How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
Finite Fields and Cyclotomic Extensions
1 · Prerequisites
- Algebraic Closure, Embeddings, and Separability
- Algebraic Extensions, Extension Degree, and Finite Fields
- Binary Operations, Monoids, Groups and Subgroups
- Congruences, the Integers Modulo n and the Chinese Remainder Theorem
- Construction of the Natural Numbers
- Construction of the Real Numbers via Cauchy Sequences
- Construction of the Real Numbers via Dedekind Cuts
- Cosets, Index and Lagrange's Theorem
- Countability and Uncountability
- Cyclic Groups and Direct Products
- Determinants of Matrices over a Commutative Ring
- Diagonalisation and the Minimal Polynomial
- Divisibility, Euclidean Domains, Principal Ideal Domains and Unique Factorisation
- Divisibility, Greatest Common Divisors and Bézout's Identity
- Eigenvalues, Eigenvectors and the Characteristic Polynomial
- Finite Counting, Factorials and Binomial Coefficients
- Foundations of the Real Numbers for Analysis
- Group Actions, Orbits, Stabilisers and Cayley's Theorem
- Group Homomorphisms and the Isomorphism Theorems
- Ideals, Quotient Rings and the Isomorphism Theorems for Rings
- Inclusion–Exclusion, the Pigeonhole Principle and Double Counting
- Linear Independence, Bases and Dimension
- Linear Transformations, Rank-Nullity and Quotient Spaces
- Matrices, the Matrix of a Linear Map, and Change of Basis
- Normal Subgroups and Quotient Groups
- Order, Zorn's Lemma, and the Axiom of Choice
- Polynomial Rings, the Division Algorithm and Roots
- Primes, Euclid's Lemma and the Fundamental Theorem of Arithmetic
- Relations, Functions, and Quotients
- Rings, Subrings, Integral Domains and Fields
- Roots, Rational Powers, and Classical Inequalities
- Simple Field Extensions and the Construction of the Complex Numbers
- Splitting Fields
- Symmetric Groups, Cycle Decomposition and the Sign Homomorphism
- The Determinant of a Linear Operator, Cofactors and Cramer's Rule
- The Fundamental Theorem of Finite Abelian Groups
- The Galois Correspondence
- The ZFC Axioms and the Basic Set Constructions
- Triangularisation, Generalised Eigenspaces and Jordan Canonical Form
- Vector Spaces, Linear Subspaces, Span and Direct Sums
2 · Summary
Finite fields on this page are governed by the Frobenius endomorphism, Artin's fixed-field theorem, and the finite Galois correspondence. The page uses the published algebraic-extension and Galois pages for degree, splitting, and automorphism machinery, then adds the relative Frobenius, the finite cyclic-group lemmas it needs, Dedekind independence for normal bases, and the polynomial and unit-group facts that control cyclotomic extensions and finite-field factorisations.
The development begins with finite fields: the -power map identifies the fixed field, determines the full Galois group, describes every intermediate field, and controls Frobenius conjugates and normal bases. It then turns to roots of unity and cyclotomic extensions, defines by the divisor recursion, proves the primitive-root and irreducibility theorems, and uses the resulting Galois groups to study finite-field factorisation, composita and intersections over , primes congruent to modulo , and finite abelian Galois groups over .
3 · Logical flowchart
4 · Definitions, theorems and proofs
A finite cyclic group has exactly one subgroup of each order dividing its own
Statement
Let be a cyclic group of finite order (The subgroup generated by a subset, the cyclic subgroup , and cyclic groups). For every positive divisor of (Divisibility in : when for some integer ):
it is the only subgroup of of order , and every subgroup of is of this form for exactly one such . Moreover, for positive divisors and of ,
The two extremes are instances rather than exceptions: gives the trivial subgroup and gives itself, and at the only divisor is , where is trivial.
Facts & Assumptions
Given: A cyclic group whose underlying set is finite of order ; divisibility of integers is that of Divisibility in : when for some integer , and orders are those of The order of a finite group and the order of an element, with when no positive power of is the identity.
, and since , [L1] gives .
For an element of finite order in a group: if and only if ; the powers are pairwise distinct; and , so is finite with (If then iff is an integer multiple of , the powers are distinct, and has exactly elements; if has infinite order then only for ).
: the cyclic subgroup generated by is exactly the set of integer powers of (, and every cyclic group is abelian).
For a finite group and one has ; consequently divides (Lagrange's theorem: for every subgroup of a finite group ).
Proof
Fix a positive divisor of and put , an integer with and .
The element has order exactly : first by [A1] and [L1]; and if then , so and hence by [L1]. Therefore by [L1].
The set of with is exactly : writing with by [L2] and [A1], the condition says by [L1], that is , that is ; and by [L2].
For positive divisors of with , write with ; then , so by [L2], and therefore , the latter being a subgroup containing .
If has , then : each has dividing by [L1] and [L3], so by [L1] and hence by step 2.2; thus , and both sets have exactly elements by step 2.1, so they are equal.
Every subgroup has this form for exactly one positive divisor of : is a subset of the finite set , so is defined and divides by [L3] and [A1], and step 3.1 gives ; the divisor is determined by , being its order.
Conversely, if then , since by step 2.1 the two subgroups have orders and and [L3] applied to the subgroup of makes divide . Together with steps 2.1, 3.1, 4.1 and 2.3 this proves every clause of the lemma.
Remarks
-
What the divisor lattice buys. The clause that matters downstream is not existence but uniqueness: a subgroup of a finite cyclic group is pinned down by its order alone, so the Galois correspondence turns "subgroups of " into "divisors of " with nothing left to choose (The intermediate fields of are the , one for each positive divisor of ).
-
Where cyclicity is used. Uniqueness fails without it. In the Klein four-group there are three distinct subgroups of order two, and the argument breaks at step 2.2, where the solutions of form the whole group rather than a single cyclic subgroup.
A cyclic group of order has exactly generators
Statement
Let be a cyclic group of finite order (The subgroup generated by a subset, the cyclic subgroup , and cyclic groups). For an integer ,
that is, generates exactly when and are coprime (Coprime integers: ). Consequently has exactly generators (The unit group and Euler's totient for ).
At the group is trivial, every integer is coprime to , and the single element is its own generator, in agreement with .
Facts & Assumptions
Given: A cyclic group whose underlying set is finite of order .
, and since , [L1] gives ; in particular .
For an element of finite order in a group: if and only if ; the powers are pairwise distinct; and , so (If then iff is an integer multiple of , the powers are distinct, and has exactly elements; if has infinite order then only for , The order of a finite group and the order of an element, with when no positive power of is the identity).
(, and every cyclic group is abelian); and is the smallest subgroup containing (The subgroup generated by a subset, the cyclic subgroup , and cyclic groups).
For not both there are integers with (Bézout's identity: for integers not both zero, is the least positive element of ; in particular has an integer solution).
is a common divisor of and , and whenever (Common divisor, and the greatest common divisor , with the convention ).
For , every class in contains exactly one integer with , and is a bijection from onto (For , every class in has one representative with , so ; while is in bijection with ).
For and , the class is a unit of if and only if (For , is a unit if and only if ).
Proof
Put . Since the pair is not , so , and by [L4].
If then generates : by [L3] there are integers with , whence using from [A1]; so by [L2], and is then a subgroup containing , so it contains by [L2] and equals .
Conversely, if generates then : from and the integer is at least and by [A1], so and hence by [L1]; but has elements, so and therefore .
By [A1] and [L1] the powers are pairwise distinct and exhaust , so the generators of are exactly the elements with and , one for each such .
By [L5] the map is a bijection from onto , and by [L6] it carries the with onto the units of ; so the number of such is by [L7], and by step 3.1 that is the number of generators of .
Remarks
- Why the criterion is stated and not only the count. The count answers "how many", but the development below repeatedly needs "which": that is again a primitive -th root of unity exactly when is coprime to is the criterion, and it is what makes the exponent of a cyclotomic automorphism a unit modulo ( is Galois and embeds its Galois group into ).
The relative Frobenius of an extension of finite fields
Definition
Let be a finite field of order (Finite fields and their order) and let be a finite field having as a subfield. The relative Frobenius of the extension is the map
It is an -automorphism of , and no separate result is needed for that. Let be the characteristic of ; the subfield has the same identity element and hence the same characteristic, so with by Every finite field has order for a unique prime characteristic and positive integer . The Frobenius map is an injective field endomorphism of , is an automorphism because is finite, and has -fold iterate (Frobenius is an injective endomorphism in characteristic , and an automorphism for finite fields); that iterate is . Every satisfies (A field with elements is the splitting field of over its prime subfield), so fixes pointwise. Hence
(Relative field automorphisms and ). Its iterates are for , with the identity.
Remarks
-
The letter. The relative Frobenius is written rather than because is Euler's totient (The unit group and Euler's totient for ) everywhere below, and the two symbols would otherwise stand side by side in the same formula.
-
Relative, not absolute. is intrinsic to ; depends on the chosen base field , and it is the identity exactly when . Taking to be the prime field returns itself.
The elements of a finite extension fixed by the -power map are exactly the base field
Statement
Let be a finite field of order and let be a finite field having as a subfield. Then
Equivalently, the fixed field of the cyclic group generated by the relative Frobenius (The relative Frobenius of an extension of finite fields) is the base field:
Facts & Assumptions
Given: Finite fields with (Finite fields and their order), and the set .
The relative Frobenius is , an -automorphism of (The relative Frobenius of an extension of finite fields).
If is a field with elements, then every satisfies (A field with elements is the splitting field of over its prime subfield).
Let be an integral domain. A nonzero polynomial of degree has at most distinct roots in (A nonzero polynomial of degree over an integral domain has at most distinct roots).
The fixed field of a subgroup of the automorphism group of is (The fixed field of a group of field automorphisms).
Proof
Applying [L2] to the field , which has exactly elements, every satisfies ; hence .
is the set of roots in of the polynomial , which is nonzero of degree ; a field is an integral domain, so [L3] gives .
Since , and , the finite sets and coincide: .
An element of is fixed by every power of precisely when it is fixed by itself, so by [L1] and [L4], and step 2.1 identifies this with .
Remarks
- What forces equality is a count, not an inclusion. The inclusion is immediate; the content is that cannot have more than roots, so the elements of already use them all up. The same count identifies the intermediate fields of an extension of finite fields (The intermediate fields of are the , one for each positive divisor of ).
For a degree- extension of a field of order , the -power map has order exactly
Statement
Let be a finite field of order and let be an extension of finite fields of degree (The degree of a finite field extension). Then
and the relative Frobenius (The relative Frobenius of an extension of finite fields) has order exactly in (The order of a finite group and the order of an element, with when no positive power of is the identity). At this says is the identity, of order one.
Facts & Assumptions
Given: Finite fields with and ; the prime subfield of is with the characteristic, and has the same characteristic, its identity element being that of .
The relative Frobenius is , an -automorphism of , with (The relative Frobenius of an extension of finite fields).
If is a field with elements, then every satisfies (A field with elements is the splitting field of over its prime subfield).
Let be an integral domain. A nonzero polynomial of degree has at most distinct roots in (A nonzero polynomial of degree over an integral domain has at most distinct roots).
If is a finite field, then there is a unique prime and a unique positive integer with ; here and (Every finite field has order for a unique prime characteristic and positive integer ).
For fields with and finite, is finite and (Tower law for finite extensions: ).
Proof
By [L4] applied to , with ; by [L4] applied to , with .
The tower and [L5] give , so .
Every satisfies , by [L2] applied to , whose order is by step 2.1; by [L1] this says .
For an integer with one has : otherwise every one of the elements of would be a root of the nonzero polynomial , whose degree is smaller than because , contradicting [L3].
The least with is therefore , that is ; for steps 3.1 and 3.2 say only that , of order one.
A finite extension of a finite field of order is Galois with cyclic Galois group generated by
Statement
Let be a finite field of order and let be a finite field having as a subfield, with (The degree of a finite field extension). Then is a finite Galois extension (Finite Galois extensions and ) and
is cyclic of order , generated by the relative Frobenius (The relative Frobenius of an extension of finite fields).
Facts & Assumptions
Given: Finite fields with and , and the subgroup of .
The relative Frobenius is an -automorphism of (The relative Frobenius of an extension of finite fields).
and has order exactly in (For a degree- extension of a field of order , the -power map has order exactly ).
If is a finite group of automorphisms of , then and (Artin's fixed-field theorem: and ).
For a finite extension with , the conditions " is Galois", " is the splitting field over of a separable polynomial", "" and "" are equivalent (Equivalent characterizations of a finite Galois extension).
Proof
is a cyclic group of automorphisms of , finite of order by [L1] and [L3].
Its fixed field is by [L2] and [L6].
Applying [L4] to the finite automorphism group of and using step 1.2, and .
Hence , so is Galois by [L5], and is cyclic of order by step 2.1 and step 1.1. At the group is trivial and .
Remarks
- Separability and normality are never argued separately. The usual route checks that is a splitting field of and that this polynomial has no repeated root. Routing through Artin's fixed-field theorem: and instead replaces both checks by one count: an automorphism group of order whose fixed field is already forces , which is one of the equivalent Galois conditions.
The intermediate fields of are the , one for each positive divisor of
Statement
Let be a finite field of order , let , and let be a finite field having as a subfield with (The degree of a finite field extension). For a positive divisor of (Divisibility in : when for some integer ) put
Then the fields , as runs over the positive divisors of , are exactly the intermediate fields of , with distinct divisors giving distinct fields;
and for positive divisors of ,
The two ends of the lattice are instances: gives the base field and gives .
Facts & Assumptions
Given: Finite fields with and , and the relative Frobenius (The relative Frobenius of an extension of finite fields), whose -th iterate is .
is Galois and is cyclic of order (A finite extension of a finite field of order is Galois with cyclic Galois group generated by ).
In a cyclic group of finite order , for each positive divisor of the subgroup has order and is the unique subgroup of that order, every subgroup has this form for exactly one such , and if and only if (A finite cyclic group has exactly one subgroup of each order dividing its own).
For finite Galois with , the assignments and are mutually inverse inclusion-reversing bijections between subgroups and intermediate fields , and , (The fundamental theorem of finite Galois theory).
For an extension of finite fields of degree one has (For a degree- extension of a field of order , the -power map has order exactly ).
For a finite group and one has (Lagrange's theorem: for every subgroup of a finite group ).
Proof
Write , cyclic of order by [L1], and for a positive divisor of put .
By [L2] applied to with generator and , the subgroup has order ; every subgroup of is for exactly one positive divisor of ; and if and only if , since reads , which by [L2] says , that is .
The fixed field of is , because an element fixed by is fixed by all its powers and conversely.
By [L3] the map is a bijection from the subgroups of onto the intermediate fields of ; composing with the bijection of step 2.1 between positive divisors of and subgroups, the fields are exactly the intermediate fields, distinct divisors giving distinct fields.
Degrees: [L3] gives , and [L7] with step 2.1 turns this into ; then [L5] gives .
Inclusions: [L3] makes the correspondence inclusion-reversing, so exactly when , which by step 2.1 holds exactly when . At one has and by [L4], and at one has and ; with steps 3.1 and 3.2 this proves every clause.
Remarks
- Why the lattice is exactly the divisor lattice. Uniqueness of the subgroup of each order in a cyclic group is what leaves no choice: had been the Klein four-group, three distinct subgroups of order two would have produced three intermediate fields of the same degree, and no indexing by divisors could exist.
The Galois description of the subfields of a finite field and the elementary divisibility criterion agree
Statement
Two statements in this library describe the subfields of a finite field, and they describe the same objects.
The subfields of are the unique fields for positive divisors of fixes a finite field of order with its characteristic and says that for each positive divisor of the set is the unique subfield of of order , and that these are all of the subfields of . Its index set is therefore the divisors of , and its base point is the prime field.
The intermediate fields of are the , one for each positive divisor of fixes a base field inside and says that the intermediate fields of are the for the positive divisors of . Its index set is therefore the divisors of , and its base point is .
The dictionary. Write , so that . For a positive divisor of the two prescriptions produce literally the same set,
which is the subfield of order named by the first statement; and runs exactly over the divisors of that are multiples of as runs over the divisors of . So the intermediate fields of are precisely those subfields of whose order is with , which is the expected answer: a subfield of contains the unique subfield of order exactly when divides , by the divisibility clause of The intermediate fields of are the , one for each positive divisor of applied over the prime field.
Neither statement is the other. The published one is elementary: it counts roots of and needs no Galois theory. The one proved here reads the lattice off the subgroup lattice of a cyclic Galois group, and it is that reading which the rest of this page uses, because the same correspondence also supplies the degrees and the automorphism groups of the intermediate fields. Recording their agreement here is what keeps the two vocabularies from drifting apart in later proofs.
A monic irreducible of degree over has the distinct roots
Statement
Let be a finite field of order , let be monic irreducible of degree , and let be a root of in some extension field of . Then is a finite field of order , the elements
are pairwise distinct roots of lying in ,
and is a splitting field of over (Polynomials that split and splitting fields of a polynomial or a family of polynomials), of degree over (The degree of a finite field extension). In particular these elements are pairwise conjugate over (Conjugate algebraic elements over a field) and form a single orbit of the relative Frobenius. The list starts at , so its first member is itself, and at it is the single element .
