the only p-th root of 1 in K is 1. Show that either f (X ) is irreducible over K or it has a root in K . 4.8. Let K be a eld of characteristic char K = p where p > 0 is a prime. Suppose that 0 = a K and f (X ) = X p a K [X ]. Show that if f (X ) has no root in K then it is irreducible over K . 72 CHAPTER 5 Galois extensions for elds of positive characteristic In this chapter we will investigate extensions of elds of positive characteristic, especially nite elds. A thorough account of nite elds and their applications can be found in [6]. Throughout this chapter we will assume that K is a eld of prime characteristic p = char K > 0, containing the prime sub eld Fp . 5.1. Finite elds If K is a nite eld, then K is an Fp -vector space. Our rst goal is to count the elements of K . Here is a more general result. 5.1. Lemma. Let F be a nite eld with q elements and let V be an F -vector space. Then dimF V < if and only if V is nite in which case |V | = q dimF V . Proof. If d = dimF V < , then for a basis v1 , . . . , vd we can express each element v V uniquely in the form v = t1 v1 + + td vd , where t1 , . . . , td F . Clearly there are exactly q d such expressions, so |V | = q d . Conversely, if V is nite then any basis has nitely many elements and so dimF V < . 5.2. Corollary. Let F be a nite eld and E/F an extension. Then E is nite if and only if E/F is nite and then |E | = |F |[E :F ] . 5.3. Corollary. Let K be a nite eld. Then K/Fp is nite and |K | = p[K :Fp ] . Our next task is to show that for each power pd there is a nite eld with pd elements. We start with the algebraic closure Fp of Fp and consider the polynomial d pd (X ) = X p X Fp [X ]. Notice that d (X ) = 1, hence by Proposition 3.55 every root of pd (X ) in Fp is simp ple. Therefore by Corollary 1.35 pd (X ) must have exactly pd distinct roots in Fp , say 0, u1 , . . . , upd 1 . Then in Fp [X ] we have d X p X = X (X u1 ) (X upd 1 ), and each root is separable over Fp . Let Fpd = {u Fp : pd (u) = 0} Fp , d 1 Notice that u F0d if and only if up p F0d = {u Fpd : u = 0}. p = 1. 5.4. Proposition. For each d 1, Fpd is a nite sub eld of Fp with pd elements and F0d = F . Furthermore, the extension Fpd /Fp is a separable splitting eld. p pd 73 Proof. If u, v Fpd then by the Idiot s Binomial Theorem 1.11, d d d d d (u + v )p (u + v ) = (up + v p ) (u + v ) = (up u) + (v p v ) = 0, d d d (uv )p uv = up v p uv = uv uv = 0. Furthermore, if u = 0 then up 1 = 1 and so u has multiplicative inverse up 2 . Hence Fpd Fp . Notice that Fp Fpd , so Fpd /Fp is a nite extension. In any eld the non-zero elements are always invertible, hence F0d = F . p pd d d 5.5. Definition. The nite sub eld Fpd Fp is called the Galois eld of order pd . The notation GF(pd ) is often used in place of Fpd . Of course, Fp1 = GF(p1 ) = GF(p) = Fp and [Fpd : Fp ] = d. 5.6. Proposition. Let d 1. (i) Fpd Fp is the splitting sub eld for each of the polynomials X p X and X p 1 1 over Fp . (ii) Fpd Fp is the unique sub eld with pd elements. (iii) If F is any eld with pd elements then there is a monomorphism F Fp with image Fpd , hence F Fpd . = d d Proof. (i) As Fpd consists of exactly the roots of pd (X ) in Fp , it is the splitting sub eld. d The non-zero elements of Fpd are the roots of X p 1 1, so Fpd is also the splitting sub eld for this polynomial. Fp have pd elements. Notice that the non-zero elements of F form a group F (ii) Let F under multiplication. This group is abelian and has pd 1 elements, so by Lagrange s Theorem, d d each element u F has order dividing pd 1, therefore up 1 = 1 and so up = u. But this means every element of F is a root of pd (X ) and so F Fpd ; equality follows since these d elements. sub elds both have p (iii) Apply the Monomorphism Extension Theorem 3.49 for K = L = Fp and M = F . By (ii), the image of the resulting monomorphism must be Fpd , therefore F Fpd . = It is worth noting the following consequence of this result and the construction of Fpd . 5.7. Corollary. Let K be a nite eld of characteristic p. Then K/Fp is a nite Galois extension. 5.8. Example. Consider the polynomial X 4 X F2 [X ]. By inspection, in the ring F2 [X ] we nd that X 4 X = X 4 + X = X (X 3 + 1) = X (X + 1)(X 2 + X + 1). Now X 2 + X + 1 has no root in F2 so it must be irreducible in F2 [X ]. Its splitting eld is a quadratic extension F2 (w)/F2 where w is one of the roots of X 2 + X + 1, the other being w + 1 since the sum of the roots is the coe cient of X . This tells us that every element of F4 = F2 (w) can be uniquely expressed in the form a + bw with a, b F2 . To calculate products we use the fact that w2 = w + 1, so for a, b, c, d F2 we have (a + bw)(c + dw) = ac + (ad + bc)w + bdw2 = (ac + bd) + (ad + bc + bd)w. 74 5.9. Example. Consider the polynomial X 9 X F3 [X ]. Let us nd an irreducible polynomial of degree 2 in F3 [X ]. Notice that X 2 + 1 has no root in F3 , hence X 2 + 1 F3 [X ] is irreducible; so if u F3 is a root of X 2 + 1 then F3 (u)/F3 has degree 2 and F3 (u) = F9 . Every element of F9 can be uniquely expressed in the form a + bu with a, b F3 . Multiplication is carried out using the relation u2 = 1 = 2. By inspection, in the ring F3 [X ] we nd that X 9 X = X (X 8 1) = (X 3 X )(X 2 + 1)(X 2 + X 1)(X 2 X 1). So X 2 + X 1 and X 2 X 1 are also quadratic irreducibles in F3 [X ]. We can nd their roots in F9 using the quadratic formula since in F3 we have 2 1 = ( 1) 1 = 1. The discriminant of X 2 + X 1 is 1 4( 1) = 5 = 2 = u2 , so its roots are ( 1)( 1 u) = 1 u. Similarly, the discriminant of X 2 X 1 is 1 4( 1) = 5 = 2 = u2 and its roots are ( 1)(1 u) = 1 u. Then we have F9 = F3 (u) = F3 (1 u) = F3 ( 1 u). There are two issues we can now clarify. 5.10. Proposition. Let Fpm and Fpn be two Galois elds of characteristic p. Then Fpm Fpn if and only if m | n. Proof. If Fpm Fpn , then by Corollary 5.2, pn = (pm )[Fpn :Fpm ] = pm[Fpn :Fpm ] , so m | n. If m | n, write n = km with k n mk up = up m 1. Then for u Fpm we have up = u, so m = (up )p Hence u Fpn and therefore Fpm m(k 1) = up m(k 1) m = = up = u. Fpn . This means that we can think of the Galois elds Fpn as ordered by divisibility of n. The diagram of sub elds for Fp24 can be seen in Figure 5.1 which shows extensions with no intermediate subextensions. 5.11. Theorem. The algebraic closure of Fp is the union of all the Galois elds of characteristic p, Fp = Fpn . n1 Furthermore, each element u Fp is separable over Fp . Proof. Let u Fp . Then u is algebraic over Fp and the extension Fp (u)/Fp is nite. Hence by Corollary 5.2, Fp (u) Fp is a nite sub eld. Proposition 5.10 now implies that Fp (u) = Fpn for some n. The separability statement follows from Corollary 5.7. We will require a useful fact about Galois elds. 5.12. Proposition. The group of units F in Fpd is cyclic. pd 75 Fp8 Fp4 Fp2 { {{ {{ { {{ { {{ {{ { {{ Fp24 Fp12 {{ {{ {{ {{ Fp6 gg gg gg gg Fp gg gg gg gg { {{ {{ { {{ Fp3 Figure 5.1. The sub elds of Fp24 This is a special case of a more general result about arbitrary elds. 5.13. Proposition. Let K be a eld. Then every nite subgroup U K is cyclic. Proof. Use Corollary 1.35 and Lemma 1.46. 5.14. Definition. w F is called a primitive root if it is a primitive (pd 1)-th root of pd unity, i.e., its order in the group F is (pd 1), hence w = F . pd pd 5.15. Remark. Unfortunately the word primitive has two confusingly similar uses in the context of nite elds. Indeed, some authors use the term primitive element for what we have called a primitive root, but that con icts with our usage, although as we will in the next result, every primitive root is indeed a primitive element in our sense! 5.16. Proposition. The extension of Galois elds Fpnd /Fpd is simple, i.e., Fpnd = Fpd (u) for some u Fpnd . Proof. By Proposition 5.12, Fpnd has a primitive root w say. Then every element of Fpnd can be expressed as a polynomial in w, so Fpnd Fpd (w) Fpnd . This implies that Fpnd = Fpd (w). 5.17. Remark. This completes the proof of the Primitive Element Theorem 3.74 which we had previously only established for in nite elds. 5.18. Example. In Example 5.8 we nd that F4 = F2 (w) has the two primitive roots w and w + 1. 5.19. Example. In Example 5.9 we have F9 = F3 (u) and F is cyclic of order 8. Since 9 (8) = 4, there are four primitive roots and these are the roots of the polynomials X 2 + X 1 and X 2 X 1 which we found to be 1 u. We record a fact that is very important in Number Theory. 5.20. Proposition. Let p > 0 be an odd prime. (i) If p 1 (mod 4), the polynomial X 2 + 1 Fp [X ] has two roots in Fp . 76 (ii) If p 3 (mod 4) the polynomial X 2 +1 Fp [X ] is irreducible, so Fp2 Fp [X ]/(X 2 +1). = Proof. (i) We have 4 | (p 1) = |F |, so if u F is a generator of this cyclic group, the p p order of u|Fp |/4 is 4, hence this is a root of X 2 + 1 (the other root is u|Fp |/4 ). (ii) If v Fp is a root of X 2 + 1 then v has order 4 in F . But then 4 | (p 1) = |F |, which is p p impossible since p 1 2 (mod 4). Here is a generalization of Proposition 5.20. 5.21. Proposition. Fpd contains a primitive n-th root of unity if and only if pd 1 (mod n) and p n. 5.2. Galois groups of nite elds and Frobenius mappings Consider an extension of Galois elds Fpnd /Fpd . By Proposition 5.6(i), Corollary 5.7 and Proposition 3.73, this extension is Galois and | Gal(Fpnd /Fpd )| = [Fpnd : Fpd ] = n. We next introduce an important element of the Galois group Gal(Fpnd /Fpd ). 5.22. Definition. The (relative ) Frobenius mapping for the extension Fpnd /Fpd is the funcd tion Fd : Fpnd Fpnd given by Fd (t) = tp . 5.23. Proposition. The relative Frobenius mapping Fd : Fpnd Fpnd is an automorphism of Fpnd that xes the elements of Fpd , so Fd Gal(Fpnd /Fpd ). The order of Fd is n, so Gal(Fpnd /Fpd ) = Fd , the cyclic group generated by Fd . Proof. For u, v Fpnd , we have the identities d d d d Fd (u + v ) = (u + v )p = up + v p , d d Fd (uv ) = (uv )p = up v p , so Fd is a ring homomorphism. Also, for u Fpd we have d Fd (u) = up = u, so Fd xes the elements of Fpd . To see that Fd is an automorphism, notice that the composition power Fn = Fd Fd (with n factors) satis es d nd Fn (t) = tp d =t n for all t Fpnd , hence Fn = id. Then Fd is invertible with inverse F 1 = Fd 1 . This also shows d d that the order of Fd in the group AutFpd (Fpnd ) is at most n. Suppose the order is k with k n; kd then every element u Fpnd satis es the equation Fk (u) = u which expands to up d u Fpkd . But this can only be true if k = n. Frobenius mappings exist on the algebraic closure Fp . For d Fd : Fp Fp ; 5.24. Proposition. Let d = u, hence 1, consider the function d Fd (t) = tp . 1. (i) Fd : Fp Fp is an automorphism of Fp which xes the elements of Fpd . In fact for u Fp , Fd (u) = u if and only if u Fpd . 77 (ii) The restriction of Fd to the Galois sub eld Fpdn agrees with the relative Frobenius mapping Fd : Fpnd Fpnd . (ii) If k 1, then Fk = Fkd . Hence in the automorphism group AutFpd (Fp ), Fd has in nite d order, so AutFpd (Fp ) is in nite. Proof. This is left as an exercise. The Frobenius mapping F = F1 is often called the absolute Frobenius mapping since it exists as an element of each of the groups AutFp (Fp ) and AutFp (Fpn ) = Gal(Fpn /Fp ) for every n 1. In Gal(Fpnd /Fpd ) = Fd , for each k with k | n there is the cyclic subgroup Fk of order d | Fk | = n/k . d Fk d 5.25. Proposition. For k | n, the xed sub eld of Fk in Fpnd is Fpnd = Fpdk . d Fpnd n/k F Fk d pnd = Fpdk k Fpd dk Proof. For u Fpnd we have Fk (u) = up , hence Fk (u) = u if and only if u Fpdk . d d Figure 5.2 shows the subgroup diagram corresponding to the lattice of sub elds of Fp24 shown in Figure 5.1. Gal(Fp24 /Fp ) = F Z/24 = l lll lll ll ll lll 3 2 l F F l l lll ll lll lll llll 4 F F6 F12 F8 24 F = {id} Figure 5.2. The subgroups of the Galois groups of Fp24 /Fp 78 5.3. The trace and norm mappings For an extension of Galois elds Fpnd /Fpd , consider the function TFpnd /Fpd : Fpnd Fpnd de ned by d 2d + + up TFpnd /Fpd (u) = u + up + up (n 1)d = u + Fd (u) + F2d (u) + + F(n 1)d (u). Notice that d 2d + up 3d + + up nd d 2d + up 3d + + up (n 1)d Fd (TFpnd /Fpd (u)) = up + up = up + up + u = TFpnd /Fpd (u). So by Proposition 5.24(i), TFpnd /Fpd (u) Fpd . If we modify TFpnd /Fpd to have codomain Fpd , we obtain the relative trace TrFpnd /Fpd : Fpnd Fpd ; d (n 1)d 2d TrFpnd /Fpd (u) = u + up + up + + up . 