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.
Algebraic Differentials Separability and Smooth Local Presentations
1 · Prerequisites
- Abelian Categories
- Adjunctions Units and Counits
- Affine Schemes and the Structure Sheaf
- Algebraic Closure, Embeddings, and Separability
- Algebraic Extensions, Extension Degree, and Finite Fields
- Artinian Rings and Length
- Associated Primes and Primary Decomposition
- Binary Operations, Monoids, Groups and Subgroups
- Cardinal Arithmetic, Cofinality and the Alephs
- Categories, Functors and Natural Transformations
- Chain Complexes and Homology
- Chain Conditions, Semisimple Modules and the Wedderburn–Artin Theorem
- Chain Homotopy and the Homotopy Category
- Compactness
- Compactness in Metric Spaces
- 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
- Dedekind Domains and Ideal Classes
- Delta Functors and Universality
- Depth and Cohen Macaulay Modules
- Derived Functors
- Determinants of Matrices over a Commutative Ring
- Divisibility, Euclidean Domains, Principal Ideal Domains and Unique Factorisation
- Divisibility, Greatest Common Divisors and Bézout's Identity
- Eigenvalues, Eigenvectors and the Characteristic Polynomial
- Exactness and the Member Calculus
- Ext and Balanced Resolutions
- Fibre Products Base Change and Scheme Theoretic Fibres
- Finite Counting, Factorials and Binomial Coefficients
- Finite Fields and Cyclotomic Extensions
- Flatness and Faithful Flatness
- Foundations of the Real Numbers for Analysis
- Free Modules, Exact Sequences, Projective and Injective Modules
- Group Actions, Orbits, Stabilisers and Cayley's Theorem
- Group Homomorphisms and the Isomorphism Theorems
- Ideals, Quotient Rings and the Isomorphism Theorems for Rings
- Integral Extensions and Going Up
- Koszul Complexes and Regular Sequences
- Krull Dimension and Height Theorems
- Limits and Colimits
- Linear Independence, Bases and Dimension
- Localisation of Modules and Support
- Long Exact Sequences in Homology
- Mapping Cones Cylinders and Chain Triangles
- Metric Spaces
- Modules, Submodules, Quotient Modules and the Isomorphism Theorems
- Noether Normalisation and Nullstellensatz
- Noetherian Rings and Hilbert Basis
- Normal Subgroups and Quotient Groups
- Order, Zorn's Lemma, and the Axiom of Choice
- Ordinal Arithmetic and the First Uncountable Ordinal
- Ordinals, Cardinals, and Transfinite Recursion
- Polynomial Rings, the Division Algorithm and Roots
- Preadditive and Additive Categories and Biproducts
- Presheaves Sheaves Stalks and Sheafification
- Prime Spectra and Radicals
- Primes, Euclid's Lemma and the Fundamental Theorem of Arithmetic
- Projective and Injective Resolutions
- Rees Modules Artin Rees and Hilbert Samuel Theory
- Reflective Subcategories and the Adjoint Functor Theorems
- Regular Local Rings and Homological Dimension
- Relations, Functions, and Quotients
- Rings, Subrings, Integral Domains and Fields
- Roots, Rational Powers, and Classical Inequalities
- Schemes Subschemes and Morphisms Locally of Finite Type
- Sheaf Operations Exactness Ringed Spaces and Module Pullback
- Simple Field Extensions and the Construction of the Complex Numbers
- Splitting Fields
- Subobject Lattices Generators and the Grothendieck Axioms
- Suprema and Infima
- Symmetric Groups, Cycle Decomposition and the Sign Homomorphism
- Tensor Products of Modules
- The Determinant of a Linear Operator, Cofactors and Cramer's Rule
- The Diagram Lemmas in an Abelian Category
- The Field of Fractions and Localisation
- The Fundamental Theorem of Finite Abelian Groups
- The ZFC Axioms and the Basic Set Constructions
- Topological Spaces and Continuity
- Tor Flatness and Global Dimension
- Universal Coefficients and Kunneth Theorems
- Universal Properties, Representables and the Yoneda Lemma
- Valuation Rings and Discrete Valuation Rings
- Vector Spaces, Linear Subspaces, Span and Direct Sums
- Yoneda Extensions and Homological Dimension
- Zariski Topology on Prime Spectra
2 · Summary
Standard smoothness is characterised here by flatness together with geometrically regular fibres, and the criterion is reached through Kähler differentials: the universal property and its localisation, transitivity and conormal sequences, separating transcendence bases, the separable-residue cotangent sequence and the fact that a smooth presentation has an invertible Jacobian minor. Over a field, locally standard smooth maps are exactly the flat maps whose fibres stay regular under every field extension, which is geometric regularity; the page proves that equivalence, the descent from flatness plus a geometrically regular fibre to a local standard smooth chart, stability under base change and composition, and the submersion criterion identifying local standard smoothness of a finite-type morphism at k-rational points with injectivity of the pullback on cotangent spaces. The Axiom of Choice is carried explicitly where the local flatness criterion, regular parameters and geometric regularity require it, and the relative dimension of a presentation is kept distinct from the dimension of an individual fibre.
3 · Logical flowchart
4 · Definitions, theorems and proofs
Universal algebraic differentials and A-derivations
Definition
Let be a ring homomorphism of commutative rings (Commutative ring), so that is an -algebra. Let be the free -module on the set underlying (Unital left and right modules over a ring; unqualified module means left module), with basis symbol for , and let be the -submodule generated by all elements
with and . The module of algebraic differentials of over , also called the module of Kähler differentials, is the quotient -module
and the map , , is the universal -derivation of over . By construction is additive, satisfies the Leibniz rule
and is -constant, meaning for every ; these are exactly the three families of relators above.
Now let be a -module. An -derivation of into is a map that is additive, is -constant in the sense that for all , and satisfies the Leibniz rule . The set of such maps is written ; it is a -module under the pointwise operations, since is a -module.
Three conventions are part of the definition. First, is presented by the whole set of generators, so no finiteness of over and no finite generation, finite presentation or Noetherian hypothesis is assumed anywhere in this definition. Second, , because makes one of the relators; consequently , for by induction on the Leibniz rule, and for , . Third, the extreme case (with the identity) gives : every is the relator , so the quotient is the zero module and the only -derivation of into any -module is the zero map.
Universal property of algebraic differentials
Statement
Let be a ring homomorphism of commutative rings and let with its universal derivation be as in Universal algebraic differentials and A-derivations. For every -module the assignment is a -module isomorphism
natural in : for every -linear the diagram of the two assignment maps commutes, and the isomorphism is additive in and .
Facts & Assumptions
Given: A ring homomorphism of commutative rings, the presented module with universal derivation , and a -module .
Universal algebraic differentials and A-derivations: , where is the free -module on the basis symbols , , and is generated by , and ; ; an -derivation into a -module is a map that is additive, -constant and satisfies the Leibniz rule.
Proof
Every gives : additivity, -constancy and the Leibniz rule for are those of transported by the additive map , and has them by construction. The assignment is additive and -linear: and pointwise. It is injective, because the classes generate as a -module, so forces .
Conversely let . Since is free on the , there is a unique -linear with . Each relator is killed: by additivity of , similarly the Leibniz relator, and by -constancy. Hence factors through , giving a -linear with , that is, .
The two assignments are mutually inverse: the derivation attached to is , and the homomorphism attached to agrees with on every generator , hence equals because those generators span. Each assignment is additive, so the bijection is a -module isomorphism; and for -linear one has , which is exactly naturality in .
Differentials of a polynomial quotient and the Jacobian cokernel
Statement
Let be a commutative ring, let be the polynomial ring on finitely many variables , and let for an ideal . Then:
- is a free -module with basis .
- With regarded as a -module, the sequence of -modules is exact, where the first map sends the class of to and the second is induced by .
- If , then is the cokernel of the -linear map given by the Jacobian matrix , that is, .
The first map of (2) need not be injective; it is not claimed to be.
Facts & Assumptions
Given: A commutative ring , the polynomial ring , an ideal and the quotient .
Universal property of algebraic differentials: for every -module , is an isomorphism , naturally in .
Tensoring is right exact: for a commutative ring and an exact sequence of -modules, the sequence is exact for every -module .
The polynomial ring as finitely supported coefficient families on monomials: is the set of finitely supported functions from monomials to , written as formal sums ; so each element of has a unique expression as a finite -linear combination of monomials , and the product of monomials is .
Proof
Define on the monomial basis of [F3] by when , and when , extended -linearly. Since exponents add under multiplication and -multiplication distributes, for monomials and hence, by -bilinearity of multiplication, for all ; also . So each is an -derivation of . By [F1] there are -linear with . The -linear map , , is surjective: by additivity and the Leibniz rule , and an arbitrary element of is a finite -linear combination of monomials, so every lies in , and these elements generate by construction. The -linear endomorphism of fixes each generator , hence is the identity; therefore is injective as well, and are a basis.
Let where the map sends to . This is well defined: the map , , is -linear, and the class of depends only on , so extension of scalars gives the displayed -linear map. There is a -linear surjection with , induced by the -derivation , ; it kills the image of because maps to the class of with . Hence it factors through a surjection . Conversely define by the class of . This is well defined because makes lie in the image defining ; it is additive, -constant, and satisfies Leibniz because does and the -module structure on is that of . By [F1] applied to the -algebra and the -module , induces a -linear map , and the two displayed maps are inverse on the generating classes of and of . Therefore , which is exactness of the sequence of (2); right exactness of is the statement of [F2] applied to , and it is what makes the receptacle of this cokernel presentation.
Suppose . Since , every class in is a -linear combination of the classes of : from and (terms with two factors of ) one gets modulo . Hence the image of the first map of (2) is generated by the classes of , and by step 1.1 and the Leibniz rule. Under the basis identification of step 1.1, corresponds to the -th column of the Jacobian matrix, so the cokernel of , , is exactly of step 2.1.
The first map of (2) is not injective in general: take , and . Then , while in because for the structure map , so the first map has nonzero kernel. This proves the final clause and, with steps 2.1 and 3.1, the whole statement.
Localization, base change and functoriality of differentials
Statement
Let be a homomorphism of commutative rings, with universal derivation of .
- (Base change.) Let be a ring homomorphism and put . Then there is a -module isomorphism which is natural in the base-change data.
- (Localization.) Let be multiplicative and let be multiplicative with the image of in contained in . Then there is a -module isomorphism
- (Functoriality.) For an arbitrary -algebra homomorphism the -derivation of into induces a canonical -linear map For a general algebra map this map is neither asserted injective nor asserted an isomorphism.
Facts & Assumptions
Given: A ring homomorphism of commutative rings with universal derivation of .
Universal property of algebraic differentials: for every -module , the assignment is an isomorphism , natural in .
Universal mapping property of the tensor product of commutative algebras: for commutative -algebras and -algebra maps , there is a unique -algebra map with and ; so carries the -algebra structure with structure maps , .
Localisation of modules is extension of scalars: for a commutative ring , multiplicative and an -module , the map , , is an isomorphism with inverse .
Proof
For (1), let be given on the generating tensors of by . This is a well-defined additive map by the defining property of the tensor product of -modules, since is -balanced, and it satisfies Leibniz because and ; it is -constant since and -linear as the written scalar action shows, the -algebra structure on being the one of [F2]. By [F1] for the -algebra there is a -linear with . In the other direction is an -derivation of into the -module , so [F1] gives a -linear , and extension of scalars gives the -linear displayed in the statement. Both composites are -linear and fix the generators: , using and Leibniz; and . Hence and are mutually inverse and is the isomorphism of (1).
