Galois Extension¶

An algebraic field extension that is both normal and separable over its stated base field.

Version: v1 · 2026-10-17 · History
Domain-specific #: 14162
Domain group: Formal Sciences
Origin domain: Mathematics
Subdomains: Field Theory, Galois Theory → Mathematics

Core Idea¶

A Galois extension is an algebraic extension of fields \(E/F\) that is both normal and separable over the stated base \(F\). Normality means that an irreducible polynomial over \(F\) with one root in \(E\) splits completely in \(E\); separability means that the relevant minimal polynomials have distinct roots. Both conditions are necessary. They make \(E/F\) a particular kind of field extension, rather than a general method for solving polynomial equations.^[1]^[2]

The conditions explain why base-fixing automorphisms can adequately reflect this field relation. In the finite case, \(E/F\) is Galois exactly when the group \(\operatorname{Aut}(E/F)\) has \([E:F]\) elements. The base \(F\) is then precisely the set of elements fixed by every such automorphism. The group is written \(\operatorname{Gal}(E/F)\). This numerical equality is a finite-extension test; an infinite algebraic Galois extension instead calls for a topological Galois group and appropriately qualified correspondence theorems.^[1]^[2]

The entry's reusable structure is the conjunction stated base + algebraic extension + normality + separability. Cyclotomic number fields and finite-field towers realize it with different fields, characteristics and automorphisms. The subgroup/intermediate-field correspondence and Artin's fixed-field construction are powerful results about Galois extensions, but neither is an additional defining condition for this type. Those results belong to the neighboring Galois Theory framework.^[3]^[4]^[1]

Structural Signature¶

Base field \(F\). It determines the coefficients, pointwise fixed elements and relative tests. The same upper field can be considered over different bases; a statement without its base is incomplete.^[1]
Algebraic extension \(E/F\). \(F\) is embedded as a subfield of \(E\), and every element of \(E\) satisfies a nonzero polynomial over \(F\). An arbitrary ring inclusion or a transcendental field extension does not meet this entry's definition.^[1]
Normality. Every irreducible \(f\in F[X]\) with a root in \(E\) has all its roots in \(E\). Removing this condition admits separable extensions such as \(\mathbb Q(\sqrt[3]{2})/\mathbb Q\) that are not Galois.^[2]
Separability. Every element's minimal polynomial over \(F\) has distinct roots. Removing it admits normal but inseparable extensions such as \(\mathbb F_p(T)/\mathbb F_p(T^p)\).^[2]

The base-fixing automorphism group is a diagnostic and mathematical consequence of these four roles. It is not a substitute for silently assuming normality or separability, and its finite order formula must not be used as an unqualified description of infinite extensions.^[1]

What It Is Not¶

Not every field extension. A subfield embedding alone supplies the broader Field Extension relation; the Galois type adds normality and separability.^[1]
Not every splitting field. A splitting field is normal, but in positive characteristic it can arise from an inseparable polynomial. The splitting field of a separable polynomial is finite Galois.^[2]
Not Galois Theory as a whole. The theory studies automorphism groups, fixed fields, subgroup correspondence and applications such as solvability.^[2] A Galois extension is the field relation to which those theorems apply.
Not guaranteed by an impressive symmetry label. The field, base and both defining conditions must be established; a group acting on some unrelated object does not make that object a Galois extension.

Scope of Application¶

The literal scope is algebraic field theory. In characteristic zero, cyclotomic extensions \(\mathbb Q(\zeta_n)/\mathbb Q\) supply familiar finite examples. In positive characteristic, \(\mathbb F_{q^n}/\mathbb F_q\) supplies finite examples with cyclic Frobenius Galois groups. Finite normal separable splitting fields of polynomials are another source of instances. An infinite algebraic extension can also be Galois when it is normal and separable, but its group must be treated topologically when using infinite correspondence results.^[3]^[4]^[1]^[2]

The title does not extend by metaphor to any situation with “symmetry,” “closure” or “extension.” A proposed instance must specify two fields and the algebraic, normal and separable relation between them. The base matters: the cyclotomic group is \((\mathbb Z/n\mathbb Z)^\times\) over \(\mathbb Q\); that same formula is not asserted for an arbitrary base field.^[2]

Clarity¶

Write the extension as \(E/F\) and state the embedding of \(F\) into \(E\). Separate the three questions: Is \(E/F\) algebraic? Does every relevant irreducible polynomial split in \(E\) once one root lies there? Are those minimal polynomials separable? A positive answer to just one of the last two is insufficient. If a polynomial is supplied, distinguish the field generated by one root from its complete splitting field, and check separability before calling that splitting field Galois.^[1]^[2]

