Chapter 8 · Introduction

Chapter 8 — Fields and Galois Theory

Where the roots of polynomials live: field extensions and their degrees, splitting fields, the finite fields $\mathbb{F}_{p^n}$, and Galois theory, which turns questions about roots into questions about groups.

For two thousand years geometers tried to double the volume of a cube, to trisect an angle and to square the circle using only a straightedge and a compass, and for three hundred years algebraists tried to solve the general quintic equation by radicals, as Cardano's Ars Magna (1545) had solved the cubic and the quartic. Both quests ended in the nineteenth century with proofs of impossibility, and both proofs rest on the same idea: to study the numbers that can be reached from $\mathbb{Q}$ by certain operations, look at the fields they generate and measure their size. Niels Henrik Abel proved in 1824 that the general quintic cannot be solved by radicals; Évariste Galois, before his death in a duel in 1832 at the age of twenty, found the exact criterion — an equation is solvable by radicals iff a certain group of permutations of its roots is solvable — and in 1830 he also discovered the finite fields that now bear his name. His manuscripts were published by Joseph Liouville only in 1846. In 1837 Pierre Wantzel used the degree of an extension to prove that the cube cannot be doubled nor $60^\circ$ trisected; in 1882 Ferdinand von Lindemann proved that $\pi$ is transcendental, which settled the squaring of the circle. Richard Dedekind gave the word "field" (Körper) in 1871, Leopold Kronecker showed how to build a root of any irreducible polynomial as a quotient ring, and Ernst Steinitz's memoir of 1910 created the abstract theory of fields we use today.

Fields are not only a nineteenth-century triumph. Finite fields are the arithmetic of the digital age: the AES cipher that protects internet and banking traffic inverts elements of the field with $256$ elements, the Reed–Solomon codes on QR codes, CDs and deep-space links correct errors by evaluating polynomials over the same field, and elliptic-curve cryptography runs over large prime fields. In this chapter Section 8.1 develops field extensions, degrees and the tower law, minimal polynomials, algebraic closures, ruler-and-compass constructions, splitting fields and separability; Section 8.2 classifies the finite fields completely; and Section 8.3 presents Galois theory — automorphism groups of fields, normal and Galois extensions, the fundamental theorem, and solvability by radicals.

Extensions & degree→Tower law→Minimal polynomials→F(α) ≅ F[x]/⟨m⟩→Constructibility→Splitting fields→Separability→Finite fields 𝔽_q→Frobenius→Galois groups→Fundamental theorem
🎯 By the end of this chapter you can…
  • compute degrees of extensions such as $\mathbb{Q}(\sqrt2,\sqrt[3]2,i)/\mathbb{Q}$ with the tower law, and find minimal polynomials over $\mathbb{Q}$ and over intermediate fields;
  • work in $F(\alpha)\cong F[x]/\langle m_\alpha\rangle$ — write inverses in the basis $1,\alpha,\dots,\alpha^{n-1}$ — and decide when two simple extensions are isomorphic;
  • prove that the cube cannot be doubled, that $60^\circ$ cannot be trisected, and decide whether a given number or regular polygon is constructible;
  • find the degree of a splitting field (cubics via the discriminant, $x^n-a$, cyclotomic polynomials) and test polynomials for separability;
  • build and compute in $\mathbb{F}_4$, $\mathbb{F}_8$, $\mathbb{F}_9$, $\mathbb{F}_{16}$; count subfields, primitive elements, primitive and irreducible polynomials, and solutions of $x^n=a$ in $\mathbb{F}_q$; use the Frobenius map;
  • compute Galois groups of standard extensions, match subgroups with intermediate fields by the fundamental theorem of Galois theory, and state the criterion for solvability by radicals.
Teaser — Which Regular Polygons Can Be Constructed?

The vertices of the regular $n$-gon are the $n$-th roots of unity $\zeta_n^k$, and $[\mathbb{Q}(\zeta_n):\mathbb{Q}]=\varphi(n)$. A constructible number has degree a power of $2$ (§8.1), so the $n$-gon can be constructible only if $\varphi(n)$ is a power of $2$; Gauss and Wantzel showed that this is also sufficient (§8.3). Press Play to run through $n=3,\dots,40$, or drag the slider. Green: constructible; red: impossible with ruler and compass.

17
n—
φ(n) = [ℚ(ζₙ) : ℚ]—
[ℚ(cos 2π/n) : ℚ]—
constructible?—

How this is examined

CSIR NET tests field theory in both parts of the mathematics paper. Part B asks for a single number: the degree of an extension generated by a few radicals and roots of unity; the degree of the splitting field of $x^n-a$, of a cubic or of a cyclotomic polynomial; the number of subfields of $\mathbb{F}_{p^n}$, of generators of $\mathbb{F}_q^\times$, of elements of a given order, of monic irreducible or primitive polynomials of a given degree over $\mathbb{F}_p$; the order of a Galois group or the number of intermediate fields of a Galois extension. Part C, where more than one option may be correct, offers statements to be judged: "$\mathbb{Q}(\sqrt[3]2)/\mathbb{Q}$ is normal", "every extension of degree $2$ is Galois", "this polynomial is separable", "this number is constructible", "$\mathbb{F}_{p^m}$ is a subfield of $\mathbb{F}_{p^n}$", "the Galois group is cyclic / abelian / isomorphic to $S_3$". The false options come from a short list of standard examples — $\mathbb{Q}(\sqrt[3]2)$, $\mathbb{Q}(\sqrt2,\sqrt8)$, $x^4-2$, $x^p-t$ over $\mathbb{F}_p(t)$, $\mathbb{F}_8\not\subseteq\mathbb{F}_{16}$, irreducible but not primitive polynomials — which the ⚠️ boxes of the three sections collect. GATE MA asks closely related questions, often as numerical-answer items.

Sections in this chapter

✅ Before you start

This chapter builds directly on Chapter 7. From §7.3 you need the division algorithm in $F[x]$, the fact that $F[x]/\langle f\rangle$ is a field iff $f$ is irreducible, the irreducibility tests (root test, rational root test, Eisenstein with shifts, reduction mod $p$), Gauss's formula for the number of irreducible polynomials over $\mathbb{F}_p$, and cyclotomic polynomials; from §7.1 and §7.2 the characteristic, quotient rings, the first isomorphism theorem and maximal ideals. Group theory enters through cyclic groups and Lagrange's theorem (§6.1) and, for Galois theory, normal subgroups and quotients (§6.2), permutation groups (§6.3) and Sylow theory (§6.4). Euler's function, the Möbius function and multiplicative orders come from §5.3, roots of unity from §1.1, and the fundamental theorem of algebra from §2.3. Bases and dimension of vector spaces are taken from the linear algebra chapters of the sister course CSIR NET Unit 1 — Analysis and Linear Algebra. Read the sections in order: §8.2 uses the splitting fields of §8.1, and §8.3 uses both.