For (2), write and let be , an element of . This is well defined: if in , there is with for , and applying to gives ; multiplying by and writing , this yields . In the class of is zero, because and becomes invertible, so the class of is zero; since also becomes invertible as a scalar, the class of is zero, which is exactly . The map is additive and kills ; for the Leibniz rule one uses the relations implied by , namely inside , after which the product rule for fractions follows from the product rule for . Hence is a -derivation and [F1] for the -algebra gives a -linear with . Conversely is an -derivation of into , so [F1] gives a -linear with , and under the identification of [F3] the balanced assignment yields the -linear with . On generators, , the last equality being the Leibniz rule applied to together with ; and . Both composites are module maps fixing generating sets, so they are inverse and is the isomorphism of (2).
For (3) let be an -algebra map. The map from to the -module is additive and -constant and satisfies Leibniz, since is a ring homomorphism and is an -derivation; here denotes the image of in when is written as a -algebra. So it is an -derivation, [F1] turns it into a -linear map , and extension of scalars along gives the -linear map of (3). For , composing this map with the canonical map gives the base-change isomorphism of step 1.1: both send to . The map before this composition need not be an isomorphism. For , step 1.2 with does identify the functorial map with a localization isomorphism. No injectivity or surjectivity is claimed for a general .
Transitivity sequence for differentials
Statement
Let be homomorphisms of commutative rings. Then the sequence of -modules
is exact, where the first map is the extension of scalars of along (so that ) and the second is induced by the universal property of from the -derivation . The first arrow is not asserted injective, and in general it is not injective.
Facts & Assumptions
Given: Ring homomorphisms of commutative rings, with universal derivations of , of and of .
Universal property of algebraic differentials: for a ring map and an -module , is an isomorphism , for the pairs , and alike.
Tensoring is right exact: if is exact, then is exact; in particular an extension of scalars of a surjection is surjective and is generated as a -module by the elements .
Differentials of a polynomial quotient and the Jacobian cokernel: for , is free on , and for the quotient formula holds.
Proof
The second map exists by [F1]: is in particular an -derivation, so it induces a -linear with . It is surjective because the elements generate . The composite is zero: on the generators of the first map sends to , and sends that to , since lies in the image of so that is killed by the universal -derivation of . Hence .
For the reverse inclusion put , with quotient map . The -derivation , , kills , since kills the image of the first map; hence by [F1] it induces a -linear with . The map of step 1.1 kills the image of the first map, so it factors as for a -linear , and then is a -linear endomorphism of fixing the generators , so . Symmetrically is a -linear endomorphism of , and the elements generate because the elements generate and is onto, so from we get . Thus is injective, and for we get , hence and . Hence and, with step 1.1, ; moreover .
The first arrow is not injective in general: take a field, , with . By [F3] the source is , while the target is , the universal derivation of a ring over itself being zero. So the first map is zero on a nonzero module, and with steps 1.1 and 2.1 the asserted sequence is exact with a noninjective first arrow.
Separating transcendence basis and separably generated extensions
Definition
Let be a field extension and let be algebraically independent over (Algebraic and transcendental elements and algebraic extensions). If is algebraic over the subfield generated by , then is a transcendence basis of over ; by A maximal algebraically independent set is a transcendence basis an algebraically independent subset that is maximal for inclusion among algebraically independent subsets automatically has this property, so that lemma and this definition describe the same notion.
Now assume that is finitely generated (Finitely generated field extensions ), say . A finite tuple of pairwise distinct elements is a separating transcendence basis of when is a transcendence basis of over and the extension is finite separable (Separable algebraic elements and separable extensions). The extension is separably generated over , or simply separably generated, when it admits a separating transcendence basis; in this terminology a finitely generated extension is separably generated precisely when it has a transcendence basis over which its residual extension is separable.
Finiteness of the residual extension is automatic: is finitely generated over , hence over , and algebraic there by the definition of a transcendence basis, so is finite by An extension generated by finitely many algebraic elements is finite. Three conventions are part of the definition.
- The empty tuple is allowed: is a separating transcendence basis exactly when is algebraic and separable, equivalently (under the finite-generation hypothesis) when is finite separable.
- The notion concerns and says nothing about the intermediate field being perfect. If is perfect, the rational function field is nevertheless imperfect in characteristic as soon as , so separability over it is a genuine restriction and cannot be replaced by separability of algebraic extensions of .
- All transcendence bases of a finitely generated extension have the same finite cardinality , namely the transcendence degree, so the number of elements in a separating transcendence basis is determined by the extension.
Finitely generated extensions of a perfect field are separably generated
Statement
Let be a perfect field and let be a finitely generated field extension. Then has a separating transcendence basis over (Separating transcendence basis and separably generated extensions): there are , algebraically independent over , such that is finite separable.
The argument in characteristic uses -power linear independence and exchanges of finitely many generators; it does not infer that is perfect, and no perfectness of any intermediate field is asserted.
Facts & Assumptions
Given: A perfect field and a finitely generated field extension , say .
Separating transcendence basis and separably generated extensions: a finite tuple is a separating transcendence basis when its entries are algebraically independent over and the residual extension is finite separable, and then is the common cardinality of all transcendence bases.
Algebraic and transcendental elements and algebraic extensions: is algebraic over a subfield when it satisfies a nonzero polynomial equation over , and transcendental otherwise; is algebraic when every element of is algebraic over .
An extension generated by finitely many algebraic elements is finite: if are algebraic over , then is finite.
A field is perfect exactly when it has characteristic zero or its Frobenius map is surjective: is perfect if and only if , or and the Frobenius map of is surjective.
Every algebraic extension of a perfect field is separable: every algebraic extension of a perfect field is separable.
Separable algebraic elements and separable extensions: an extension is separable when each of its elements has separable minimal polynomial.
The inseparable degree of a finite extension: for a finite extension , .
Separable degree is multiplicative in finite towers: : for a finite tower .
Tower law for finite extensions: : for a finite tower.
A finite extension is separable if and only if : a finite extension is separable exactly when its separable degree equals its degree, so by [F7] a finite extension is separable exactly when its inseparable degree is .
The evaluation kernel and the unique monic irreducible minimal polynomial of an algebraic element: for algebraic over , implies , where is the monic irreducible minimal polynomial; and generates the evaluation kernel.
Gauss lemma over a UFD with For every field , is a unique factorisation domain: a polynomial that is irreducible and of positive degree in one variable over a polynomial ring over a field remains irreducible over the fraction field, and a polynomial ring over a field is a unique factorisation domain.
A nonzero polynomial over a field is separable exactly when its gcd with its derivative is : for over a field, is separable if and only if .
An algebraic extension generated by separable elements is separable: an algebraic extension generated by separable elements is separable.
One element of a transcendence basis can be exchanged for a suitable rival: given transcendence bases of and there is with a transcendence basis; consequently two transcendence bases of a finitely generated extension have the same finite cardinality.
A finite extension generated by elements all but possibly one of which are separable is simple: a finite extension generated by elements all but possibly one of which are separable is simple.
The binomial theorem over an arbitrary commutative ring, A prime divides for : the binomial expansion holds in every commutative ring, and divides its intermediate coefficients for exponent . Hence in characteristic .
Proof
Running through the finite list and adjoining an element exactly when it is transcendental over the field generated by the elements already adjoined produces, after finitely many steps, an algebraically independent over which every is algebraic; then is algebraic over and finitely generated over it, hence finite over by [F3], so is a transcendence basis and is defined by [F7]. The set of values , taken over the finite, algebraically independent with finite, is a nonempty set of natural numbers and therefore has a least element; fix attaining it, and note that each such is a transcendence basis of by [F2]. If , then also has characteristic zero, so is perfect by [F4], and is separable by [F5], so is already a separating transcendence basis. Henceforth assume . Then Frobenius is surjective on by [F4], and -linearly independent elements have -linearly independent -th powers: from with and one gets , hence because is additive by [F17] and injective on the field , so all , hence all , vanish.
Suppose is not separable. By [F6] there is that is not separable over , and is algebraic over by [F2] since is algebraic. Choose nonzero of least total degree with . Then is irreducible: a factorisation with nonconstant satisfies , and either or vanishes at with strictly smaller total degree, contradicting minimality.
Suppose every monomial exponent of were divisible by . Since is perfect, write each coefficient using [F4]. Then for the nonzero polynomial , by Frobenius additivity [F17]. Evaluation gives in the field , hence , contradicting minimality because . Thus some variable occurs with an exponent not divisible by ; fix such .
Set and . Write with and . The total degree of is strictly less than that of , so by the minimality in step 2.1. Thus is a nonconstant polynomial in vanishing at , which proves algebraic over . This makes a transcendence basis of : if were dependent, a maximal independent subset would be a transcendence basis of over , and since is algebraic over by the preceding coefficient argument, the same finite set would be a transcendence basis of over with , contradicting that is a transcendence basis of that field and that all transcendence bases of a finitely generated extension have the same cardinality [F15]; so is independent, is algebraic, and is a rational function field over in the variables , . Consequently is the image of the polynomial , which has positive -degree and is irreducible in ; its coefficients are primitive, since a nonunit common factor would factor the irreducible of positive -degree. Thus Gauss' lemma [F12] makes its image irreducible in . Since some -exponent of is not divisible by , the same exponent occurs in , so and because is irreducible and ; thus is separable by [F13]. As and is irreducible, is a nonzero scalar multiple of the monic minimal polynomial of over by [F11], so is separable over and is separable by [F14].
By [F8] and [F9] the inseparable degree is multiplicative, for finite ; and by [F10]. With and we get and , while by [F10] because is not separable. Therefore , and is a transcendence basis of by step 4.1 with a strictly smaller inseparable degree than , contradicting the minimality of step 1.1. Hence is separable and is a separating transcendence basis, which proves the theorem; moreover, by [F16] the residual finite separable extension is simple, so for a single element separable over .
Differentials of a separably generated field extension
Statement
Let be a finitely generated field extension that is separably generated over by (Separating transcendence basis and separably generated extensions). Then are a -basis of ; in particular is a free -module of rank .
Assume moreover the Axiom of Choice. Let and let be a tower of fields with finitely generated. Then the natural map induced by is injective. The Axiom of Choice is used exactly to choose maximal algebraically independent subsets, that is transcendence bases, of over and of over ; every subsequent step is choice-free. The assumption is declared as The Axiom of Choice and is inherited by the consumers of this theorem.
Facts & Assumptions
Given: A finitely generated field extension separably generated by , so that is finite separable for ; and, for the second part, an extension with , finitely generated, together with the Axiom of Choice.
Separating transcendence basis and separably generated extensions: is algebraically independent over and the residual extension is finite separable; every finitely generated extension is separably generated exactly when it possesses such a tuple.
A finite extension generated by elements all but possibly one of which are separable is simple: a finite extension generated by elements all but possibly one of which are separable over the base is simple; in particular every finite separable extension is simple, so for an element separable over .
The evaluation kernel and the unique monic irreducible minimal polynomial of an algebraic element: for algebraic over a field there is a unique monic irreducible minimal polynomial with , and holds if and only if .
A nonzero polynomial over a field is separable exactly when its gcd with its derivative is : for over a field, is separable over that field if and only if .
Differentials of a polynomial quotient and the Jacobian cokernel: for a commutative ring and the module is free with basis .
Localization, base change and functoriality of differentials: for a multiplicative set inside an -algebra the canonical map is an isomorphism, with inverse sending to ; and for every -algebra map there is a natural -linear map .