For a finite extension, one may corroborate the answer by counting the automorphisms fixing \(F\) and comparing with \([E:F]\). For an infinite extension, do not write this finite numerical test or pair arbitrary subgroups with intermediate fields; the relevant correspondence uses closed subgroups of a topological Galois group. Keep a theorem's scope beside the conclusion it supports.^[1]^[2]

Manages Complexity¶

The Galois criterion compresses several failures of symmetry information into two field properties. If an extension is nonnormal, some conjugate roots lie outside it, so its internal automorphisms cannot permute the complete root set. If it is inseparable, distinct-root information collapses. Testing normality and separability first tells the analyst whether finite automorphism counts and correspondence theorems may be used, instead of discovering a missing hypothesis after a group calculation.^[2]^[1]

This classification does not calculate the group for free. Cyclotomic extensions require an argument that the permitted root-of-unity maps actually yield every automorphism; finite-field extensions use Frobenius. The shared criterion makes these unlike computations comparable without pretending that their groups or applications are the same.^[3]^[4]

Abstract Reasoning¶

State \(F\subseteq E\) and whether the extension is algebraic and finite or infinite.
Test normality relative to \(F\): track all roots of each relevant irreducible polynomial once one root enters \(E\).
Test separability relative to \(F\): inspect the distinctness of roots of minimal polynomials.
Conclude that \(E/F\) is Galois only when algebraicity, normality and separability all hold.
If finite, use \(|\operatorname{Aut}(E/F)|=[E:F]\) or the fixed-field test as an equivalent check. If infinite, state the topological qualification before using a correspondence theorem.
Apply group or intermediate-field results as later reasoning steps, keeping the definition of the extension separate from those results.^[1]^[2]

Knowledge Transfer¶

The same entry test works for a root-of-unity number field and a finite field, although their elements and characteristic differ. In both, a specified base sits inside an algebraic upper field, every conjugate required by normality is present, and separability prevents repeated-root loss. What transfers is the field-theoretic test, not a single concrete group: \((\mathbb Z/n\mathbb Z)^\times\) in the cyclotomic example and a cyclic Frobenius group in the finite-field example come from different constructions.^[3]^[4]

The broader insight that transformations preserving a structure reveal invariants belongs to the Symmetry Prime. Here the transformations are specifically field automorphisms fixing \(F\), and the Galois designation still requires algebraic normality and separability. Carrying only the word “symmetry” into another domain does not carry this entry with it.

Examples¶

Canonical: cyclotomic extension over the rationals¶

Let \(\zeta_5\) be a primitive fifth root of unity and set \(E=\mathbb Q(\zeta_5)\) and \(F=\mathbb Q\). \(E\) is the splitting field of \(X^5-1\) over \(\mathbb Q\), so it is normal; characteristic zero makes this polynomial separable. The base-fixing automorphisms send \(\zeta_5\) to \(\zeta_5^a\) for \(a\in(\mathbb Z/5\mathbb Z)^\times\), yielding a group of order four, equal to the extension degree.^[3]^[2]

Mapped back: \(\mathbb Q\) is the stated base; adjoining \(\zeta_5\) makes an algebraic upper field; complete splitting supplies normality; distinct roots supply separability. The automorphism group is a finite diagnostic, not a fifth defining role.

Applied formal construction: a finite-field tower¶

Take \(E=\mathbb F_8\) and \(F=\mathbb F_2\). The upper field is the splitting field of \(X^8-X\) over \(\mathbb F_2\); the derivative is \(-1\), so its roots are distinct. The map \(x\mapsto x^2\) is a base-fixing Frobenius automorphism and generates a cyclic group of order three, matching \([\mathbb F_8:\mathbb F_2]=3\).^[4]^[2]

Mapped back: \(\mathbb F_2\) is the base; \(\mathbb F_8\) is its degree-three algebraic extension; complete splitting gives normality; distinct roots give separability. Frobenius supplies a different automorphism realization of the same Galois-extension type.

Structural Tensions¶

Normality and separability are jointly necessary conditions.^[1] Finite and infinite extensions demand different correspondence formulations.^[2] The inspected sources establish conditions and theorem scopes, but no opposed design pressures with an evidenced cost trade-off. This entry therefore records no invented tension. The useful diagnostic is whether a proposed example or inference silently drops a defining condition or carries a finite theorem into an infinite setting.