5.26. Proposition. The relative trace TrFpnd /Fpd is a surjective Fpd -linear mapping and whose kernel is an Fpd -vector subspace of dimension n 1. d Proof. Clearly TrFpnd /Fpd is additive. For t Fpd we have tp = t, so Fpd -linearity follows from the formula d (n 1)d 2d tu + (tu)p + (tu)p + + (tu)p d 2d = tu + tup + tup + + tup (n 1)d . To see that TrFpnd /Fpd is surjective, notice that TrFpnd /Fpd (u) = 0 if and only if u is a root of the polynomial d 2d X + Xp + Xp (n 1)d + + Xp Fpd [X ] which has degree p(n 1)d and so has at most p(n 1)d < pnd roots in Fpnd . This means that ker TrFpnd /Fpd cannot be the whole of Fpnd . TrFpnd /Fpd is surjective since its codomain has dimension 1. There is a multiplicative version of this construction. Consider the function N : F F pnd pnd for which d 2d (n 1)d N(u) = uup up up = u Fd (u) F2d (u) F(n 1)d (u). Then we have d 2d 3d nd d 2d 3d (n 1)d Fd (N(u)) = up up up up = up up up up pd p2d = uu u u p3d u p(n 1)d u = N(u). So by Proposition 5.24(i), N(u) Fpd . By rede ning the codomain we obtain the relative norm NormFpnd /Fpd : F F ; pnd pd d 2d (n 1)d NormFpnd /Fpd (u) = uup up up . 5.27. Proposition. The relative norm NormFpnd /Fpd is a surjective group homomorphism. 79 Proof. Multiplicativity is obvious. The kernel of NormFpnd /Fpd consists of the roots in Fpnd of the polynomial d + +p(n 1)d X 1+p 1 Fpd [X ], so | ker NormFpnd /Fpd | 1 + pd + + p(n 1)d = Hence | im NormFpnd /Fpd | = Since im NormFpnd /Fpd pnd 1 | ker NormFpnd /Fpd | pnd 1 . pd 1 pd 1. F , we also have pd | im NormFpnd /Fpd | pd 1, therefore im NormFpnd /Fpd = F . pd Exercises on Chapter 5 5.1. Show that Proposition 5.13 also applies to an integral domain in place of a eld. 5.2. What happens to Theorem 5.20 if we try to take p = 2. 5.3. Let f (X ) Fpd [X ] be an irreducible polynomial with deg f (X ) = n. Find the splitting eld of f (X ). Deduce that for any other irreducible polynomial g (X ) Fpd [X ] with deg g (X ) = n, the splitting elds of f (X ) and g (X ) over Fpd agree. 5.4. Find the smallest Galois elds containing all the roots of the following polynomials, in each case nd a primitive root of this Galois eld: (a) X 8 1 F41 [X ]; (b) X 8 1 F5 [X ]; (c) X 8 1 F11 [X ]; (d) X 8 1 F2 [X ]. 5.5. Let w F be a primitive root. If < d, show that w F . Deduce that degFp w = d / p pd and d | (pd 1). 1, and K/Fpd is an extension. For a K , let 5.6. Let p > 0 be a prime. Suppose that d d ga (X ) = X p X a K [X ]. (a) If the polynomial ga (X ) is irreducible over K , show that the splitting eld E of ga (X ) over K is separable and Gal(E/K ) Fpd . [Hint: show that if u E is a root of ga (X ) = in an extension E/K , then so is u + t for every t Fp .] (b) If d = 1, show that ga (X ) is irreducible over K if and only if it has no root in K . (c) If K is a nite eld and d > 1, explain why ga (X ) can never be irreducible over K . 5.7. Let p be an odd prime, d 1 and write q = pd . (a) Consider { 1} = {1, 1} as a group under multiplication. Show that there is a unique group homomorphism q : F { 1} which is characterized by the requirement q , (u) = 1 if and only if u = v 2 for some v F . Is always that for every u Fq q q q surjective? 80 (b) Consider the set of all squares in Fq , q = {u2 Fq : u Fq } Fq . Show that the number of elements of q is | q | = (q + 1)/2. Deduce that if t Fq then the set t q = {t u2 Fq : u Fq } has |t q | = (q + 1)/2 elements. (c) If t Fq , show that | q (t q )| 1. Deduce that every element of Fq is either a square or can be written as the sum of two squares. (d) Deduce that the equation x2 + y 2 + z 2 = 0 has at least one non-trivial solution in Fq . (e) What can you say about the case p = 2? 81 CHAPTER 6 A Galois Miscellany In this chapter we will explore some miscellaneous topics in Galois Theory. Historically, Galois Theory has always been an important tool in Number Theory and Algebra, stimulating the development of subjects such as Group Theory, Ring Theory and such diverse areas as Differential Equations, Complex Analysis and Algebraic Geometry. Many of the ideas introduced in this chapter are of great importance in these and other mathematical areas. 6.1. A proof of the Fundamental Theorem of Algebra We will prove the Fundamental Theorem of Algebra for the complex numbers C. This proof is essentially due to Gauss but he did not use the historically more recent Sylow theory. It is interesting to compare the proof below with others which use the topology of the plane and circle or Complex Analysis; our proof only uses the connectivity of the real line (via the Intermediate Value Theorem) together with explicit calculations in C involving square roots. 6.1. Theorem (The Fundamental Theorem of Algebra). The eld of complex numbers C is algebraically closed and R = C. Proof. We know that [C : R] = 2, so C/R is algebraic. Let p(X ) C[X ] be irreducible. Then any root u of p(X ) in the algebraic closure C is algebraic over R, so in C[X ] we have p(X ) | minpolyR,u (X ). The splitting eld of p(X ) over C is contained in the splitting eld E of minpolyR,u (X )(X 2 + 1) over R. Since C E , we have 2 | [E : R] and so 2 | | Gal(E/R)|. Now consider a 2-Sylow subgroup P Gal(E/R) and recall that | Gal(E/R)|/|P | is odd. For the xed sub eld of P , we have [ E P : R] = | Gal(E/R)| , |P | which shows that E P /R has odd degree. The Primitive Element Theorem 3.74 allows us to write E P = R(v ) for some v whose minimal polynomial over R must also have odd degree. But by the Intermediate Value Theorem, every real polynomial of odd degree has a real root, so irreducibility implies that v has degree 1 over R and therefore E P = R. This shows that Gal(E/R) = P , hence Gal(E/R) is a 2-group. As C/R is a Galois extension, we can consider the normal subgroup Gal(E/C) Gal(E/R) for which | Gal(E/R)| = 2 | Gal(E/C)|. We must show that | Gal(E/C)| = 1, so suppose not. From the theory of 2-groups, there is a normal subgroup N Gal(E/C) of index 2, so we can consider the Galois extension E N /C of degree 2. But from known properties of C (see Proposition 3.29), every quadratic aX 2 + bX + c C[X ] has complex roots (because we can nd square roots of every complex number). So we cannot have an irreducible quadratic polynomial in C[X ]. Therefore | Gal(E/C)| = 1 and E = C. 83 6.2. Cyclotomic extensions We begin by discussing the situation for cyclotomic extensions over Q using material discussed in Section 1.3. Let n = e2 i/n , the standard primitive n-th root of 1 in C. In Theorem 1.43, it was claimed that the irreducible polynomial over Q which has n as a root was the n-th cyclotomic polynomial t n (X ) = (X n ). t=1,...,n 1 gcd(t,n)=1 6.2. Theorem. Let n 2. Then Q( n ) = Q[X ]/( n (X )); [Q( n ) : Q] = (n); t Gal(Q( n )/Q) (Z/n) , where the element tn (Z/n) acts on Q( n ) by tn n = n . = t Proof. Since the complex roots of n (X ) are the powers n with t = 1, . . . , n 1 and t gcd(t, n) = 1, Q( n ) is the splitting eld of n (X ) over Q and indeed Q( n ) = Q( n ) whenever t t is a primitive n-th root of unity. The main step in the proof has the above properties and so n t is to show that n (X ) Z[X ] is irreducible. To do this we will show that every power n as above is actually a Galois conjugate of n over Q, therefore n (X ) = minpolyQ, n (X ) = minpolyQ, n (X ) t and hence n (X ) is irreducible. Consider r Z( n ) = {a0 + a1 n + + ar n : r 0, aj Z} Q( n ). Then Z( n ) is a subring of Q( n ) and so is an integral domain. Its group of units contains the cyclic subgroup n of order n. Let p > 0 be a prime which does not divide n. Let P Z( n ) be a maximal ideal which contains p; then the quotient ring Z( n )/P is a eld of characteristic p. In fact, it is a nite eld, say Fpd for some d. Let : Z( n ) Fpd be the quotient homomorphism. Inside the group of units of Z( n ) is the subgroup of powers of n , n Z( n ) ; this is a cyclic subgroup of order n. We claim that when restricted to n , gives an injective r group homomorphism, : n F . To see this, suppose that ( n ) = 1 for some r = pd r 1, 2, . . . , n 1; then n 1 P . By elementary Group Theory we can assume that r | n and so p r. On factoring we have r ( n 1)( n 1 + + n + 1) ( n 1)r (mod P ), so n 1 P or r P since maximal ideals are prime. But Z P = (p) and so r P , hence / n 1 P . Recalling that n n 1 + + n + 1 = 0, we see that n P and hence p | n, thus contradicting our original assumption on n. So is injective. Writing u = (u), we can consider the e ect of the absolute Frobenius map F : Fpd Fpd t t on n = n , t t tp F( n ) = ( n )p = n . 84 t tp This shows that in the Galois extension Fpd /Fp , n is conjugate to n ; by iterating this we nd t k tp that n is conjugate to every power of the form n . Now let t = 1, . . . , n 1 and gcd(t, n) = 1. Suppose there is a factorization n (X ) = f (X ) minpolyQ, n (X ) t for some monic polynomial f (X ) Z[X ] and f ( n ) = 0. Consider the prime power factorization r1 rm , where the p are primes with 2 t = p1 pm p1 < < pm and rj 1 with. Since j gcd(t, n) = 1 we also have pj n s. Now consider a maximal ideal P1 Z[ n ] containing p1 . Reducing modulo P1 and working r1 p in the resulting extension Fpd1 /Fp1 , we nd that n is conjugate to n1 . By separability and 1 the fact that the reduction map 1 : Z[ n ] Fpd1 is injective on the powers of n , we nd that 1 r p1 f ( n1 ) = 0 and so conjugate to n . r p1 f ( n1 ) p r1 r1 p = 0 in Z[ n ]. This shows that minpolyQ, n ( n1 ) = 0 and so n1 is r1 p Repeating this argument starting with n1 and using the prime p2 we nd that p r1 r2 p2 minpolyQ, n ( n1 rr p11 p22 and so n )=0 is conjugate to n . Continuing in this fashion, for each j = 1, . . . , m we have r r1 r2 p2 pj j p minpolyQ, n ( n1 p )=0 r r1 p j t and so n1 j is conjugate to n . When j = m, this shows that minpolyQ, n ( n ) = 0. Hence t n is conjugate to n in the extension Q( n )/Q. 6.3. Theorem. For n > 2, consider the cyclotomic extension Q( n )/Q where n = e2 i/n . Then Q( n )R = Q( n ). Furthermore, Q( n )R = Q( n ) ( ) = Q( n + n ) = Q(cos(2 /n)), and [Q(cos(2 /n)) : Q] = Proof. Recall that (n) . 2 Gal(Q( n )/Q) Z/n , = r where the residue class of r acts by sending n to n . Complex conjugation corresponds to the residue class of 1 n 1 (mod n). Making use of the identities e i = cos + sin i, 1 cos = (e i + e i ), 2 we obtain 1 1 cos(2 /n) = ( n + n ) = ( n + n 1 ). 2 2 Complex conjugation xes each of the real numbers cos(2 k/n) for k = 1, 2, . . . , n 1. The residue class of r acts by sending cos(2 /n) to cos(2 r/n); it is elementary to show that cos(2 r/n) = cos(2 /n) unless r 1 (mod n). Hence ( ) = {id, ( )} = Gal(Q(cos(2 /n))/Q). 85 Thus we have Q( n ) ( ) = Q(cos(2 /n)), and so [Q(cos(2 /n)) : Q] = (n)/2. Notice that n is a root of the polynomial X 2 2 cos(2 /n)X + 1 Q(cos(2 /n))[X ], so we also have (6.1) minpolyQ(cos(2 /n)), n (X ) = X 2 2 cos(2 /n)X + 1. 