Transitivity sequence for differentials: for the sequence is exact.
Universal property of algebraic differentials: for every -module , composition with is an isomorphism , naturally in .
A field is perfect exactly when it has characteristic zero or its Frobenius map is surjective: a field is perfect if and only if its characteristic is , or its characteristic is and its Frobenius map is surjective.
Every algebraic extension of a perfect field is separable: every algebraic extension of a perfect field is separable.
A maximal algebraically independent set is a transcendence basis: if is maximal for inclusion among the subsets of algebraically independent over a subfield , then every element of is algebraic over .
An extension generated by finitely many algebraic elements is finite: if are algebraic over a field , then is finite.
The Axiom of Choice: every family of nonempty sets has a choice function; this is what licenses the maximal algebraically independent subsets chosen below.
Proof
Set and let with be separable over , as supplied by [F2]. Let be the minimal polynomial of over ; by [F3] it is monic and irreducible, and by [F1] and the definition of a separable element is separable over . Hence by [F4], , and the class of is invertible modulo , so in the field .
has -basis . Indeed , , identifies with the polynomial ring on the algebraically independent elements by [F1], so by [F5] (with and ) the module is free on ; since a nonzero element of maps to a nonzero element of , the localisation isomorphism of [F6] applies with and exhibits with the images of as a -basis, and those images are exactly .
: for every -module and every -derivation we have , because with the coefficients of and , and by step 1.1; then on all of , since a derivation vanishing on and on vanishes on every polynomial in . By [F8] this says for all , hence . Applying the transitivity sequence of [F7] to , the first map is therefore surjective, and by step 1.2 the elements generate as a -module.
Independence of the generators. For each the -linear functional on with exists by step 1.2 and corresponds by [F8] to a -derivation with . Write and put , which is defined by step 1.1. On the polynomial ring define for , so that and for all and : these identities follow from additivity of and the product rules for and for the formal derivative, and they say that is a derivation along the evaluation , , with . Moreover , so vanishes on the ideal ; since by [F3], descends to a well-defined -derivation extending , and by [F8] to a -linear map sending to . If now with , applying that map gives for each , so are linearly independent over and, with step 2.1, form a -basis of .
The second part. Assume and choose, using [F13], a maximal algebraically independent subset over and a maximal algebraically independent subset over . By [F11] the extensions and are algebraic; by [F9] every field of characteristic is perfect and by [F10] every algebraic extension of a perfect field is separable, so and are separable algebraic. Because is finitely generated, [F12] makes finite, so is a separating transcendence basis of in the sense of [F1], and step 3.1 exhibits a finite -basis of .
Extension of derivations. Let be a -derivation. First extend to : each element of lies in for some finite , and on the polynomial ring , for which is free on the , , by [F5], the prescription for , together with on , defines a unique -derivation of extending ; it extends uniquely to the fraction field by the quotient rule of [F6], and for the extension on restricts to the one on by uniqueness, so a derivation extending is well defined on the union of the fields . Next let ; then is finite separable by step 4.1, hence simple, with minimal polynomial of over that is separable by [F10], so by [F3] and [F4]. Substituting for , for and for in the construction of step 3.1 produces an extension of to , and any two extensions of to agree, because their difference vanishes on and takes at a value killed by . Declaring the value at to be that unique value defines for every ; it is a derivation because for the field is finite separable over by [F10] and [F12], carries an extension of by the same construction, and on it the derivation laws hold while its restrictions to and agree with the unique extensions, so the values assigns are additive and satisfy Leibniz.
Injectivity. Keep the notation of step 4.1, let be the natural map of [F6], and let , the elements being a -basis of by step 4.1. For each the functional is a -linear map , hence by [F8] equals for a -derivation ; by step 5.1 there is a -derivation extending , and by [F8] it induces an -linear with . Then for every , so . Hence is injective, which is the second assertion, and the theorem is proved.
Separable residue and the cotangent sequence of a local algebra
Statement
Let be a field and let be a Noetherian local -algebra with maximal ideal and residue field . Assume that is finitely generated and separably generated over in the sense of Separating transcendence basis and separably generated extensions. Then is a short exact sequence, the first map sending the class of to . If in addition is finite separable, then , so the first map is an isomorphism ; this applies in particular at a closed point of a finite-type -algebra whose residue field is a finite separable extension of .
Facts & Assumptions
Given: A field , a Noetherian local -algebra whose residue field is finitely generated and separably generated over , and, for the final clause, finite separable.
Universal property of algebraic differentials: for a ring map and every -module , composition with is an isomorphism , naturally in .
Separating transcendence basis and separably generated extensions: a finitely generated extension that is separably generated has algebraically independent elements over the base whose residual extension is finite separable; so with finite separable.
A finite extension generated by elements all but possibly one of which are separable is simple: every finite separable extension is simple; this is applied both to and, in the final clause, to .
The evaluation kernel and the unique monic irreducible minimal polynomial of an algebraic element: for algebraic over a field the evaluation map has kernel generated by a monic irreducible , so and implies .
Separable algebraic elements and separable extensions: is separable over when it is algebraic over and its minimal polynomial over is separable.
A nonzero polynomial over a field is separable exactly when its gcd with its derivative is : for over a field, is separable if and only if .
Proof
A conormal computation. Let be any commutative -algebra, let be an ideal and . Then is exact, the first map sending the class of to . The second map exists by [F1] applied to , and it is surjective because the elements generate ; the composite is zero because maps to . The first map is well defined because for one has , which lies in the image of in , so the classes of and of in have the same image. Let be the cokernel of the first map. A -linear map is the same thing as a -derivation with for all , because -linear maps out of correspond by [F1] to -derivations of , and the quotient imposes exactly the vanishing on the classes , . Such a factors through a -derivation : it is constant on cosets, since for , and it satisfies Leibniz on classes, since for the identity holds because for the -module . Conversely every -derivation of pulled back along is such a . By [F1] the functor is therefore isomorphic to , so the canonical map is an isomorphism.
Preparation of the section. Write with as in [F2] and [F3], and let be the minimal polynomial of over ; by [F4] and [F5] the polynomial is monic, irreducible and separable, so by [F6], since is nonzero and makes its class a unit of . Put , a local ring with maximal ideal and residue field , and write for the quotient map. Choose with and , and let be the -algebra map with . It is injective because the are algebraically independent over , so is a polynomial ring and every nonzero element of it is a unit of the local ring , since places it outside the maximal ideal. By the universal property of localisation, extends to a -algebra map , and is the inclusion because they agree on the generators .
Finite separable residue. If is finite separable, then by [F3] there is with ; its minimal polynomial is separable by [F5], so by [F4] and [F6]. Every -derivation into a -module satisfies with and , so because is a field, and then on . By [F1] this forces .
Applying step 1.1 with , and gives exactness of ; it remains to prove that the first map is injective.
Correction of the lift. With the notation of step 1.2 form , where is with coefficients transported by . Then , so , and in because . Also , so is a unit of . Set , an element of with the same residue . Taylor expansion in the commutative ring terminates after the linear term because has square zero, so . Hence the -algebra map with and kills , and by [F4] it descends along to a -algebra map with .
A derivation inverse to . Define ; its image lies in , and for because . For write and ; since is a ring map, and have square zero and their product with any element of vanishes, so expanding gives and hence . Thus is a -derivation of into the -module , and composing with gives a -derivation whose value at is the class of modulo .
Conclusion. By [F1] the derivation of step 3.1 corresponds to an -linear map , which factors through because the target is annihilated by , giving a -linear with for . The first map of step 2.1 sends the class of to , so is a left inverse of it and that map is injective; with step 2.1, the displayed sequence is exact.
Finite separable residue concluded. By step 1.3 the hypothesis of the final clause gives , so the exact sequence of step 4.1 reads , that is, via the first map.
Extension of scalars of a scheme along a field extension
Statement
Assume the Axiom of Choice (The Axiom of Choice). Let be a field, let be a scheme (Schemes) with a morphism , and let be a field extension. Then:
- Construction. For every affine open the ring is a -algebra, the -algebra map gives a morphism (Affine schemes are contravariantly equivalent to commutative rings), and for a principal open of the canonical identifications (Sections and restrictions on distinguished opens of an affine scheme) identify the corresponding open subschemes of and . These data satisfy the identity and cocycle conditions, so Gluing affine schemes along compatible open isomorphisms glues the affine schemes , over all affine opens of , to a -scheme with a morphism over .
- Independence of the cover. If and arise from two affine covers of by this construction, there is a unique isomorphism commuting with both the morphisms to and the structure morphisms to .
- Affine restrictions and fibres. For every affine open the restriction of over is canonically . For with residue field (The residue field at a point of an affine scheme) and any affine open containing , the fibre of over , computed as , is independent of up to canonical isomorphism and is canonically .
- Transitivity. For a tower of fields the canonical morphism is an isomorphism, canonically over .
The Axiom of Choice is used only to present the affine cover as a family of affine opens indexed by the points of ; the same construction runs on the set of all affine opens of , which needs no choice. The assumption is declared and is inherited by consumers, since the statement promises it.
Facts & Assumptions
Given: A field , a scheme with a morphism to , a field extension (and a tower for clause 4), and the Axiom of Choice.
Gluing affine schemes along compatible open isomorphisms: affine schemes equipped with open subschemes and isomorphisms on overlaps satisfying the identity and cocycle conditions glue to a scheme, uniquely up to unique isomorphism respecting the chart identifications, and the given affine schemes become an open affine cover.
Affine schemes are contravariantly equivalent to commutative rings: naturally, and is a contravariant equivalence with quasi-inverse global sections.
Sections and restrictions on distinguished opens of an affine scheme: for , , and if the restriction is the canonical localisation .
Intersections of affine opens admit principal affine covers: if are affine open subschemes of a scheme, then is covered by open subschemes that are principal opens in and principal opens in affine open charts of .
Localisation of modules is extension of scalars: for a multiplicative set and an -module the map , , is an isomorphism.
Universal property of localisation: maps that invert factor uniquely through : a unital ring homomorphism carrying every element of to a unit factors uniquely through .
Universal mapping property of the tensor product of commutative algebras: is the coproduct of the commutative -algebras and , with the universal map for each pair of -algebra maps into a common -algebra .
Associativity of tensor products for compatible bimodules: there is a canonical isomorphism , , respecting compatible outer module actions.
The residue field at a point of an affine scheme: , and for in an affine spectrum the canonical isomorphisms hold.
Schemes: a scheme is a locally ringed space every point of which has an open neighbourhood that is an affine scheme with the restricted structure sheaf.
The Axiom of Choice: every family of nonempty sets has a choice function.
Proof
Principal opens under base change. Let be a -algebra and . The -algebra map , , is well defined by [F6], and after tensoring with gives a -linear map , because is an -algebra by [F7]. Conversely is a -algebra map carrying to a unit, so by [F6] it factors through a map . The two maps are inverse on the generators and ; composing with by [F2], the principal open of is canonically , compatibly with further principal localisations and with the restriction maps of [F3].
Fibre rings. Let be a -algebra and with residue field as in [F9]; then , and the composite of the canonical isomorphisms of [F5] and [F8] sends to . Hence there is a canonical -linear isomorphism .
Common principal neighbourhoods. For affine opens , and , first take using the principal-open basis. Then take . The restriction of to is by [F3], and its nonvanishing locus there is both and : the equality follows by applying the residue-field maps of the open immersion to this section. Thus is principal in both original affines. This supplies the common principal refinements needed below, with independent denominators in the two coordinate rings.