Facts & Assumptions
Given: A finite field of order , a monic irreducible of degree (Degree, leading coefficient and monic polynomial, with the zero polynomial having no degree), a root of in an extension field, and the field .
If is a field extension and is algebraic, there is a unique monic irreducible with , and for every one has if and only if (The evaluation kernel and the unique monic irreducible minimal polynomial of an algebraic element).
If is algebraic over with minimal polynomial of degree , then (An element is algebraic over if and only if its simple extension is finite).
If , then has an ordered -basis of length , and every element of has a unique coordinate list in that basis (The degree of a finite field extension, Basis of a vector space: a linearly independent spanning subset; and ordered basis: an injective finite list whose image is a basis, A finite list is an ordered basis if and only if every equals for exactly one ; those scalars are the coordinates of in that ordered basis).
An extension of finite fields of degree is Galois with cyclic of order , where (A finite extension of a finite field of order is Galois with cyclic Galois group generated by , The relative Frobenius of an extension of finite fields).
Let be a commutative ring, and . Then if and only if divides in (Factor theorem over a commutative ring).
A nonzero polynomial of degree over an integral domain has at most distinct roots in that domain (A nonzero polynomial of degree over an integral domain has at most distinct roots).
If is a field with elements, then every satisfies (A field with elements is the splitting field of over its prime subfield).
Two elements algebraic over are conjugate over when they have the same minimal polynomial over (Conjugate algebraic elements over a field).
Proof
is the minimal polynomial of over : since , [L1] gives , and is irreducible while is monic of degree at least one, so . Hence by [L2].
Fix the length- basis supplied by [L3]. Unique coordinates give a bijection , so and is a finite field. Now [L4] applies: is Galois with cyclic of order , where .
Each fixes the coefficients of , which lie in , so applying the field homomorphism to the equation gives : every is a root of lying in .
The elements are pairwise distinct. Suppose with and apply the automorphism : since for every by [L7] and step 2.1, this yields with and . The set is the fixed set of the automorphism , hence a subfield of ; it contains by [L7] and contains , so . But is the root set in of the nonzero polynomial , so by [L6], giving because and . This is impossible.
The product divides in . Indeed, listing the distinct roots as , [L5] writes ; for the equation and in the field give , so the same step applies to with the remaining distinct roots, and after such steps for some .
Both and are monic of degree , so is monic of degree , that is and .
Consequently splits over , and is generated over by the root ; since the subfield of generated over by all the roots of contains , it contains and hence equals , so is a splitting field of over . All roots share the minimal polynomial by step 1.1 and [L1], so they are pairwise conjugate over by [L8], and step 3.1 exhibits them as one orbit of .
Remarks
-
The index starts at zero. The orbit is , so the factor is present in the product; dropping the term would leave a polynomial of degree that is not .
-
Where irreducibility is used. It enters twice: to identify with the minimal polynomial of in step 1.1, and through that identification to force , which is what makes the count in step 4.1 tight. For a reducible the conclusion fails outright, as over shows: its roots and are not a Frobenius orbit.
for the counts of monic irreducibles of degree over
Statement
Let be a finite field of order and, for an integer , let denote the number of monic irreducible polynomials of degree in . Then each is finite, and for every
the sum being over the positive divisors of (Divisibility in : when for some integer , The sum over a finite index set, and its product form). At the identity reads .
Facts & Assumptions
Given: A finite field of order and an integer ; monic polynomials are as in Degree, leading coefficient and monic polynomial, with the zero polynomial having no degree.
In one has , the product being over the monic irreducible whose degree divides , each such occurring once (Over , is the product of all monic irreducibles whose degrees divide ).
If is an integral domain and are nonzero, then and (Over an integral domain, degrees add under multiplication of nonzero polynomials).
If is an integral domain, then is an integral domain (A polynomial ring over an integral domain is an integral domain).
Proof
For each the monic polynomials of degree in are the with , so there are exactly of them and is finite.
Consequently the family of monic irreducible with is finite, having at most members, so the product in [L1] is a finite product of nonzero polynomials in the integral domain ([L3]).
Taking degrees in [L1] and applying [L2] repeatedly to that finite product gives , where the sum runs over the same finite family; and because .
Splitting that sum according to the degree of : the possible degrees are exactly the positive divisors of , there are monic irreducibles of degree , and each contributes ; hence .
At the only positive divisor is , so the identity reads , in agreement with the fact that the monic polynomials of degree one are the for and each is irreducible.
Remarks
- What the identity does not give. It determines only once every for proper divisors of is known, so it is a recursion rather than a formula. Inverting it into a closed form is a separate matter and is not carried out here.
Normal bases of a finite Galois extension
Definition
Let be a finite Galois extension (Finite Galois extensions and ) of degree (The degree of a finite field extension), and list its Galois group as
which has exactly elements because for a finite Galois extension (Equivalent characterizations of a finite Galois extension). Scalar multiplication by makes an -vector space of dimension .
An element is a normal basis generator for when the list
is an ordered basis of as an -vector space (Basis of a vector space: a linearly independent spanning subset; and ordered basis: an injective finite list whose image is a basis). Such a list is called a normal basis of over : a basis that is a single orbit of the Galois group.
Two conditions, not one. A list is an ordered basis when it is injective and its image is a basis, so a normal basis generator must in particular have distinct conjugates . Neither half implies the other: a basis of over need not be a Galois orbit, and a Galois orbit of size need not be a basis.
The list is indexed by the group, not ordered by it. Reordering permutes the list and leaves the property of being a basis unchanged, since a basis is a property of the underlying set together with injectivity of the list (Basis of a vector space: a linearly independent spanning subset; and ordered basis: an injective finite list whose image is a basis). The automorphisms are those of Relative field automorphisms and .
A polynomial vanishing at every tuple from an infinite subdomain is the zero polynomial
Statement
Let be an integral domain, let be a subring whose underlying set is infinite, let , and let (Polynomial rings in finitely many commuting indeterminates by iteration). If
then in .
Here is the iterated evaluation of Evaluation and roots of a polynomial in a commutative target ring, carried out one indeterminate at a time along the construction of .
Facts & Assumptions
Given: An integral domain , an infinite subring , and the polynomial rings built by iteration (Polynomial rings in finitely many commuting indeterminates by iteration).
A nonzero polynomial of degree over an integral domain has at most distinct roots in (A nonzero polynomial of degree over an integral domain has at most distinct roots).
If is an integral domain, then is an integral domain for every , including (A polynomial ring in finitely many indeterminates over an integral domain is an integral domain).
A nonzero has a largest index with , its degree; the zero polynomial has no degree (Degree, leading coefficient and monic polynomial, with the zero polynomial having no degree).
Proof
Base case : let vanish at every element of . If it has a degree by [L3], so by [L1] it has at most distinct roots in ; but every one of the infinitely many elements of is a root, and an infinite set has more than elements. Hence .
Inductive hypothesis: fix and assume that every vanishing at all tuples from is zero.
Let vanish at every tuple from , and write with . Fix ; then is an element of vanishing at every element of , so it is zero by step 1.1, and therefore for every .
Since was arbitrary, each vanishes at every tuple from , so by step 1.2 and hence . This completes the induction, and the statement holds for every .
Remarks
- Infinite, not merely large. The hypothesis cannot be weakened to a finite of any size: over the nonzero polynomial vanishes at every element, and in indeterminates so does . This is exactly why the normal basis theorem needs a separate argument over a finite base field (Every finite cyclic extension has a normal basis).
For a finite Galois extension, is a base-field basis exactly when the matrix is invertible
Statement
Let be a finite Galois extension of degree , list its Galois group as , and let . Let be the matrix with entries
Then is an ordered basis of as an -vector space (Basis of a vector space: a linearly independent spanning subset; and ordered basis: an injective finite list whose image is a basis) if and only if is invertible over (Invertible matrices and the general linear group ).
Facts & Assumptions
Given: A finite Galois extension of degree with ; elements ; the matrix with (Finite rectangular matrices over a commutative ring, their entries, rows and columns); and the -linear map given by . Each is an -automorphism of (Relative field automorphisms and ), hence additive and -linear.
For a finite Galois extension with one has (Equivalent characterizations of a finite Galois extension, Finite Galois extensions and ); so (The degree of a finite field extension).
Let be a group and a field. Every finite family of distinct group homomorphisms is linearly independent over as a family of functions (Dedekind's linear independence theorem for distinct characters).
For a linear map of -vector spaces with finite-dimensional, (Rank-nullity: ).
For and on : is invertible if and only if is a linear isomorphism (A square matrix is invertible exactly when its multiplication map is a linear isomorphism; matrices preserve inverses of linear isomorphisms).
Let be a commutative ring, , . Then is invertible if and only if is a unit of (A positive-sized square matrix over a commutative ring is invertible if and only if its determinant is a unit).
for every over a commutative ring (For every square matrix over a commutative ring, ); the transpose is (Entrywise ring-matrix operations, rectangular matrix products, identity matrices and transpose).
Proof
By [L1] the -vector space has dimension , and is a map between -vector spaces of dimension ; so by [L3] it is injective if and only if it is surjective, and is an ordered basis of over exactly when is bijective.
For the implication that a basis has an invertible matrix, suppose is an ordered basis, and let satisfy , that is for every . The map , , is -linear because each is, and it vanishes at every , hence on their -span, which is .
For the implication that a non-basis has a singular matrix, suppose is not an ordered basis. By step 1.1 the map is not injective, so there is with and . Applying and using for gives for every , that is with in . Were invertible with inverse , this would force ; so is not invertible.
So is the zero function on , in particular on . The restrictions are group homomorphisms and are pairwise distinct, since two automorphisms of agreeing on agree on ; so [L2] forces for every .
Hence the -linear map on has zero kernel, so by [L3] over it is also surjective and therefore a linear isomorphism; by [L4] the matrix is invertible, so is a unit of by [L5], and by [L6] is a unit, whence is invertible by [L5].
Step 2.1 gives one implication, that a list which is not a basis has a matrix that is not invertible, and step 3.1 gives the other, that a list which is a basis has an invertible matrix; together they are the stated equivalence.
Remarks
-
Why the transpose appears. The dependence relation among the produces a null vector on the right of , while the Dedekind relation among the produces one on the right of . Only the determinant sees both, which is why the two halves are joined through For every square matrix over a commutative ring, rather than by a single rank computation.
-
Where the Galois hypothesis is used. Twice: to know that the group has exactly elements, so that is square, and to know that the are -linear, which is what lets step 2.1 pull the scalars through.
Over an infinite base field, no nonzero polynomial vanishes at the conjugate tuple of every element
Statement
Let be a finite Galois extension of degree whose base field is infinite, and list . Then for every nonzero (Polynomial rings in finitely many commuting indeterminates by iteration) there exists with
Facts & Assumptions
Given: A finite Galois extension of degree with infinite and ; by [L4] the -vector space has dimension , so an ordered -basis of exists (Basis of a vector space: a linearly independent spanning subset; and ordered basis: an injective finite list whose image is a basis).
Let be an integral domain and an infinite subring. If satisfies for all , then (A polynomial vanishing at every tuple from an infinite subdomain is the zero polynomial).
For a finite Galois extension of degree with and , the list is an ordered -basis of if and only if the matrix with is invertible (For a finite Galois extension, is a base-field basis exactly when the matrix is invertible).
For commutative rings , a unital ring homomorphism and , there is a unique unital ring homomorphism extending on constants and sending to (Universal property of : a coefficient homomorphism and the image of determine a unique ring homomorphism). Iterating this along Polynomial rings in finitely many commuting indeterminates by iteration gives, for any commutative -algebra and any , a unique -algebra homomorphism sending to .
For a finite Galois extension with one has (Equivalent characterizations of a finite Galois extension, Finite Galois extensions and , The degree of a finite field extension).
is invertible when some satisfies ; such a is unique and written (Invertible matrices and the general linear group , Entrywise ring-matrix operations, rectangular matrix products, identity matrices and transpose).
Proof
It suffices to prove the contrapositive: if satisfies for every , then . Assume that hypothesis on .
Fix an ordered -basis of and put ; by [L2] the matrix is invertible, with inverse as in [L5].
For write ; then is a bijection because the form a basis, and since each fixes pointwise and is additive.
Let be the unique -algebra homomorphism with , and the unique one with ; both exist by [L3].
For , the evaluation homomorphism at composed with sends to , so by the uniqueness clause of [L3] it is evaluation at ; hence evaluated at equals , which is by the hypothesis of step 1.1.
So vanishes at every tuple from the infinite subring of the integral domain , and [L1] gives .
The composite is a -algebra endomorphism sending to , so it is the identity by the uniqueness clause of [L3]; applying to step 4.1 therefore gives , which is the contrapositive.
Remarks
- Where infiniteness of enters. Only in step 4.1, through [L1]. Over a finite base field the conclusion is false: with and , the nonzero polynomial vanishes at every conjugate tuple, since every element of satisfies . The finite case of the normal basis theorem is therefore proved by a different argument (Every finite cyclic extension has a normal basis).
Every finite Galois extension of an infinite field has a normal basis
Statement
Let be a finite Galois extension whose base field is infinite. Then has a normal basis (Normal bases of a finite Galois extension): there is such that is an ordered -basis of , where .
Facts & Assumptions
Given: A finite Galois extension (Finite Galois extensions and ) of degree with infinite, and numbered so that ; the matrix over with where is the index determined by ; and (For , the determinant over a commutative ring by the Leibniz formula, and for a real matrix).
For every nonzero there is with (Over an infinite base field, no nonzero polynomial vanishes at the conjugate tuple of every element).
is an ordered -basis of if and only if the matrix with is invertible (For a finite Galois extension, is a base-field basis exactly when the matrix is invertible).
(For , the determinant over a commutative ring by the Leibniz formula, and for a real matrix); a matrix over a commutative ring is invertible if and only if its determinant is a unit (A positive-sized square matrix over a commutative ring is invertible if and only if its determinant is a unit).
A matrix is invertible when some satisfies (Invertible matrices and the general linear group ); matrix products, the identity and the transpose are as in Entrywise ring-matrix operations, rectangular matrix products, identity matrices and transpose, and obey the arithmetic laws of Matrix arithmetic over a commutative ring is associative, unital and distributive, and transpose reverses products.
is an integral domain (A polynomial ring in finitely many indeterminates over an integral domain is an integral domain, Polynomial rings in finitely many commuting indeterminates by iteration), and for any commutative -algebra and there is a unique -algebra homomorphism with (Universal property of : a coefficient homomorphism and the image of determine a unique ring homomorphism).
Proof
is a well-defined matrix over : for each pair the product lies in the group and so equals for exactly one index .
Let be the -algebra homomorphism with and for , supplied by [L5]. Applying entrywise to gives the matrix with when , that is when , and otherwise.
is invertible, with as an inverse: the entry of is , and a term is nonzero exactly when and , which happens for exactly one when and for no otherwise; so , and the same computation on gives . Hence is a unit of by [L3], and in particular .
Since is a polynomial expression in the entries by [L3] and is a ring homomorphism, ; hence in .
By [L1] there is with . Substituting in replaces the entry , where , by ; since substitution is a ring homomorphism, it carries to the determinant of the matrix with for . So .
A nonzero element of the field is a unit, so is invertible by [L3], and [L2] makes an ordered -basis of ; that is a normal basis.
Remarks
-
What the specialisation is for. The matrix of indeterminates is a device for showing that one determinant polynomial is not the zero polynomial, and the cheapest way to see that is to send it to a permutation matrix. Nothing about the particular substitution survives into the conclusion: the element produced in step 5.1 has no relation to it.
-
Why is the identity. Only so that the specialised matrix is the permutation matrix of ; any other choice of which indeterminate to set to would give the permutation matrix of a different bijection, with the same conclusion.
Every finite cyclic extension has a normal basis
Statement
Let be a finite Galois extension whose Galois group is cyclic, say of order . Then has a normal basis (Normal bases of a finite Galois extension): there is for which
is an ordered -basis of (Basis of a vector space: a linearly independent spanning subset; and ordered basis: an injective finite list whose image is a basis).
Facts & Assumptions
Given: A finite Galois extension with cyclic of order ; by [L6] , and is an -linear endomorphism of the -vector space , written when regarded as such. Evaluation of a polynomial at is that of Polynomial evaluation at an endomorphism: .
Let be a group and a field. Every finite family of distinct group homomorphisms is linearly independent over as a family of functions (Dedekind's linear independence theorem for distinct characters).
For every endomorphism of a finite-dimensional -vector space, is a nonzero ideal with a unique monic generator , and if and only if (The annihilator ideal is nonzero and has a unique monic generator; if and only if , The annihilator set ; once existence is proved, its unique monic generator is the minimal polynomial).
For the polynomial is monic of degree ( is monic of degree ; for its coefficient is and its constant coefficient is , while ); is defined as for any ordered basis and is independent of it (The basis-independent characteristic polynomial of an endomorphism of a finite-dimensional space, including in dimension zero).
for every endomorphism of a finite-dimensional vector space (The minimal polynomial divides the characteristic polynomial, ).