6.4. Example. We have [Q( 24 ) : Q] = (24) = 8 and Gal(Q( 24 )/Q) Z/2 Z/2 Z/2. = Proof. By Theorem 1.43 we have [Q( 24 ) : Q] = 8. Also, 2 2 1 3 8 6 3 + i, 24 = + i, 24 = i, 24 = 2 2 2 2 and all of these numbers are in Q( 24 ), hence Q( 2, 3, i) Q( 24 ). It is easy to check that [Q( 2, 3, i) : Q] = 8, which implies that Q( 24 ) = Q( 2, 3, i). Using this we nd that Gal(Q( 24 )/Q) Z/2 Z/2 Z/2. = We also have cos(2 /24) = cos( /12) Q( 24 ). Since 3 cos(2 /12) = cos( /6) = , 2 we have 3 2 cos ( /12) 1 = 2 2 and so 3 4 cos4 ( /12) 4 cos2 ( /12) + 1 = , 4 giving 16 cos4 ( /12) 16 cos2 ( /12) + 1 = 0. Then 16X 4 16X 2 + 1 = 16 minpolyQ,cos( /12) (X ). Note that case (i) of Kaplansky s Theorem 4.28 applies to the polynomial minpolyQ,cos( /12) (X ). For this example, Gal(Q( 24 )/Q) has 23 1 = 7 subgroups of each of the orders 2 and 4; it is an interesting exercise to nd them all together with their xed sub elds. 86 6.5. Remark. The minimal polynomial for cos( /12) can also be found as follows. We have 24 ( 24 ) = 0, hence since 24 (X ) = X 8 X 4 + 1, we obtain 8 4 24 24 + 1 = 0. Then after multiplying by 244 we have 4 24 1 + 244 = 0, giving 4 ( 24 + 244 ) 1 = 0. Now 4 2 ( 24 + 241 )4 = ( 24 + 244 ) + 4( 24 + 242 ) + 6, hence 4 2 24 + 244 = ( 24 + 241 )4 4( 24 + 242 ) 6. Similarly, 2 ( 24 + 241 )2 = 24 + 242 + 2, so 2 24 + 242 = ( 24 + 241 )2 2. Combining these we have ( 24 + 241 )4 4( 24 + 241 )2 + 1 = 0, and so 16 cos4 ( /12) 16 cos2 ( /12) + 1 = 0. This method will work for any n where (n) is even, i.e., when n > 2. 6.6. Remark. The polynomial that expresses cos n as a polynomial in cos is the nth Chebsyhev polynomial of the rst kind Tn (X ) Z[X ]. Here are the rst few of these polynomials: T2 (X ) = 2X 2 1, T3 (X ) = 4X 3 3X, T4 (X ) = 8X 4 8X 2 + 1, T5 (X ) = 16X 5 20X 3 + 5X, T6 (X ) = 32X 6 48X 4 + 18X 2 1, T7 (X ) = 64X 7 112X 5 + 56X 3 7X. These form a system of orthogonal polynomials which can be computed in Maple using the command orthopoly[T](n,X). Now let K be a eld with characteristic char K n. The polynomial n (X ) has integer coe cients, so we can view it as an element of K [X ] since either Q K or Fp K and we can reduce the coe cients modulo p. In either case it can happen that n (X ) factors in K [X ]. However, we can still describe the splitting eld of X n 1 over K and its Galois group. 6.7. Theorem. If char K n, then the splitting eld of X n 1 over K is K ( ), where K is a primitive n-th root of unity. The Galois group Gal(K ( )/K ) is isomorphic to a subgroup of (Z/n) , hence it is abelian with order dividing (n). 87 Proof. Working in K , we know that n ( ) = 0, hence the roots of minpolyK, (X ) K [X ] are primitive roots of 1. So X n 1 splits over K ( ) and each element Gal(K ( )/K ) has the action ( ) = r , where gcd(r , n) = 1. Hence Gal(K ( )/K ) is isomorphic to a subgroup of Gal(Q( n )/Q) (Z/n) which implies that it is abelian and its order divides (n). = 6.8. Remark. When p = char K > 0, this Galois group only depends on the largest sub eld of K which is algebraic over Fp . For example, if K = Fpd (T ) then the value of d is the crucial factor. The precise outcome can be determined with the aid of Proposition 5.21. 6.9. Example. We have the following splitting elds and Galois groups. (i) The splitting eld of X 4 1 over F3 (T ) is F9 (T ) and Gal(F9 (T )/F3 (T )) (Z/4) Z/2. = = (ii) By Proposition 5.20, X 4 1 splits over F5 (T ) and the Galois group Gal(F5 (T )/F5 (T )) is trivial. Proof. (i) By Proposition 5.20, X 4 1 is separable over F3 (T ) and has irreducible factors (X 1), (X + 1) and (X 2 + 1). The splitting eld of (X 2 + 1) over F3 is F9 = F3 ( ), where 2 + 1 = 0, so (X 2 + 1) splits over F9 (T ). Also, Gal(F9 /F3 ) (Z/4) Z/2, = = with generator satisfying ( ) = 1 = . This generator clearly extends to an automorphism of F9 (T ) which xes T . (ii) By Proposition 5.20, X 4 1 splits over F5 . 6.3. Artin s Theorem on linear independence of characters Let G be a group and K a eld. 6.10. Definition. A group homomorphism : G K is called a character of G with values in K . 6.11. Example. Given any ring homomorphism : R K we obtain a character of R in K by restricting to a map : R K . 6.12. Example. Given an automorphism : K K , : K K is a character of K in K . 6.13. Example. Let E/K be a Galois extension and Gal(E/K ). Then : E E is a character. 6.14. Definition. Let 1 , . . . , n be characters of a group G in a eld K . Then 1 , . . . , n are linearly independent if for t1 , . . . , tn K , t1 1 + + tn n = 0 = t1 = = tn = 0. If 1 , . . . , n are not linearly independent then they are linearly dependent. In this de nition, the functional equation means that for all g G, t1 1 (g ) + + tn n (g ) = 0. 88 6.15. Theorem (Artin s Theorem). Let 1 , . . . , n be distinct characters of a group G in a eld K . Then 1 , . . . , n are linearly independent. Proof. We proceed by induction on n. For n = 1 the result is easily veri ed. For the inductive assumption, suppose that it holds for any n k . Let 1 , . . . , k+1 be a set of k + 1 distinct characters for which there are t1 , . . . , tk+1 K not all zero and such that t1 1 + + tk+1 k+1 = 0. (6.2) If one of the ti is zero, say tr = 0, then 1 , . . . , r 1 , r+1 , . . . , k+1 is linearly dependent, contradicting the inductive assumption. Hence all of the ti must be non-zero. As 1 = 2 , there must be an element g0 G for which 1 (g0 ) = 2 (g0 ). So for all g G, Equation (6.2) applied to g0 g yields t1 1 (g0 g ) + + tk+1 k+1 (g0 g ) = 0, and therefore since j (g0 g ) = j (g0 ) j (g ), we see that t1 1 (g0 ) 1 + + tk+1 k+1 (g0 ) k+1 = 0. Multiplying Equation (6.2) by 1 (g0 ) and subtracting gives t2 ( 2 (g0 ) 1 (g0 )) 2 + t3 ( 3 (g0 ) 1 (g0 )) 3 + + tk+1 ( k+1 (g0 ) 1 (g0 )) k+1 = 0, in which the coe cient t2 ( 2 (g0 ) 1 (g0 )) is not zero. Hence 2 , . . . , k+1 is linearly dependent, again contradicting the inductive assumption. So 1 , . . . , k+1 is linearly independent, which demonstrates the inductive step. 6.16. Corollary. Suppose that 1 , . . . , n are distinct automorphisms of the eld K . Let t1 , . . . , tn K be a sequence of elements, not all of which are 0. Then there is a z K for which t1 1 (z ) + + tn n (z ) = 0. Hence the K -linear transformation t1 1 + + tn n : K K is non-trivial. 6.17. Corollary. Let E/K be a nite Galois extension of degree n and let 1 , . . . , n be the distinct elements of Gal(E/K ). Then the function 1 + + n : E E is a nontrivial K -linear transformation whose image is contained in K . Hence the associated K -linear transformation TrE/K : E K ; TrE/K (x) = 1 (x) + + n (x) is surjective. The function TrE/K : E K is called the trace mapping of E/K . Proof. First note that for x E and Gal(E/K ), ( 1 (x) + + n (x)) = 1 (x) + + n (x) = 1 (x) + + n (x), since the list 1 , . . . , n is the same as 1 , . . . , n apart from its order. Hence, 1 (x) + + n (x) E Gal(E/K ) = K. The rest of the statement follows directly from Corollary 6.16. 89 Suppose that E/K is a nite Galois extension with cyclic Galois group Gal(E/K ) = of order n. For each u E , the element u (u) n 1 (u) E satis es (u (u) n 1 (u)) = (u) n 1 (u) n (u) = (u) n 1 (u)u, hence in u (u) n 1 (u) E = K . Now using this we de ne a group homomorphism NE/K : E K ; NE/K (u) = u (u) n 1 (u). NE/K is called the norm mapping for E/K and generalizes the norm mapping for nite elds of Section 5.3. There is another homomorphism E/K : E E ; E/K (u) = u (u) 1 . Notice that for u E , NE/K ( E/K (u)) = (u (u) 1 )( (u) 2 (u) 1 n 1 (u) n (u) 1 ) = 1, since n (u) = u. So im E/K Proposition 5.27. ker NE/K . Our next result is an important generalization of 6.18. Theorem (Hilbert s Theorem 90). Let E/K be a nite Galois extension with cyclic Galois group Gal(E/K ) = of order n. Then im E/K = ker NE/K . Explicitly, if u E and u (u) n 1 (u) = 1, then there is a v E such that u = v (v ) 1 . Proof. Let u ker NE/K . The characters k : E E with k = 0, 1, . . . , n 1 are distinct and linearly independent by Artin s Theorem 6.15. Consider the function id +u + u (u) 2 + + u (u) n 2 (u) n 1 : E E. This cannot be identically zero, so for some w E , the element v = w + u (w) + u (u) 2 (w) + + u (u) n 2 (u) n 1 (w) is non-zero. Notice that u (v ) = u (w) + u (u) 2 (w) + u (u) 2 (u) 3 (w) + + u (u) 2 (u) n 1 (u) n (w) = v, since u (u) 2 (u) n 1 (u) n (w) = w. Thus we have u = v (v ) 1 as required. 6.4. Simple radical extensions In this section we will investigate splitting elds of polynomials of the form X n a, where char K n. We call these simple radical extensions and later in De nition 6.33 we introduce a more general notion of radical extension. 6.19. Proposition. Let f (X ) = X n a K [X ] be irreducible and separable over K . Then the splitting eld of f (X ) over K has the form K (u, ), where u is a root of f (X ) and is a primitive n-th root of 1. 90 6.20. Corollary. If K contains a primitive n-th root of 1, , then the splitting eld of f (X ) = X n a over K has the form K (u), where u is a root of f (X ). The Galois group Gal(K (u)/K ) is cyclic of order n with a generator for which (u) = u. In the more general situation of Proposition 6.19, {id} Gal(K ( , u)/K ( )) Gal(K ( , u)/K ), where Gal(K ( , u)/K ( )) is cyclic and Gal(K ( )/K )) Gal(K ( , u)/K ))/ Gal(K ( , u)/K ( )) = is abelian. The Galois Correspondence identi es the following towers of sub elds and subgroups. K ( , u) Gal(K ( , u)/K ) a 7 K ( ) K f Gal(K ( , u)/K ( )) 8 } ' {id} 6.21. Definition. Let K be a eld with char K n and which contains a primitive n-th root of 1, say. Then L/K is a simple n-Kummer extension if L = K (u) where un = a for some a K . L/K is an (iterated ) n-Kummer extension if L = K (u1 , . . . , uk ) where un = a1 , . . . , un = ak for some elements a1 , . . . , ak K . 1 k Note that in this de nition we do not require the polynomials X n aj K [X ] to be irreducible. 6.22. Proposition. Let K (u)/K be a simple n-Kummer extension. Then K (u)/K is a Galois extension and Gal(K (u)/K ) is cyclic with order dividing n. Proof. Suppose that un = a K . Then in K [X ] we have X n a = (X u)(X u) (X n 1 u). Clearly the roots of X n a are distinct and so K (u)/K is separable over K ; in fact, K (u) is a splitting eld of X n a over K . This means that K (u)/K is Galois. For each Gal(K (u)/K ) we have (u) = r u for some r = 0, 1 . . . , n 1. Notice that for Gal(K (u)/K ), (u) = ( r u) = r (u) = r r u = r +r u, and so r = r + r . Hence the function : Gal(K (u)/K ) ; ( ) = r , is a group homomorphism. As is cyclic of order n, Lagrange s Theorem implies that the image of has order dividing n. Since every element of Gal(K (u)/K ) is determined by its e ect on u, is injective, hence | Gal(K (u)/K )| divides n. In fact, Gal(K (u)/K ) is cyclic since every subgroup of a cyclic group is cyclic. 6.23. Example. Let n 1 and q Q. Then Q( n , n q )/Q( n ) is a simple n-Kummer extension. 91 6.24. Example. Q(i, 2)/Q(i) is a simple 4-Kummer extension with Gal(Q(i, 2)/Q(i)) cyclic of order 2. Proof. We have ( 2)4 4 = 0, but X 4 4 = (X 2 2)(X 2 + 2), and X 2 2 = minpolyQ(i), 2 (X ). The corresponding group homomorphism : Gal(Q(i)( 2)/Q(i)) i has image im = {1, 1} i . Here is a converse to Proposition 6.22. 6.25. Proposition. Suppose that char K n and there is an element K which is a primitive n-th root of unity. If E/K is a nite Galois extension with cyclic Galois group of order n, then there is an element a E such that E = K (a) and a is a root of a polynomial of the form X n b with b K . Hence E/K is a simple n-Kummer extension. Proof. We have NE/K ( 1 ) = n = 1, so by Hilbert s Theorem 6.18, there is an element a E for which 1 = a (a) 1 . Then (a) = a and the elements k (a) = k a for k = 0, 1, . . . , n 1 are distinct, so they must be the n conjugates of a. Also note that X n an = (X a)(X a) (X n 1 a) = (X a)(X (a)) (X n 1 (a)), hence an K since it is xed by . Since K (a) E , this shows that n = [K (a) : K ] [E : K ] = n and therefore [K (a) : K ] = [E : K ] = n, whence K (a) = E . 