Gluing. Let be a scheme over and let be a family of affine opens covering ; each is a -algebra because restricts to . For each the -algebra map gives by [F2] a morphism , and the affine pieces over distinct are to be identified over the principal opens. Whenever is an open subscheme of which is principal in and in , say , the rings and are both by [F3] and hence canonically equal, and the identity of rings induces by step 1.1 an identification of the corresponding open subschemes and . These identifications are induced by identities of section rings and are therefore compatible: the identity and cocycle conditions hold on triple overlaps because all the identifications are the canonical comparison of with itself. Since step 1.3 covers every overlap by such common principal opens, the data satisfy the hypotheses of [F1], which glues the schemes to a scheme with open affine cover , and the morphisms to glue to . The maps to induced by also agree on overlaps, so they glue to the -scheme structure on .
Affine restriction. Let be any affine open of . By step 1.3 cover each by common principal opens . Their section rings and are canonically identified by [F3]. Step 1.1 identifies the corresponding base-changed opens with on either side. These opens cover the inverse image of in and cover : a principal cover remains a cover under inverse image, since is precisely the inverse image of . The identifications agree on common refinements by step 1.1, so glue to an isomorphism over both and by [F1]. Hence the restriction is the asserted affine base change.
Independence of the cover. Let and be two affine covers of . By step 1.3 the family of open subschemes of that are principal in some and in some covers . For such a , with by [F3], the construction of step 2.1 attaches to the affine scheme in the glueing over and, by the same computation, in the glueing over : in both cases arises as a principal open of an affine chart, and the attached piece is of the localisation of the chart ring tensored with , which is by step 1.1. Both glued schemes are therefore obtained by glueing the same family along the same canonical identifications over principal opens of . By the uniqueness clause of [F1], applied to the two open affine covers of and , the canonical chart identifications glue to an isomorphism over and . Any other such isomorphism must preserve each inverse image of ; on its ring , the induced map fixes both factors because it is over and over . It is therefore the identity by [F7]. These opens cover, proving uniqueness with both compatibilities.
Fibres. Let and let be an affine open containing . By step 3.1 the preimage of in is over , and the scheme over attached to the point of that affine piece is , which by step 1.2 is canonically over . If is a second affine open containing , choose principal in and in with , say , using step 1.3 and [F3]; then and are canonically the same ring, so tensoring the identification with over the common ring identifies with canonically. Hence the fibre is independent of the affine neighbourhood and is as asserted.
Transitivity. Let be a tower, and write for the construction applied over the base field to the extension . The affine pieces constructed in step 2.1 form an affine cover of , so the construction of glues the schemes . The canonical isomorphisms of [F7] and [F8] are compatible with the transition identifications of step 2.1, because those are induced by identities of section rings; hence , which is glued from the pieces , and are glued from corresponding pieces with corresponding identifications. By step 3.2 (applied to the two covers of the same scheme, or directly by the uniqueness clause of [F1]) the displayed chart isomorphisms glue to an isomorphism over and . It is unique with both compatibilities, by the same two-factor argument as step 3.2, proving clause 4.
Geometrically regular algebras and geometrically regular fibres
Definition
Let be a field. A -algebra of finite type (Finitely presented modules and finitely presented algebras) is geometrically regular over when for every finitely generated field extension (Finitely generated field extensions ) the -algebra (Universal mapping property of the tensor product of commutative algebras) is a regular Noetherian ring (regular noetherian ring). Since a finitely generated extension makes a finite-type -algebra, the Noetherian condition in that clause is automatic; the regularity is the content. The quantifier is over the finitely generated extensions only, and the later field-test lemma shows that regularity for those is equivalent to regularity after every field extension of .
Now let be a homomorphism of commutative rings with finitely presented as an -algebra (Finitely presented modules and finitely presented algebras), let and let . The fibre of over is , a -algebra, and it is geometrically regular at when for every field extension and every prime of lying over the image of in , the local ring is regular. A fibre is geometrically regular when it is geometrically regular at each of its points, and the condition is imposed on every field extension , not only on the finitely generated ones; an empty fibre satisfies the pointwise condition vacuously, and the affine line over the residue field is the model case.
Three conventions belong to the definition. First, geometric regularity of a finite-type -algebra is a statement about all scalar extensions , so it is strictly stronger than regularity of and is defined without using smoothness of the structural morphism; the equivalence with local standard smoothness is a theorem, not part of the definition. Second, the fibre condition is pointwise at a prime and ranges over all field extensions of the residue field , so that a rational point over an algebraic closure of is covered without appealing to any later theorem. Third, the hypothesis that is finitely presented over is part of the definition of the fibre condition, since it is what lets the standard smooth presentations and the local presentation theorem apply to it.
Remarks
The field-case equivalence with local standard smoothness is proved in Locally standard smooth iff flat with geometrically regular fibres, clause 3, under the Axiom of Choice assumed there. Standard smooth presentations and locally standard smooth maps supplies the presentation terminology, rather than the equivalence theorem.
Standard smooth presentations and locally standard smooth maps
Definition
Let be a commutative ring. A standard smooth presentation of an -algebra consists of integers , elements of the polynomial ring (The polynomial ring as finitely supported coefficient families on monomials) and an element such that as -algebras, and such that the Jacobian matrix of (Differentials of a polynomial quotient and the Jacobian cokernel) has a minor whose image in is a unit, that is, an invertible element. The integer is the relative dimension of the presentation. The case is allowed and is exactly a localisation of a polynomial ring, ; the case presents for , that is a localisation of itself.
For a homomorphism of commutative rings that is finitely presented as an -algebra (Finitely presented modules and finitely presented algebras) and a prime , the map is standard smooth at , or has a standard smooth presentation at , when there is such that admits a standard smooth presentation over . It is locally standard smooth when this holds at every prime of ; equivalently, when every point of has an affine open neighbourhood on which is presented by a single standard smooth presentation.
Two conventions are part of the definition. First, the invertible minor may be assumed to sit in the first columns: if a minor on columns is a unit, the automorphism of permuting the variables so that these become the first variables carries the presentation to one whose minor in the first columns is that unit, and the relative dimension is unchanged. Second, a further principal localisation can be absorbed into the presentation: adjoining a variable with the single equation to a presentation produces a standard smooth presentation of the localisation, the new equation contributing a diagonal entry that keeps a block minor invertible, and it changes and by the same amount, so that the relative dimension is again unchanged.
Base change of standard smooth presentations
Statement
Let be a homomorphism of commutative rings and let be an -algebra carrying a standard smooth presentation (Standard smooth presentations and locally standard smooth maps) of relative dimension , with and with the leading Jacobian minor (in the sense of Differentials of a polynomial quotient and the Jacobian cokernel) mapping to a unit of ; the conventions of the definition allow the invertible minor to be assumed in the first columns. Let be the images of under the induced map and put Then:
- there is a unique -algebra isomorphism with , where is the image of in and its image in ;
- is standard smooth over with the same , and relative dimension ; explicitly, the image of is a unit of .
No hypothesis is placed on , and the relative dimension is unchanged.
Facts & Assumptions
Given: A ring homomorphism and a standard smooth presentation over whose leading minor maps to a unit in .
Standard smooth presentations and locally standard smooth maps: a standard smooth presentation consists of integers , elements and with , such that the Jacobian matrix has a minor whose image in is a unit; is the relative dimension, and the invertible minor may be assumed to lie in the first columns.
The polynomial ring as finitely supported coefficient families on monomials: is the commutative -algebra of polynomials in the indeterminates , generated as an -algebra by them.
Universal property of a polynomial ring on an arbitrary family of indeterminates: for a ring homomorphism and any family in there is a unique ring homomorphism restricting to on with .
Universal property of localisation: maps that invert factor uniquely through : if is a unital homomorphism of commutative rings carrying a multiplicative set into the units of , there is a unique unital ring homomorphism with , namely .
A ring homomorphism whose kernel contains a two-sided ideal factors uniquely through the quotient ring: a ring homomorphism whose kernel contains a two-sided ideal factors uniquely through the quotient .
Universal mapping property of the tensor product of commutative algebras: for commutative -algebras and -algebra homomorphisms , there is a unique -algebra homomorphism with and , namely .
The tensor product of -algebras has multiplication : is a commutative -algebra with and unit ; in particular and are ring homomorphisms.
Multiplicative subsets and the localisation as equivalence classes of fractions: the localisation of a commutative ring at a multiplicative subset is a commutative ring, , , is a ring homomorphism and each maps to a unit.
Differentials of a polynomial quotient and the Jacobian cokernel: is free on for , the partial derivatives are computed on the monomial basis by and extended -linearly, , and for the module is the cokernel of the Jacobian matrix .
Proof
Notation. Put , , , so that ; put , and . All four are commutative rings by [F2] and [F8], and the coefficient-change map , , is a ring homomorphism by [F3] applied to .
The map . The composite kills and carries to , which is a unit of by [F8]; by [F5] it factors uniquely through , and by [F4] the resulting map factors uniquely through . This gives a unique ring homomorphism with for , .
The map . By [F7] the assignment is a ring homomorphism , and are elements; by [F3] there is a unique ring homomorphism restricting to and sending . Its kernel contains , because , and it sends to , which is a unit with inverse in view of [F7] and invertible in . Applying [F5] and then [F4] gives a unique ring homomorphism over .
The map . The identity map of and the map of step 2.1 are -algebra maps into that agree on ; the latter is induced by the coefficient-change map . Hence [F6] provides a unique -algebra homomorphism with and , that is, ; on the elements it is .
and . Both composites are -algebra homomorphisms. The -algebra is generated by the images of the and by : every element of is a class with , and is an -linear combination of monomials in the . A ring homomorphism out of is determined by its restriction to and the images of the , by [F3], [F5] and [F4] applied in that order, so and force . Similarly is generated as an -algebra by and , by [F7], and fixes these elements: and , the last because is a ring homomorphism sending to . Hence as well, and is an isomorphism with inverse .
The minor maps to a unit of . The element maps to a unit of by hypothesis, hence is a unit with inverse by [F7], and carries it to 's image, namely ; as a ring isomorphism carries units to units, so is a unit of .
The Jacobian of the changed polynomials. By the monomial formula of [F9] the partial derivative is linear over the coefficient ring, so is the image of under for all ; hence is the leading minor of the Jacobian matrix of . By step 5.1 its image in is a unit, so is a standard smooth presentation over of relative dimension , the same parameters as the given presentation. With step 4.1 this proves both assertions.
Invertible Jacobian minor gives regular parameters in a polynomial fibre
Statement
Assume the Axiom of Choice (The Axiom of Choice). Let be a field, let , let , let be a prime ideal and put , with maximal ideal and residue field (Localisation at a prime ideal: , is local with unique maximal ideal ). Let and suppose that the leading minor of the Jacobian matrix of Differentials of a polynomial quotient and the Jacobian cokernel satisfies . Then:
- the classes of in are -linearly independent;
- is a regular local ring, is a regular sequence in , and is a regular local ring with .
This is the fibre computation used when a standard smooth presentation is examined over a field.
Facts & Assumptions
Given: A field , the polynomial ring , a prime , the localisation with maximal ideal and residue field , and elements whose leading Jacobian minor is not in ; and the Axiom of Choice.