An endomorphism of a finite-dimensional vector space has a cyclic vector if and only if (A cyclic vector exists exactly when the minimal and characteristic polynomials agree); is a cyclic vector when is all of (Cyclic subspaces, cyclic vectors, and vector annihilators).
For a finite Galois extension with one has (Equivalent characterizations of a finite Galois extension, Finite Galois extensions and , The degree of a finite field extension).
With the unique monic generator of (The vector annihilator is the unique monic generator of and divides the minimal polynomial) and , the list is an ordered basis of (A vector annihilator gives a power basis and its companion matrix).
Proof
, because has order in ; so the polynomial lies in and by [L2]. In particular .
The maps are pairwise distinct elements of , and their restrictions to are pairwise distinct group homomorphisms , since two field automorphisms of agreeing on agree on .
If with satisfies , then is the zero function on , hence on , so [L1] applied to the family of step 1.2 forces every to be ; thus no nonzero polynomial of degree less than annihilates , and .
Combining steps 1.1 and 2.1, , and since is monic and divides the monic of the same degree, .
By [L3] the polynomial is monic of degree , and by [L4]; two monic polynomials of the same degree, one dividing the other, are equal, so .
By [L5] there is a cyclic vector for , that is .
Let . By [L7] the list is an ordered basis of , which has dimension , so and is an ordered -basis of .
Since , that list is exactly the family of conjugates of , so it is a normal basis.
Remarks
-
Why the minimal polynomial is forced to be . The divisibility is cheap; the content is the lower bound on its degree, and that is exactly Dedekind's independence of characters. Without it the minimal polynomial could be a proper divisor of and no cyclic vector would be available.
-
The hypothesis is on the group, not on the base field. No finiteness or infiniteness of is used, so this proof also covers cyclic extensions of infinite fields, for which Every finite Galois extension of an infinite field has a normal basis gives a second and quite different argument.
Every finite Galois extension has a normal basis
Statement
Every finite Galois extension has a normal basis (Normal bases of a finite Galois extension): there is whose family of conjugates , indexed by , is an ordered -basis of .
Facts & Assumptions
Given: A finite Galois extension of degree , so that is an -vector space of dimension (The degree of a finite field extension, Equivalent characterizations of a finite Galois extension).
Every finite Galois extension of an infinite field has a normal basis (Every finite Galois extension of an infinite field has a normal basis).
Every finite Galois extension whose Galois group is cyclic has a normal basis (Every finite cyclic extension has a normal basis).
An extension of finite fields of degree is Galois with cyclic of order , generated by (A finite extension of a finite field of order is Galois with cyclic Galois group generated by ).
A finite list of length is an ordered basis of if and only if every has exactly one coordinate list with (A finite list is an ordered basis if and only if every equals for exactly one ; those scalars are the coordinates of in that ordered basis, Basis of a vector space: a linearly independent spanning subset; and ordered basis: an injective finite list whose image is a basis).
A finite field is a field whose underlying set is finite, and its order is that cardinality (Finite fields and their order).
Proof
In the case that is infinite, [L1] applies directly and has a normal basis.
In the case that is finite, fix an ordered -basis of of length ; by [L4] the map sending a coordinate list to is a bijection onto , so is finite and is a finite field.
In that same finite case, is therefore an extension of finite fields of degree , so is cyclic by [L3], and [L2] gives a normal basis.
The two cases are exhaustive, a field being finite or infinite and not both, so a normal basis exists in either case.
Remarks
-
Two genuinely different proofs, not one proof with a case split. The infinite case runs on a determinant that is a nonzero polynomial (Every finite Galois extension of an infinite field has a normal basis); the finite case runs on a cyclic vector for the Frobenius acting linearly (Every finite cyclic extension has a normal basis). Neither argument covers the other case: the first fails because a polynomial can vanish on all of a finite field, the second because a Galois group need not be cyclic.
-
The finite case is not a hypothesis on the group. It is a hypothesis on the base field, which forces the group to be cyclic through [L3]. That is the whole reason the split is by the base field rather than by the group.
The group of -th roots of unity in a field, and primitive -th roots of unity
Definition
Let be a field and let be an integer. An element is an -th root of unity when , that is when is a root of (Evaluation and roots of a polynomial in a commutative target ring). Write
This is a subgroup of (Subgroup). Every is invertible, with inverse (Left inverse, right inverse, and invertible element of a monoid), so ; it contains ; it is closed under multiplication, since ; and it is closed under inverses, since .
An element is a primitive -th root of unity when its order in the group is exactly (The order of a finite group and the order of an element, with when no positive power of is the identity):
Equivalently, and no exponent with has .
An -th root of unity need not be primitive. In the element is a fourth root of unity of order two, not four; and may consist of alone, as does. The two notions are separated deliberately, and the exact circumstances under which a primitive -th root of unity exists are the content of is cyclic of order dividing , and has a primitive -th root of unity exactly when its order is and is separable over exactly when the characteristic does not divide , and then a splitting field carries distinct -th roots of unity.
Remarks
- The concrete complex picture arrives later in the reading order. Over the -th roots of unity are the numbers , and that description is developed in The -th roots of a complex number and the distinct roots of unity for every ↗. Nothing on this page uses it: the definition above is purely algebraic and applies to every field, including those of positive characteristic, where the count of -th roots of unity can be smaller than .
is cyclic of order dividing , and has a primitive -th root of unity exactly when its order is
Statement
Let be a field and . Then (The group of -th roots of unity in a field, and primitive -th roots of unity) is a finite cyclic subgroup of whose order divides (Divisibility in : when for some integer ). It contains a primitive -th root of unity if and only if , and in that case the primitive -th roots of unity in are exactly the generators of , of which there are (The unit group and Euler's totient for ).
Facts & Assumptions
Given: A field , an integer , and the subgroup of (The group of -th roots of unity in a field, and primitive -th roots of unity).
Let be an integral domain. Every finite subgroup of the unit group of is cyclic (Every finite subgroup of the unit group of an integral domain is cyclic).
A nonzero polynomial of degree over an integral domain has at most distinct roots in that domain (A nonzero polynomial of degree over an integral domain has at most distinct roots).
For an element of finite order in a group: if and only if ; and (If then iff is an integer multiple of , the powers are distinct, and has exactly elements; if has infinite order then only for , The order of a finite group and the order of an element, with when no positive power of is the identity).
In a cyclic group of finite order , the element generates the group if and only if , and the group has exactly generators (A cyclic group of order has exactly generators).
Proof
is the set of roots in of the nonzero polynomial , of degree ; a field is an integral domain, so by [L2] and is finite.
Being a finite subgroup of , is cyclic by [L1]; write and fix a generator , so that by [L3].
The order divides : gives , and [L3] turns this into .
If contains a primitive -th root of unity , that is an element of order , then puts in , and has elements by [L3], so ; with step 1.1 this forces .
Conversely, if then the generator of step 2.1 has order and so is a primitive -th root of unity in .
Suppose . An element of order lies in and satisfies by [L3], so and is a generator; conversely a generator has order by [L3]. So the primitive -th roots of unity in are exactly the generators of , and [L4] counts them as .
Remarks
- Order dividing , not equal to . The two failures are different in kind. Over the polynomial simply does not split, so ; over a field of characteristic three it cannot split into distinct factors at all, since there (In characteristic the only -th root of unity is , and ). Only the first failure is repaired by passing to a splitting field.
is separable over exactly when the characteristic does not divide , and then a splitting field carries distinct -th roots of unity
Statement
Let be a field and . Then
(Repeated roots in extension fields and separable polynomials, The characteristic of a ring: the least with when one exists, and otherwise; divisibility of integers being that of Divisibility in : when for some integer , under which holds only for , so characteristic never divides ).
When and is a splitting field of over , the group is cyclic of order exactly , it contains exactly primitive -th roots of unity, and
for every primitive -th root of unity .
Facts & Assumptions
Given: A field , an integer , the polynomial , and the element obtained by adding to itself times.
for (The formal derivative of a polynomial); for and constants have derivative , and the derivative is additive (Linearity, power rule, Leibniz rule and the degree bound for the formal derivative). Hence .
For a field and : is separable over if and only if in (A nonzero polynomial over a field is separable exactly when its gcd with its derivative is ).
For not both zero, is monic, divides both and , and every common divisor of and divides (Bézout identity and the Euclidean algorithm for polynomials over a field).
For a ring and , one has if and only if divides (The characteristic of a ring is the additive order of , with recording infinite order; holds exactly when ; and in an integral domain every nonzero element has the same additive order as , The characteristic of a ring: the least with when one exists, and otherwise); the characteristic of a field is or a prime (The characteristic of a field is zero or a prime number).
Every nonzero has a splitting field over (Every nonzero polynomial over a field has a splitting field); of degree splits over when with and , repetitions allowed, and a splitting field is generated over by the roots (Polynomials that split and splitting fields of a polynomial or a family of polynomials), the notation for the subfield generated by and a finite set being that of Finitely generated field extensions .
is a finite cyclic subgroup of of order dividing ; it contains a primitive -th root of unity exactly when its order is , and there are then of them, namely the generators ( is cyclic of order dividing , and has a primitive -th root of unity exactly when its order is , The group of -th roots of unity in a field, and primitive -th roots of unity, The unit group and Euler's totient for ).
is separable over when it has no repeated root in any extension field of , where is a repeated root of in when divides the image of in (Repeated roots in extension fields and separable polynomials).
Proof
In the case , put . By [L1] with a unit of , so by [L3], hence ; and , so divides and, being monic, . By [L2] the polynomial is separable over .
In the case , [L1] gives , so every polynomial dividing is a common divisor of and ; by [L3] the monic is divisible by every such divisor and divides , so it is itself, of degree . Thus and is not separable over by [L2].
The two cases are exhaustive and, by [L4], says exactly ; so is separable over if and only if .
Assume now and let be a splitting field of over , which exists by [L5]. Over one has with , and comparing leading coefficients of the monic gives .
The are pairwise distinct: if for then divides in , making a repeated root of in the extension of , which contradicts the separability supplied by step 2.1 through [L7].
Hence has exactly distinct roots in , that is ; by [L6] the group is cyclic of order and contains exactly primitive -th roots of unity.
Fix such a . Every root of in lies in , so is a power of ; since is generated over by those roots by [L5], . At the polynomial is , , , and .
Remarks
- The failing direction is inseparability, not a shortage of roots. When divides , the derivative vanishes identically and no extension can separate the roots: writing with , one has in every field of characteristic (In characteristic the only -th root of unity is , and ). Passing to a larger field does not help, which is why the hypothesis is carried on every later statement rather than removed by enlarging .
In characteristic the only -th root of unity is , and
Statement
Let be a field of characteristic (The characteristic of a ring: the least with when one exists, and otherwise; is prime by The characteristic of a field is zero or a prime number) and let . Then in (The polynomial ring over a commutative ring as finitely supported coefficient sequences with convolution)
and consequently (The group of -th roots of unity in a field, and primitive -th roots of unity).
More generally, if and with (Divisibility in : when for some integer ), and if , then
Facts & Assumptions
Given: A field of characteristic , so that and hence for every and every integer divisible by ; and an integer .
For a commutative ring , all and , , the natural-number coefficients acting by repeated addition (The binomial theorem over an arbitrary commutative ring).
If is prime and , then (A prime divides for ).
Proof
For every one has : by [L1] applied in the commutative ring , ; for the coefficient is a multiple of by [L2], so that term vanishes by the hypothesis on ; the surviving terms are and , and for odd while for one has in , so in either case.
Hence for every , by induction on : at this is step 1.1 with ; and if it holds at , then , the last equality being step 1.1 with .
Therefore : an with is a root of , so by step 2.1, and a field has no nonzero element with a vanishing power, so ; and .
Let and with , and put . If then , so , which is by step 3.1 when and is trivially when ; either way . Conversely gives . Hence .
Remarks
- This is why every later hypothesis reads "the characteristic does not divide ". Nothing is lost by it: the -part of contributes no roots of unity at all in characteristic , so a statement about there is already a statement about for the prime-to- part . The hypothesis excludes a degenerate case rather than a genuine one.
The cyclotomic extension as a splitting field of
Definition
Let be a field and . A cyclotomic extension of of order is a splitting field of over (Polynomials that split and splitting fields of a polynomial or a family of polynomials); one exists by Every nonzero polynomial over a field has a splitting field. It is written
The notation is accurate. The roots of in are exactly the elements of (The group of -th roots of unity in a field, and primitive -th roots of unity), and a splitting field is generated over by the roots, so
in the sense of Finitely generated field extensions : is the smallest subfield of itself containing and the -th roots of unity it holds.
When the characteristic does not divide the extension has a single generator: by is separable over exactly when the characteristic does not divide , and then a splitting field carries distinct -th roots of unity the group is then cyclic of order and
for any primitive -th root of unity . Without that hypothesis the notation still names a splitting field, but can be much smaller than and no primitive -th root of unity need exist (In characteristic the only -th root of unity is , and ).
Remarks
- Which splitting field. Any two splitting fields of over are -isomorphic (Any two splitting fields of a polynomial are isomorphic over the base field), and every statement made below about is invariant under a -isomorphism, so the definite article is harmless; where a fixed ambient field matters, as in the compositum and intersection results, the statement says so and works inside one chosen extension of .
is Galois and embeds its Galois group into
Statement
Let be a field and with (The characteristic of a ring: the least with when one exists, and otherwise), and let (The cyclotomic extension as a splitting field of ). Then is a finite Galois extension, and there is an injective group homomorphism
where is any integer with for a primitive -th root of unity . The class does not depend on which primitive -th root of unity is used, and holds for every .
Facts & Assumptions
Given: A field , an integer with , the extension (The cyclotomic extension as a splitting field of ), and a primitive -th root of unity (The group of -th roots of unity in a field, and primitive -th roots of unity).
For , is separable over when ; in a splitting field the group is then cyclic of order , has primitive -th roots of unity, and for every primitive -th root of unity ( is separable over exactly when the characteristic does not divide , and then a splitting field carries distinct -th roots of unity).
For a finite extension with , the conditions " is Galois", " is the splitting field over of a separable polynomial", "" and "" are equivalent (Equivalent characterizations of a finite Galois extension, Finite Galois extensions and , Relative field automorphisms and ).
is a finite cyclic subgroup of of order dividing , and when its order is its generators are exactly the primitive -th roots of unity ( is cyclic of order dividing , and has a primitive -th root of unity exactly when its order is ).
In a cyclic group of finite order , generates the group if and only if (A cyclic group of order has exactly generators).
For an element of finite order : if and only if (If then iff is an integer multiple of , the powers are distinct, and has exactly elements; if has infinite order then only for ).
For and , the class (The congruence class and the quotient set ) is a unit of if and only if (For , is a unit if and only if ); the units form the group of order (The unit group and Euler's totient for ).
If is algebraic over a field , then the simple extension is finite, with degree equal to the degree of the minimal polynomial of (A simple algebraic extension is its minimal-polynomial quotient and has power basis and degree ).
Proof
By [L1] the polynomial is separable over , the group is cyclic of order generated by , and . The element is algebraic because it is a root of , so [L7] makes finite. Since is the splitting field of the separable polynomial , [L2] now makes finite Galois.
Each maps into itself, since ; being injective on the finite set it restricts to a bijection, and it is multiplicative, so it restricts to a group automorphism of .
Hence generates , so for some integer , and [L4] gives , so by [L6]. The class is well defined: means , which by [L5] and says , that is . Write for this class.
For every one has for some , so .
The map is a group homomorphism: , so by the well-definedness of step 3.1.
It is injective: if then by step 3.1, and since and fixes pointwise, is the identity on .
The class is independent of the chosen primitive -th root of unity: any other one is a generator of by [L3], so for some , and , which exhibits the same exponent class. With steps 4.1, 4.2 and 4.3 this proves the theorem.
Remarks
- The embedding need not be onto. Surjectivity is exactly irreducibility of the -th cyclotomic polynomial over ( is irreducible over exactly when , exactly when the embedding into is onto), and it fails over many base fields: over the image is the cyclic subgroup generated by the class of (For the image of in is generated by ).
The Galois group of a cyclotomic extension is abelian
Statement
Let be a field and with (The characteristic of a ring: the least with when one exists, and otherwise). Then (The cyclotomic extension as a splitting field of ) is abelian.
Facts & Assumptions
Given: A field and with .
is finite Galois and is an injective group homomorphism ( is Galois and embeds its Galois group into ).
For every , is a commutative monoid (For every natural , is an abelian group, multiplication is a commutative monoid operation, and both distributive laws hold), and is the group of its invertible elements (The unit group and Euler's totient for ).
Proof
Multiplication on is commutative by [L2], so the group , whose operation is that multiplication restricted to the units, is abelian.
By [L1] the group is isomorphic to its image in , a subgroup of an abelian group; a subgroup of an abelian group is abelian, and a group isomorphic to an abelian group is abelian, so is abelian.
For the image of in is generated by
Statement
Let be a finite field of order (Finite fields and their order) and let with (Coprime integers: ). Then the image of the embedding
of is Galois and embeds its Galois group into is the cyclic subgroup generated by , and
the multiplicative order of in (The order of a finite group and the order of an element, with when no positive power of is the identity).