6.5. Solvability and radical extensions We begin by recalling some ideas about groups, see [3, 5] for further details. 6.26. Definition. A group G is solvable, soluble or soluable if there is a chain of subgroups (called a subnormal series ) {1} = G G 1 G1 G0 = G in which Gk+1 Gk and each composition factor Gk /Gk+1 is abelian; we usually write {1} = G G 1 G1 G0 = G. If each composition factor is a cyclic group of prime order the subnormal series is called a composition series. A group which is not solvable is called insolvable. 92 6.27. Remark. It is a standard result that we can always re ne (i.e., add extra terms) a subnormal series of a solvable group to obtain a composition series. The primes appearing as well as the number of times each occurs are all determined by |G|, only their order varying for di erent composition series. 6.28. Example. Let G be a nite abelian group. Then G is solvable. 6.29. Example. Let G be a nite p-group, where p is a prime. Then G is solvable. In fact, for a nite p-group G, there is always a normal subgroup of a p-group with index p, so in this case we can assume each quotient Gk /Gk+1 is cyclic of order p. 6.30. Proposition. Let G be a group. (i) If G is solvable then every subgroup H G and every quotient group G/N is solvable. (ii) If N G and G/N are solvable then so is G. In the opposite direction we can sometimes see that a group is insolvable. Recall that a group is simple if it has no non-trivial proper normal subgroups. 6.31. Proposition. Let G be a nite group. Then G is insolvable if any of the following conditions holds: (i) G contains a subgroup which is a non-abelian simple group. (ii) G has a quotient group which is a non-abelian simple group. (iii) G has a composition series in which one of the terms is a non-abelian simple group. 6.32. Example. For n 5, the alternating and symmetric groups An and Sn are insolvable. Proof. This follows from the fact that if n quotient group Sn /An Z/2. = 5, An is a simple group and An Sn with Now we explain how this relates to elds and their extensions. Let K be a eld and L/K a nite extension. For simplicity, we assume also that char K = 0. 6.33. Definition. L/K is a radical extension of K if it has the form L = K (a1 , a2 , . . . , an ) with adk K (a1 , a2 , . . . , ak 1 ) k for some dk of K . 1. Thus every element of L is expressible in terms of iterated roots of elements We will need the following Lemma and its Corollary. According to [4], several text books make subtle errors or omissions related to this result, so beware when reading other sources! 6.34. Lemma. Let L/K be a nite Galois extensions and let L(u)/L be a radical extension. Let E/L be an extension where E is a splitting eld for the polynomial minpolyK,u (X ) L[X ]. Then E/L is a radical Galois extension. In particular, if L/K is a radical Galois extension then so is E/K . Proof. Suppose that ud = a L with a = 0. Then X d a has a d distinct roots in E , and if v is any other root then (v/u)d = 1, so there are d distinct d-th roots of unity in E . Hence there is a primitive d-th root of unity E and the sub eld L( , u) E is normal over 93 L, so L( , u)/L is a radical Galois extension. But L( , u)/K need not be Galois. However, if u = u1 , . . . , ut E are the distinct roots of minpolyK,u (X ) in E , then E = L( , u, u1 , . . . , ut ). But this is clearly a radical extension of L. If L/K is a radical Galois extension, say L = K (a1 , . . . , an ), then E = L(a1 , . . . , an , , u, u1 , . . . , ut ), which is a radical Galois extension of K . 6.35. Corollary. If L/K is a radical extension then it is contained in a radical Galois extension L /K . Proof. Writing L = K (a1 , a2 , . . . , an ) as in De nition 6.33, this is proved by induction on n using Lemma 6.34. In the next de nition, the word Galois is super uous because of the preceding results. 6.36. Definition. If L is the splitting eld of a polynomial f (X ) K [X ], then f (X ) is solvable by radicals over K if L is contained in a radical (Galois) extension of K . 6.37. Definition. L/K is solvable if L of K . L where L /K is a nite radical Galois extension 6.38. Theorem. Let E/K be a nite Galois extension. Then E/K is solvable if and only if the group Gal(E/K ) is solvable. Proof. Suppose that E E where E /K is a nite radical Galois extension, so E = K ( , u1 , . . . , um ), where d = 1, ud1 K ( ) and udr K ( , u1 , . . . , ur 1 ) for r = 2, . . . , m with d1 dm | d. If r 1 Gr Gal(E /K ) and (E )Gr = K ( , u1 , . . . , ur ), with (E )G0 = K ( ), then {1} = Gm Gm 1 G0 Gal(E /K ) and Gr 1 /Gr Gal(K ( , u1 , . . . , ur )/K ( , u1 , . . . , ur 1 )), = which is abelian by Proposition 6.22. Hence Gal(E /K ) is solvable, and since Gal(E/K ) is a quotient group of Gal(E /K ), is also solvable by Proposition 6.30. Now suppose that Gal(E/K ) is solvable and let n = | Gal(E/K )|. Let E be the splitting eld of X n 1 over E , so E contains a primitive n-th root of unity and therefore it contains a primitive d-th root of unity for every divisor d of n. Now Gal(E /E ) Gal(E /K ) and by Theorem 6.7, Gal(E /E ) is abelian. Also, Gal(E /K )/ Gal(E /E ) Gal(E/K ) which is = /K ) is solvable by Proposition 6.30. We will now show that E /K is a solvable, so Gal(E radical extension. 94 Clearly K ( )/K is radical. Then Gal(E /K ( )) Gal(E /K ) is solvable. Let {1} = G G 1 G1 G0 = Gal(E /K ( )) be a composition series. The extension (E )G1 /K ( ) is radical by Proposition 6.25. Similarly, each extension (E )Gk+1 /(E )Gk is radical. Hence E /K ( ) is radical, as is E /K . 6.39. Example. The Galois group of the extension Q( 3 , 3 2)/Q is solvable. Proof. We have already studied this extension in Example 3.30 and 4.20. Clearly Q( 3 , 3 2) is a radical extension of Q and 3 3 Q( 3 , 2) = Q( 3 )( 2). We know that Gal(Q( 3 , 3 2)/Q) S3 , where we identify each element of the Galois group with = a permutation of the three roots of X 3 2 in Q( 3 , 3 2) which we list in the order 2 3 3 3 2, 2 3 , 2 3 . We have the following towers of sub elds and subgroups related under the Galois Correspondence. Q( 3 , 3 2) 5 S3 e 2 3 Q( 3 ) = Q( 3 , 3 2)A3 A3 = Gal(Q( 3 , i 5 2 Q 3 2)/Q( 3 )) 3 y ) {id} Here Q( 3 )/Q is itself a Galois extension and A3 S3 . Notice that A3 Z/3 and S3 /A3 Z/2, = = so we have the following composition series for S3 : {id} A3 S3 . It is also interesting to reverse the question and ask whether there are extensions which are not solvable. This was a famous problem pursued for several hundred years. To nd examples, we rst recall that the smallest non-abelian simple group is A5 which has order 60. We should therefore expect to look for a polynomial of degree at least 5 to nd a Galois group for a splitting eld to be simple or occur as a composition factor of such a Galois group. Here is an explicit example over Q. 6.40. Example. The splitting eld of the polynomial f (X ) = X 5 35X 4 + 7 Q[X ] is not solvable. Proof. Let E C be the splitting eld of f (X ) over Q. Using the Eisenstein Test 1.38 with p = 7, we nd that f (X ) is irreducible over Q. By Theorem 4.8(iii), 5 divides the order of Gal(E/Q), so by Cauchy s Lemma this group contains an element of order 5. Now observe that f (X ) = 5X 4 140X 3 = 5X 3 (X 28), f (X ) = 20X 4 420X 2 = 20X 2 (X 21). There are two turning points, namely a maximum at x = 0 and a minimum at x = 28. Then f (0) = 7 > 0 > f (28) = 4302585, 95 hence there are three real roots of f (X ) and two non-real complex ones. Then complex conjugation restricts to an element of order 2 in Gal(E/Q) which interchanges the non-real roots and xes the others. If we list the roots of f (X ) as u1 , u2 , u3 , u4 , u5 with u1 , u2 being the non-real roots, then the transposition (1 2) S5 corresponds to this element. Furthermore, the only elements of S5 of order 5 are 5-cycles; by taking an appropriate power we can assume that there is a 5-cycle of the form (1 2 3 4 5) corresponding to an element of Gal(E/Q) which we can view as a subgroup of S5 . The next lemma shows that Gal(E/Q) S5 . = 6.41. Lemma. Let n 1. Suppose that H (1 2 n). Then H = Sn . Sn and H contains the elements (1 2) and The proof is left as an exercise. This completes the veri cation of Example 6.40. It is worth remarking that the most extreme version of this occurs when we ask for a Galois group which is simple. There has been a great deal of research activity on this question in the past few decades, but apparently not all simple groups are known to occur as Galois groups of extensions of Q or other nite subextensions of C/Q. Here is an example whose Galois group is A5 ; this is veri ed using Proposition 4.26. 6.42. Example. The Galois group of f (X ) = X 5 + 20X + 16 over Q is Gal(Q(f (X ))/Q) = A5 , hence it is not solvable. 6.6. Symmetric functions Let k be a eld. Consider the polynomial ring on n indeterminates k[X1 , . . . , Xn ] and its eld of fractions K = k(X1 , . . . , Xn ). Each permutation Sn acts on k[X1 , . . . , Xn ] by f (X1 , . . . , Xn ) = f (X1 , . . . , Xn ) = f (X (1) , . . . , X (n) ). Viewed as a function : k[X1 , . . . , Xn ] k[X1 , . . . , Xn ] is a ring isomorphism; this extends to a ring isomorphism : k(X1 , . . . , Xn ) k(X1 , . . . , Xn ). Varying we obtain actions of the group Sn on k[X1 , . . . , Xn ] and k(X1 , . . . , Xn ) by ring isomorphisms xing k and in the latter case it is by eld automorphisms xing k. 6.43. Definition. The eld of symmetric functions on n indeterminates is Symn (k) = k(X1 , . . . , Xn )Sn k(X1 , . . . , Xn ). So if f (X1 , . . . , Xn ) k(X1 , . . . , Xn ), then f (X1 , . . . , Xn ) Symn (k) Sn f (X1 , . . . , Xn ) = f (X (1) , . . . , X (n) ). 6.44. Theorem. The extension k(X1 , . . . , Xn )/ Symn (k) is a nite Galois extension for which Gal(k(X1 , . . . , Xn )/ Symn (k)) Sn . = Proof. There are elements of k[X1 , . . . , Xn ] k(X1 , . . . , Xn ) called elementary symmetric functions, ek = Xi1 Xi2 Xik , i1 <i2 < <ik where 1 k n. It is easy to see that for every Sn , e = ek , so ek Symn (k). Working in k the ring k(X1 , . . . , Xn )[Y ] we have fn (Y ) = Y n e1 Y n 1 + + ( 1)n 1 en 1 Y + ( 1)n en = 0, 96 hence the roots of this polynomial are the Xi . So k(X1 , . . . , Xn ) is the splitting eld of fn (Y ) over Symn (k). Now Sn Gal(k(X1 , . . . , Xn )/ Symn (k)), hence [k(X1 , . . . , Xn ) : Symn (k)] = | Gal(k(X1 , . . . , Xn )/ Symn (k))| |Sn | = n!. But as every element of Gal(k(X1 , . . . , Xn )/ Symn (k)) permutes the roots of fn (Y ) and is determined by this permutation, we also have n! | Gal(k(X1 , . . . , Xn )/ Symn (k))|. Combining these inequalities we obtain | Gal(k(X1 , . . . , Xn )/ Symn (k))| = n! and therefore Gal(k(X1 , . . . , Xn )/ Symn (k)) = Sn . 6.45. Remark. In fact, this proof shows that the extension k(X1 , . . . , Xn )/k(e1 , . . . , en ) is Galois of degree n!. Since k(e1 , . . . , en ) Symn (k) we can also deduce that k(e1 , . . . , en ) = Symn (k). Hence every element of Symn (k) is a rational function in the ei . Analogous results are true for polynomials, i.e., k[X1 , . . . , Xn ]Sn = k[e1 , . . . , en ]. 6.46. Corollary. If n 5, the extension k(X1 , . . . , Xn )/ Symn (k) is not solvable. Exercises on Chapter 6 6.1. Let p > 0 be a prime and G a group of order |G| = pn for some n 1. Show by induction on n that there is a normal subgroup N G with |N | = pn 1 . [Hint: what do you know about the centre of G? Use this information to produce a quotient group of smaller order than G.] 6.2. Let K be a eld for which char K = 2 and n 1 be odd. If K contains a primitive n-th root of unity, show that then K contains a primitive 2n-th root of unity. 