Structural–Framed Character¶

The entry is mostly structural within field theory. Vocabulary travel: “normal,” “separable” and “extension” have precise algebraic meanings here, so an everyday use of those words does not transport the identity. Evaluative weight: Galois is a classification, not a claim that the extension is better. Institutional origin: a mathematician's naming convention does not determine which field pairs satisfy the criterion. Human-practice dependence: analysts choose the base and prove the two conditions, but the resulting relation is formal. Import versus recognition: cyclotomic and finite-field towers are literal instances; a social system with “symmetries” is an analogy. Its character: structural but domain-bound. The same formal test recurs across unlike fields, while the field axioms and relative polynomial conditions remain indispensable.

Structural Core vs. Domain Accent¶

Base-fixing field automorphisms preserve field operations and leave each element of \(F\) fixed; normality and separability complete the Galois criterion.^[1] The liftable skeleton is that a structure-preserving transformation group can reveal what remains fixed. Symmetry names that broad pattern in the Prime catalog; Embedding names a faithful structure-preserving inclusion. The Galois-extension core adds two fields in an algebraic extension, every required conjugate root present, and no inseparable minimal polynomial. Remove the field and polynomial conditions and there may still be symmetry or embedding, but there is no Galois extension.

This boundary explains why the entry stays domain-specific even though its automorphism reasoning informs other subjects. Whether the residual fixed-structure idea warrants any further Prime is a separate cross-domain evidence question; the current field examples alone cannot establish one. Here the admission test is irreducibly field-theoretic.

This entry is a kind of Field Extension.

\(\operatorname{Gal}(E/F)\) consists of field automorphisms fixing every element of \(F\).^[1] Symmetry and Embedding name broader patterns visible in those automorphisms and in the subfield inclusion. Their definitions alone do not establish a strict parent relation for the whole Galois-extension identity. Field Extension is its strict domain-specific genus: every Galois \(E/F\) satisfies that two-field relation, while normality and separability restrict it further. A nonnormal field extension shows that the genus is broader than this child.

Relationships to Other Abstractions¶

Current abstraction Galois Extension Domain-specific

Parents (1) — more general patterns this builds on

Galois Extension is a kind of Field Extension Domain-specific

Every Galois extension is a field extension, with normality and separability adding strict conditions relative to its base.

Hierarchy path (1) — routes to 1 parentless root

Galois Extension → Field Extension

Neighborhood in Abstraction Space¶

Galois Extension sits in a sparse region of the domain-specific corpus (78^th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Field Extensions & Galois-Theoretic Structures (19 abstractions)

Nearest neighbors

Galois Theory — 0.84
Algebraic Extension — 0.84
Norm Form — 0.83
Cyclic Algebra — 0.82
Ternary Quartic — 0.82

Computed from structural-signature embeddings · 2026-10-17

Not to Be Confused With¶

Field Extension. Every Galois extension is a field extension; the converse fails when normality or separability fails.^[2]
Splitting Field. Its minimal root-generated identity guarantees normality but not separability for arbitrary polynomials. A separable polynomial's splitting field is Galois.^[2]
Galois Theory. The theory organizes automorphism groups and intermediate fields of suitable extensions.^[2] This entry classifies the extension itself.
A separable nonnormal extension. \(\mathbb Q(\sqrt[3]{2})/\mathbb Q\) lacks the two nonreal conjugate roots and is not Galois.^[2]
A normal inseparable extension. \(\mathbb F_p(T)/\mathbb F_p(T^p)\) does not become Galois merely because it is normal.^[2]

References¶

[1] The Stacks Project, Fields Section 9.21 Galois theory, Definition 9.21.1, Lemma 9.21.2, Definition 9.21.3, Lemma 9.21.6 and Theorem 9.21.7. The printed page heading is “9.21 Galois theory” within Fields. Author-maintained mathematical reference. Definition and finite theorems are cited at their respective scopes; its note after Definition 9.21.3 flags the topology needed for infinite extensions. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o

[2] Milne, J. S. (2022), Fields and Galois Theory, version 5.10 dated September 2022, Example 3.8, Definition 3.9, Theorem 3.10, Proposition 4.23, Lemma 5.9, Theorem 5.10 and Theorem 7.13. Author-maintained full course monograph. Definition 3.9 is finite-scope; Theorem 7.13 concerns infinite extensions and closed subgroups. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t

[3] Brian Conrad, Math 210B Galois group of cyclotomic fields over Q, §1 opening remarks and §2 Theorem 2.1. The printed heading is “Math 210B. Galois group of cyclotomic fields over Q”. Original Stanford mathematics handout; its group identification is over \(\mathbb Q\). registry ↩a ↩b ↩c ↩d ↩e

[4] Andrew V. Sutherland, MIT 18.782 Introduction to Arithmetic Geometry, Fall 2013 Lecture 3, §3.2, printed and PDF p.2. Original lecture notes for the finite-field splitting-field and Frobenius example. registry ↩a ↩b ↩c ↩d ↩e