Facts & Assumptions
Given: A finite field of order , of characteristic with a power of (Every finite field has order for a unique prime characteristic and positive integer ), and an integer with ; the extension (The cyclotomic extension as a splitting field of ).
For a field and with , the extension is finite Galois and , determined by on a primitive -th root of unity, is an injective homomorphism into with for every ( is Galois and embeds its Galois group into ).
An extension of finite fields of degree is Galois with cyclic of order , where (A finite extension of a finite field of order is Galois with cyclic Galois group generated by , The relative Frobenius of an extension of finite fields).
For and , the class is a unit of if and only if (For , is a unit if and only if , The unit group and Euler's totient for ).
For a finite Galois extension one has (Equivalent characterizations of a finite Galois extension, The degree of a finite field extension).
Proof
The characteristic divides , and , so ; hence [L1] applies to and is finite Galois over . Also is a unit of by [L3].
is a finite field: it is a finite extension of the finite field by step 1.1, so it is a finite-dimensional -vector space over a finite field and therefore has finitely many elements. By [L2], with .
The exponent attached to by [L1] is , since for a primitive -th root of unity . Because the embedding is a homomorphism and is generated by , the image is the subgroup of generated by .
The embedding is injective, so equals the order of , which is ; and by [L4]. Hence .
Remarks
- What the coprimality hypothesis does. It makes a unit, so that the subgroup it generates is defined, and it puts the extension inside the scope of [L1] by ruling out . Without it the group collapses onto for the prime-to- part of (In characteristic the only -th root of unity is , and ), and the statement is about rather than .
The cyclotomic polynomials , defined by
Definition
The cyclotomic polynomials (The polynomial ring over a commutative ring as finitely supported coefficient sequences with convolution) are defined by recursion on :
the divisors being the positive divisors of (Divisibility in : when for some integer ). The product is formed in the commutative ring (Polynomial convolution makes a commutative ring containing as its constant subring): multiply the finitely many factors in any enumeration of the divisor set. Associativity and commutativity make the result independent of that enumeration.
What the fraction means. The denominator is a product of monic polynomials in , hence itself monic (Degree, leading coefficient and monic polynomial, with the zero polynomial having no degree), so Division by a monic polynomial over a commutative ring supplies a unique pair with and or . The definition sets , and asserts . That assertion, together with the consequences that each is monic of degree and that
for every , is discharged by The recursion defines a unique monic , of degree ↗, which is why that theorem is a numbered result and not a parenthesis: the division is carried out over , not over a field, so exactness is a statement about integer coefficients and does not follow from the division algorithm.
Remarks
-
Why not define by its roots. The usual definition takes to be the monic polynomial whose roots are the primitive -th roots of unity in , or the minimal polynomial over of a primitive -th root of unity. Neither is available at this point in the reading order, and neither would serve: the minimal-polynomial version makes is irreducible in for every a tautology rather than a theorem, and over a base field where the reduction of is reducible it no longer controls the factorisation there (For the reduction of in is a product of distinct monic irreducibles, each of degree the order of modulo ). The recursion above is a statement about alone, and the description of the roots is then a theorem (Over a field whose characteristic does not divide , the roots of are exactly the primitive roots of unity).
-
Reduction into another field. For a field the image of under the coefficientwise ring homomorphism induced by the canonical map and fixing is again written when no confusion arises; the identity is preserved, since a ring homomorphism preserves finite products.
The recursion defines a unique monic , of degree
Statement
The recursion of The cyclotomic polynomials , defined by is well posed: for every the division defining is exact, is a monic element of ,
the product being over the positive divisors of , and
(The unit group and Euler's totient for ). Moreover is the only family of monic polynomials in satisfying the displayed product identity for every .
Facts & Assumptions
Given: The recursion of The cyclotomic polynomials , defined by ; the field , which is an ordered field (The rationals form a totally ordered field), so that and in particular for every , whence (The characteristic of a ring: the least with when one exists, and otherwise) and divides no (Divisibility in : when for some integer ).
Let be a commutative ring and monic. For every there are unique with and or (Division by a monic polynomial over a commutative ring).
is separable over exactly when ; and then a splitting field has cyclic of order ( is separable over exactly when the characteristic does not divide , and then a splitting field carries distinct -th roots of unity, The group of -th roots of unity in a field, and primitive -th roots of unity, is cyclic of order dividing , and has a primitive -th root of unity exactly when its order is ).
Every nonzero has a splitting field over (Every nonzero polynomial over a field has a splitting field); monic of degree splits over when with , repetitions allowed (Polynomials that split and splitting fields of a polynomial or a family of polynomials).
is separable over when it has no repeated root in any extension field, a repeated root of in being an with dividing the image of in (Repeated roots in extension fields and separable polynomials).
For an element of finite order in a group, if and only if (If then iff is an integer multiple of , the powers are distinct, and has exactly elements; if has infinite order then only for , The order of a finite group and the order of an element, with when no positive power of is the identity).
for every positive integer (For every positive integer , , The sum over a finite index set, and its product form).
If is an integral domain then so is (A polynomial ring over an integral domain is an integral domain), and for nonzero one has and (Over an integral domain, degrees add under multiplication of nonzero polynomials, Degree, leading coefficient and monic polynomial, with the zero polynomial having no degree).
Proof
The assertion to be proved by strong induction on is: the division defining is exact, is monic, , and . At the recursion sets outright, the only positive divisor of is so the product is , and .
Inductive hypothesis: fix and assume the assertion for every with .
Since does not divide , [L2] and [L3] supply a splitting field of over in which is separable and is cyclic of order . For a positive divisor of put and , a monic polynomial of degree .
For every positive divisor of one has with : the polynomial divides in , since when , so it splits over by [L3], and a repeated root of in an extension would be a repeated root of there, which [L4] excludes; hence its roots are distinct and they are by definition the elements of .
For every positive divisor of , is the disjoint union of the over positive divisors of : an element has finite order dividing , and holds exactly when by [L5]. Hence by step 2.1.
For every positive divisor of with one has , by induction on through the divisors of : at both equal , since ; and if for every positive divisor of with , then step 1.2 and step 3.1 give , and cancelling the nonzero left factor in the integral domain ([L7]) yields .
Write , monic in by step 1.2 and [L7]. By step 4.1 and step 3.1 applied with , in one has .
The division of by in is exact and its quotient is : by [L1] over there are unique with and or ; this is also a division by the monic in , where step 5.1 exhibits the division with quotient and remainder , so the uniqueness clause of [L1] over forces and . Hence is monic in and .
Degrees: taking degrees in with [L7] gives , and by step 1.2 and [L7]; so by [L6].
This is the assertion at , so the strong induction is complete and the assertion holds for every . Uniqueness of the family follows by the same induction: if is monic in with for all , then , and if for all then with in the integral domain ([L7]), so .
Remarks
-
Why a splitting field over appears in a statement about . The exactness of the division is an identity between integer polynomials, but the only cheap reason for it is that the roots of partition by order. The passage to produces that partition; the passage back is the uniqueness clause of monic division, which holds over and over and pins the two computations to the same quotient.
-
The degree computation is where enters. Nothing before step 7.1 mentions Euler's totient; it appears only through For every positive integer , , and the same identity is what makes the count of primitive -th roots of unity match in Over a field whose characteristic does not divide , the roots of are exactly the primitive roots of unity.
and for
Statement
For the cyclotomic polynomials of The cyclotomic polynomials , defined by ,
Facts & Assumptions
Given: The cyclotomic polynomials and their defining identity; evaluation at is as in Evaluation and roots of a polynomial in a commutative target ring.
For every one has , the product being over the positive divisors of (The recursion defines a unique monic , of degree , The sum over a finite index set, and its product form, Divisibility in : when for some integer ).
Evaluation at an element of a commutative ring is a unital ring homomorphism (Universal property of : a coefficient homomorphism and the image of determine a unique ring homomorphism), so it carries finite products to finite products.
Proof
by The cyclotomic polynomials , defined by , so ; this is also the base of the induction below, taken at , where [L1] and [L2] give , hence and .
Inductive hypothesis: fix and assume for every with .
Evaluating the identity of [L1] at and using [L2] gives ; separating the factor , which is by step 1.1, leaves .
Every factor with equals by step 1.2, so the product reduces to , which is the assertion at ; the induction is complete and for every .
Remarks
- What this is for. The value at the origin makes modulo for every integer , so no prime dividing divides . That is the whole mechanism behind For every there are infinitely many primes with , and the exceptional value at is why that theorem treats separately.
Over a field whose characteristic does not divide , the roots of are exactly the primitive roots of unity
Statement
Let be a field and with (The characteristic of a ring: the least with when one exists, and otherwise, Divisibility in : when for some integer ), and let be a splitting field of over , that is (The cyclotomic extension as a splitting field of ). Write also for the image of the integer polynomial (The cyclotomic polynomials , defined by ) in . Then is separable over (Repeated roots in extension fields and separable polynomials), it splits over , and its roots in are exactly the primitive -th roots of unity in (The group of -th roots of unity in a field, and primitive -th roots of unity, The unit group and Euler's totient for ).
Facts & Assumptions
Given: A field , an integer with , a splitting field of over , and the convention that for every finite subset the product means the finite product along any enumeration of ; because is a commutative ring, the value is independent of the enumeration (The product of a finite list in a monoid, by recursion, with the empty product () equal to the identity, Generalised associativity: in a monoid the product of a finite list does not depend on the bracketing, and in a commutative monoid it does not depend on the order of the factors either, Polynomial convolution makes a commutative ring containing as its constant subring). In particular, for each positive divisor of , let and .
is separable over when , and then is cyclic of order with exactly primitive -th roots of unity ( is separable over exactly when the characteristic does not divide , and then a splitting field carries distinct -th roots of unity, is cyclic of order dividing , and has a primitive -th root of unity exactly when its order is ).
For every one has in , each monic of degree (The recursion defines a unique monic , of degree ); reduction into preserves this identity.
A monic of degree splits over when with , repetitions allowed, and a splitting field is generated over by the roots (Polynomials that split and splitting fields of a polynomial or a family of polynomials).
is separable over when no extension field of contains an with dividing the image of (Repeated roots in extension fields and separable polynomials).
is an integral domain when is (A polynomial ring over an integral domain is an integral domain); and if and only if divides (Factor theorem over a commutative ring).
Proof
By [L1] the polynomial is separable over and is cyclic of order ; so has distinct roots in , namely the elements of , and by [L3].
Each has order dividing by [L5], and for a positive divisor of the condition says exactly ; so is the disjoint union of the over positive divisors of .
For every positive divisor of the polynomial divides in , since ; hence it splits over with distinct roots, which are the elements of , and with .
Consequently for every positive divisor of .
For every positive divisor of the image of in is , by induction on through the divisors of : at both are , since ; and if the claim holds for every positive divisor of with , then [L2] and step 3.1 give , and cancelling the nonzero left factor in the integral domain ([L6]) gives .
Taking : the image of in is , so splits over and its roots there are exactly the elements of , which are the elements of order in , that is the primitive -th roots of unity in ; there are of them by [L1], in agreement with from [L2].
is separable over . The product identity [L2] shows that the image of divides in . If an extension field contained an for which divided the image of , then the same square would divide the image of in , making a repeated root of . This contradicts the separability of supplied by [L1]. Thus [L4] applies.
Remarks
-
Both hypotheses are used, and neither can be dropped. If the characteristic divides then is not separable and the roots of do not separate into orders at all; and the statement is about the roots in a splitting field, not in itself, since may have no root in at all, as over shows.
-
Irreducibility is a separate question. The theorem says what the roots of are; it says nothing about whether factors over . Over it does not ( is irreducible in for every ), over a finite field it usually does (For the reduction of in is a product of distinct monic irreducibles, each of degree the order of modulo ), and in both cases the root description above is the same.
is irreducible over exactly when , exactly when the embedding into is onto
Statement
Let be a field and with (The characteristic of a ring: the least with when one exists, and otherwise), let (The cyclotomic extension as a splitting field of ), and let be a primitive -th root of unity (The group of -th roots of unity in a field, and primitive -th roots of unity). The following are equivalent.
- The image of in (The cyclotomic polynomials , defined by ) is irreducible (Irreducible and prime elements of an integral domain).
- (The degree of a finite field extension, The unit group and Euler's totient for ).
- The embedding of is Galois and embeds its Galois group into is an isomorphism.
Facts & Assumptions
Given: A field , an integer with , the extension , a primitive -th root of unity , and the minimal polynomial of over .
The image of in has as its roots in exactly the primitive -th roots of unity in (Over a field whose characteristic does not divide , the roots of are exactly the primitive roots of unity).
is monic of degree (The recursion defines a unique monic , of degree , Degree, leading coefficient and monic polynomial, with the zero polynomial having no degree); reduction into preserves both.
For algebraic over there is a unique monic irreducible with if and only if (The evaluation kernel and the unique monic irreducible minimal polynomial of an algebraic element).
If is algebraic over with minimal polynomial of degree , then (A simple algebraic extension is its minimal-polynomial quotient and has power basis and degree ).
For this , is finite Galois and is an injective homomorphism ( is Galois and embeds its Galois group into ); moreover for every primitive -th root ( is separable over exactly when the characteristic does not divide , and then a splitting field carries distinct -th roots of unity).
(The unit group and Euler's totient for ), and for a finite Galois extension (Equivalent characterizations of a finite Galois extension).
In an integral domain, an irreducible element is a nonzero nonunit every one of whose factorisations has a unit factor (Irreducible and prime elements of an integral domain).
Proof
is a root of the image of in by [L1], so divides that image by [L3]; both are monic, and has positive degree because is not a root of a nonzero constant. Write the image of as with monic.
For the equivalence of clauses 2 and 3: by [L5] one has and the embedding is injective into a group of order by [L6], so it is surjective if and only if ; and by [L6]. An injective homomorphism onto its target is an isomorphism.
For the implication from clause 1 to clause 2: if the image of is irreducible, then in the factorisation of step 1.1 one factor is a unit by [L7], and is not, so is a nonzero constant; both and being monic forces and . Hence by [L2] and [L4].
For the implication from clause 2 to clause 1: if then by [L2] and [L4], so in step 1.1 is monic of degree , that is and the image of equals , which is irreducible by [L3].
Steps 2.1 and 2.2 give the equivalence of clauses 1 and 2, and step 1.2 the equivalence of clauses 2 and 3; so all three are equivalent.
Remarks
- This is a criterion, not a theorem about . Over all three clauses hold for every ( is irreducible in for every ), but over they hold only when the class of generates (The reduction of is irreducible over exactly when generates ).
, and is Eisenstein at
Statement
Let be a prime (Prime and composite integers: is prime when and its only positive divisors are and ) and . Then
the polynomial satisfies the Eisenstein criterion at (Eisenstein criterion over the integers), and consequently is irreducible in (Irreducible and prime elements of an integral domain).
The sum starts at : its first term is the constant , and evaluation at (Evaluation and roots of a polynomial in a commutative target ring) gives .
Facts & Assumptions
Given: A prime and an integer ; the cyclotomic polynomials of The cyclotomic polynomials , defined by and the substitution homomorphisms of Universal property of : a coefficient homomorphism and the image of determine a unique ring homomorphism, under which is a ring automorphism of and of with inverse .
For every , with each monic in of degree (The recursion defines a unique monic , of degree , Degree, leading coefficient and monic polynomial, with the zero polynomial having no degree).
Let be primitive with . If a prime satisfies , for every , and , then is irreducible in (Eisenstein criterion over the integers).
A nonzero integer polynomial is primitive exactly when no prime divides all of its coefficients (Content is the positive common divisor of the coefficients divisible by every common divisor, Content and primitive integer polynomials).
For a fixed integer , in a field of characteristic one has in (In characteristic the only -th root of unity is , and ); is such a field (For every prime , the two operations on make it a field, The characteristic of a ring: the least with when one exists, and otherwise).
for every prime and (For a prime and , ).
Every integer greater than has a prime divisor (Every integer has a prime divisor; indeed the least divisor of that exceeds is prime); and if a prime divides a finite product of integers it divides one of the factors (If a prime divides a finite product of integers then for some ; at the product is and the hypothesis cannot hold).
is an integral domain when is (A polynomial ring over an integral domain is an integral domain).
Proof
The positive divisors of are exactly : a positive dividing with has a prime divisor by [L6], and forces by [L6], hence since is prime and (Prime and composite integers: is prime when and its only positive divisors are and , Divisibility in : when for some integer ); writing gives by cancellation, and repeating reduces to a power of not exceeding .
By [L1] at and at , using step 1.1, and ; dividing, .
With the elementary identity gives ; comparing with step 2.1 and cancelling the nonzero factor in the integral domain ([L7]) yields .
Reducing step 2.1 modulo and applying [L4] twice in gives , and cancelling in the integral domain ([L7]) gives .
Evaluating step 3.1 at gives , since the sum has terms each equal to ; so the constant term of is , which is divisible by and not by .
Substituting , which commutes with reduction modulo because both are ring homomorphisms fixing the coefficients appropriately, gives by [L5]. So every coefficient of other than the leading one is divisible by , while the leading coefficient is because is monic of degree by [L1] and the substitution being degree preserving.
is primitive by [L3], no prime dividing its leading coefficient , and its degree is at least by [L5]; steps 4.1 and 4.2 supply the three Eisenstein conditions at , so [L2] makes it irreducible in .