Differentials of a polynomial quotient and the Jacobian cokernel: is free on ; the partial derivatives are defined on the monomial basis by and extended -linearly, satisfy the Leibniz rule, and ; for the module is the cokernel of the Jacobian matrix .
localisation and polynomial extension of regular rings: under the Axiom of Choice, localizations and finite polynomial extensions of a commutative regular Noetherian ring are regular, and regularity can equivalently be tested at maximal ideals.
embedding dimension and regular local ring: for a nonzero commutative Noetherian local ring one has , and is regular local exactly when .
regular noetherian ring: a commutative Noetherian ring is regular when every prime localisation is a regular local ring.
regular system of parameters equivalent basis: under the Axiom of Choice, for a nonzero Noetherian local ring of dimension and , the tuple is a regular system of parameters if and only if its classes form a -basis of ; in particular every lift of a cotangent basis generates and is a system of parameters.
regular local rings are domains and cohen macaulay: under the Axiom of Choice, a regular local ring of dimension is a domain and Cohen–Macaulay, and for every regular system of parameters the tuple is -regular and is regular local of dimension for every .
regular system of parameters: in a regular local ring of dimension , a regular system of parameters is an ordered minimal generating tuple of the maximal ideal, of length ; the empty tuple when .
Zorn's lemma gives a basis between any linearly independent set and any spanning set containing it: if with independent and , there is a basis of with : under the Axiom of Choice, a linearly independent subset of a vector space extends to a basis.
Localisation of modules is exact: localisation at a multiplicative set preserves short exact sequences.
is an integral domain if and only if is a prime ideal: is an integral domain if and only if is a prime ideal; in particular is prime in a domain.
A commutative ring is Noetherian exactly when every ideal is finitely generated, exactly when its ideals satisfy the ascending chain condition, and exactly when every nonempty set of ideals has a maximal member: a commutative ring is Noetherian if and only if every ideal is finitely generated.
Field: a field is a commutative ring with in which every nonzero element has a multiplicative inverse.
Krull dimension of a nonzero ring: the Krull dimension of a nonzero commutative ring is the supremum of the lengths of strict chains of prime ideals.
A local ring is a nonzero commutative ring with a unique maximal ideal: a local ring is a commutative ring with exactly one maximal ideal; its residue field is the quotient by that ideal.
Localisation at a prime ideal: : is the localisation at the multiplicative set .
is local with unique maximal ideal : is a local ring with maximal ideal .
is the residue field at : the residue field of is the fraction field of .
The Axiom of Choice: every family of nonempty sets has a choice function.
Proof
The field is a regular Noetherian ring. Its ideals are and : a nonzero ideal contains a nonzero element, which is a unit by [F12], hence contains and equals . Both ideals are finitely generated, so is Noetherian by [F11]. The only prime ideal of is : it is prime because is a domain and every nonzero ideal equals , which is not prime; so has exactly one maximal ideal, namely , and is a local ring in the sense of [F14] with residue field . There is no strict chain of primes, so by [F13]; and , so is regular local by [F3]. Every prime localisation of is itself, hence regular local, so is a regular Noetherian ring by [F4].
The local ring is regular. By [F2] applied to the regular Noetherian ring of step 1.1, the polynomial ring is regular, and its localisation at the prime is regular as well; by [F4] this says that every prime localisation of the Noetherian ring is a regular local ring, in particular itself, whose only maximal ideal is by [F16]. Hence is a regular local ring with residue field [F17], and by [F3].
The differential map . Localising the exact sequence of -modules at and using [F9] identifies . The assignment is -balanced: for and the Leibniz rule of [F1] gives , and for one has by the same rule applied to a product of two elements of . Hence it induces an -linear map , that is, a -linear map on , which sends the class of to the -th Jacobian column .
The classes of are linearly independent. Let satisfy in . Applying of step 3.1 and using its -linearity gives in . The first coordinates are the matrix equation , where is the image in of the matrix ; its determinant is the image of , which is nonzero because and has kernel exactly on . Hence is invertible over the field and , that is, every . Therefore no nontrivial -linear relation exists among the classes of .
A regular system of parameters. Put by step 2.1. By step 4.1 the family of classes in the -vector space is linearly independent, so by [F8] it extends to a -basis with ; here because the independent family has at most members. Define for and for . The classes of form a -basis of , so is a regular system of parameters of by [F5], of length as required by [F7].
Conclusion. By [F6] applied to the regular local ring of step 2.1 and its regular system of parameters of step 5.1, the tuple is an -regular sequence and is a regular local ring of dimension . Since for , the quotient is and the initial segment is a regular sequence. This proves both assertions.
Local flatness criterion by regular parameters
Statement
Assume the Axiom of Choice (The Axiom of Choice). Let be a local homomorphism of Noetherian local rings and let be a finite -module. If , then is flat over . The module is not assumed finite over .
Consequently, if and are regular local rings and the images in of a regular system of parameters of extend to a regular system of parameters of , then is flat over .
Facts & Assumptions
Given: A local homomorphism of Noetherian local rings and a finite -module with ; for the second assertion regular local and a regular system of parameters of whose images extend to one of ; and the Axiom of Choice.
The long exact Tor sequence in the right-module variable: under Dependent Choice, a short exact sequence of right -modules and a left module with a supplied projective resolution give the natural long exact sequence , with the usual tensor-product tail.
Flatness is equivalent to preserving injections and to the ideal and finitely generated ideal tests: is flat over if and only if is injective for every finitely generated ideal .
Artin-Rees controls intersections of submodules with high ideal powers: for a Noetherian ring , an ideal , a finite -module and a submodule there is with for every .
The Krull intersection is the -torsion submodule, and it vanishes in the Jacobson-radical case: for a Noetherian ring , an ideal and a finite -module , the intersection is zero.
Composition series and length of a module: a composition series of a module is a finite chain with simple factors, the length is the number of factors, and the zero module has length .
Module length is additive in short exact sequences: for , the module has finite length if and only if and do, and then .
Simple module: a nonzero module with no proper nonzero submodule: a module is simple when it is nonzero and has no proper nonzero submodule.
A local ring is a nonzero commutative ring with a unique maximal ideal: a local ring has exactly one maximal ideal.
Left and right Noetherian rings: in a Noetherian ring every ideal is finitely generated.
Tor from a projective resolution of the right module: for a right module with a specified projective resolution one sets .
The balanced Tor bifunctor: under Dependent Choice, is the balanced bifunctor obtained from either a resolution of or one of , identified by the left-right comparison theorem.
The left and right projective constructions of Tor are naturally isomorphic: under Dependent Choice there is a natural isomorphism for supplied projective resolutions.
The recursion theorem: for a set , an element and a function there is a unique with and .
The axiom of dependent choice: a relation in which every element is related to something admits an -indexed chain: for every nonempty set , every entire relation on and every there is with and for all .
The Axiom of Choice: every family of nonempty sets has a choice function.
regular local residue field koszul resolution: under the Axiom of Choice, for a regular local ring of dimension the Koszul complex on any regular system of parameters is a minimal free resolution of of length .
Koszul Complex Of A Sequence With Coefficients: has degree- term and differential .
Regular Sequences Give Acyclic Koszul Complexes: every finite -regular sequence is -Koszul-regular, that is for .
regular local rings are domains and cohen macaulay: under the Axiom of Choice, for a regular local ring of dimension every regular system of parameters is a regular sequence and is regular local of dimension for ; in particular an initial segment of a regular system of parameters is a regular sequence.
Tensoring is right exact: tensoring an exact sequence with a module preserves exactness at the right.
Proof
AC gives DC, so the Dependent-Choice suppliers [F1], [F11] and [F12] are available. Given a nonempty set , an entire relation on and , apply [F15] to the family of nonempty subsets of to obtain with for every nonempty , and put , a function because is entire. By [F13] there is with and ; then for all because . This is exactly the statement of [F14]. Supply a free resolution of by mapping the free module on its underlying set onto , then repeating this construction on each successive kernel; recursion gives the required resolution for [F1].
Ideals of finite colength. Let be an ideal containing for some . For , and . For , has finite length over : the ring carries the finite chain whose successive quotients are ; each is finitely generated over the Noetherian ring by [F9], so each is a finitely generated module over the field , hence a finite-dimensional -vector space, which has a finite composition series with simple factors and therefore finite length by [F5]; a finite extension of modules of finite length has finite length with additive length by [F6], so , and , being a quotient of , has finite length as well.
Vanishing on finite length. Suppose with . Then for every -module of finite length : prove this by induction on . For we have . If , choose a proper submodule that is maximal for inclusion, which exists because has finite length; then is simple by [F7], so choosing presents it as , and is a maximal ideal of , because for a proper ideal the submodule is nonzero and hence all of , forcing and . By [F8] the only maximal ideal is , so . By [F6] , so the induction hypothesis applies to , and the exact sequence from [F1] has vanishing outer terms, whence .
Injectivity for finite colength ideals. Let be any ideal. Since the free module with the resolution concentrated in degree satisfies by [F10], the long exact sequence of [F1] for exhibits as the image of . Hence if and , then has finite length by step 1.2 and by step 2.1, so is injective.
The diagram chase. Let be a finitely generated ideal and let . For every the sequence , with maps and , is exact, so after tensoring with and using [F20] the sequence is exact; the vertical maps to are the multiplication maps, which are injective on and on because both ideals contain and step 3.1 applies. Given , its image in the middle maps to zero in and hence, by exactness at the middle, equals for some ; then maps to in and to in , so and lies in the image of .
Concluding . Apply Artin–Rees [F3] over the Noetherian ring to the finite module , its submodule and the ideal . It gives such that for every . Put , a finite -module because is finite over and is finite over . By step 4.1 and this inclusion, for every : an elementary tensor with equals . The local homomorphism gives , so Krull intersection [F4] on the finite -module gives . Hence .
Since was an arbitrary finitely generated ideal and is injective, [F2] shows that is flat over . This proves the first assertion.
The regular-parameter case. Let be a regular system of parameters of and let be their images, extending to a regular system of parameters of . By [F19] the tuple is -regular, hence so is its initial segment . By [F16] the Koszul complex is a free resolution of , and tensoring its defining formulas, [F17], with replaces each by and reproduces the Koszul complex of [F17] term by term, so as complexes. Therefore, using [F10] for the right-resolution construction and [F12] (available by step 1.1) to identify it with the balanced Tor of [F11], by [F18], as is an -regular sequence. The module is a finite -module, so the first assertion of this lemma, applied to , gives that is flat over .
Standard smooth algebras are finitely presented and flat
Statement
Assume the Axiom of Choice (The Axiom of Choice). Let be a commutative ring and let be a standard smooth -algebra (Standard smooth presentations and locally standard smooth maps), so that for some , some , and with the leading Jacobian minor mapping to a unit of . Then:
- is a finitely presented -algebra (Finitely presented modules and finitely presented algebras);
- is flat over .
No hypothesis is placed on : it may be non-Noetherian, and it may have zero divisors.
Facts & Assumptions
Given: A commutative ring , a standard smooth presentation whose leading minor maps to a unit of , and the Axiom of Choice. Write and .
Standard smooth presentations and locally standard smooth maps: a standard smooth presentation consists of integers , elements and with , such that the Jacobian matrix has a minor whose image in is a unit; is the relative dimension, and the invertible minor may be assumed to lie in the first columns.
Base change of standard smooth presentations: for any ring map and a standard smooth presentation as above, there is a unique -algebra isomorphism sending to , and the target is standard smooth over with the same and relative dimension, its minor still a unit.
Invertible Jacobian minor gives regular parameters in a polynomial fibre: under the Axiom of Choice, if is a field, is prime and have leading Jacobian minor , then the classes of in are linearly independent, is regular local, is a regular sequence in it and the quotient is regular local of dimension .