6.3. Find all values of n in the following elds: 1 for which (n) | 4. Using this, determine which roots of unity lie Q(i), Q( 2 i), Q( 3 i), Q( 5 i). 6.4. (a) Describe the elements of (Z/24) explicitly and verify that this group is isomorphic to Z/2 Z/2 Z/2. Describe the e ect of each element on Q( 24 ) and Q(cos( /12)) under the action described in Theorem 6.2. (b) Determine the group (Z/20) and describe the e ect of each of its elements on Q( 20 ) and Q(cos( /10)) under the action described in Theorem 6.2. 6.5. Let n 1. (a) What can you say about sin(2 /n) and Gal(Q(sin(2 /n))/Q))? (b) Determine sin( /12) and Gal(Q(sin( /12))/Q)). 6.6. In this question, work in the cyclotomic eld Q( 5 ) where 5 = e2 i/5 . (a) Describe the Galois group Gal(Q( 5 )/Q) and its action on Q( 5 ). 97 (b) Determine the minimal polynomial of cos(2 /5) over Q. Hence show that 1 + 5 cos(2 /5) = . 4 For which other angles is cos a root of this minimal polynomial? What is the value of sin(2 /5) ? (c) Find the tower of sub elds of Q( 5 ) and express them as xed elds of subgroups of Gal(Q( 5 )/Q). 6.7. In this question, let p be an odd prime and let p = e2 i/p Q( p ) C. (a) Consider the product (p 1)/2 = r ( p p r ) Q( p ). r=1 Show that = ( 1) 2 (p 1)/2 p 1 r (1 p ). r=1 (b) Deduce that p 2 = p (c) Conclude that p = p i if p 1 (mod 4), if p 3 (mod 4). if p 1 (mod 4), if p 3 (mod 4). and also p Q( p ) if p 1 (mod 4) and p i Q( p ) if p 3 (mod 4). 6.8. Prove Lemma 6.41. [Hint: show that every 2-cycle of the form (i i + 1) is in H by considering elements of the form (1 2 n)r (1 2)(1 2 n)n r .] 6.9. This question is about an additive version of Hilbert s Theorem 90, see Theorem 6.18. Let E/K be a Galois extension with cyclic Galois group Gal(E/K ) = of order n. (a) Show that the function T : E E ; T (u) = u + (u) + 2 (u) + + n 1 (u), takes values in K and use this to de ne a K -linear mapping TrE/K : E K . (b) If v E has TrE/K (v ) = 0, show that there is a w E such that v = w (w). [Hint: Show that there is an element t E for which TrE/K t = 0, then consider w= 1 (TrE/K t) ( ) v (t) + (v + (v )) 2 (t) + + (v + (v ) 2 (t) + + n 2 (v )) n 1 (t) and adapt the proof of Hilbert s Theorem 90 in Theorem 6.18, using TrE/K in place of NE/K .] 6.10. (a) For n 1 and 1 k n, the k -th power sum sk k[X1 , . . . , Xn ]Sn is de ned by sk = Xik . 1in Prove the formula sk = e1 sk 1 e2 sk 2 + + ( 1)k 1 ek 1 s1 + ( 1)k kek . 98 (b) For n 1 and 1 k n, the total symmetric function is de ned by hk = Xj1 Xj2 Xjk , j1 j2 jk i.e., the sum of all the monomials in the Xi of degree k . (i) For large values of n, express h1 , h2 , h3 in terms of the elementary symmetric functions e1 , e 2 , e 3 . (ii) Show that the power sum functions sk of the previous question satisfy sk = (h1 sk 1 + h2 sk 2 + + hk 1 s1 ) + khk . 99 Bibliography [1] E. Artin, Galois Theory, Dover Publications (1998); ISBN 0 486 62342 4. [2] J-P. Esco er, Galois theory, Springer-Verlag, New York (2001); ISBN 0-387-98765-7. [Highly recommended, especially for its historical notes ] [3] J. B. Fraleigh, A First Course in Abstract Algebra, Addison Wesley (1999); ISBN 0 201 33596 4. [Highly recommended ] [4] T. W. Hungerford, A counterexample in Galois theory, American Mathematical Monthly 97 (1997), 54 57. [5] S. Lang, Algebra, Addison Wesley (1993); ISBN 0 201 55540 9. [6] R. Lidl & H. Niederreiter, Finite Fields, Cambridge University Press (1997); ISBN 0 521 39231 4. [7] J. Rotman, Galois Theory, Springer-Verlag (1998); ISBN 0 387 98541 7. [8] I. Stewart, Galois Theory, Chapman and Hall (1989); ISBN 0 412 34550-1. [Very highly recommended.] 101 Solutions Chapter 1 1.1. Clearly {n Z : n > 0 and nr = 0 for all r R} {n Z : n > 0 and n1 = 0}. If 0 < n Z and n1 = 0, then for every r R, nr = r + + r = (1 + + 1)r = (n1)r = 0r = 0, n n so {n Z : n > 0 and n1 = 0} {n Z : n > 0 and nr = 0 for all r R}. Hence these sets are in fact equal. When char R = p > 0 they must both be non-empty. Now by de nition of characteristic, char R = min{n Z : n > 0 and n1 = 0} = min{n Z : n > 0 and nr = 0 for all r R}. 1.2. (a) Let u, v S and suppose that uv = 0; then u = 0 or v = 0 since u, v R and R is an integral domain. Consider the unit homomorphisms : Z R and : Z S . Then for n Z, (n) = (n), so ker = ker and therefore char S = char R. (b) Q is a eld and Z Q is a subring which is not a eld. 1.3. (a) For any subring R C, R is an integral domain with characteristic subring Z and char R = 0. (b) The characteristic subring of A[X ] is the same as that of A and char A[X ] = char A. A[X ] is an integral domain if and only if A is an integral domain. (c) If we identify A with the subring of scalar matrices in Matn (A), then the characteristic subring of Matn (A) is the same as that of A and char Matn (A) = char A. If n > 1 then Matn (A) is not commutative, in any case it always has zero-divisors since any singular matrix is a zero-divisor. 1.4. The main thing to check is that (u + v ) = (u) + (v ) which is a consequence of the Idiot s Binomial Theorem. For R = Fp [X ], is not surjective, while for R = Fp [X ]/(X 2 ), is not injective. 1.5. (a) Recall from the Isomorphism Theorems of basic Ring Theory that 1 Q R; we need to show it is a prime ideal. Suppose that u, v R with uv 1 Q; then (u) (v ) = (uv ) Q and so (u) Q or (v ) Q, hence u 1 Q or v 1 Q. (b) Consider the inclusion function inc : R S ; then inc 1 Q = Q R, so this result follows from (a). (c) Consider Z Q; then the zero-ideal (0)Q Q has (0)Q Z = (0)Z Z but this is not maximal in Z since for any prime p > 0, (p)Z Z is a (maximal) ideal that properly contains (0)Z . (d) We have P Q R R with P R maximal; so P Q R. In fact Q only needs to be a proper ideal of S for this argument to work. 1.6. The only proper ideal of k is the zero ideal (0), so ker = (0). 1.7. (a) Addition and multiplication follow from the obvious formulae (u1 + v1 i)+(u2 + v2 i) = (u1 + u2 )+(v1 + v2 )i, (u1 + v1 i)(u2 + v2 i) = (u1 u2 v1 v2 )+(u1 v2 + u2 v1 )i, with Z[i] and Q[i] both closed under these operations and containing 1 = 1 + 0i as a unity, so they are subrings of the eld C; by Qu. 1.1, they are both integral domains. To see that Q[i] is a eld, notice that if u + vi = 0 with u, v Q, (u vi)(u + vi) = (u + vi)(u vi) = u2 + v 2 = 0, so u v +2 i Q(i) 2 +v u + v2 is the inverse of u + vi. Hence every non-zero element of Q[i] has an inverse, therefore Q[i] is a eld. 1 (b) & (c) The crucial point is that every element of Q[i] can be written as (u + vi) with n n, u, v Z and n = 0. Then ) ( ) ( (u + vi) 1 (u + vi) = inc = (u + vi), inc n n + 0i n u2 so the latter element is in the image of inc which must therefore be a surjection. 1.8. (a) Existence and uniqueness of such an a,b follow from the Homomorphism Extension Property 1.22 and its e ect on f (X ) = n ri X i R[X ] where ri R is i=0 a,b (f (X )) = f (aX + b) = n ri (aX + b)i . i=0 We have a,b c,d (X ) = a,b (cX + d) = c(aX + b) + d = caX + (cb + d) = ca,cb+d (X ). By the uniqueness part of the Homomorphism Extension Property, we have a,b c,d = ca,cb+d . If a is a unit then a 1 , ba 1 : R[X ] R[X ] has the property that a 1 , ba 1 (aX + b) = X and a,b (a 1 X ba 1 ) = X , so by the uniqueness part of the Homomorphism Extension Property, a,b a 1 , ba 1 = id = a 1 , ba 1 a,b . 1 Therefore these are inverse isomorphisms, a 1 , ba 1 = a,b . n (b) (i) If f (X ) = i=0 ci X i k[X ] with ci k and cn = 0, then deg f (X ) = n. Now a,b (f (X )) = n ci (aX + b)i i=0 = cn an X n + terms of lower degrees in X. Since cn an = 0, this shows that deg a,b (f (X )) = deg f (X ). (ii) Suppose that a,b (p(X )) | g (X )h(X ) for g (X ), h(X ) k[X ]. Choose k (X ) k[X ] so that g (X )h(X ) = k (X ) a,b (p(X )). Since a,b is an isomorphism, we have 1 1 1 a,b (g (X )) a,b (h(X )) = a,b (k (X ))p(X ) 104 1 1 and as p(X ) is prime, p(X ) | a,b (g (X )) or p(X ) | a,b (h(X )). Hence a,b (p(X )) | g (X ) or a,b (p(X )) | h(X ) and so a,b (p(X )) is prime. (iii) This follows from (ii) and Proposition 1.30. 1.9. (a) Addition and multiplication are given by the usual formulae k k ak X ) + ( bk X ) = (ak + bk )X k , ( k=0 k=0 k k k ( ak X )( bk X ) = ( a bk )X k . k=0 k=0 k=0 k=0 =0 Clearly k[X ] k[[X ]] is a subring. Given two non-zero elements a, b k[[X ]] we may write a= k ak X , b= k=k0 b X 0 with ak0 = 0 = b 0 . Then the lowest degree term in ab is ak0 b 0 X k0 + 0 with ak0 b 0 = 0. Hence ab = 0. So k[[X ]] is an integral domain. k (b) Let a = k=0 ak X k[[X ]]. Then a has an inverse in k[[X ]] only if there is a b = k=0 b X k[[X ]] with ab = 1, in particular this forces a0 = 0 since otherwise the lowest term in X in ab would be of degree greater than 0. Conversely, if a0 = 0, then we can inductively solve the system of equations n a0 b0 = 1, a bn = a0 bn + a1 bn 1 + + an bn = 0 (n 1), =0 to ensure that ab = 1. (c) We can de ne make the set k((X )) of all such nite tailed Laurent series into a ring with addition and multiplication de ned by ( k ak X ) + ( k=k1 ( bk X ) = ak X k )( k=k0 (ak + bk )X k , k=min{k1 ,k2 } k = k2 k b X ) = ( k a bk j )X k . k=min{k0 , 0 } j =0 = 0 Clearly k[[X ]] k((X )) is a subring. Notice that every element k0 < 0 can be written as ( ar+k0 X r )X k0 . k=k0 ak X k k((X )) with r=0 The inclusion inc : k[[X ]] k((X )) extends to the monomorphism inc : Fr(k[[X ]]) k((X )) for which ( r) r=0 ar+k0 X inc =( ar+k0 X r )X k0 , X k0 r=0 so inc is surjective. 1.10. Here f (X ) = (3X 3)d(X ) + ( 9X + 7). 1.11. Here f (X ) = X 3 X 2 + X + 1 and d(X ) = X 3 X with f (X ) = d(X ) + ( X X 2 + 1) = d(X ) + (2X 2 + 2X + 1). 1.12. The reduction modulo p function : Z[X ] Fp [X ]; 105 (f (X )) = f (X ), is a ring homomorphism. If f (X ) = g (X )h(X ) with g (X ), h(X ) Z[X ], deg g (X ) < deg f (X ) and deg h(X ) < deg f (X ), then f (X ) = (g (X )h(X )) = (g (X )) (h(X )) = g (X )h(X ), where deg g (X ) < deg f (X ) = deg f (X ) and deg h(X ) < deg f (X ) = deg f (X ). But this is impossible since f (X ) is irreducible. So f (X ) must be irreducible. X 3 X + 1 reduces modulo 3 to an irreducible since it has no roots modulo 3. So X 3 X + 1 is irreducible. X 3 + 2X + 1 X 3 X + 1 (mod 3) so this polynomial reduces modulo 3 to an irreducible and so is irreducible. X 3 + X 1 reduces modulo 2 to an irreducible since it has no roots modulo 2. So X 3 + X 1 is irreducible. X 5 X + 1 is irreducible modulo 3 and 5 so is itself irreducible. X 5 + X 1 = (X 3 + X 2 1)(X 2 X + 1) and 5X 3 10X + X 2 2 = (5X + 1)(X 2 2) so neither of these is irreducible. 1.13. I1 = (X 2 +1), I2 = (X 2 +2), I3 = (X 2 2), I4 = (X 2), I5 = (X 2 +2), I6 = X 2 + X +1. 1.14. The image is The image of 2 2 Q[X ] = Q[ 2] = {a + b 2 : a, b Q}. is 2 Q[X ] = Q[ 2] = 2 Q[X ]. We have ker 2 = ker 2 = (X 2 2) Q[X ] which is a maximal ideal. 1.15. Notice that = ( 1 + 3i)/2 = 3 is a primitive 3-rd root of unity and is a root of the irreducible polynomial X 2 + X + 1 Q[X ]. Then Q[X ] = Q[ ] = {a + b : a, b Q}, ker = (X 2 + X + 1) Q[X ], where (X 2 + X + 1) Q[X ] is a maximal ideal. The other complex root of X 2 + X + 1 is 2 , so the evaluation homomorphism 2 has 2 Q[X ] = Q[X ] and ker 2 = ker . 1.16. We have Q[X ] = Q[ ] = {a + b + c 2 + d 3 : a, b, c, d Q}, ker = (X 4 2) Q[X ], and the latter ideal is maximal. The other complex roots of X 4 2 are , i, i (notice that two of these are real while the other two are not). Then ker = ker i = ker i = (X 4 2) Q[X ] but although Q[X ] = Q[ ], we have i Q[X ] = i Q[X ] = Q[ i] = {a + b i + c 2 + d 3 i : a, b, c, d Q} = Q[ ], so i Q[X ] = Q[X ] since one of these is a subset of R but the other is not. If we replace Q by R, then in R[X ], 4 4 X 4 2 = (X 2 2)(X 2 + 2) = (X 2)(X + 2)(X 2 + 2). Let be a root of X 4 2. If = 4 2, then 4 R[X ] = R[ ] = {a + b : a, b R} = R, ker = (X 2) R[X ]. 