The substitution is a ring automorphism of carrying to ; a ring automorphism preserves units and factorisations, so it carries irreducible elements to irreducible elements, and is irreducible in .
Remarks
-
The term is load bearing. Dropping it would change from to , and the Eisenstein constant-term condition would fail. The highest exponent would be unchanged, so the degree alone would not detect the wrong polynomial.
-
A self-contained route for prime powers. This gives irreducibility over for a prime power without the general argument of is irreducible in for every , and unlike that argument it exhibits an explicit polynomial to which a named criterion applies. The general theorem covers every and does not supersede this computation; the companion page works out the case in is Eisenstein at seven ↗.
is irreducible in for every
Statement
For every the cyclotomic polynomial (The cyclotomic polynomials , defined by ) is irreducible in (Irreducible and prime elements of an integral domain).
Facts & Assumptions
Given: An integer ; is an ordered field (The rationals form a totally ordered field), so and in particular for , whence (The characteristic of a ring: the least with when one exists, and otherwise) and divides no (Divisibility in : when for some integer ); a splitting field of over (Every nonzero polynomial over a field has a splitting field); a primitive -th root of unity (The group of -th roots of unity in a field, and primitive -th roots of unity); and the minimal polynomial of over .
is separable over and is cyclic of order with exactly primitive -th roots of unity, which are its generators ( is separable over exactly when the characteristic does not divide , and then a splitting field carries distinct -th roots of unity, is cyclic of order dividing , and has a primitive -th root of unity exactly when its order is , The unit group and Euler's totient for ).
The roots of in are exactly the primitive -th roots of unity in ; is monic of degree (Over a field whose characteristic does not divide , the roots of are exactly the primitive roots of unity, The recursion defines a unique monic , of degree , Degree, leading coefficient and monic polynomial, with the zero polynomial having no degree).
For algebraic over there is a unique monic irreducible with if and only if (The evaluation kernel and the unique monic irreducible minimal polynomial of an algebraic element).
If is a primitive -th root of unity in , is its minimal polynomial and is a prime with , then (If is a prime not dividing , a rational minimal polynomial of a primitive -th root of unity also kills its -th power).
Every integer is a finite product of primes, the empty product being (The fundamental theorem of arithmetic: every integer is a product of primes, and the factorisation is unique up to order — if with every and prime, then and for some , Prime and composite integers: is prime when and its only positive divisors are and ).
In a cyclic group of finite order , generates the group if and only if , that is when and are coprime (A cyclic group of order has exactly generators, Coprime integers: ).
A nonzero polynomial of degree over an integral domain has at most distinct roots in it (A nonzero polynomial of degree over an integral domain has at most distinct roots).
Proof
is a root of by [L2], so in by [L3], and is monic and irreducible.
Let be any primitive -th root of unity. By [L1] it generates , so for some integer , which may be taken with after adding a multiple of ; and by [L6]. Write as a product of primes by [L5], with when . No divides , since and .
Put and for , so that . By induction on , each is a primitive -th root of unity and : at this is the hypothesis on and step 1.1; and given it at , the polynomial is monic irreducible and vanishes at the primitive -th root of unity , so is the minimal polynomial of by [L3], whence by [L4], while is primitive by [L6] because .
So vanishes at every primitive -th root of unity in , of which there are distinct ones by [L1]; hence by [L7].
On the other hand with by [L2], so ; therefore , and and are monic with , so is irreducible. At this reads , of degree .
Remarks
-
What the prime factorisation is doing. The lemma advances one prime at a time, and step 2.1 chains those advances along a factorisation of the exponent. The chain works only because each intermediate is again primitive and has the same minimal polynomial, which is what lets the lemma be reapplied rather than merely applied once.
-
Over other base fields the theorem is false. Irreducibility of depends on the base field, and over a finite field it usually fails (For the reduction of in is a product of distinct monic irreducibles, each of degree the order of modulo ). The criterion that separates the cases is is irreducible over exactly when , exactly when the embedding into is onto.
and
Statement
Let and let be a primitive -th root of unity (The group of -th roots of unity in a field, and primitive -th roots of unity) in (The cyclotomic extension as a splitting field of ). Then
(The degree of a finite field extension, The unit group and Euler's totient for ), and the embedding
of is Galois and embeds its Galois group into is an isomorphism.
Facts & Assumptions
Given: An integer ; is an ordered field (The rationals form a totally ordered field), so (The characteristic of a ring: the least with when one exists, and otherwise) and divides no (Divisibility in : when for some integer ).
is irreducible in for every ( is irreducible in for every ).
For a field with and a primitive -th root of unity in a splitting field, irreducibility of the image of in , the equality , and surjectivity of the embedding are equivalent ( is irreducible over exactly when , exactly when the embedding into is onto).
Proof
Since does not divide , [L2] applies with .
The image of in is itself, irreducible by [L1]; so the first clause of [L2] holds, and therefore so do the other two: , and the embedding is onto, hence an isomorphism, being injective.
Remarks
- The isomorphism is canonical. It sends to the class of the exponent with , and is Galois and embeds its Galois group into shows that class does not depend on which primitive -th root of unity is chosen. So every subgroup of names an intermediate field of without any choice being made.
For the reduction of in is a product of distinct monic irreducibles, each of degree the order of modulo
Statement
Let be a finite field of order and let with (Coprime integers: ). Put , the multiplicative order of in (The order of a finite group and the order of an element, with when no positive power of is the identity, The unit group and Euler's totient for ). Then the image of in (The cyclotomic polynomials , defined by ) is a product of pairwise distinct monic irreducible polynomials, each of degree , and there are of them.
Facts & Assumptions
Given: A finite field of order , of characteristic with a power of (Every finite field has order for a unique prime characteristic and positive integer , The characteristic of a ring: the least with when one exists, and otherwise), and with ; hence (Divisibility in : when for some integer ). Write (The cyclotomic extension as a splitting field of , Finitely generated field extensions ) and for the image of in .
Over a field with and a splitting field of , the image of in is separable, splits over , and its roots in are exactly the primitive -th roots of unity in (Over a field whose characteristic does not divide , the roots of are exactly the primitive roots of unity, The group of -th roots of unity in a field, and primitive -th roots of unity).
is monic of degree , and reduction into preserves both (The recursion defines a unique monic , of degree , Degree, leading coefficient and monic polynomial, with the zero polynomial having no degree).
For monic irreducible of degree with a root in an extension field, is a finite field of order and (A monic irreducible of degree over has the distinct roots ).
is a unique factorisation domain for every field (For every field , is a unique factorisation domain); irreducible elements are as in Irreducible and prime elements of an integral domain.
is separable over when no extension field contains an with dividing the image of (Repeated roots in extension fields and separable polynomials).
Over an integral domain, for nonzero (Over an integral domain, degrees add under multiplication of nonzero polynomials).
In any field , the group is cyclic of order dividing ; if it contains a primitive -th root of unity, then its order is , and in that case its generators are exactly the primitive -th roots of unity ( is cyclic of order dividing , and has a primitive -th root of unity exactly when its order is ).
Proof
Since is a power of and , the prime does not divide ; so [L1] applies with and the splitting field , and [L3] gives .
By [L2] the polynomial is monic of degree , so by [L5] it is a product of monic irreducible polynomials of , say with each monic irreducible.
Each has a root in . By [L1] the polynomial splits over into distinct linear factors, and divides it; since is a unique factorisation domain by [L5], is, up to a unit, a product of some of those linear factors. Thus it has a root , and is a primitive -th root of unity by [L1].
No two of the coincide. If for , then divides . By step 2.1 the polynomial has a linear factor in , so divides there, contradicting the separability supplied by [L1] and [L6].
Every has degree . Writing , [L4] gives . Since is primitive, [L8] makes a cyclic group of order with generators exactly the primitive -th roots, so generates . Hence and therefore ; and gives . Thus and by step 1.1.
Comparing degrees with [L7] and [L2], , so ; with steps 3.1 and 3.2 this is the assertion.
Remarks
- The degree of every factor is the same, and that is the content. A polynomial can factor into irreducibles of different degrees; here it cannot, because adjoining any primitive root produces the same field . The primitive roots need not form a single Frobenius orbit: for over they split into two orbits of size three, one for each irreducible cubic factor on the companion page.
The reduction of is irreducible over exactly when generates
Statement
Let be a finite field of order and with (Coprime integers: ). The image of in (The cyclotomic polynomials , defined by ) is irreducible (Irreducible and prime elements of an integral domain) if and only if generates (The subgroup generated by a subset, the cyclic subgroup , and cyclic groups, The unit group and Euler's totient for ). In particular this can happen only when is cyclic.
Facts & Assumptions
Given: A finite field of order and with ; write (The order of a finite group and the order of an element, with when no positive power of is the identity) and for the image of in .
is a product of pairwise distinct monic irreducible polynomials, each of degree , and there are of them (For the reduction of in is a product of distinct monic irreducibles, each of degree the order of modulo ).
(The unit group and Euler's totient for ), and for an element of finite order (If then iff is an integer multiple of , the powers are distinct, and has exactly elements; if has infinite order then only for ).
Proof
By [L1] the number of monic irreducible factors of , counted without repetition and with none repeated, is .
If then is itself one of those monic irreducible polynomials, hence irreducible; if then is a product of polynomials each of degree , none of them a unit, so it is not irreducible. Hence is irreducible exactly when , that is exactly when .
By [L3] the subgroup has order and has order , so holds exactly when , that is exactly when generates the unit group. With step 2.1 this proves the equivalence, and a group with a generator is cyclic.
Remarks
- When the criterion cannot be met at all. If is not cyclic then no class generates it, so the reduction of is reducible over every finite field of order coprime to ; the smallest such is , where has three elements of order two and no element of order four.
For finite Galois and finite inside a common field,
Statement
Let be a finite Galois extension (Finite Galois extensions and ) and a finite extension (The degree of a finite field extension), both subfields of a common field. Then the compositum is finite over and
Only one of the two extensions is required to be Galois.
Facts & Assumptions
Given: Subfields and of a common field, both containing , with finite Galois and finite; is a subfield containing and contained in , hence an intermediate field of .
For finite Galois and an extension inside a common overfield, is finite Galois and restriction gives (The Galois translation theorem).
If is finite Galois and , then is finite Galois (A finite Galois extension is Galois over every intermediate field).
For fields with and finite, is finite and (Tower law for finite extensions: ).
For finite subextensions and of a common field, the compositum is finite and (For finite subextensions in a common field, ).
For a finite Galois extension one has (Equivalent characterizations of a finite Galois extension).
Proof
is an intermediate field of , so is finite Galois by [L2].
By [L4] the compositum is finite over .
Applying [L3] to gives , so .
By [L1] the extension is finite Galois with , so [L5] applied to both sides gives .
Applying [L3] to and substituting steps 2.1 and 1.3 gives .
Remarks
- The Galois hypothesis is not decoration. Without it the formula fails: over take and inside a splitting field of , where is a primitive cube root of unity. Both have degree three over , since is irreducible there. The compositum contains , hence contains the splitting field and equals it, so . And : its degree over divides by the tower law, and it cannot be , since would put in and force . The formula would predict . Neither nor is Galois over .
Statement
Let be a field and let be integers such that divides neither nor (The characteristic of a ring: the least with when one exists, and otherwise, Divisibility in : when for some integer ). Put (Common multiple, and the least common multiple , taken to be when or ) and let be a splitting field of over (Every nonzero polynomial over a field has a splitting field). Then ; the subfields and of are cyclotomic extensions of of orders and (The cyclotomic extension as a splitting field of ); and their compositum inside is
Facts & Assumptions
Given: A field , integers with dividing neither, , and a splitting field of over ; the characteristic of a field is or a prime (The characteristic of a field is zero or a prime number), and divides no positive integer.
is a common multiple of and (Common multiple, and the least common multiple , taken to be when or ); every common multiple of and is a multiple of , and (Every common multiple of and is a multiple of , and ).
If a prime divides then or (Euclid's lemma: if is prime and then or , Prime and composite integers: is prime when and its only positive divisors are and ).
is separable over exactly when , and then a splitting field has cyclic of order and for any primitive -th root of unity ( is separable over exactly when the characteristic does not divide , and then a splitting field carries distinct -th roots of unity, is cyclic of order dividing , and has a primitive -th root of unity exactly when its order is , The group of -th roots of unity in a field, and primitive -th roots of unity).
A polynomial is separable over when no extension field contains a repeated root (Repeated roots in extension fields and separable polynomials); a splitting field is generated over by the roots (Polynomials that split and splitting fields of a polynomial or a family of polynomials, Finitely generated field extensions ).
For a finite group and , divides (Lagrange's theorem: for every subgroup of a finite group ); is the smallest subgroup containing (The subgroup generated by a subset, the cyclic subgroup , and cyclic groups); and for an element of finite order (If then iff is an integer multiple of , the powers are distinct, and has exactly elements; if has infinite order then only for , The order of a finite group and the order of an element, with when no positive power of is the identity).
Proof
. If this is immediate; if is a prime dividing , then divides because by [L1] and , so and [L2] gives or , contrary to hypothesis.
By [L3] and step 1.1 the polynomial is separable over , the group is cyclic of order , and .
For every positive divisor of one has and is a splitting field of over , hence a cyclotomic extension of of order : indeed divides , since , so it splits over , and a repeated root of it would be a repeated root of , excluded by step 2.1 through [L4]; so its roots are distinct and they are the elements of , which generate over .
and are positive divisors of by [L1], so step 3.1 applies to both: and are cyclotomic extensions of of orders and , and each contains a primitive root of unity of its order by [L3].
Both are contained in , since and are subsets of by and ; hence their compositum inside is contained in .
For the reverse inclusion, fix a primitive -th root of unity and a primitive -th root of unity , and let . By [L5] the orders and both divide , so is a common multiple of and and therefore a multiple of by [L1]; and divides by [L5]. Hence and .
Both and lie in the compositum , which is a field, so and therefore by step 5.2; hence . With step 5.1 this gives .
Remarks
- The intersection is not the mirror image of this. The compositum identity holds over every base field of admissible characteristic, but the corresponding identity does not: it is proved here only over (), and the companion page gives a finite-base-field failure in is larger than although five and seven are coprime ↗.
Statement
For all integers ,
(The unit group and Euler's totient for , Common divisor, and the greatest common divisor , with the convention , Common multiple, and the least common multiple , taken to be when or ). At both sides are ; at the identity is the multiplicativity .
Facts & Assumptions
Given: Integers ; write and , both because and are nonzero. For a prime (Prime and composite integers: is prime when and its only positive divisors are and ) and an integer put when and . For write for the set of primes dividing (Divisibility in : when for some integer ); it is finite, being contained in by If and then and ; hence the set of divisors of a nonzero integer is bounded above by . Put . Finite products over such sets are those of The sum over a finite index set, and its product form.
Let and let be an injective finite list consisting exactly of the prime divisors of ; put , so . Then (Euler's product formula for , stated through a finite injective list of its prime divisors).
For a prime and a nonzero integer , is the greatest with (The -adic valuation of a nonzero integer: the greatest with ).
Proof
For a prime and an integer : if and only if . If then divides , which divides by [L3]; conversely says , so the greatest such exponent is at least .
For each write and ; then and by [L2], and the unordered pair is , so .
Consequently and : by [L2] and step 1.1, says , that is and ; and says , that is or .
The set is a finite set of primes containing , , and by step 2.1. For every one has : applying [L1] with the list gives , and for outside step 1.1 gives , so the extra factors are .
Multiplying the equalities of step 1.2 over the finite set and using step 3.1 four times gives .
Remarks
- Why the identity is not simply multiplicativity. For coprime and it reduces to , but the general case is what the intersection theorem needs: the degrees of and multiply to the degree of the compositum times the degree of the intersection, and it is the gcd–lcm form of the identity that turns that into ().
Statement
Let , put and (Common divisor, and the greatest common divisor , with the convention , Common multiple, and the least common multiple , taken to be when or ), and let be a splitting field of over (Every nonzero polynomial over a field has a splitting field), inside which the cyclotomic extensions for are taken (The cyclotomic extension as a splitting field of ). Then
Facts & Assumptions
Given: Integers with and ; is an ordered field (The rationals form a totally ordered field), so (The characteristic of a ring: the least with when one exists, and otherwise) and divides no positive integer (Divisibility in : when for some integer ); a splitting field of over ; and, for each positive divisor of , the subfield of . Write .
For a positive divisor of , the subfield generated by the -th roots of unity in is a cyclotomic extension of of order (The cyclotomic extension as a splitting field of , The group of -th roots of unity in a field, and primitive -th roots of unity); because divides no positive integer, is separable over exactly when the characteristic does not divide , and then a splitting field carries distinct -th roots of unity gives cyclic of order , with exactly primitive -th roots of unity.
for every ( and , The unit group and Euler's totient for , The degree of a finite field extension).
is finite Galois ( is Galois and embeds its Galois group into ).
For finite Galois and finite inside a common field, (For finite Galois and finite inside a common field, ).
For fields with and finite, (Tower law for finite extensions: ).