Local flatness criterion by regular parameters: under the Axiom of Choice, for a local homomorphism of Noetherian local rings and a finite -module with , the module is flat over ; the module is not assumed finite over .
Finitely presented modules and finitely presented algebras: a commutative -algebra is finitely presented when for some and a finitely generated ideal ; the boundary values and are admitted.
Universal property of a polynomial ring on an arbitrary family of indeterminates: for a ring homomorphism and a family in there is a unique ring homomorphism restricting to with .
A ring homomorphism whose kernel contains a two-sided ideal factors uniquely through the quotient ring: a ring homomorphism whose kernel contains an ideal factors uniquely through .
Universal property of localisation: maps that invert factor uniquely through : a unital homomorphism carrying a multiplicative set into the units of factors uniquely through .
Multiplicative subsets and the localisation as equivalence classes of fractions: the localisation of a commutative ring at a multiplicative set consists of the classes , and with every a unit.
A module is flat if and only if all prime localizations are flat, equivalently all maximal localizations are flat: for an -module , is flat over if and only if is flat over for every prime , equivalently for every maximal ideal.
Flatness is equivalent to preserving injections and to the ideal and finitely generated ideal tests: is flat over if and only if is injective for every finitely generated ideal .
Every localization is flat, and localizing a flat module preserves flatness: is a flat -algebra, and if is flat over then is flat over .
Under the stated choice boundary, free modules are projective and hence flat: for every commutative ring and free -module , is flat over regardless of choice.
Localisation of modules is extension of scalars: for a commutative ring , a multiplicative set and an -module there is an isomorphism , .
Tensoring is right exact: tensoring an exact sequence with a module preserves exactness; tensoring preserves cokernels and surjections.
Symmetry and associativity isomorphisms for tensor products over a commutative ring: tensor products of modules over a commutative ring are commutative and associative, so .
For a finite module, support is the set of primes containing the annihilator: for a finitely generated module over a commutative ring , ; in particular a finitely generated module whose localisations at all maximal ideals vanish is zero, because a proper ideal lies in a maximal ideal.
In a nonzero commutative ring, every proper ideal is contained in a maximal ideal: every proper ideal of a nonzero commutative ring is contained in a maximal ideal.
Finitely generated modules over a left Noetherian ring are Noetherian: submodules of finitely generated modules over a Noetherian ring are again finitely generated.
Every quotient and every localisation of a Noetherian ring is Noetherian: quotients and localisations of a Noetherian commutative ring are Noetherian.
If is Noetherian then is Noetherian for every : for a Noetherian commutative ring and the polynomial ring is Noetherian.
A commutative ring is Noetherian exactly when every ideal is finitely generated, exactly when its ideals satisfy the ascending chain condition, and exactly when every nonempty set of ideals has a maximal member: a commutative ring is Noetherian if and only if every ideal is finitely generated.
Every subgroup of is for exactly one natural number : every subgroup of is generated by one element.
Every algebra of finite type over a Noetherian ring is a Noetherian ring: a commutative algebra of finite type over a Noetherian commutative ring is a Noetherian ring.
The long exact Tor sequence in the left-module variable: under the Axiom of Dependent Choice, a short exact sequence of modules and a module give a natural long exact sequence .
Localisation commutes with kernels images and cokernels: localisation commutes with kernels, images and cokernels of module homomorphisms.
Extension of scalars carries flat modules to flat modules: if is a flat -module and is a ring homomorphism, then is a flat -module.
Equality, vanishing, and the kernel of the localisation map: in a localisation, if and only if for some , and if and only if for some .
Localisation commutes with quotient rings: : for an ideal of and multiplicative , where is the image of in .
A local ring is a nonzero commutative ring with a unique maximal ideal: a local ring is a commutative ring with exactly one maximal ideal; a local homomorphism of local rings is one carrying the maximal ideal of into that of .
The axiom of dependent choice: a relation in which every element is related to something admits an -indexed chain: for every nonempty set , every entire relation on and every there is with and for all .
The Axiom of Choice: every family of nonempty sets has a choice function.
The polynomial ring as finitely supported coefficient families on monomials: is the commutative -algebra of polynomials in the indeterminates; a ring map induces a ring map sending each coefficient, and this map is injective when is injective.
The Krull intersection is the -torsion submodule, and it vanishes in the Jacobson-radical case: if is a Noetherian commutative ring, is an ideal and is a finite -module, then .
Proof
The encoded presentation. Let . The composite sends to a unit of , so by [F6] and [F7], and then [F8], there is a unique -algebra homomorphism with and ; here is the new variable and the relations , hold. Conversely, the substitution , gives a ring homomorphism whose kernel contains because each maps to , and which sends to a unit of with inverse ; hence [F7] and [F8] produce a unique -algebra homomorphism with . The two composites fix the generators of over and the generators of , so by the uniqueness clauses of [F6], [F7] and [F8] they are the respective identities; thus .
Reduction of flatness to the Noetherian local case. Assume first that is Noetherian. By [F10] it suffices to prove that is flat over for every prime ; by [F2] the algebra is standard smooth over with the same and a minor that is still a unit. Hence it suffices to prove: if is a Noetherian local ring and is standard smooth over , then is flat over .
Dependent choice is available. Given the Axiom of Choice [F33], let be a nonempty set with an entire relation and let ; choosing an element of each nonempty subset of and setting to be the chosen element of gives a function , so recursion produces with and . Thus the Dependent Choice supplier [F26] is available throughout this proof.
The local criterion, prepared. Let now be Noetherian local with maximal ideal and residue field , and let be standard smooth over . Then is Noetherian: is Noetherian by [F21], is Noetherian by [F20] and so is its localisation by [F20]. Fix a finitely generated ideal and put ; this is a finitely generated -module by [F19], since is a finitely generated -module. For every maximal ideal the localisation equals the kernel of , by [F27] and [F14] applied to the localisation . Hence if is flat over , then ; and if that holds for every maximal , then by [F17] and [F18]. Therefore, by [F11], it is enough to prove that is flat over for every maximal ideal .
Set-up at the contracted prime. Fix a maximal ideal and put . Before the local computation, replace the base and presentation by and , using [F2]. The ideal induces a maximal ideal there with the same local ring . In steps 1.6, 2.2, 3.1 and 4.1 only, write for this localized base, its maximal ideal and residue field, and its polynomial ring. Let be the prime over , so , and . Put and ; then by [F30]. The module is flat over this base: tensoring an injection with the free -module preserves injectivity by [F13], and localizing the result preserves it by [F27], with the tensor identifications of [F14, F16]. Each is Noetherian local, and makes local. Once the computation proves flat over , it is flat over the original base by clause 2 of Every localization is flat, and localizing a flat module preserves flatness.
The fibre at . The quotient map tensored with has cokernel by [F15]; localising this at the prime induced by and applying [F14] twice together with [F16] gives , where denotes the image of in .
is Noetherian. Every ideal of is an additive subgroup, hence generated by one element by [F23] and therefore finitely generated; so is Noetherian by [F22].
The unit witness for the minor. Since maps to a unit of , there are and with in , that is, in ; by [F29] applied to the localisation there is with . Putting and gives , so there are with in .
Finite presentation. By step 1.1 the -algebra is isomorphic to the quotient of the polynomial ring by the ideal generated by the finitely many elements ; by [F5] this exhibits as a finitely presented -algebra.
The minor survives in the fibre. The element maps to a unit of , hence its image in the localisation is a unit, hence its image in is a unit. By step 1.6 with this ring is , and the image of there is the image of ; since a unit of a local ring does not lie in the maximal ideal, . So [F3] applies over the field : the images form a regular sequence in the regular local ring , and consequently is a nonzerodivisor on for every . By step 1.6 this says that is a nonzerodivisor on for every .
Descent to a finitely generated subring. Let be the -subalgebra generated by the finitely many coefficients occurring in the polynomials , , , , . Then is a finitely generated -algebra, hence Noetherian by steps 1.7 and [F24]. The polynomial identity of step 1.8 has all its coefficients in , and is injective by [F34], so holds already in ; therefore is a standard smooth -algebra with the same and the minor a unit.
Induction: a fibre nonzerodivisor lifts across a flat map. Suppose is flat over for some , put , and let . By step 2.2, multiplication by the image of on is injective. For every , flatness of over applied to gives the natural isomorphism After choosing a -basis of , this is a direct sum of copies of , so multiplication by is injective on each graded piece . If in , induction on now gives for every : injectivity modulo starts the induction, and injectivity on advances it. The ring is Noetherian local and lies in its maximal ideal by step 1.5, so [F35] gives and therefore . Thus is a nonzerodivisor on , and is a short exact sequence.
Induction: the Tor vanishing and flatness pass to the next quotient. In the situation of step 3.1 the long exact Tor sequence of [F26] for , together with from the flatness of , exhibits as the kernel of , which is zero by step 2.2. The ring is a Noetherian local ring, is a local homomorphism by step 1.5 and is a finite -module, so [F4] gives that is flat over .
Conclusion of the induction and of the local case. Steps 3.1 and 4.1, starting with of step 1.5, prove that every is flat over the localized base . In particular the original local ring is flat over , hence over the original by step 1.5. As was arbitrary, step 1.4 gives that is flat over the original base. This proves flatness over every Noetherian local base, and step 1.2 proves it over every Noetherian base.
Flatness of and base change back to . By step 2.3 the algebra is standard smooth over the Noetherian ring , so is flat over by step 5.1; and by [F2] applied to the ring map there is an -algebra isomorphism . Hence is flat over by [F28]. This proves the second assertion for arbitrary , and step 2.1 proves the first.
Fibres of standard smooth algebras are regular of relative dimension
Statement
Assume the Axiom of Choice (The Axiom of Choice). Let be a commutative ring and let be a standard smooth -algebra (Standard smooth presentations and locally standard smooth maps), presented as with leading Jacobian minor mapping to a unit of . Let , put , and let be a field extension. Write so that, by base change of the presentation, , where are the images of in . Then:
- every local ring of at a prime is a regular local ring; if is the prime corresponding to , then (The height of a prime ideal);
- every irreducible component of has dimension ; equivalently, for every minimal prime of one has .
Both clauses are vacuous when is the zero ring, and no hypothesis is placed on or on the field extension . The relative dimension is the dimension of the components, not the dimension of every local ring: a local ring at the generic point of a component has dimension .
Facts & Assumptions
Given: A commutative ring , a standard smooth presentation with leading minor a unit of , a prime , a field extension , and the Axiom of Choice.
Standard smooth presentations and locally standard smooth maps: a standard smooth presentation consists of integers , elements with , such that the Jacobian matrix has a minor whose image in is a unit; is the relative dimension, and the invertible minor may be assumed to be the leading one, in the first columns.
Base change of standard smooth presentations: for a ring map and a standard smooth presentation as above, with the same , and the image of is again a unit.
Invertible Jacobian minor gives regular parameters in a polynomial fibre: under the Axiom of Choice, if is a field, is prime and have leading Jacobian minor , then is regular local, is a regular sequence in it, and the quotient is regular local of dimension .
Minimal primes are exactly the primes of height zero: a minimal prime ideal of a commutative ring has height .
Equality, vanishing, and the kernel of the localisation map: for a multiplicative set and , the class is zero in if and only if for some ; a fraction equals if and only if for some .
Prime ideals of a localization are exactly the primes disjoint from the denominator set: for a multiplicative set , contraction along is an inclusion-preserving bijection from onto the primes of disjoint from , with inverse .
Localising twice is localising once at the multiplicative set generated by both denominator sets: for multiplicative sets with images in and the multiplicative set generated by , there is a unique -algebra isomorphism .