106 Similarly, if = 4 2, then R[X ] = R[ ] = {a b : a, b R} = R, ker = (X + 4 2) R[X ]. If 2 + 2 = 0, then R and / R[X ] = R[ ] = {a + b : a, b R} = C, ker = (X 2 + 2) R[X ]. 1.17. First change variable to obtain g (X ) = f (X + 3) = X 3 6X + 4. Using Cardan s method we have to solve the quadratic equation U 2 + 4 U + 8 = 0, which has roots 2 2i = ( 2)3 e3 i/4 . Thus we can take 2 u = 2e = (1 + i) r = (1 + i) r (r = 0, 1, 3). 2 For the roots of g (X ) we obtain 2, 3 1, 3 1, while for f (X ) we have 5, 3 + 2, 3 + 2. i/4 r 1.18. Work backwards with Cardan s method. For , take q = 10, 2 27q 2 + 4p3 = 108, 108 so q = 20 and p = 6. Thus is a real root of f (X ) = X 3 + 6X 20. Notice that 2 is a real root of this polynomial and f (X ) = (X 2)(X 2 + 2X + 10), where X 2 + 2X + 10 has no real roots. Therefore = 2. For , take q 27q 2 + 4p3 28 = 1, =, 2 108 27 3 + X 2 for which 1 is also a root and so q = 2 and p = 1. Thus is a real root of g (X ) = X g (X ) = (X 1)(X 2 + X + 2), where X 2 + X + 2 has no real roots. Therefore = 1. 1.19. To see that the homomorphism A 1 (k) Autk (k[X ]); A A 1 , described in the Proof of Example 1.60 is surjective, suppose that Autk (k[X ]) is any automorphism. Let (X ) = a0 + a1 X + + an X n with ai k and an = 0. If n = 0 then k[X ] = k k[X ] so would not be surjective, hence we must have n 1. Suppose that show that n > 1. Then k[X ] = {c + 0 + c1 (X ) + + ck (X )k : c0 , c1 , . . . , ck k} = k[X ]. 107 But if k > 0 and ck = 0 then deg(c + 0 + c1 (X ) + + ck (X )k ) = kn > 1, so X k[X ], / which gives a contradiction. So we must have n = 1. Therefore (X ) = a0 + a1 X and so = A for some A A 1 (k). 1.20. Calculation. 1.21. We have deg 20 (X ) = (20) = (4) (5) = 2 4 = 8 and X 20 1 = (X 10 1)(X 10 + 1) = (X 10 1)(X 2 + 1)(X 8 X 6 + X 4 X 2 + 1). Since the roots of X 10 1 are the 10-th roots of unity, we nd that 20 (X ) | (X 2 + 1)(X 8 X 6 + X 4 X 2 + 1); since cyclotomic polynomials are irreducible, we must have 20 (X ) = X 8 X 6 + X 4 X 2 + 1. 1.22. (a) We have k 1 k X p 1 = (X p so by (1.5), k 1 )p 1 = (X p k 1 pj (X ) = p (X p k 1 1) p (X p ) ), pj (X ), 0 j k 1 0jk k 1 (X p ). and therefore pk (X ) = p The complex roots of p (X ) are the primitive p-th roots of 1, so the roots of pk (X ) are their pk 1 -st roots which are the primitive pk -th roots of 1. (b) Using the formula of Equation 1.4, we have pk (X ) = p (X p k 1 k 1 ) = (X p k 1 1)p 1 + cp 2 (X p k 1 1)p 2 + + c1 (X p 1) + c0 , where cr 0 (mod p) and c0 = p. The Idiot s Binomial Theorem gives k 1 Xp k 1 1 (X 1)p (mod p) so k 1 pk (X ) = (X 1)(p 1)p + c 2 (X 1)(p 2)p p k 1 k 1 + + c (X 1)p 1 + c , 0 where c 0 (mod p). In fact, r c = pk (1) = p (1) = c0 = p, 0 so the Eisenstein Test can be applied to show that pk (X ) is irreducible over Q. (c) First notice that r r deg n (X ) = (n) = (p1 1) (pk 1)p11 1 pkk 1 , and r 1 r1 1 pkk deg p1 pk (X p1 r r ) = (p1 pk )pr1 1 prk 1 = (p1 1) (pk 1)p11 1 pkk 1 , 1 k r1 1 so deg n (X ) = deg p1 pk (X p1 r 1 pkk ). Also, each root of n (X ), r 1 r1 1 pkk ( p1 r1 1 and no smaller power of ( p1 r 1 pkk )p1 pk = n = 1, r 1 r1 1 pkk ) has this property, hence ( p1 p1 pk (X ). This shows that n (X ) | p1 pk (X the same degree they are equal. r 1 r 1 p11 pkk 108 ) is a root of ). As these are monic polynomials of 1.23. By Theorem 1.43, n (X ) = t (X n ), so t=1,...,n 1 gcd(t,n)=1 n (X 1 ) = t (X 1 n ) t=1,...,n 1 gcd(t,n)=1 = X (n) t (1 X n ) t=1,...,n 1 gcd(t,n)=1 = X (n) n (1 X n t ) t=1,...,n 1 gcd(t,n)=1 = X (n) (1 X n t ) t=1,...,n 1 gcd(t,n)=1 = X (n) t ( n X ) t=1,...,n 1 gcd(t,n)=1 = ( 1) (n) X (n) t (X n ) t=1,...,n 1 gcd(t,n)=1 = ( 1) (n) X (n) n (X ). Since 2 | (n) when n > 2 and the result is immediate when n = 2, we see that desired equation always holds. 1.24. We have n + n 1 = e2 i/n + e 2 i/n = (cos(2 /n) + sin(2 /n) i) + (cos(2 /n) sin(2 /n) i) = 2 cos(2 /n). Now we have 5 + 5 1 = 2 cos(2 /5), 2 5 + 5 2 = ( 5 + 5 1 )2 2 = 4 cos2 (2 /5) 2. We also have 5 (X ) = X 4 + X 3 + X 2 + X + 1, so 4 3 2 5 + 5 + 5 + 5 + 1 = 0. 4 3 Rearranging and using the formulae 5 = 5 1 , 5 = 5 2 , we have 2 ( 5 + 5 2 ) + ( 5 + 5 1 ) + 1 = 0, hence 4 cos2 (2 /5) + 2 cos(2 /5) 1 = 0. Thus a suitable polynomial is 4X 2 + 2X 1 Q[X ]. 1.25. (a) In K [X ], by the Idiot s Binomial Theorem 1.11, X p 1 = X p + ( 1)p = (X + ( 1))p = (X 1)p . By the Unique Factorization Property 1.33, the only root of this polynomial in K must be 1. Similarly, m m X np 1 = (X n 1)p 109 and the only roots of this must be n-th roots of 1. (b) If u K is a root of this polynomial then up = a. As in (a) we have X p a = X p up = (X u)p , so u is the only root in K . Chapter 2 2.1. This is similar to Example 2.4. 2.2. It is obvious that [Q( p, q ) : Q( p)] 2; if [Q( p, q ) : Q( p)] = 1 then q Q( p), say q = a + b p for some a, b Q. Then q = (a + b p)2 = (a2 + b2 p) + 2ab p, giving the simultaneous pair of equations a2 + b2 p = q, 2ab = 0. If b = 0 then q Q which contradicts the result of Qu. 2.1. If a = 0 then q = b p. Writing b = b1 /b2 with b1 , b2 Z and gcd(b1 , b2 ) = 1, we obtain b2 q = b2 p 2 1 and so p | b2 and q | b1 . Writing b1 = b q and b2 = b q for suitable b , b Z, we obtain 1 2 12 (b )2 p2 q = (b )2 q 2 p, 2 1 hence (b )2 p = (b )2 q. 2 1 From this we obtain p | b and q | b ; but then p | b1 as well as p | b2 , contradicting the fact 1 2 that gcd(b1 , b2 ) = 1. So q Q( p). / 2.3. Arrange the induction carefully. 2.4. Notice that if v = u then b = v 2 = u2 = a which is impossible; so v = u. Then u v = (u v )(u + v ) u2 v 2 a b = = K (u + v ). u+v u+v u+v Hence u= So K (u, v ) 1 ((u + v ) + (u v )) K (u + v ), 2 K (u + v ) v= 1 ((u + v ) (u v )) K (u + v ). 2 K (u, v ) and therefore K (u + v ) = K (u, v ). 2.5. Since 1, i span the Q-vector space Q(i), we have [Q(i) : Q] 2. But also if x, y R, then x + yi = 0 x = y = 0, so 1, i is a basis for Q(i) over Q. Hence [Q(i) : Q] = 2. 2.6. First notice that [Q( 3) : Q] = 2 (with Q-basis 1, 3) and Q( 3) R. Also, i Q( 3) / and since i2 + 1 = 0, Q( 3, i) = Q( 3)(i) has [Q( 3, i) : Q( 3)] = 2. By Theorem 2.6(ii), [Q( 3, i) : Q] = [Q( 3, i) : Q( 3)] [Q( 3) : Q] = 2 2 = 4. 110 The following three sub elds of Q( 3, i) are distinct and are extensions of Q having degree 2: L1 = Q( 3), L2 = Q(i), L3 = Q( 3 i). Then [Lr Ls : Q] > 1 Lr Ls = Lr = Ls , so Lr Ls = Q whenever r = s. The only real sub eld amongst these is L1 . C 2 R Q( 3, i) tt tt tt t tt 2 Q( 3) uu uu 2 uu uu u 2 Q(i) uu uu uu u 2 uuu u 2 Q Q ( 3 i) rr rr rr r rr 2 rr 2.7. (a) Since 5 is a prime, [Q( 5 ) : Q] = [Q[X ]/( 5 (X )) : Q] = (5) = 5 1 = 4. (b) We have 5 = cos(2 /5) + sin(2 /5) i Q( 5 ). But also 5 1 Q( 5 ) and 5 1 = cos(2 /5) sin(2 /5) i Q( 5 ). Hence we have ) ) 1( 1( cos(2 /5) = 5 + 5 1 Q( 5 ), sin(2 /5) i = 5 5 1 Q( 5 ). 2 2 (c) This can be found by repeated use of the double angle formula cos(A + B ) = cos A cos B sin A sin B. The polynomial Tn (X ) expressing cos n in terms of cos is called the n-th Chebyshev polynomial, see Remark 6.6. (d) For k = 0, 1, 2, 3, 4, cos(5(2k /5)) = cos(2k ) = 1, so T5 (cos 2k /5) 1 = 0. So each of the numbers cos(2k /5) is a root of the polynomial T5 (X ) 1 = (X 1)(4X 2 + 2X 1)2 . For k = 1, 2, 3, 4, cos(2k /5) is a root of 4X 2 + 2X 1, therefore Q(cos(2 /5)) Q[X ]/(4X 2 + 2X 1), = (e) [Q(cos(2k /5)) : Q] = 2. Q( 5 ) 2 Q(cos(2 /5)) 2 Q 2.8. This is similar to the previous question. 2.9. (a) If AutQ (En ) then (21/n )n = (2) = 2, so (21/n ) En is also a real nth root of 1. If n is odd, the only possibility is (21/n ) = 21/n , so = id. If n is even, the possibilities are (21/n ) = 21/n . We can realize this automorphism starting with the evaluation 111 homomorphism 21/n : Q[X ] En and precomposing with the isomorphism : Q[X ] Q[X ] for which (X ) = X to form 1/n = 21/n . On passing to the quotient homomorphism of 2 1/n we obtain an automorphism n of En under which n (21/n ) = 21/n . 2 (b) Since E R, an automorphism AutQ (E ) has the e ect 21/n if n is odd, 1/n (2 ) = 21/n if n is even. If for some n we have (21/n ) = 21/n then 21/n = (21/n ) = (21/2n )2 > 0 since (21/2n ) R. This contradiction shows that (21/n ) = 21/n for every n, so = id. (c) Assuming there are only 6 such sub elds, they form the following tower. E12 { {{ {{ { {{ 3 E4 2 E2 { {{ {{ { {{ 3 gg gg gg 2 gg 2 E6 Q gg gg 2 gg gg { {{ {{ { {{ 3 E3 (d) This element is a root of the polynomial (X (21/2 + 21/3 ))(X ( 21/2 + 21/3 )) = X 2 2(21/3 )X + 22/3 2 E3 [X ], so it is certainly an element of E6 which is the only degree 2 extension of E3 . If 21/2 + 21/3 E3 then 21/2 E3 , which would imply 2 = [E2 : Q] | [E3 : Q] = 3 which is false, so 21/2 + 21/3 E3 ; / 1/2 + 21/3 E . Writing = e2 i/3 , 21/2 + 21/3 is a root of a similar argument shows that 2 /2 (X (21/2 + 21/3 ))(X (21/2 + 21/3 ))(X (21/2 + 21/3 2 )) = X 3 3(21/2 )X 2 + 6X (2 + 2(21/2 )) E2 [X ], so it cannot lie in E4 since 21/2 + 21/3 E2 and 3 [E4 : E2 ] = 2. So 21/2 + 21/3 is in E6 and / E12 and none of the others. Chapter 3 3.1. Clearly, t is algebraic over K if and only if ker t = (0), i.e., (i) (ii). By Theorem 2.9, (ii) (iii). Hence these three conditions are indeed equivalent. 3.2. The diagrams at the bottom indicate useful sub elds of the splitting elds occurring in each of these examples. 1 3i p1 (X ) = X 4 X 2 + 1: The polynomial X 2 X + 1 has the complex roots e i/3 = , so 2 i/6 . Explicitly the four roots of p1 (X ) are the complex square roots of these numbers, i.e., e these are 31 31 31 31 + i, i, i, + i. 2 2 2 2 2 2 2 2 112 The splitting eld is E = Q( 3, i) and [E : Q] = 4. p2 (X ) = X 6 2: The roots are the six complex 6-th roots of 2, i.e., 6 2e2k i/6 = 6 2ek i/3 for k = 0, 1, 2, 3, 4, 5. Explicitly, these are 6 6 6 6 6 6 6 6 2 23 2 23 2 23 2 23 6 6 2, + i, + i, 2, i, i. 2 2 2 2 2 2 2 2 The splitting eld is E = Q( 6 2, 3i) = Q( 6 2)( 3i) which has degree [E : Q] = 12. p3 (X ) = X 4 + 2: The roots are the four 4-th roots of 2, i.e., 4 2e(2k+1) i/4 for k = 0, 1, 2, 3. Explicitly these are 1 1 1 1 1 1 1 1 + i, + i, i, i. 4 4 4 4 4 4 4 4 2 2 2 2 2 2 2 2 4 The splitting eld is E = Q( 2, i) and [E : Q] = 8. p4 (X ) = X 4 + 5X 3 + 10X 2 + 10X + 5: Notice that p4 (Y 1) = Y 4 + Y 3 + Y 2 + Y + 1 = 5 (Y ), so the splitting eld of p4 (X ) over Q is the same as that of 5 (Y ) over Q and this is the cyclotomic eld Q( 5 ) where 5 = cos(2 /5) + sin(2 /5)i with [Q( 5 ) : Q] = 4; in fact we have Q( 5 ) = Q(cos(2 /5), sin(2 /5)i). Q( 6 2, 3, i) Q ( 3 , i) 2 Q( 6 2, 3) 2 Q( 3) 2 Q 6 Q( 3) Q( 4 2, i ) 2 Q( 4 2) 4 2 Q Q(cos(2 /5), sin(2 /5)i) Q 2 Q(cos(2 /5)) 2 Q 2 3.3. List the three roots of X 3 2 as u1 = 3 2, u2 = 3 2 3 , u3 = 3 2 3 . Then each automorphism AutQ (Q( 3 2, 3 )) permutes these roots, so can be identi ed with the unique permutation S3 for which (ui ) = u (i) (i = 1, 2, 3). We nd that (using cycle notation for permutations) = (1 2), 2 = (1 3 2), 1 These are the six elements of S3 , therefore AutQ (Q( 3 2, 3 )) S3 . = id = id, 0 = (2 3), 1 = (1 2 3), = (1 3). 2 3.4. Irreducibility is a consequence of the polynomial version of the Eisenstein Test 1.48. Suppose that t k(T ) is a root of g (X ); then using the Idiot s Binomial Theorem we have (X t)p = X p tp = X p T, so t is in fact a root of multiplicity p, hence it is the only root of g (X ) in k(T ). This also gives the factorization of g (X ) into linear factors over k(T ). 3.5. Q( 5, 10)/Q: Here [Q( 5, 10) : Q] = 4 and the element 5 + 10 has degree 4 with minimal polynomial X 4 30X 2 + 25 which has roots 5 10. 