Proof
divides and , and all divide . For each positive divisor of , write ; then , so splits over the splitting field of , and [L1] applies to . In particular all four cyclotomic extensions for sit inside .
For the inclusion : since , every with satisfies , so and hence ; the same argument with gives .
By [L3] the extension is finite Galois and is finite, both inside , so [L4] and [L5] give , using [L2] twice.
Hence by [L6].
By step 2.1 the tower is defined, and [L7] with [L2] gives , so and .
Remarks
-
Where the base field is used. Only through [L2]: the equality for every , which is irreducibility of over . Over a base field where some becomes reducible the degrees drop unevenly and the degree count in step 2.2 no longer forces the intersection down to .
-
The base field really matters. Over a general base field the same formula can fail; the companion page gives a finite-field witness in is larger than although five and seven are coprime ↗.
For an odd prime , has exactly one intermediate field of degree two over
Statement
Let be an odd prime (Prime and composite integers: is prime when and its only positive divisors are and ) and let be a primitive -th root of unity (The group of -th roots of unity in a field, and primitive -th roots of unity) in (The cyclotomic extension as a splitting field of ). Then there is exactly one intermediate field with
(The degree of a finite field extension).
Intermediate, not proper. At the Galois group has order two, the unique subgroup of index two is the trivial one, and the field it names is itself. Reading the statement as "proper subfield" would make it false at the smallest case in scope.
Facts & Assumptions
Given: An odd prime and a primitive -th root of unity in the cyclotomic extension ; write .
is finite Galois ( is Galois and embeds its Galois group into ) and the embedding is an isomorphism ( and ).
Every finite subgroup of the unit group of an integral domain is cyclic (Every finite subgroup of the unit group of an integral domain is cyclic); is a field (For every prime , the two operations on make it a field, Field), hence an integral domain, and is its group of units (The unit group and Euler's totient for ).
for a prime (, and for every prime ), and (The unit group and Euler's totient for ).
In a cyclic group of finite order there is exactly one subgroup of each order dividing , and every subgroup has that form (A finite cyclic group has exactly one subgroup of each order dividing its own).
For finite Galois with , the maps and are mutually inverse bijections between subgroups and intermediate fields, and (The fundamental theorem of finite Galois theory).
For a finite group and , (Lagrange's theorem: for every subgroup of a finite group ).
Proof
By [L1] the group is isomorphic to , which is cyclic by [L2] and has order by [L3]; so is cyclic of order .
Since is odd, is even, so divides and is a positive divisor of (Divisibility in : when for some integer ). By [L4] there is exactly one subgroup with .
By [L5] and [L6], an intermediate field of has for its corresponding subgroup , so holds exactly when .
The correspondence of [L5] is a bijection, so the intermediate fields of degree two over are in bijection with the subgroups of order , of which there is exactly one by step 2.1. Hence there is exactly one such field.
Remarks
-
Which field it is, is a different question. The argument counts intermediate fields; it produces no generator of the one it counts, and no claim is made here about identifying it. Naming that field concretely requires a computation this proof does not carry out.
-
Oddness is needed. At the field is itself, of degree , and it has no intermediate field of degree two at all; the step that fails is step 2.1, where is odd.
For every there are infinitely many primes with
Statement
For every integer , the set
(Prime and composite integers: is prime when and its only positive divisors are and , Congruence modulo an integer: when , including the moduli and ) is not finite (Finite, countably infinite, countable, uncountable).
Facts & Assumptions
Given: An integer and the cyclotomic polynomial (The cyclotomic polynomials , defined by ); evaluation at an integer is the ring homomorphism of Universal property of : a coefficient homomorphism and the image of determine a unique ring homomorphism and Evaluation and roots of a polynomial in a commutative target ring.
for every ( and for ).
Over a field with and a splitting field of over , the roots of the image of in are exactly the primitive -th roots of unity in , that is the elements of order in (Over a field whose characteristic does not divide , the roots of are exactly the primitive roots of unity, The group of -th roots of unity in a field, and primitive -th roots of unity, The order of a finite group and the order of an element, with when no positive power of is the identity).
is a field of characteristic for every prime (For every prime , the two operations on make it a field, The characteristic of a ring: the least with when one exists, and otherwise, The characteristic of a field is zero or a prime number, The congruence class and the quotient set ), and every nonzero polynomial over a field has a splitting field (Every nonzero polynomial over a field has a splitting field).
for a prime (, and for every prime , The unit group and Euler's totient for ); and for a finite group with , divides (Lagrange's theorem: for every subgroup of a finite group ), while (If then iff is an integer multiple of , the powers are distinct, and has exactly elements; if has infinite order then only for ).
Every integer greater than has a prime divisor (Every integer has a prime divisor; indeed the least divisor of that exceeds is prime, Divisibility in : when for some integer ).
A nonzero polynomial of degree over an integral domain has at most distinct roots in it (A nonzero polynomial of degree over an integral domain has at most distinct roots).
The set of primes is not finite (Euclid's theorem: for every and every list of primes there is a prime not among ; consequently the set of primes is not finite).
Proof
In the case , every integer is congruent to modulo , since divides every integer, so is the set of all primes, which is not finite by [L8].
In the case , let be any finite set and put , an integer with ; the goal is to produce a prime in outside .
There is an integer with : the three polynomials , and are nonzero of degree by [L1], so by [L7] over the integral domain at most integers satisfy ; the integers are pairwise distinct, so some with avoids that finite set.
Since by [L2], there is with ; evaluating at gives .
By step 2.1 the integer exceeds , so it has a prime divisor by [L6], and . That prime does not divide : otherwise would divide and hence , which is impossible for a prime. In particular , so and , both and every member of dividing .
Reduce modulo . Since , the class is a root of the image of in , and because by step 3.1. Let be a splitting field of over , which exists by [L4]; as does not divide , [L3] applies and , lying in , has order exactly in .
The order of in the subgroup of is the same , so divides by [L5]; that is , so and .
In the case , then, no finite subset of exhausts it, so is not finite; with step 1.1 the two cases are exhaustive and cover every .
Remarks
-
What replaces the archimedean estimate. The usual proof chooses large enough that by a growth estimate. Step 2.1 replaces that by a root count, which needs no order structure on beyond the distinctness of the multiples of , and gives exactly the same conclusion.
-
The case is not a degenerate instance. For one has , not ( and for ), so step 2.2 is unavailable; the congruence is vacuous there and the statement is Euclid's theorem.
Every finite abelian group is a quotient of for some and
Statement
For every finite abelian group there are positive integers and and a surjective group homomorphism
where denotes the set of -tuples of classes in , with componentwise addition, for the additive group (The congruence class and the quotient set , For every natural , is an abelian group, multiplication is a commutative monoid operation, and both distributive laws hold). Equivalently, for a subgroup (The quotient group and coset product , First isomorphism theorem for groups: ).
Facts & Assumptions
Given: A finite abelian group .
For every finite abelian group there is a unique list with ; the trivial group corresponds to the empty list (Fundamental theorem of finite abelian groups: invariant-factor form, Invariant-factor data for a finite abelian group).
A cyclic group of finite order is isomorphic to (Every cyclic group is isomorphic to or to for its finite order ).
For every natural number , one has exactly when (The congruence class and the quotient set , Congruence modulo an integer: when , including the moduli and , Divisibility in : when for some integer ), addition of classes is represented by addition of representatives, and is an abelian group (For every natural , is an abelian group, multiplication is a commutative monoid operation, and both distributive laws hold).
For a homomorphism , the rule is an isomorphism (First isomorphism theorem for groups: , The quotient group and coset product ).
Proof
In the case that the invariant-factor list of [L1] is empty, is trivial; take and , so that is a one-element group and the unique map to is a surjective homomorphism.
In the case , put and . Each divides , the list being a divisibility chain by [L1].
For each the rule is a well-defined surjective group homomorphism : if then , hence because , so by [L3]; it respects addition by [L3]; and every class is the image of .
Let and , both with componentwise addition. By [L3] each coordinate and is an abelian group, so and are abelian groups under these coordinatewise operations. Define by . Step 2.1 gives each coordinate map as a well-defined surjective homomorphism, so is a well-defined surjective group homomorphism. Fact [L2] identifies each coordinate group with , and [L1] identifies with up to isomorphism. Composing with that isomorphism gives a surjective homomorphism .
The two cases are exhaustive, the invariant-factor list being empty or not, so such and exist for every finite abelian ; and [L4] turns any such surjection into an isomorphism with its kernel.
Remarks
- Why the invariant-factor form is convenient. The invariant factors form a divisibility chain, so the single modulus works immediately. A primary decomposition also proves the statement: take to be the least common multiple of the finitely many prime-power orders and reduce onto each cyclic factor. The invariant-factor form simply avoids that extra choice of modulus.
Every finite abelian group is the Galois group of some finite Galois extension of
Statement
For every finite abelian group there is a finite Galois extension (Finite Galois extensions and ) with
and may be taken inside a cyclotomic field (The cyclotomic extension as a splitting field of ).
Facts & Assumptions
Given: A finite abelian group ; is an ordered field (The rationals form a totally ordered field), so (The characteristic of a ring: the least with when one exists, and otherwise) and divides no positive integer (Divisibility in : when for some integer ).
There are positive integers and a surjective group homomorphism (Every finite abelian group is a quotient of for some and , The external direct product with componentwise multiplication).
For a finite pairwise-coprime list of positive integers with , the map is a bijection preserving addition, multiplication, and (Chinese remainder theorem for a finite pairwise-coprime list: simultaneous residues determine one class modulo the product, and the resulting bijection preserves addition and multiplication, The congruence class and the quotient set , For every natural , is an abelian group, multiplication is a commutative monoid operation, and both distributive laws hold, Coprime integers: , Common divisor, and the greatest common divisor , with the convention ).
is a finite subgroup of the unit group of the field , hence cyclic (Every finite subgroup of the unit group of an integral domain is cyclic, For every prime , the two operations on make it a field, Field), of order (, and for every prime , The unit group and Euler's totient for ).
A cyclic group of finite order is isomorphic to (Every cyclic group is isomorphic to or to for its finite order ).
( and ), and this group is abelian (The Galois group of a cyclotomic extension is abelian).
For finite Galois with group and , the field is an intermediate field (The fundamental theorem of finite Galois theory); it is Galois over exactly when is normal (Normal subgroup: invariance under conjugation), and then (Normal subgroups, conjugate fields, and quotient groups in the Galois correspondence, The quotient group and coset product ).
For a homomorphism the rule is an isomorphism (First isomorphism theorem for groups: ).
Proof
Fix and a surjection by [L1].
Choose pairwise distinct primes with : the set of such primes is not finite by [L2], so at each of the steps one may pick a prime outside the finitely many already chosen. Put .
Distinct primes are coprime: a positive common divisor of and is or , and is or , so if it is not then . Hence is a pairwise-coprime list.
For each there is a surjective homomorphism : by [L4] the group is cyclic of order , which divides by step 2.1; [L5] identifies it with , and is well defined because , is a homomorphism, and is onto.
By [L3] the map is a bijection preserving multiplication and , so it carries units to units bijectively and restricts to a group isomorphism .
Taking the product of the maps of step 3.2 and composing with step 4.1 and with gives a surjective group homomorphism ; composing with the isomorphism of [L6] gives a surjective homomorphism .
Put and . The group is abelian by [L6], so every subgroup is normal, holding for all ; hence is an intermediate field, is finite Galois, and by [L7].
By [L8] applied to , which is surjective, ; hence , with inside .
Remarks
-
What is produced is a subfield, not a cyclotomic field. The construction realises as the Galois group of an intermediate field of , and it must: the Galois group of itself is , whose order is even for , so most finite abelian groups are not of that form. The companion page spells out that failure in FALSE: every finite abelian group is for some ↗.
-
The distinctness of the primes is needed twice. It makes the list pairwise coprime so that the Chinese remainder theorem applies, and it makes the product have exactly the intended unit group. Repeating a prime would collapse two factors into one.
Every intermediate field of is Galois over with abelian Galois group
Statement
Let and let be an intermediate field of (The cyclotomic extension as a splitting field of ). Then is a finite Galois extension (Finite Galois extensions and ) and is abelian.
Facts & Assumptions
Given: An integer and an intermediate field ; is an ordered field (The rationals form a totally ordered field), so (The characteristic of a ring: the least with when one exists, and otherwise) and divides no positive integer (Divisibility in : when for some integer ). Write .
is finite Galois ( is Galois and embeds its Galois group into ) with ( and ), and is abelian (The Galois group of a cyclotomic extension is abelian).
For finite Galois with group , the maps and are mutually inverse bijections between subgroups of and intermediate fields (The fundamental theorem of finite Galois theory).
With finite Galois, , and : the extension is Galois exactly when is normal in (Normal subgroup: invariance under conjugation), and then restriction gives (Normal subgroups, conjugate fields, and quotient groups in the Galois correspondence, The quotient group and coset product ).
Proof
By [L1] the extension is finite Galois and is abelian.
By [L2] there is a subgroup with .
Since is abelian, for every , so is normal in ; hence is Galois and by [L3].
A quotient of an abelian group is abelian, since the images of two commuting elements commute and every element of is such an image; so is abelian.
Remarks
- This is the proved half of the Kronecker–Weber picture on this page. Every subfield of a rational cyclotomic field is abelian over ; the converse is recorded separately as Recorded, not proved: every finite abelian extension of lies in a cyclotomic field ‡.
Recorded, not proved: every finite abelian extension of lies in a cyclotomic field
Statement
Kronecker–Weber theorem. Let be a finite Galois extension whose Galois group is abelian. Then there is an integer with
(The cyclotomic extension as a splitting field of ).
The statement fails over larger base fields. Over the extension obtained by adjoining a fourth root of is abelian over and is contained in no cyclotomic extension of .
Remarks
What this library does prove. The converse half is Every intermediate field of is Galois over with abelian Galois group: every intermediate field of is Galois over with abelian Galois group. Together with and , which computes exactly, that half says the subfields of cyclotomic fields are abelian; Kronecker–Weber says there are no others.
What would prove it, and which track that belongs to. The standard argument reduces the global statement to a local one at each prime that ramifies in , and then analyses the higher ramification groups of the -adic rationals to show that the local extension is contained in a local cyclotomic extension. Every ingredient of that reduction — valuations, local fields, the ring of integers of a number field, decomposition and inertia groups — belongs to algebraic number theory, and none of it is developed anywhere in this library. An alternative route through class field theory needs strictly more. Neither source consulted here proves the theorem: Conrad calls it deep and states it without proof, and Milne states it only in a footnote to an exercise.
Why it is recorded rather than omitted. The arithmetic consequences of the theorem are stated elsewhere in terms of it, so a page that builds the cyclotomic machinery and then says nothing about its sharpest classical application would leave that seam pointing at nothing. Recording it with the fuchsia not-proved-here marking is the honest form: the reader is told, at the point of contact, that the proof belongs to a later algebraic-number-theory track.
5 · Examples, counterexamples and false statements
is cyclic of order three with no proper intermediate field
Example
Let and let be the class of . Then is a field of order , the squaring map generates
cyclic of order three, its two orbits on are
and has no intermediate field other than and .
Facts & Assumptions
Given: The polynomial , the ring and the class of , so that because and in characteristic two.
A polynomial of degree or over a field is irreducible if and only if it has no root in that field (A polynomial of degree two or three over a field is irreducible exactly when it has no root in the field).
For a field and nonconstant , is irreducible if and only if is a field (For a nonconstant in , the ideal is maximal and is a field exactly when is irreducible).
If is algebraic over with minimal polynomial of degree , then has power basis and (A simple algebraic extension is its minimal-polynomial quotient and has power basis and degree , The degree of a finite field extension); a monic irreducible vanishing at is that minimal polynomial (The evaluation kernel and the unique monic irreducible minimal polynomial of an algebraic element).
An extension of finite fields of degree is Galois with cyclic of order , where , and (A finite extension of a finite field of order is Galois with cyclic Galois group generated by , The relative Frobenius of an extension of finite fields, For a degree- extension of a field of order , the -power map has order exactly , Finite fields and their order).
The intermediate fields of are the for the positive divisors of , one for each divisor (The intermediate fields of are the , one for each positive divisor of , Divisibility in : when for some integer ).
Verification
has no root in : and . By [L1] it is irreducible, so is a field by [L2].
is monic irreducible with , so it is the minimal polynomial of over and with power basis by [L3]; hence by [L4].
By [L4] the extension is Galois with Galois group generated by and of order three.
The orbit of : is ; ; and , using that squaring is additive in characteristic two. So is one orbit of size three.
The orbit of : , , and . So is the other orbit of size three, and together with and these account for all eight elements.
The positive divisors of three are and , so by [L5] the intermediate fields are exactly two: and itself. There is no field strictly between them.
Remarks
- Why the two nontrivial orbits have the same size. Each is an orbit of a group of prime order acting without fixed points outside : a fixed point of is an element with , and those are exactly the two elements of (The elements of a finite extension fixed by the -power map are exactly the base field).