Localisation commutes with quotient rings: : for an ideal and a multiplicative set , the image of in gives a canonical isomorphism , with both sides zero when .
Height plus quotient dimension equals ambient dimension in an affine domain: under the Axiom of Choice, for a field , a finite-type -domain and one has .
A polynomial ring in n variables over a field has dimension n: for a field and , .
Irreducible components of the spectrum correspond to minimal prime ideals: under the Axiom of Choice, the irreducible components of are exactly the closed sets for minimal primes of , each minimal prime giving a unique component.
The spectrum of a quotient is a closed subspace: for an ideal , contraction along is a homeomorphism from onto .
The height of a prime ideal: the height of a prime ideal is .
Affine-domain dimension equals transcendence degree: a finite-type domain over a field has dimension .
Proof
Set and for the image of in . By [F2] applied to the fibre is , and applying [F2] again to the ring map shows that , with the image of in a unit; the relative dimension is unchanged throughout.
Primes of correspond under [F8] to the primes with , and these in turn correspond to the primes with and . For such a one has : since the image of in is a unit, there are and with in , so by [F7] there is with in ; if lay in , hence in , the contradiction with would follow.
Let be a prime of , let be the prime it contracts to and let be the corresponding prime; then . Indeed , localising further at gives by [F9], and [F10] identifies with the quotient of by the ideal generated by the .
Regularity of local rings. In the situation of step 3.1 the elements lie in the prime and, by step 2.1, have leading minor ; so [F3] applies over the field and shows that is a regular local ring with , the last equality by [F15]. This proves clause 1, including for , where the empty leading minor is and the quotient is .
Height of the ambient prime at a component. Let be a minimal prime of , let be its contraction to , and let be the corresponding prime of . The prime is minimal in : a prime strictly below it avoids and would localize to a prime strictly below by [F8]. The local ring has dimension zero by [F5, F15], so clause 1, already proved in step 4.1, gives . Thus , also when .
Component dimension. Put , a finite-type -domain, and let be the image of . Then , and [F10] gives . The localization is again a finite-type -domain (adjoin with ), and has the same fraction field as . Applying [F16] to both rings gives . Now [F11, F12] and step 5.1 give . By [F13, F14], the component is homeomorphic to , so has dimension .
Both clauses hold: every local ring of is regular local of dimension by step 4.1, and every irreducible component has dimension by step 6.1; if is the zero ring there are no primes and no components, so both clauses are vacuous. ∎
Regularity ascends and descends along a flat local homomorphism
Statement
Assume the Axiom of Choice (The Axiom of Choice). Let be a flat local homomorphism (Flat and faithfully flat modules and ring homomorphisms) of Noetherian local rings, so that is again a Noetherian local ring. Then:
- ascent. if is a regular local ring and the closed fibre is regular, then is regular;
- descent. if is regular, then is regular.
No regularity of the closed fibre is assumed in the descent statement, and no finiteness of the field extension over is assumed anywhere.
Facts & Assumptions
Given: A flat local homomorphism of Noetherian local rings and the Axiom of Choice.
embedding dimension and regular local ring: for a nonzero commutative Noetherian local ring one has , and is regular local exactly when .
regular system of parameters: a regular system of parameters of a regular local ring of dimension is an ordered minimal generating tuple of its maximal ideal, of length ; the empty tuple when .
quotient and lifting regularity across a regular element: under the Axiom of Choice, if is nonzero Noetherian local, is a nonzerodivisor and is regular, then is regular.
A flat ring map is faithfully flat exactly when it detects proper ideals and is surjective on spectra: under the Axiom of Choice, a flat ring homomorphism is faithfully flat if and only if for every proper ideal the extended ideal is proper.
Localisation And Faithfully Flat Base Change Of Regular Sequences: a regular sequence remains regular after faithfully flat base change.
regular local residue field projective dimension dimension: under the Axiom of Choice, for a regular local ring of dimension one has .
finite local modules admit minimal free resolutions: under the Axiom of Choice, every finite module over a nonzero Noetherian local ring has a resolution by finite-rank free modules.
Flat and faithfully flat modules and ring homomorphisms: an -module is flat when preserves exact sequences, and a ring map is flat when the target is flat as a module over the source.
Tensoring is right exact: is right exact, so applying it to gives .
Projective dimension at most n iff the nth syzygy is projective: for and a projective resolution , one has if and only if the -th syzygy is projective.
Every projective module over a commutative ring is flat: every projective module over a commutative ring is flat, with no use of the Axiom of Choice.
Finitely generated modules over a left Noetherian ring are Noetherian: every finitely generated module over a Noetherian ring is Noetherian, so its submodules are finitely generated.
Flatness descends along faithfully flat base change: for a faithfully flat ring map and an -module , the module is flat over if and only if is flat over .
A finite flat module over a Noetherian ring is finite projective: a finite flat module over a Noetherian commutative ring is finite projective.
auslander buchsbaum serre regularity criterion: under the Axiom of Choice, a nonzero Noetherian local ring is regular if and only if its global dimension is finite, and then the global dimension equals the projective dimension of the residue field and equals the dimension.
Every faithfully flat ring map is injective: a faithfully flat ring homomorphism is injective.
regular local rings are domains and cohen macaulay: under the Axiom of Choice, a regular local ring is a domain, and every regular system of parameters is a regular sequence.
A Noetherian local domain has dimension zero exactly when it is a field: a Noetherian local domain of dimension zero is a field.
In a nonzero commutative ring, every proper ideal is contained in a maximal ideal: under the Axiom of Choice every proper ideal of a commutative ring is contained in a maximal ideal.
A local ring is a nonzero commutative ring with a unique maximal ideal: a local ring has a unique maximal ideal, which contains every proper ideal.
Field: a field is a nonzero commutative ring in which every nonzero element is a unit.
Proof
The map is faithfully flat. Since the homomorphism is local, ; for every proper ideal the ideal is contained in the maximal ideal by [F20], so is proper. Hence is faithfully flat by [F4].
Descent, the case : setting up the resolution. Assume regular of dimension and put . By [F7] choose a resolution by finite free -modules and let be its -th syzygy, a finitely generated -module by [F12]. Tensoring with the flat -module preserves exactness by [F8], and [F9] identifies the tensor of the augmentation with , so is a free resolution of the -module whose -th syzygy is .
Ascent, set-up. Assume regular of dimension with regular system of parameters , so that by [F2]; if then and is regular by hypothesis. For , the parameters are an -regular sequence by [F17]; hence [F5] and the faithful flatness of step 1.1 make an -regular sequence, and .
Descent, the case . Assume now that is regular of dimension . Then is a domain by [F17] and hence a field by [F18]. Because is faithfully flat by step 1.1, the extension is proper by [F4], and is an ideal of the field , so ; moreover is injective by [F16], so . A nonzero element of the local ring is then a unit: otherwise would be a proper ideal, hence would lie in a maximal ideal by [F19], necessarily the unique maximal ideal , forcing . Thus is a field by [F21], in particular regular.
Descent, the case : the syzygy is projective. Since is regular local of dimension , its global dimension is by [F15], so every -module, in particular , has projective dimension at most ; by [F10] applied to the resolution of step 1.2, the -module is projective, hence flat over by [F11].
Ascent, induction. For put , a nonzero Noetherian local ring, so that , and for with the image of a nonzerodivisor on . If is regular for some , then [F3] applied to the nonzero Noetherian local ring and the nonzerodivisor makes regular. Since is regular by hypothesis when , downward induction gives that is regular; together with the case of step 2.1 this proves the ascent claim.
Descent, conclusion. By [F13] and the faithful flatness of step 1.1, the finite -module is flat over , hence finite projective by [F14]. Therefore the resolution of step 1.2 has projective -th syzygy, so by [F10], and [F15] makes the Noetherian local ring regular. Combined with step 2.2 this proves the descent claim for every .
Both claims are proved: the ascent in step 3.1 and the descent in steps 2.2 and 3.2; the case was separated out in steps 2.2 and 3.2 because [F10] requires . ∎
Separable generation after finite purely inseparable extensions
Statement
Assume the Axiom of Choice (The Axiom of Choice). Let be a finitely generated field extension (Finitely generated field extensions ) whose characteristic is . There are finite purely inseparable extensions and fitting into a commutative square of field embeddings such that is separably generated (Separating transcendence basis and separably generated extensions for the separability terminology). No perfectness of is assumed.
In characteristic the same conclusion holds with and , since a finitely generated extension of a perfect field is separably generated; the statement above is the positive-characteristic case, where and may both be nontrivial.
Facts & Assumptions
Given: A finitely generated field extension of characteristic and the Axiom of Choice.
Separating transcendence basis and separably generated extensions: for a finitely generated extension , a finite tuple is a separating transcendence basis when it is a transcendence basis and is finite separable; is separably generated when it admits such a tuple.
Algebraic and transcendental elements and algebraic extensions: an element is algebraic over a subfield when it satisfies a nonzero polynomial over it, transcendental otherwise, and a set is algebraically independent when it satisfies no nonzero polynomial relation.
A maximal algebraically independent set is a transcendence basis: an algebraically independent subset maximal for inclusion is a transcendence basis.
An extension generated by finitely many algebraic elements is finite: a finitely generated algebraic field extension is finite.
Separable algebraic elements and separable extensions: is separable over when it is algebraic with separable minimal polynomial, and is separable when every element of is separable over .
The separable closure of the base inside an algebraic extension: for algebraic , the separable closure is the largest intermediate field separable over .
An algebraic extension is purely inseparable over its separable closure: is purely inseparable.
Pure inseparability and its conjugate, embedding, and separable-degree criteria: for algebraic of characteristic , is purely inseparable if and only if every has for some ; a finite extension is purely inseparable exactly when .
, and in positive characteristic the inseparable degree is a power of : for finite , and is a power of in characteristic .
For a finite extension, : for finite .
If is not a th power in a characteristic- field, then is irreducible for every : if has characteristic and is not a -th power, then is irreducible in for every .
Repeated roots in extension fields and separable polynomials: a polynomial is separable over when it has no repeated root in any extension field.
A nonzero polynomial over a field is separable exactly when its gcd with its derivative is : is separable if and only if .
The evaluation kernel and the unique monic irreducible minimal polynomial of an algebraic element: if and only if the minimal polynomial of over divides .
The binomial theorem over an arbitrary commutative ring: in every commutative ring.
A prime divides for : for , so in characteristic the binomial theorem gives and, more generally, .
Pure inseparability is transitive in towers and stable under composita: purely inseparable extensions compose, and the compositum of purely inseparable subextensions of a common algebraic extension is purely inseparable over the base.
Tower law for finite extensions: : for with and finite, .
Perfect fields: every irreducible polynomial is separable, A field is perfect exactly when it has characteristic zero or its Frobenius map is surjective and Fields of characteristic zero, finite fields, and algebraically closed fields are perfect: a field of characteristic is perfect, and in characteristic perfectness is equivalent to every element being a -th power.
Finitely generated extensions of a perfect field are separably generated: every finitely generated extension of a perfect field is separably generated.
Finitely generated field extensions : is finitely generated when for finitely many elements.
Proof
Reduction and set-up. If the characteristic is , then is perfect by [F19], so [F20] makes separably generated and , are finite purely inseparable over their bases by [F8] (in characteristic the only purely inseparable extension is the trivial one). Assume from now on that the characteristic is . Fix a finite generating set of over by [F21] and choose a maximal algebraically independent subset of it, which is a transcendence basis of by [F3, F2]. Put ; every is algebraic over by maximality, so is finitely generated algebraic, hence finite by [F4]. Let be the separable closure of in ; then is finite separable by [F6, F4] and is purely inseparable by [F7], finite by [F18], so that is a nonnegative integer by [F9, F8].