113 Q( 2, i)/Q: Here [Q( 2, i) : Q] = 4 and the element 2 + i has degree 4 with minimal polynomial X 4 2X 2 + 9 which has roots 2 i. Q( 3, i)/Q: Here [Q( 3, i) : Q] = 4 and the element 3 + i has degree 4 with minimal polynomial X 4 4X 2 + 16 which has roots 3 i. Q( 4 3, i)/Q: Here [Q( 4 3, i) : Q] = 8 and the element 4 3 + i has degree 8 with minimal polynomial X 8 + 4X 6 + 40X 2 + 4 which has roots 4 3 i and 4 3i i. 3.6. The induction is straightforward. Here is the argument that K (u, v )/K is simple. We assume that K is in nite since otherwise the result will be proved in Proposition 5.16. Consider the sub elds K (u + tv ) K (u, v ) with t K . Then there are only nitely many of these, so there must be s, t K such that s = t and K (u + sv ) = K (u + tv ). Then (s t)v = (u + sv ) (u + tv ) K (u + tv ), hence v K (u + tv ). This implies that u = (u + tv ) tv K (u + tv ), hence K (u, v ) K (u + tv ) K (u, v ) and so K (u, v ) = K (u + tv ). 3.7. If E/K is a quadratic extension then for any u E K we have 1 < [K (u) : K ] 2, so [K (u) : K ] = 2 = [E : K ] and therefore K (u) = E . Then minpolyK,u (X ) must factor into linear factors over E , so both its roots in K lie in E . This shows that E is normal over K . The example F2 (Z )/F2 (Z 2 ) is not separable since X 2 Z 2 F2 (Z 2 )[X ] is irreducible but not separable (see Qu. 3.4). If char K = 2 then all quadratic polynomials over K are separable. 3.8. Let E C be a splitting sub eld for f (X ) over Q. Then if v C is a non-real root of f (X ) we have v Q(u), so f (X ) does not split over Q(u) even though it has a root in this eld. / This means that there is a monomorphism MonoQ (Q(u), C) = MonoQ (Q(u), Q) for which (u) = v , hence Q(u) = Q(u) and so Q(u)/Q is not normal. Chapter 4 4.1. By Theorem 3.80 we know that splitting elds are always normal, so it is only necessary to show that the splitting eld E of p(X ) over K is separable over K . Since E is obtained by repeatedly adjoining roots of p(X ), the result follows from Proposition 3.73 together with the fact that if L/K E /K is separable and v E is a root of p(X ), then L(v )/K is separable. 4.2. (a) Suppose that f (X ) = c3 X 3 + c2 X 2 + c1 X + c4 with c3 = 0. Then f (uX + v ) = c3 u3 X 3 + (3c3 vu2 + c2 u2 )X 2 + (3c3 uv 2 + c1 u + 2c2 uv )X + (c3 v 3 + c4 + c1 v + c2 v 2 ), so if we take u to be any cube root of c3 and u = c2 /3c3 then f (uX + v ) has the desired form. Notice that v K (u) and then f (uX + v ) K (u), so provided that we can nd a cube root of 1/c3 in K , we have f (uX + v ) K . (b) Viewing Gal(E/K ) as a subgroup of S3 , by Theorem 4.8 we know that 3 divides | Gal(E/K )|; but the only subgroups of S3 with this property are S3 and A3 . (c) This is a tedious calculation! See Section 4.7 for the rest of this question. 114 4.3. If a/b is a rational root of f (X ), we may assume that gcd(a, b) = 1. Now a3 3ab2 + b3 = 0, which easily implies that a, b = 1; but 1 is certainly not a root. Hence there are no rational roots and so no proper rational factors. By the formula following Proposition 4.25, the discriminant of f (X ) is = 27 4( 3)3 = 81 = 92 . If the distinct roots of f (X ) in C are u, v, w, the splitting sub eld K (v, w) = Q(u, v, w) C satis es 3 | [Q(u, v, w) : Q] and [Q(u, v, w) : Q] | 3! = 6. The Galois group Gal(Q(u, v, w)/Q) is a subgroup of S3 (viewed as the permutation group of {u, v, w}). Since the discriminant is a square in Q, Proposition 4.26 implies that Gal(Q(u, v, w)/Q) A3 Z/3. So | Gal(Q(u, v, w))/Q)| = 3 = and Gal(Q(u, v, w)) is cyclic of order 3 whose generator is a 3-cycle which cyclically permutes u, v, w. 4.4. (a) This should be a familiar result. (b) The centre of D8 is 2 which has order 2, and there are three normal subgroups of order 4, namely 2 2 = { , , 2 , 3 }, , = { , 2 , , 2 }, , = { , 2 , , 3 }. Notice that there are also four non-normal subgroups of order 2, 2 3 = { , }, = { , }, = { , 2 }, = { , 3 }. 4.5. This is an example of case (iii) of Kaplansky s Theorem and we use the notation of the proof. The discriminant here is 2 = 12, so we can take = 2 3i. The roots of X 2 + 3 are 3i, so we may assume 1 243 243 4 4 u = 3 8 = (1 + i), v = 3 8 = (1 i), 2 2 where as usual 8 = e2 i/8 = (1 + i)/ 2. Hence we have uv = 3 and uv = 6i. This gives the diagram of sub elds of E E = Q( 4 3 8 , 4 3 8 1 ) = Q( 4 3, 8 ) Q( 3, i) j jjjj jjjj j jjjj jjj jj Q( 3i) Q( 3) Q(i) jjjj jjjj jj 2 jjjj2 2 jjjj jjjj Q Then is the restriction of complex conjugation to E , while ( 3i) = 3i and ( 3) = 3, hence also (i) = i. Using the choices of the proof, we have 4 4 4 4 4 ( 3 8 ) = 3 8 , ( 3 8 1 ) = ( 3 8 i) = 3 8 i. The e ects of and on the four roots 4 3 8 , 4 3 8 1 , 4 3 8 , 4 3 8 1 of f (X ) are given in permutation notation by = (1 4 3 2) and = (1 2)(3 4), and these generate a dihedral 115 subgroup of S4 . Using the previous question (but beware that the notation there is inconsistent with that of the present situation!) we have the normal subgroups 2 2 2 , , , , , , and these have xed elds 2 E = Q( 3, i), E = Q(i), E 2 , = Q( 3), E 2 , = Q( 3 i), each of which is a normal extension of Q. 4.6. Q(X 3 10)/Q: This is similar to Example 4.20, with splitting eld Q( 3 10, 3 ) and Gal(Q( 3 10, 3 )/Q) S3 . = Q( 2)(X 3 10)/Q( 2): The splitting eld is Q( 2, 3 10, 3 ), [Q( 2, 3 10) : Q( 2)] = 3 and 3 3 Q( 2, 10) Q( 2, 10, 3 ). Since 3 is not real, [Q( 2, 3 10, 3 ) : Q( 2)] = 6. The Galois group is isomorphic to S3 . Q( 3 i)(X 3 10)/Q( 3 i): Here Q( 3 i) = Q( 3 ), with [Q( 3 ) : Q] = 2. The splitting eld is Q( 3 10, 3 ) and [Q( 3 10, 3 ) : Q( 3 )] = 3, hence Gal(Q( 3 10, 3 )/Q( 3 )) Z/3 with generator = for which ( 3 10) = 3 10 3 . Q( 23 i)(X 3 X 1)/Q( 23 i): First note that X 3 X 1 Z[X ] must be irreducible since its reduction modulo 2, X 3 + X + 1 F2 [X ], has no root in F2 and hence has no linear factor (see Qu. 1.10). To proceed further we can use the ideas of Qu. 4.2 above (see also Section 4.7). The discriminant of the polynomial X 3 X 1 is = 23 and so = 23 i. Then if E = Q( 23 i)(X 3 X 1) is the splitting eld of X 3 X 1 over Q, Gal(E/Q) S3 and = A3 . Gal(E/Q( 23 i)) = K (X 3 X 1)/K for K = Q, Q( 5), Q( 5 i), Q(i): Continuing the preceding discussion, / notice that [E R : Q] = 3, so 5 E , hence Q( 5)(X 3 X 1) = Q(X 3 X 1)( 5) and [Q( 5)(X 3 X 1) : Q( 5)] = [Q(X 3 X 1) : Q] = 6, hence Gal(Q( 5)(X 3 X 1)/Q( 5)) S3 . Similarly, 5 i E and i E , hence / / = Gal(Q( 5 i)(X 3 X 1)/Q( 5 i)) S3 Gal(Q(i)(X 3 X 1)/Q(i)). = = 4.7. (a) Since char K = 0, f (X ) = pX p 1 = 0, so if u L is any root of f (X ) then f (u) = pup 1 = 0. By Proposition 3.55, there are no multiple roots, hence p distinct roots. If u, v L are distinct roots, then (vu 1 )p = 1, so v = u for K a p-th root of 1 with = 1. (b) If there is a root u K , the Galois group Gal(L/K ) acts in the following way. By The/ orem 4.8, there must be an element Gal(L/K ) with (u) = u. We can write (u) = u r where = 1 is a p-th root of 1. Since ( ) = , for r 1 we have r (u) = u , which can only equal u if p | r. So u must have at least p conjugates which are all roots of f (X ). Since deg f (X )p , every root of f (X ) is conjugate to u, so f (X ) must be irreducible over K . (c) Suppose that f (X ) = g (X )h(X ) with g (X ) K [X ] monic irreducible and 0 < d = deg g (X ) < p. Let L/K with L a splitting eld for f (X ) over K and let w L be a root of g (X ). Arguing as in (a), we know that each root of g (X ) has the form w where is some 116 p-th root of 1; moreover, L must contain p distinct p-th roots of 1. Now the constant coe cient of g (X ) is g (0) = ( 1)d 0 wd K where 0 is a p-th root of 1. So p g (0)p = ( 1)dp 0 (wp )d = ( 1)dp ad , from which it follows that ad is a p-th power in K . As gcd(p, d) = 1, there are integers r, s such that rp + sd = 1, so we have a = (ar )p (ad )s = a p-th power in K. Hence if f (X ) is not irreducible in K [X ] it has a root in K . 4.8. If u L is a root of f (X ) in an extension L/K then by the Idiot s Binomial Theorem 1.11 X p a = X p + ( u)p = (X u)p , so u is the only such root in L and f (X ) splits over L. If (X u)d K [X ] for some d with 1 < d < p then ud K . Since gcd(d, p) = 1, there are integers r, s such that rd + sp = 1. Hence (ud )s (up )r = u, where the left hand side is in K . This shows that u K . Hence either f (X ) has a root in K or it must be irreducible over K . Chapter 5 5.1. By Theorem 1.17, an integral domain D always admits a monomorphism into a eld j : D F (e.g., F can be taken to be the eld of fractions of D), so any subgroup U D becomes isomorphic to a subgroup jU F , and if U is nite so is jU . Therefore jU and U are cyclic. 5.2. The only root of X 2 + 1 in F2 is the multiple root 1. 5.3. The eld Fpd [X ]/(f (X )) is an extension of Fpd which has degree n, hence it is a nite eld with pdn elements, hence Proposition 5.6 implies that it is isomorphic to Fpdn . Since the extension Fpdn /Fpd is normal, Fpdn is a splitting eld for f (X ) over Fpd . 5.4. (a) Here 41 is prime. Since 8 | (41 1), there is a primitive 8-th root of unity in F41 . 6 is a primitive root for F41 and 65 27 (mod 4)1 has order 8. (b) Here 5 is prime 4 | (5 1), so there is a primitive 4-th root of unity in F , but no primitive 5 8-th root of unity. In fact, 2 and 3 have order 4, so these are primitive roots for F5 . Notice that in F5 [X ], X 8 1 = (X 4 1)(X 4 + 1) = (X 4 1)(X 2 2)(X 2 3), where the polynomials X 2 2 and X 2 3 are irreducible. Therefore F25 is the splitting eld for X 8 1 over F5 and we have F25 F5 (u) = F5 (v ), where u2 = 2 and v 2 = 3, so u and v are = primitive 8-th roots of unity. To nd an element of order 24 in F , we rst nd one of order 3. 25 Consider the polynomial X 2 + X + 1 F5 [X ]; in F5 , this has roots which have order 3. These roots are given by ( 1 w)/2, where w2 = (1 4) = 3 = 2, hence they are ( 1 u) = 3 3u. 2 Now the elements (2 2u)u = ( 4 + 2u) = 4 2u all have order 8 3 = 24. (c) Here 11 is prime and 8 | (121 1) = 120, so F121 is the splitting eld of X 8 1 over F11 . The polynomial X 2 + 1 is irreducible over F11 so F121 = F11 (z ) where z 2 = 1. Since 120 = 8 3 5, it is su cient to nd elements of order 8, 3 and 5 whose product will have order 120. 117 Suppose that a + bz F121 with a, b F11 . If this element has order 8, then (a + bz )2 = z . So let us solve (a2 b2 ) + 2abz = z. Then 2ab = 1 and b2 = a2 , hence b = a. Now we have 2a2 = 1 and so a2 = 1/2 = 6. Now 6 is not a square in F11 but 72 6 42 (mod 11), so we have a = 4, b = 4 and a = 7, b = 7. Therefore the elements of order 8 in F are 121 4 4z and 7 7z . By the same approach as in (b), the elements of order 3 in F121 are ( 1 5z )/2 = 5 8z . 2 is a primitive root for F11 so 4 = 22 has order 5. Combining these we obtain the following primitive roots for F121 : 7 z , 10 4z . (d) In F2 [X ] we have X 8 1 = (X 1)8 , whose only root in F2 is 1. So the splitting eld is F2 . 5.5. Notice that Fp (w) is a splitting eld of the separable polynomial X p 1 1 over Fp , so if w F then Fp (w) Fp . Since Fp (w) = Fpd we have d ; we also have degFp w = d. p The number of conjugates of w is d, hence each primitive root of Fpd has d conjugates and the total number of these is the number of generators of the cyclic group F Z/(pd 1), i.e., pd = d (pd 1). Hence d | (pd 1). This can also be interpreted in terms of the evident action of Gal(Fpd /Fp ) Z/d on the set of all primitive roots of Fpd ; each orbit has exactly d elements, = so the number of orbits is (pd 1)/d which is an integer. 