The intermediate fields of match the divisors of twelve
Example
Let be a field with as a subfield and , so that . Its intermediate fields over are exactly
one for each of the six positive divisors of twelve, with exactly when divides . Neither of and contains the other, and
Facts & Assumptions
Given: A field with subfield and (The degree of a finite field extension); divisibility as in Divisibility in : when for some integer , with and as in Common divisor, and the greatest common divisor , with the convention and Common multiple, and the least common multiple , taken to be when or .
is Galois with cyclic Galois group generated by (A finite extension of a finite field of order is Galois with cyclic Galois group generated by ), and (For a degree- extension of a field of order , the -power map has order exactly , Finite fields and their order).
The intermediate fields of are exactly the for the positive divisors of , one for each divisor, with and if and only if (The intermediate fields of are the , one for each positive divisor of ).
A field of order has, for each positive divisor of , exactly one subfield of order , namely , and these are all of its subfields (The subfields of are the unique fields for positive divisors of ).
Every common divisor of two integers divides their greatest common divisor, and their least common multiple divides every common multiple (Every common divisor of and divides ; consequently exactly when , , , and every common divisor of and divides — a characterisation that holds at as well, Every common multiple of and is a multiple of , and ).
Verification
The positive divisors of twelve are , six in all, since a positive divisor of satisfies and direct inspection of leaves exactly these.
By [L1] and [L2] the intermediate fields of are the for those six , one for each, with exactly when .
Neither nor , so by step 2.1 neither of and contains the other.
Their intersection is an intermediate field of , being a subfield of containing , so it is for a unique divisor of by step 2.1; from and one gets and , so [L4] gives ; and and put inside both, so . Hence and the intersection is .
Their compositum is likewise an intermediate field , and it contains both, so and by step 2.1. Thus [L4] gives ; since , and the compositum is .
The same six fields are what [L3] produces for , whose order is : its subfields are the for the divisors of , and these are the sets named in [L2]. So the Galois indexing and the elementary one agree here.
Remarks
- Where the two descriptions coincide. Because the base field is the prime field, the divisors of and the divisors of are the same list; over a larger base field the two indexings differ by the factor , and The Galois description of the subfields of a finite field and the elementary divisibility criterion agree records the translation.
The divisor-sum identity at , finds exactly two monic irreducible cubics
Example
Over the divisor-sum identity ( for the counts of monic irreducibles of degree over ) at reads
and , so . The two monic irreducible cubics in are
Facts & Assumptions
Given: The field with two elements and the counts of monic irreducible polynomials of degree in (Degree, leading coefficient and monic polynomial, with the zero polynomial having no degree).
for every , the sum over the positive divisors of ( for the counts of monic irreducibles of degree over , Divisibility in : when for some integer ).
A polynomial of degree or over a field is irreducible if and only if it has no root in that field (A polynomial of degree two or three over a field is irreducible exactly when it has no root in the field).
For a commutative ring , and : if and only if divides (Factor theorem over a commutative ring, Evaluation and roots of a polynomial in a commutative target ring).
Verification
The monic polynomials of degree one in are and , and each is irreducible, having degree one; so .
A monic cubic over is with , so there are eight of them. Such an has no root in exactly when and , that is exactly when and .
The positive divisors of are and , so [L1] at and reads ; with step 1.1 this gives and .
The pairs with in are and , so exactly two monic cubics have no root in , namely and ; by [L2] these two are irreducible and by [L2] and [L3] the other six are not, each having a root and hence a linear factor. This agrees with the count of step 2.1.
Remarks
- The identity is a recursion, not a formula. It determines only because is already known; at it would read and would need first.
The four roots of over are the Frobenius powers of any one of them
Example
Let and let be the class of , so . Then is a field of order , and the four conjugates of over ,
are pairwise distinct, are exactly the roots of in , and satisfy , so the Frobenius orbit closes at length four.
Facts & Assumptions
Given: The polynomial , the ring , and the class of , so since in characteristic two; squaring is additive there.
A polynomial of degree or over a field is irreducible if and only if it has no root in that field (A polynomial of degree two or three over a field is irreducible exactly when it has no root in the field).
if and only if divides (Factor theorem over a commutative ring); and over an integral domain for nonzero (Over an integral domain, degrees add under multiplication of nonzero polynomials, Degree, leading coefficient and monic polynomial, with the zero polynomial having no degree).
For a field and nonconstant , is irreducible if and only if is a field (For a nonconstant in , the ideal is maximal and is a field exactly when is irreducible).
A monic irreducible vanishing at is the minimal polynomial of (The evaluation kernel and the unique monic irreducible minimal polynomial of an algebraic element), and then has power basis with its degree (A simple algebraic extension is its minimal-polynomial quotient and has power basis and degree , The degree of a finite field extension).
An extension of finite fields of degree over is Galois with cyclic Galois group generated by , and has elements (A finite extension of a finite field of order is Galois with cyclic Galois group generated by , The relative Frobenius of an extension of finite fields, For a degree- extension of a field of order , the -power map has order exactly ).
A monic irreducible of degree over with a root has the distinct roots and (A monic irreducible of degree over has the distinct roots ).
Verification
has no root in , since and ; so by [L2] it has no factor of degree one.
The only monic irreducible quadratic in is : the four monic quadratics are , , and , and the first three have the root , and respectively, so [L1] leaves only the last.
is irreducible. A factorisation of into two nonconstant factors has degrees summing to four by [L2], so it is either , excluded by step 1.1, or ; and every monic quadratic factor would have to be irreducible, hence equal to by step 1.2, giving , which is not .
By [L3] the ring is a field; is monic irreducible with , so with power basis by [L4], and by [L5].
Compute the conjugates in that basis: by hypothesis, and . So the four elements have coordinate lists , , and , which are pairwise different, so the four elements are pairwise distinct.
, so the orbit closes after four steps.
By [L6] applied to , of degree four, the elements are exactly the roots of and in ; steps 4.1 and 5.1 verify the distinctness and the closing of the orbit directly.
A normal basis of over
Example
Let , a field of order , let be the class of , and let generate . Put
Then the conjugate list
is a normal basis of over (Normal bases of a finite Galois extension).
Not every element works. The generator itself does not: its conjugate list is , whose three members sum to , so they are linearly dependent over and are not a basis.
Facts & Assumptions
Given: The field with the class of , so and ; squaring is additive in characteristic two.
has no root in , so it is irreducible (A polynomial of degree two or three over a field is irreducible exactly when it has no root in the field) and is a field (For a nonconstant in , the ideal is maximal and is a field exactly when is irreducible); it is the minimal polynomial of (The evaluation kernel and the unique monic irreducible minimal polynomial of an algebraic element), so with power basis (A simple algebraic extension is its minimal-polynomial quotient and has power basis and degree , The degree of a finite field extension).
A list of length is an ordered basis of if and only if every has exactly one coordinate list with (A finite list is an ordered basis if and only if every equals for exactly one ; those scalars are the coordinates of in that ordered basis, Basis of a vector space: a linearly independent spanning subset; and ordered basis: an injective finite list whose image is a basis).
For a linear map with finite-dimensional, (Rank-nullity: ).
Every finite Galois extension with cyclic Galois group has a normal basis (Every finite cyclic extension has a normal basis).
Verification
By [L1] and [L2] the space is a three-dimensional -vector space with basis , and acts by .
The conjugate list of is with ; hence . A vanishing combination with all coefficients is nontrivial, so this list is linearly dependent over and is not a basis.
The conjugates of are , and .
The seven nonzero -combinations of are nonzero: the three single terms are , and ; the three pairwise sums are , and ; and the total sum is . None of these seven is , as each has a nonzero coordinate list in the basis .
So the -linear map sending to has trivial kernel by step 4.1; both spaces have dimension three by step 1.1, so [L4] makes surjective as well, hence bijective, and [L3] makes an ordered basis of over .
That list is the family of conjugates of under by step 3.1, so it is a normal basis, as [L5] guarantees exists for this cyclic extension.
Remarks
- A conjugate family of the right size can still fail. The list has three distinct members and is a single Galois orbit, yet it is not a basis; what fails is independence, not the orbit condition. The normal basis theorem asserts that some element works, never that every element does (FALSE: every basis of a finite field over a subfield is a normal basis).
is a normal basis of while is not
Example
The extension (The complex numbers as , with the real embedding and imaginary unit ) is finite Galois of degree two with (Real and imaginary parts, complex conjugation, and modulus). For it:
- the conjugate list is a normal basis (Normal bases of a finite Galois extension);
- is an -basis of that is not a conjugate list of any element;
- is the conjugate list of but is not a basis.
The last two show that the two conditions in the definition of a normal basis are independent of each other.
Facts & Assumptions
Given: The complex field with and conjugation for (Real and imaginary parts, complex conjugation, and modulus); in one has .
is a simple algebraic extension with power basis and ( has power basis and degree , The degree of a finite field extension).
Every field automorphism of fixing pointwise is either the identity or complex conjugation, and these two are distinct (The only real-field automorphisms of are the identity and complex conjugation, Relative field automorphisms and ).
Complex conjugation is a real-field automorphism (Conjugation is an involutive real-field automorphism, , and modulus is definite, multiplicative, and subadditive).
A finite extension with is Galois exactly when (Equivalent characterizations of a finite Galois extension, Finite Galois extensions and ).
A list of length is an ordered basis of if and only if every has exactly one coordinate list with respect to it (A finite list is an ordered basis if and only if every equals for exactly one ; those scalars are the coordinates of in that ordered basis, Basis of a vector space: a linearly independent spanning subset; and ordered basis: an injective finite list whose image is a basis); and for a linear on a finite-dimensional (Rank-nullity: ).
Every finite Galois extension of an infinite field has a normal basis (Every finite Galois extension of an infinite field has a normal basis).
Verification
By [L2] and [L3] the group has exactly the two elements and conjugation, so its order is by [L1]; hence is finite Galois with that Galois group by [L4].
The conjugate list of is , whose members have coordinate lists and in the ordered basis of [L1]. For , vanishes exactly when and , hence when , that is and then .
is a basis by [L1], but no has conjugate list with underlying set : such a would lie in , and the set for is while for it is , neither of which is .
So the linear map sending to has trivial kernel; both spaces have dimension two by [L1], so [L5] makes it bijective and an ordered -basis of . Being the conjugate list of , it is a normal basis, in agreement with [L6].
is the conjugate list of , since , and its two members are distinct; but is a vanishing combination with nonzero coefficients, so the list is not independent and by [L5] is not a basis. With steps 3.1 and 2.2 this establishes all three claims.
FALSE: every basis of a finite field over a subfield is a normal basis
Statement
False claim. For every extension of finite fields, every -basis of is a normal basis (Normal bases of a finite Galois extension).
Facts & Assumptions
Given: The ring with the class of , so that because and in characteristic two.
A polynomial of degree or over a field is irreducible if and only if it has no root in that field (A polynomial of degree two or three over a field is irreducible exactly when it has no root in the field); and is a field exactly when is irreducible (For a nonconstant in , the ideal is maximal and is a field exactly when is irreducible).
A monic irreducible vanishing at is the minimal polynomial of (The evaluation kernel and the unique monic irreducible minimal polynomial of an algebraic element), and then has power basis with (A simple algebraic extension is its minimal-polynomial quotient and has power basis and degree , The degree of a finite field extension, Basis of a vector space: a linearly independent spanning subset; and ordered basis: an injective finite list whose image is a basis).
An extension of finite fields of degree over is Galois with cyclic Galois group generated by , of order (A finite extension of a finite field of order is Galois with cyclic Galois group generated by , The relative Frobenius of an extension of finite fields, For a degree- extension of a field of order , the -power map has order exactly ).
A normal basis of is an ordered -basis of the form for a single , indexed by (Normal bases of a finite Galois extension).
Every finite Galois extension has a normal basis (Every finite Galois extension has a normal basis).
Refutation
has no root in , its values at and both being , so it is irreducible and is a field by [L1]; it is the minimal polynomial of , so with ordered basis by [L2], and has four elements .
By [L3] the extension is Galois with , where .
The conjugate lists of the four elements are , , and , using and . Their underlying sets are , and .
The list is an -basis of by step 1.1, but its underlying set is none of the three sets in step 3.1, so it is not the conjugate list of any element and hence is not a normal basis by [L4]. The false claim therefore fails already for .
What is true is the existential statement: some element of generates a normal basis, and does, since is a list of two distinct elements whose only vanishing -combinations are trivial, as , and . That is the content of [L5], which asserts existence and never universality.
Remarks
- Where the false claim comes from. The normal basis theorem is an existence statement, and its proofs single out an element by a nonvanishing condition — a determinant in the infinite case, a cyclic vector in the finite case. Both conditions genuinely exclude some elements, as A normal basis of over shows over .
through computed from the divisor recursion
Example
Running the recursion of The cyclotomic polynomials , defined by gives
each monic in , with degrees
matching (The unit group and Euler's totient for ).
Facts & Assumptions
Given: The recursion and (The cyclotomic polynomials , defined by , The sum over a finite index set, and its product form, Divisibility in : when for some integer ); and the elementary identity for .
For every , is monic in with and (The recursion defines a unique monic , of degree , Degree, leading coefficient and monic polynomial, with the zero polynomial having no degree).
For a prime and , (, and is Eisenstein at ).
is an integral domain (A polynomial ring over an integral domain is an integral domain), so a nonzero factor may be cancelled; and division by a monic polynomial has a unique quotient and remainder (Division by a monic polynomial over a commutative ring).
and for a prime (, and for every prime ); (For a prime and , ); and for with prime divisors and , (Euler's product formula for , stated through a finite injective list of its prime divisors).
Verification
is the base clause of the recursion, of degree by [L4].
The prime powers among are , and [L2] gives their cyclotomic polynomials directly: , , , , , , and .
For : the positive divisors of are and those of are , so [L1] gives and ; dividing and using the given identity with , yields . Since and is nonzero, cancelling in by [L3] gives .
For : the divisors of are and those of are , so by [L1] and the given identity with , ; and , so cancelling gives .
For : the divisors of are and those of are , so by [L1] and the given identity with , ; and with , so cancelling gives .
The degrees read off the displayed polynomials are . By [L4] these are , , , , , , , , , , and , so every degree matches [L1].
Remarks
- Every division in the recursion is exact and stays over . That is not visible from the table and is not a coincidence of small : it is The recursion defines a unique monic , of degree , and it is what makes the recursion a definition rather than a computation that might fail.
is Eisenstein at seven
Example
The translated seventh cyclotomic polynomial is
Its leading coefficient is , every other coefficient is divisible by , and its constant term is not divisible by . So it satisfies Eisenstein's criterion at the prime , and therefore is irreducible over .
Facts & Assumptions
Given: The prime-power cyclotomic formula and the polynomial .
For a prime and , and is Eisenstein at (, and is Eisenstein at ).
Eisenstein's criterion: if a prime divides every non-leading coefficient of a polynomial in , does not divide the leading coefficient, and does not divide the constant term, then the polynomial is irreducible over (Eisenstein criterion over the integers).
Verification
Applying [L1] at and gives .
Therefore by the binomial theorem.
In the polynomial of step 2.1 the leading coefficient is , the remaining coefficients are all divisible by , and the constant term is not divisible by ; so [L2] applies at the prime .
Hence is irreducible over , and this is exactly the degree-one prime-power case of [L1].
Remarks
- Why this example matters later. The explicit coefficients are what the counterexample page uses when it says the Eisenstein route already proves irreducibility for prime-power cyclotomic polynomials before the general Dedekind argument is built.
has four roots in
Example
Over the fifth cyclotomic polynomial
splits into four distinct linear factors:
The four roots are exactly the primitive fifth roots of unity in .
Facts & Assumptions
Given: The field and the polynomial .
If , the reduction of in is a product of distinct monic irreducibles, each of degree the order of modulo (For the reduction of in is a product of distinct monic irreducibles, each of degree the order of modulo ).
For , the image of in is generated by , so the extension degree is the order of modulo (For the image of in is generated by ).
Verification
In one has , so the order of modulo is . Hence [L1] and [L2] say every irreducible factor of over is linear, and the factors are distinct.
The powers of in are , , and , so the four nontrivial fifth roots of unity in are .
Each of is therefore a root of and is not , so each is a root of . Since is monic of degree , it follows that
The roots all have multiplicative order by step 1.2, so they are exactly the primitive fifth roots of unity in .
Remarks
- This is the order-one case of the finite-field factorisation theorem. When has order one modulo , every irreducible factor has degree one and the whole cyclotomic polynomial splits over the base field.
factors over into the two monic irreducible cubics
Example
Over one has
These are the two monic irreducible cubic factors, and over a splitting field the two factors correspond to the Frobenius orbits
of a primitive seventh root of unity .
Facts & Assumptions
Given: The field and a primitive seventh root of unity in a splitting field of .
If , the reduction of in is a product of distinct monic irreducibles, each of degree the order of modulo (For the reduction of in is a product of distinct monic irreducibles, each of degree the order of modulo ).
For , the image of in is generated by (For the image of in is generated by ).
The roots of a monic irreducible polynomial of degree over are one Frobenius orbit (A monic irreducible of degree over has the distinct roots ).
Verification
In the class has order , since and , . So [L1] and [L2] say every irreducible factor of over is a distinct cubic.
The product of the two monic cubics is in . Since both factors are monic of degree , step 1.1 makes them the two irreducible factors of .
Frobenius acts by , so the orbit of is because , and the orbit of is because , and . These are the two size-three Frobenius orbits of primitive seventh roots, and [L3] identifies them as the respective root sets of the two irreducible cubic factors from step 2.1.
Remarks
- This is the degree-three case of the theorem, not a coincidence of cubics. The orbit size and the factor degree are both the order of modulo .
and its three quadratic subfields
Example
Let . Then
every nonidentity element has order two, and the three order-two subgroups have fixed fields
So has exactly three quadratic intermediate fields.
Facts & Assumptions
Given: A primitive twelfth root of unity and the automorphisms for .
and ( and ).
For a finite Galois extension , subgroups of correspond bijectively to intermediate fields, and the fixed field of a subgroup has degree (The fundamental theorem of finite Galois theory).
Verification
The units modulo are , since these are exactly the residue classes in coprime to . Their squares are , and , so every nonidentity element has order two.
Therefore is the Klein four-group, with three order-two subgroups: By [L1] and [L2], each fixed field has degree over .
The subgroup fixes , because . Since and the fixed field has degree by step 2.1, that fixed field is .
The subgroup fixes , because ; and since is a root of . So the fixed field contains , and again step 2.1 makes it exactly .
The subgroup fixes , because is complex conjugation. Moreover since . So the fixed field contains , and step 2.1 makes it exactly .
The three order-two subgroups of step 2.1 therefore yield the three quadratic intermediate fields , and , and there are no others because [L2] gives a bijection between subgroups and intermediate fields.
Remarks
- The same phenomenon already occurs at order eight. The field also has Klein four Galois group and three quadratic subfields. The order-twelve calculation is useful because its three fields are the familiar , and .
In characteristic three, and coincides with
Example
Let be a field of characteristic . Then
so has only the two distinct roots and rather than six.
Facts & Assumptions
Given: A field of characteristic .
For a fixed integer , in characteristic the only -th root of unity is , and (In characteristic the only -th root of unity is , and ).
Verification
Applying [L1] at and gives and .
Since and , one has . Conversely, if then , so and therefore or in the field ; hence .
If then , so by [L2]; step 1.1 gives , hence by [L2]. Thus .
Conversely, if then , so . Therefore .
Using step 1.1, so its only distinct roots are and , exactly the two elements of step 2.2.
Remarks
- This is why the characteristic hypothesis is load-bearing. The statement "" fails here for two different reasons at once: the polynomial is inseparable, and the extra cube roots never appear even after passing to a splitting field because the splitting field is already the base field.
is larger than although five and seven are coprime
Statement refuted
That the rational intersection theorem holds over every base field: that for every field and all positive integers with the characteristic of dividing neither,
The witness below takes , and , and realizes both splitting fields inside one fixed field of order . Since , the right-hand side is , while the intersection on the left is the common subfield .
Facts & Assumptions
Given: The base field and a field of order , which exists by For every prime and , a field with elements exists. The two cyclotomic splitting fields will be identified with their base-field-isomorphic copies inside .
For , the image of in is generated by , so the degree of is the order of modulo (For the image of in is generated by ).
The intermediate fields of are exactly the for the positive divisors of , one for each divisor, with exactly when (The intermediate fields of are the , one for each positive divisor of ).
is the splitting field of over (The cyclotomic extension as a splitting field of ).
Finite fields of the same order are isomorphic by an isomorphism fixing their common prime field (Finite fields of the same order are isomorphic).
Counterexample
In , the class has order , since and no smaller positive power of is congruent to modulo . So [L1] gives , hence this splitting field has order and [L5] lets us identify it over with the unique subfield of .
In , the powers of are , so has order . Thus [L1] gives , hence this splitting field has order and [L5] lets us identify it over with the unique subfield of .
Under the fixed identifications of steps 1.1 and 1.2, both and are subfields of by [L2], since and . Their intersection is then an intermediate field of , so by [L2] it is for some divisor of . Because the intersection lies in both fields, [L2] gives and , hence ; and since lies in both fields, [L2] gives . Therefore and
Since , the right-hand side of the refuted identity is , which is just because already splits over . So the claimed equality would read , which is false.
The refuted statement therefore fails over the base field , even though the rational theorem [L3] is true.
Remarks
- Why the rational hypothesis matters. Over finite fields the intersection is controlled by the gcd of the extension degrees, not by the gcd of the orders of the roots of unity.
FALSE: every cyclotomic polynomial has all coefficients in
Statement
False claim. Every coefficient of every cyclotomic polynomial lies in .
The failure first appears at : the coefficient of in is .
Facts & Assumptions
Given: The cyclotomic recursion for (The cyclotomic polynomials , defined by ).
For every , is a monic polynomial in and the displayed divisor recursion holds (The recursion defines a unique monic , of degree ).
Refutation
Running the divisor recursion for and truncating modulo gives The first congruence is the defining recursion with every factor of degree at least dropped modulo , and the second comes from expanding .
Step 1.1 shows that the coefficient of in is , and . So the false claim fails.
Remarks
- Why this is a real pattern and not a silly claim. For many small values of the coefficients do lie in , so the first counterexample is not visually obvious from the recursion alone.
FALSE: is irreducible over every field
Statement
False claim. For every field and every with the characteristic of not dividing , the image of in is irreducible.
What is true is different in two directions: over every is irreducible, but over a finite field the factor degree is governed by the order of the Frobenius class modulo .
Facts & Assumptions
Given: The rational irreducibility theorem and the finite-field factorisation theorem.
For every , the cyclotomic polynomial is irreducible in ( is irreducible in for every ).
If , the reduction of in is a product of distinct monic irreducibles, each of degree the order of modulo (For the reduction of in is a product of distinct monic irreducibles, each of degree the order of modulo ).
Refutation
In one has , so the order of modulo is . Therefore [L2] says that over every irreducible factor of has degree .
Hence the reduction of in is a product of distinct linear factors, so it is reducible there. This contradicts the false claim.
The contradiction does not touch [L1]: irreducibility over is a theorem, but it does not persist over arbitrary base fields.
Remarks
- The finite-field theorem is the correct replacement. The question over is not "irreducible or not?" in the abstract, but "what is the order of modulo ?"
FALSE: has elements in every field
Statement
False claim. For every field and every , the group of -th roots of unity in has exactly elements.
The witness below shows two different failures: over there is no primitive cube root of unity in the field at all, while in characteristic the equation is inseparable and has only one root.
Facts & Assumptions
Given: The groups of roots of unity.
is cyclic of order dividing , and it has a primitive -th root of unity exactly when its order is ( is cyclic of order dividing , and has a primitive -th root of unity exactly when its order is ).
( and ).
In characteristic the only -th root of unity is (In characteristic the only -th root of unity is , and ).
Refutation
If had three elements, then [L1] would give a primitive cube root of unity in . But then , contradicting [L2], which says this extension has degree . So does not have three elements.
If has characteristic , then [L3] gives , so again does not have three elements.
The false claim fails already at , both over and over every field of characteristic .
Remarks
- The two failures have different causes. Over the polynomial is separable but its nontrivial roots lie in a quadratic extension; in characteristic the polynomial itself collapses to .
FALSE: every finite abelian group is for some
Statement
False claim. For every finite abelian group there is an with
The obstruction is visible already at the level of cardinality: the cyclic group occurs as a Galois group over , but never as the Galois group of a cyclotomic field.
Facts & Assumptions
Given: Cyclotomic Galois groups and the theorem realising finite abelian groups over .
, so its order is ( and , The unit group and Euler's totient for ).
Every finite abelian group is the Galois group of some finite Galois extension of (Every finite abelian group is the Galois group of some finite Galois extension of ).
In a finite group, the order of every subgroup divides the order of the group (Lagrange's theorem: for every subgroup of a finite group ).
Refutation
For and , the group is trivial, so .
If , then the unit class in has order : one has , and because otherwise would divide , contrary to . Therefore [L3] makes the group order even.
Steps 1.1 and 1.2 show that is never . So no cyclotomic field has Galois group of order , and in particular none has Galois group isomorphic to .
By [L2], however, some finite Galois extension of does have Galois group . Hence the false claim fails.
Remarks
- What the true theorem says instead. The proved result is that every finite abelian group is the Galois group of a subfield of a cyclotomic field, not of the cyclotomic field itself.
A degree-three Galois extension of inside
Example
Let . Then the fixed field of the unique order-two subgroup of is
a degree-three Galois extension of with cyclic Galois group, and the element has minimal polynomial
Facts & Assumptions
Given: A primitive seventh root of unity .
and has order ( and ).
A finite cyclic group has exactly one subgroup of each order dividing its own (A finite cyclic group has exactly one subgroup of each order dividing its own).
For a finite Galois extension, subgroups correspond to intermediate fields, and the fixed field of a subgroup has degree equal to the subgroup index (The fundamental theorem of finite Galois theory).
Every finite subgroup of the unit group of an integral domain is cyclic (Every finite subgroup of the unit group of an integral domain is cyclic).
Under the finite Galois correspondence, a normal subgroup has a Galois fixed field and restriction gives (Normal subgroups, conjugate fields, and quotient groups in the Galois correspondence).
Every finite group of prime order is cyclic (A finite group of prime order is cyclic and every nonidentity element generates it).
Verification
By [L1] the Galois group of is isomorphic to the finite subgroup of the unit group of the field , so [L4] makes it cyclic; its order is . Thus [L2] gives a unique subgroup of order , and [L3] makes its fixed field have degree .
The subgroup is generated by the class , so it acts by complex conjugation. Therefore is fixed by and lies in .
Put . Then Since , dividing by gives Substituting the expressions above yields
The element is not rational: if it were, then would satisfy the quadratic polynomial , which would force , contradicting [L1]. Since , the prime degree from step 1.1 leaves only the subfields and , so . Therefore the minimal polynomial of has degree , and the cubic from step 3.1 is that minimal polynomial.
The ambient Galois group is cyclic and hence abelian, so is normal. By [L5], the fixed field is Galois and is isomorphic to the quotient by , which has order . The group is cyclic by [L6], so is a cyclic cubic Galois extension.
Remarks
- This is the smallest nontrivial case of the subfield theorem. The subgroup lattice of has one index-two subgroup, and the fixed field is already visible through the real element .
Sources
- J. S. Milne, Fields and Galois Theory, v5.10, Corollary 4.21 and its proof
- K. Conrad, Finite Fields (expository blurb), Theorem 5.2
- K. Conrad, Cyclotomic Extensions (expository blurb), Section 1
- P. L. Clark, Field Theory (course notes/monograph), Chapter 9, Section 1
- K. Conrad, Finite Fields (expository blurb), Section 5
- J. S. Milne, Fields and Galois Theory, v5.10, Chapter 4, Finite fields
- K. Conrad, Roots and Irreducibles (expository blurb), Lemma 4.2
- K. Conrad, Finite Fields (expository blurb), Theorem 5.6
- J. S. Milne, Fields and Galois Theory, v5.10, Proposition 4.20
- J. S. Milne, Fields and Galois Theory, v5.10, Proposition 4.23 and Corollary 4.21
- J. S. Milne, Fields and Galois Theory, v5.10, Corollary 4.21
- K. Conrad, Finite Fields (expository blurb), Theorem 5.2 and Example 2.9
- K. Conrad, Finite Fields (expository blurb), Theorem 5.5
- K. Conrad, Roots and Irreducibles (expository blurb), Theorem 5.4
- K. Conrad, Roots and Irreducibles (expository blurb), formula (6.2) and Example 6.2
- K. Conrad, Finite Fields (expository blurb), Section 6
- J. S. Milne, Fields and Galois Theory, v5.10, Definition 5.17
- P. L. Clark, Field Theory (course notes/monograph), Chapter 8, Section 5
- P. L. Clark, Field Theory (course notes/monograph), Lemma 8.21
- J. S. Milne, Fields and Galois Theory, v5.10, Lemma 5.19
- P. L. Clark, Field Theory (course notes/monograph), Lemma 8.23
- J. S. Milne, Fields and Galois Theory, v5.10, Corollary 5.16
- P. L. Clark, Field Theory (course notes/monograph), Theorem 8.24
- J. S. Milne, Fields and Galois Theory, v5.10, Lemma 5.19 and the normal basis theorem
- P. L. Clark, Field Theory (course notes/monograph), Theorem 8.25
- K. Conrad, Linear Independence of Characters (expository blurb), Theorem 3.6
- P. L. Clark, Field Theory (course notes/monograph), Proposition 8.22
- K. Conrad, Linear Independence of Characters (expository blurb), Theorem 3.7
- J. S. Milne, Fields and Galois Theory, v5.10, Theorem 5.18
- P. L. Clark, Field Theory (course notes/monograph), Section 9.1.1
- K. Conrad, Cyclotomic Extensions (expository blurb), Theorem 1.1
- P. L. Clark, Field Theory (course notes/monograph), Lemmas 9.1 and 9.2
- P. L. Clark, Field Theory (course notes/monograph), Proposition 9.4
- K. Conrad, Cyclotomic Extensions (expository blurb), Lemma 2.1 and Theorem 2.3
- J. S. Milne, Fields and Galois Theory, v5.10, Proposition 5.8
- P. L. Clark, Field Theory (course notes/monograph), Proposition 9.5
- K. Conrad, Cyclotomic Extensions (expository blurb), Section 2
- K. Conrad, Cyclotomic Extensions (expository blurb), Theorem 2.10
- K. Conrad, Cyclotomic Extensions (expository blurb), Section 5
- P. L. Clark, Field Theory (course notes/monograph), Section 9.1.2
- K. Conrad, Cyclotomic Extensions (expository blurb), Theorem 5.2
- P. L. Clark, Field Theory (course notes/monograph), Proposition 9.6
- P. L. Clark, Field Theory (course notes/monograph), Exercise 9.8
- P. L. Clark, Field Theory (course notes/monograph), Theorem 9.7
- J. S. Milne, Fields and Galois Theory, v5.10, Lemma 5.9
- K. Conrad, Cyclotomic Extensions (expository blurb), Theorem 5.3
- J. S. Milne, Fields and Galois Theory, v5.10, Lemma 1.42
- J. S. Milne, Fields and Galois Theory, v5.10, Theorem 5.10
- K. Conrad, Cyclotomic Extensions (expository blurb), Theorem 2.5
- P. L. Clark, Field Theory (course notes/monograph), Theorem 9.8
- P. L. Clark, Field Theory (course notes/monograph), Corollary 9.9
- K. Conrad, Cyclotomic Extensions (expository blurb), Theorem 5.4
- K. Conrad, Cyclotomic Extensions (expository blurb), Corollary 5.7
- K. Conrad, Cyclotomic Extensions (expository blurb), Section 3
- J. S. Milne, Fields and Galois Theory, v5.10, Chapter 3, composita and translation
- K. Conrad, Cyclotomic Extensions (expository blurb), formula (4.1) in Section 4
- K. Conrad, Cyclotomic Extensions (expository blurb), Theorem 3.4
- P. L. Clark, Field Theory (course notes/monograph), Exercise 9.10
- K. Conrad, Cyclotomic Extensions (expository blurb), Section 4
- J. S. Milne, Fields and Galois Theory, v5.10, Chapter 5, cyclotomic extensions
- P. L. Clark, Field Theory (course notes/monograph), Theorem 9.10
- P. L. Clark, Field Theory (course notes/monograph), Lemma 9.11
- P. L. Clark, Field Theory (course notes/monograph), Corollary 9.12
- K. Conrad, Cyclotomic Extensions (expository blurb), Remark 2.7
- J. S. Milne, Fields and Galois Theory, v5.10, footnote to Exercise 3-2
- K. Conrad, Finite Fields (expository blurb), Examples 4.3-4.5
- K. Conrad, Finite Fields (expository blurb), Example 2.9
- K. Conrad, Roots and Irreducibles (expository blurb), Example 6.2
- K. Conrad, Finite Fields (expository blurb), Example 2.10
- K. Conrad, Linear Independence of Characters (expository blurb), Section 3
- J. S. Milne, Fields and Galois Theory, v5.10, the normal basis theorem
- K. Conrad, Linear Independence of Characters (expository blurb), Examples 3.1-3.3
- K. Conrad, Linear Independence of Characters (expository blurb), Example 3.1
- J. S. Milne, Fields and Galois Theory, v5.10, Definition 5.17 and the normal basis theorem
- K. Conrad, Cyclotomic Extensions (expository blurb), Example 5.1
- K. Conrad, Cyclotomic Extensions (expository blurb), Example 2.2
- K. Conrad, Cyclotomic Extensions (expository blurb), Example 5.5
- K. Conrad, Cyclotomic Extensions (expository blurb), Example 5.6
- K. Conrad, Cyclotomic Extensions (expository blurb), Sections 2-3
- J. S. Milne, Fields and Galois Theory, v5.10, Proposition 5.8 and the fundamental theorem
- K. Conrad, Cyclotomic Extensions (expository blurb), Example 3.3
- J. S. Milne, Fields and Galois Theory, v5.10, Proposition 4.23
- K. Conrad, Cyclotomic Extensions (expository blurb), Theorem 5.4 and Corollary 5.7
- K. Conrad, Linear Independence of Characters (expository blurb), Example 3.2