5.6. (a) First note that ga (X ) = 1, so ga (X ) is separable, hence E/K is separable. If u E is a root of ga (X ), then for t Fpd , d d d d ga (u + t) = (u + t)p (u + t) a = (up u a) + (tp t) = (up u a) = 0, hence u + t is also a root of ga (X ). This means that E = K (u) since all the other roots of ga (X ) lie in K (u). As ga (X ) is irreducible over K , [E : K ] = pd = | Gal(E/K )| and so the following pd automorphisms are the elements of Gal(E/K ): t : E E ; t (u) = u + t (t Fpd ). It is easy to check that for s, t Fpd , s t = s+t . Hence there is an isomorphism Gal(E/K ) = Fpd with t corresponding to t Fpd . (b) If ga (X ) is irreducible over K then it cannot have a root in K since its degree is greater than 1. Conversely, suppose that ga (X ) has no root in K . Then if u E is any root of ga (X ) in a splitting eld over K , the other roots are the p elements u + t E (t Fp ). If u + t0 = u is a conjugate of u with 0 = t0 Fp , there must be an element t0 Gal(E/K ) for which t0 (u) = u + t0 . Then t0 must be isomorphic to a non-trivial subgroup of Fp , but this must be Fp since this group is simple. Hence, u must have p conjugates and so ga (X ) is irreducible over K . (c) If K is a nite eld and d > 1 then if ga (X ) were irreducible over K , then by (a), E would be nite and Gal(E/K ) Fpd . But Fpd is not cyclic, yet we know from Proposition 5.23 that = Gal(Fpd /Fp ) Z/d is cyclic. = 118 5.7. (a) By Proposition 5.12, F is a cyclic group. If p = 2 then |F | = 2d 1, which is odd, q 2d so every element of F is a square; we may therefore take 2d (u) = 1 for all u F . So now d 2 2d d 1, which is even. The set of squares in F is the suppose that p is odd. Then |Fpd | = p pd normal subgroup (F )2 = {u2 : u F } pd pd F pd and it is easily seen that its quotient group has order 2, hence F /(F )2 { 1}. = pd pd We may use this group isomorphism to de ne q . Clearly we have ker q = (F )2 . pd q is surjective if and only if p ( odd. is ) u Remark: when d = 1, p (u) = , the Legendre symbol of u from Number Theory. p (b) If u q , then either u = 0 or u = 0 and u = ( v )2 for some v F . Thus we have q | q | = 1 + (q 1) (q + 1) = . 2 2 Then |t q | = | q | = (q + 1) . 2 (c) Since q (t q ) Fq , we have q | q (t q )| = | q | + |t q | | q (t q )|. This implies that q (q + 1) | q (t q )| and so | q (t q )| 1. Thus for every t Fq , there are u, v Fq (possibly 0) for which u2 = t v 2 , whence t = u2 + v 2 . (d) By (c), we may write 1 = a2 + b2 for some a, b Fq , i.e., 12 + a2 + b2 = 0. Chapter 6 6.1. Now when n = 1, G Z/p, which is abelian. Suppose that the result holds whenever = k with k < n. Now if |G| = pn , recall that by Cauchy s Lemma, the centre Z of G |G| = p is non-trivial. Hence G/Z has order |G/Z | = pk with k < n. By the inductive hypothesis, there is a normal subgroup M G/Z with |M | = pk 1 . By one of the Isomorphism Theorems, there is a normal subgroup N G containing Z and satisfying N/Z = M G/Z . Clearly |N | = |Z | |M | = pn 1 . This establishes the inductive step and hence the desired result. 6.2. In this situation, for any non-zero t K , t = t (since otherwise 2t = 0 and so t = 0). If K is a primitive n-th root of unity, then ( )n = ( 1)n n = 1, while ( )2n = ( 1)2n 2n = 1. Hence K is a primitive 2n-th root of unity. 119 6.3. Write n = 2k pr1 prs , where each pj is an odd prime, p1 < p2 < < ps , rj s 1 k 0. Then 1 and r (n) = (2k ) (pr1 ) (prs ) = (2k )(p1 1)p11 1 (ps 1)prs 1 . s s 1 If s > 0 then (n) | 4 happens precisely when r1 = = rs = 1 and one of the following possibilities occurs: p1 = 5, s = 1 and k = 0 (hence n = 5); p1 = 3, s = 1 and k = 0, 1, 2 (hence n = 3, 6, 12); s = 0 and k = 0, 1, 3 (hence n = 1, 2, 4, 8). Q(i): Here degree [Q(i) : Q] = 2 and clearly the four 4-th roots of unity 1, i lie in this eld. As (5) = 4, it has no 5-th roots of unity except 1. If it contained a 3-rd root of unity then it would contain 3 and so Q( 3, i) Q(i) which is impossible since [Q( 3, i) : Q] = 4. From this we see that the only roots of unity in Q(i) are 1, i. Q( 2 i): This eld contains only the square roots of unity 1. 1 3 Q( 3 i): This contains the six 6-th roots of unity 1, i. 2 2 Q( 5 i): This eld contains only the square roots of unity 1. 6.4. (a) We have (24) = (8) (3) = 4 2 = 8. The elements of Z/24 which are invertible are the residue classes modulo 24 of the numbers 1, 5, 7, 11, 13, 17, 19, 23. For each of these numbers r, the residue class modulo 24, r, satis es r2 = 1, hence these all have order 2 except 1 which has order 1. Since (Z/24) is abelian, it is isomorphic to Z/2 Z/2 Z/2. The e ect r of these elements on Q( 24 ) is given by r 24 . Notice that 23 acts like complex conjugation. The e ect on Q(cos( /12)) is given by r cos( /12) = cos( r/12), so in particular, r cos( /12) = cos( r/12) = cos( r/12) = r cos( /12). (b) This is similar to (a). We have (20) = (4) (5) = 2 4 = 8 and the elements of (Z/20) are the residue classes modulo 20 of the numbers 1, 3, 7, 9, 11, 13, 17, 19. This time there are elements of order 4, for instance 7 and 13. Then we have (Z/20) Z/2 Z/4. = 6.5. For any n 1, let n = e2 i/n = cos(2 /n) + sin(2 /n) i. Notice that if n is odd, then Q( n ) = Q( n ) where n is a primitive 2n-th root of unity, so we might as well assume that n is even from now on. We also have n n 1 = 2 sin(2 /n) i Q( n ). (a) If 4 n then writing n = 2k with k odd, we have [Q( n ) : Q] = (2k ) = (2) (k ) = (k ), while [Q( 2n ) : Q] = (4k ) = (4) (k ) = 2 (k ). 120 k Hence, Q( n ) cannot contain 2n and by another simple argument it cannot contain i = 2n . So k we see that sin(2 /n) Q( n ) in this situation. Notice that since i = 2n , / sin(2 /n) = 2 2n 2n2 Q( 2n ), 2i and by Theorem 6.3, sin(2 /n) Q( 2n ) R = Q(cos( /n)). Also, we have [Q(cos( /n)) : Q] = 2[Q(cos(2 /n)) : Q], hence Q(cos( /n)) = Q(cos(2 /n))(sin(2 /n)) and [Q(cos(2 /n))(sin(2 /n)) : Q(cos(2 /n))] = 2, with minpolyQ(cos(2 /n)),sin(2 /n) (X ) = X 2 + cos2 (2 /n) 1. If 4 | n, we can write n = 4 . Then i = n , so i Q( n ), whence sin( /2 ) = sin(2 /n) = n n 1 Q( n ). i Clearly sin( /2 ) Q( n ) R = Q(cos(2 /n)). 2 Consider the automorphism Gal(Q( n )/Q)) for which ( n ) = n +1 = n ; it is easy to see that has order 2. Then (cos(2 /n)) = (cos( /2 )) = cos( /2 ), (cos( / )) = cos( / ), (sin(2 /n)) = (sin( /2 )) = n + n 1 2( n ) sin( /2 ) if is odd, = sin( /2 ) if is even. From this we nd that when is odd, Q(cos(2 /n)) = Q(cos( /2 )) = Q(cos( / ))(sin( /2 )) = Q(sin( /2 )), since cos( / ) = 1 2 sin2 ( /2 ) Q(sin( /2 )). Thus we have [Q(sin( /2 )) : Q] = 2 ( ) and Gal(Q(sin( /2 ))/Q) = Gal(Q(cos( /2 ))/Q) = (Z/4 ) /{1, 1}. Similarly, if is even, [Q(cos( / ))(sin( /2 )) : Q(cos( / ))] = 2 and we must have Q(cos(2 /n)) = Q(cos( /2 )) = Q(sin( /2 )) with Gal(Q(sin( /2 ))/Q) = Gal(Q(cos( /2 ))/Q) = (Z/4 ) /{1, 1} (b) We have 2 3 1 cos( /6) = , sin ( /12) = 2 4 2 121 and so sin( /12) = Then 2 2 3 6 2 = . 4 Q(sin( /12)) = Q( 6 2) = Q( 2, 3). and Gal(Q(sin( /12))/Q)) (Z/4 ) /{1, 1} Z/2 Z/2. = = Here the e ect of the coset of the residue class of r (Z/4 ) is given by r sin( /12) = r r 24 24 = sin(r /12) i1 r . ir Explicitly we have 6 2 1 sin( /12) = 1 sin( /12) = sin( /12) = , 4 6+ 2 5 sin( /12) = 5 sin( /12) = sin(5 /12) = , 4 6 2 , 7 sin( /12) = 7 sin( /12) = sin(7 /12) = 4 6+ 2 11 sin( /12) = 11 sin( /12) = sin(11 /12) = . 4 In terms of the generators 2 and 3 these act by 1 2 = 2, 1 3 = 3, 5 2 = 2, 5 3 = 3, 7 2 = 2, 5 3 = 3, 11 2 = 2, 11 3 = 3. 6.6. (a) We have | Gal(Q( 5 )/Q)| = [Q( 5 ) : Q] = deg 5 (X ) = (5) = 4, and (Z/5) is cyclic generated by the residue class 2. The action is given by 2 2 5 = 5 , 2 4 2 5 = 5 , 3 3 2 5 = 5 , 4 2 5 = 5 . 3 4 (b) We have 5 + 5 1 = 2 cos(2 /5) and 5 ( 5 ) = 0, so since 5 = 5 2 and 5 = 5 1 , 2 ( 5 + 5 2 ) + ( 5 + 5 1 ) + 1 = 0 and therefore ( 5 + 5 1 )2 + ( 5 + 5 1 ) 1 = 0. Hence 4 cos2 (2 /5) + 2 cos(2 /5) 1 = 0. The quadratic polynomial 4X 2 + 2X 1 Z[X ] has discriminant 20 which is not a square in Q, so this polynomial is irreducible over Q, therefore 1 1 minpolyQ,cos(2 /5) (X ) = X 2 + X . 2 4 122 1 5 1 + 5 The roots of this are . As cos(2 /5) > 0 we must have cos(2 /5) = . We 4 4 1 5 also have cos(4 /5) = . As sin(2 /5) > 0, 4 5+ 5 1+5 2 5 2 2 = , sin (2 /5) = 1 cos (2 /5) = 1 16 8 5+ 5 hence sin(2 /5) = . 8 Z/4 and has 3 subgroups {1} {1, 4} Gal(Q( 5 )/Q), giving the following (c) Gal(Q( 5 )/Q) = tower of sub elds. Q( 5 ) 2 Q( 5 ) 4 = Q(cos(2 /5)) = Q( 5) 2 Q 6.7. (a) We have (p 1)/2 2 = r ( p p r )2 = ( 1)(p 1)/2 r=1 (p 1)/2 r r ( p p r )( p r p ) r=1 = ( 1)(p 1)/2 (p 1)/2 2 (1 p 2r )(1 p r ) r=1 = ( 1)(p 1)/2 p 1 2 (1 p r ) r=1 = ( 1)(p 1)/2 (p 1) s (1 p ) s=1 since each congruence 2x t (mod p) has exactly one solution modulo p for each t. (b) Since 1 if p 1 (mod 4), (p 1)/2 ( 1) = 1 if p 3 (mod 4), and p 1 s (1 p ) = p (1) = p, s=1 the result follows. (c) Taking square roots we nd that p = p i As Q( p ), we see that if p 1 (mod 4), if p 3 (mod 4). p Q( p ) if p 1 (mod 4) and p i Q( p ) if p 3 (mod 4). 123 6.8. Recall the well-known formula (i1 ir ) 1 = ( (i1 ) (ir )). Then for 1 r n 2 we have (1 2 n)r (1 2)(1 2 n)n r = (1 2 n)r (1 2)((1 2 n)r ) 1 = (r + 1 r + 2), while (1 2 n)n 1 (1 2)((1 2 n)n 1 ) 1 = (1 2 n) 1 (1 2)((1 2 n) 1 ) 1 = (n 1) = (1 n). This means that every such 2-cycle (r + 1 r + 2) is in H . Also recall that every permutation Sn is a product of 2-cycles, so it su ces to show that every 2-cycle (a b) Sn is a product of 2-cycles of the form (r + 1 r + 2). Assuming that a < b, we also have (a b) = (b 1 b) (a + 2 a + 3)(a + 1 a + 2)(a a + 1)(a + 1 a + 2)(a + 2 a + 3) (b 1 b), and this is in H . Hence H = Sn . 6.9. (a) For each u E , (T (u)) = (u + (u) + 2 (u) + + n 1 (u)) = (u) + 2 (u) + + n (u) = (u) + 2 (u) + + n 1 (u) + u = T (u), so T (u) is xed by and all its powers, hence by Gal(E/K ). Therefore T (u) is in E Gal(E/K ) = K . It is straightforward to verify that the resulting function TrE/K : E K is K -linear. (b) Let v E and suppose that TrE/K (v ) = 0. By Artin s Theorem 6.15, the linear combination of characters id + + + n 1 must be linearly independent, so there is an element t E for which TrE/K t = t + (t) + + n 1 (t) = 0. Then u = v (t) + (v + (v )) 2 (t) + + (v + (v ) 2 (t) + + n 2 (v )) n 1 (t) satis es ( )( ) u (u) = v (t) + 2 (t) + + n 1 (t) (v ) + + n 1 (v ) t ( )( ) = v t + (t) + 2 (t) + + n 1 (t) v + (v ) + + n 1 (v ) t = (TrE/K t)v (TrE/K v )t = (TrE/K t)v. So we obtain v= 1 TrE/K t ( u 1 TrE/K t ) u. 6.10. (a) This can be proved by induction on n. Write e[m] = Xi1 Xir , s[m] = Xir . r r i1 <i2 < <ir m 1im Then we easily nd that [m 1] [ erm] = e[m 1] + er 1 Xm , r 124 [ r srm] = s[m 1] + Xm . r [m] Notice also that er = 0 whenever r > m. The desired result is that for all n 1 and k 1, sk = e1 sk 1 e2 sk 2 + + ( 1)k 1 ek 1 s1 + ( 1)k kek . [n] [n] [n] [n] [n] [1] [n] [n] [n] [1] r When n = 1 we have sr = X1 and e1 = X1 from which the result follows. Now suppose that [n+1] [n] k the result is true for some n 1. Then sk = sk + Xn+1 , while [n+1] [n+1] s k 1 e1 [n+1] [n+1] s k 2 e2 + + ( 1)k 1 ek 1 s1 [n+1] [n+1] [n+1] + ( 1)k kek = k 1 k 2 (e1 + Xn+1 )(sk 1 + Xn+1 ) (e2 + e1 Xn+1 )(sk 2 + Xn+1 ) + [n] [n] [n] [n] [n] + ( 1)k 1 (ek 1 + ek 2 Xn+1 )(s1 + Xn+1 ) + ( 1)k k (ek + ek 1 Xn+1 ) [n] [n] [n] [n] [n] k 1 k 2 = sk + (e1 Xn+1 e2 Xn+1 + + ( 1)k 1 ek 1 Xn+1 ) [n] [n] [n] [n] + (sk 1 e1 sk 2 + + ( 1)k 1 ek 2 s1 + ( 1)k kek 1 )Xn+1 [n] [n] [n] [n] [n] [n] k 1 k 2 + (Xn+1 e1 Xn+1 + + ( 1)k 1 ek 2 Xn+1 ) [n] [n] [n] [n+1] k = sk + Xn+1 = sk which demonstrates the inductive step. (b)(i) We have h1 = e1 , h2 = e2 e2 and h3 = e3 2e1 e2 + e3 . 1 1 (ii) This can be done by induction on n in a similar way to part (a). 125 ,

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

Formatting page ...

 

  Print intermediate debugging step

Show debugging info


 

 


© 2010 - 2026 ResPaper. Terms of ServiceContact Us Advertise with us

 

ranjitsran89 chat