An element of with a -th root of the right shape. Assume , so . By [F8] and [F7] applied to an element of there is and a minimal with ; then satisfies by minimality of and . Since is separable over by [F5, F6], its minimal polynomial over is separable by [F12, F5], hence by [F13] and has pairwise distinct roots; write with .
A finite purely inseparable base change making the coefficients -th powers. Write each with and , and let be the finite set of all coefficients occurring in the finitely many polynomials . Choose an algebraic closure of and inside it put , so that is finite purely inseparable by [F17, F8]; put , the compositum of and . Each monomial with is a -th power in : writing with one has . Each equals , so every and is a finite sum of -th powers, hence a -th power by [F16], and therefore each is a -th power in . Write with and put .
The -th root of is separable. By [F15] and [F16], in . Let be the distinct roots of (so , and ), and for each choose with , which exists because is algebraically closed. Then , so , and the are distinct because are. Hence has distinct roots in , so is separable over by [F12] and [F13] applied to its distinct-root factorisation. Since is a root of , the element satisfies , hence ; and , a compositum which is finitely generated over . As is separable with the root , the minimal polynomial of over divides by [F14] and is separable, so is separable over by [F5] and lies in the separable closure of in by [F6].
The case . If then , so and is finite separable by [F6]; then is a separating transcendence basis of by [F1], and with , the conclusion holds trivially.
The separable closure of in and the degree drop. is finite purely inseparable and is purely inseparable. First, is separable: is finite separable by step 1.1, and a compositum of a separable algebraic extension with a further extension is separable because the minimal polynomial over the larger field divides the separable minimal polynomial over the smaller one by [F14]. Second, is purely inseparable by [F17]. Hence : one inclusion holds because is separable and is the largest separable intermediate field by [F6], and for the other, an element is separable over and has for some by [F8], so it is simultaneously separable and purely inseparable over and therefore already lies in . Since and , the field satisfies : the polynomial vanishes at while is not a -th power in (a relation would give , hence by [F16]), so is irreducible by [F11] and is the minimal polynomial of over by [F14]. Put , an intermediate field of containing , and choose with , possible since is finite by [F18]. Then equals , and for each the minimal polynomial of over divides its minimal polynomial over by [F14], so those two extensions satisfy the degree inequality ; multiplying over and applying [F18] twice yields , the last equality being [F18] for the tower .
Induction. The extension is finitely generated by [F21] and has characteristic , and by step 2.2 its invariant is strictly smaller than . Applying the induction hypothesis (on the nonnegative integer , with the same statement for the pair ) produces finite purely inseparable extensions and with separably generated. Then is finite purely inseparable by [F17, F9] and is finite purely inseparable by [F17] since is, so and satisfy the conclusion for . The base case of the induction is step 2.1, so the assertion holds for every .
Conclusion. In characteristic steps 1.2–1.4, 2.1, 2.2 and 3.1 produce the required finite purely inseparable and with separably generated, the induction being on the integer of step 1.1; the characteristic case is step 1.1. ∎
Field tests for geometric regularity
Statement
Assume the Axiom of Choice (The Axiom of Choice). Let be a field and let be a finite-type -algebra (Geometrically regular algebras and geometrically regular fibres). Then:
- is geometrically regular over if and only if is a regular ring (regular noetherian ring) and is regular for every finite purely inseparable field extension ;
- if is geometrically regular over , then is a regular ring for every field extension , not only for the finitely generated ones appearing in the definition;
- conversely, if is a field extension such that is geometrically regular over , then is geometrically regular over .
Clause 1 is the finite purely inseparable test, clause 2 removes the finite generation from the scalar extension, and clause 3 is descent of geometric regularity along the faithfully flat field extension . The zero algebra is regular vacuously and all clauses hold for it.
Facts & Assumptions
Given: A field , a finite-type -algebra , and the Axiom of Choice.
Geometrically regular algebras and geometrically regular fibres: is geometrically regular over when is a regular Noetherian ring for every finitely generated field extension ; the condition on a single scalar extension is tested at primes, and geometric regularity of over does not presuppose that is regular, which is why clause 1 of the statement carries regularity of as a separate hypothesis.
regular noetherian ring: a commutative Noetherian ring is regular when every prime localisation is a regular local ring, the zero ring being regular vacuously; the maximal-ideal test is proved with the localisation and polynomial-extension theorem.
Separable generation after finite purely inseparable extensions: under the Axiom of Choice, for a finitely generated field extension of characteristic there are finite purely inseparable extensions and with separably generated, and in characteristic one may take , .
localisation and polynomial extension of regular rings: under the Axiom of Choice, localisations and finite polynomial extensions of a commutative regular Noetherian ring are regular, and regularity may be tested at maximal ideals.
Separating transcendence basis and separably generated extensions: is separably generated when it has a finite transcendence basis with finite separable.
A finite extension generated by elements all but possibly one of which are separable is simple: a finite extension generated by elements all but at most one of which are separable is simple; in particular a finite separable extension is simple.
Regularity ascends and descends along a flat local homomorphism: under the Axiom of Choice, for a flat local homomorphism of Noetherian local rings: if and are regular then is regular, and if is regular then is regular.
A nonzero polynomial over a field is separable exactly when its gcd with its derivative is : is separable if and only if .
Every nonzero nonunit polynomial over a field factors into irreducible polynomials and For a nonconstant in , the ideal is maximal and is a field exactly when is irreducible: a nonzero nonunit polynomial over a field is a product of irreducibles, and the quotient by a nonconstant polynomial is a field exactly when that polynomial is irreducible.
Tensoring is right exact: is right exact, so for a field extension .
auslander buchsbaum serre regularity criterion: under the Axiom of Choice, a nonzero Noetherian local ring is regular exactly when every finite module over it has finite projective dimension, and then its global dimension equals its dimension.
Noether normalisation yields module finiteness over a polynomial subring: a nonzero finite-type algebra over a field is module-finite over a polynomial ring .
Injective integral extensions preserve Krull dimension and A polynomial ring in n variables over a field has dimension n: an injective integral extension of nonzero commutative rings preserves dimension, and .
Localisation does not increase Krull dimension: localising a commutative ring does not increase its dimension.
Projective dimension of an object and Left and right global dimension of a ring: is the infimum of the lengths of projective resolutions of , and the global dimension of a ring is the supremum of the projective dimensions of its modules.
Extension of scalars carries flat modules to flat modules and Every localization is flat, and localizing a flat module preserves flatness: extension of scalars along a flat ring map preserves flatness, and localisations are flat.
Proof
The forward implication of clause 1 is immediate: taking in [F1] exhibits itself as regular, and every finite purely inseparable is a finitely generated field extension, so [F1] makes regular.
A base-change fact used below: if is a regular Noetherian -algebra and is a separably generated field extension, then is regular. Indeed, choose a separating transcendence basis and put , so that is finite separable by [F5] and by [F6] with minimal polynomial satisfying by [F8]. The ring is a localisation of the polynomial extension , hence regular by [F4]; let be a prime of and , so that is a flat local homomorphism of Noetherian local rings whose closed fibre is a localisation of by [F10]. Reducing the Bezout identity for modulo shows , so is separable by [F8] and factors into pairwise distinct irreducibles by [F9], whence is a product of fields and its further localisation is a field, in particular regular; [F7] then makes regular. As was arbitrary, is regular by [F2].
The finite purely inseparable test implies the finitely generated case of clause 1: assume is regular and is regular for every finite purely inseparable , and let be finitely generated. By [F3] there are finite purely inseparable extensions and with separably generated (in characteristic take both extensions trivial). The ring is regular by hypothesis, so step 1.2 applied to and the separably generated extension makes regular. Since is finite purely inseparable, the field extension is faithfully flat and local, and for every prime of and every prime of over it the induced map is a flat local homomorphism of Noetherian local rings with regular target, so [F7] descends regularity; as was arbitrary, is regular by [F2]. With step 1.1 this proves clause 1, since [F1] defines geometric regularity by the regular rings over finitely generated .
Arbitrary field extensions. Assume is geometrically regular over and let be any field extension, written as the filtered union of its finitely generated subextensions . Then is a filtered union, and each is regular by clause 1 as proved in step 2.1. Let be a prime of and put and with , so that is a filtered union of regular local rings. There is a uniform bound on dimensions: for , if is module-finite over by [F12], then base change along presents as a quotient of a finite free -module by [F10], so it is integral over and has dimension by [F13]; hence by [F14]. For the ring is the zero ring, regular by [F2].
The filtered union is regular. In the notation of step 3.1 let be a finitely generated -module; presenting by finitely many generators and relations, all structure constants lie in some stage, so for a finitely generated -module and some . By [F11] the regular local ring of dimension at most has , so has a projective resolution of length at most by [F15]; tensoring it with , which is flat over because it is a localisation of the flat base change by [F16], gives an exact sequence of projective -modules of length at most (a direct summand of a free module stays such after tensoring), hence by [F15]. As every finitely generated -module has finite projective dimension, [F11] makes regular; therefore is regular by [F2], which is clause 2.
Descent, clause 3. Let be a field extension with geometrically regular over , and let be a finitely generated field extension; we show is regular. The -algebra is nonzero, so choose a prime of it and let be its residue field, a field receiving both and for which is a finitely generated field extension. Then is regular by clause 2, applied over to the geometrically regular -algebra ; and is faithfully flat and local at corresponding primes because is a field extension, so [F7] descends regularity to every local ring of , making it regular by [F2]. As was arbitrary, is geometrically regular over by [F1].
Clause 1 is step 2.1, clause 2 is step 4.1 and clause 3 is step 5.1; the zero algebra is covered by the vacuous regularity of [F2]. ∎
Regular algebras over a perfect field are geometrically regular
Statement
Assume the Axiom of Choice (The Axiom of Choice). Let be a perfect field (Perfect fields: every irreducible polynomial is separable) and let be a finite-type -algebra that is regular (regular noetherian ring). Then is a regular ring for every field extension , not only for the finitely generated ones. In particular a regular finite-type algebra over a perfect field is geometrically regular (Field tests for geometric regularity).
No assumption is made on the extension field : it need not be perfect. In characteristic the hypothesis that is perfect is automatic, and the statement says that regular finite-type algebras over such fields stay regular after arbitrary scalar extension.
Facts & Assumptions
Given: A perfect field , a regular finite-type -algebra , and the Axiom of Choice.
A field is perfect exactly when it has characteristic zero or its Frobenius map is surjective: is perfect if and only if either , or and the Frobenius map is surjective; iterating, every element of a perfect field of characteristic is a -th power for every .
Perfect fields: every irreducible polynomial is separable: a field is perfect when every algebraic extension of it is separable, equivalently (in characteristic ) when its Frobenius endomorphism is surjective.
Pure inseparability and its conjugate, embedding, and separable-degree criteria: for algebraic of characteristic , is purely inseparable if and only if every has for some ; in characteristic a purely inseparable extension is trivial.
The binomial theorem over an arbitrary commutative ring and A prime divides for : in a commutative ring of characteristic one has and hence for every ; in a field forces .
Field tests for geometric regularity: under the Axiom of Choice, a finite-type -algebra is geometrically regular over if and only if is regular and is regular for every finite purely inseparable ; and then is regular for every field extension .
regular noetherian ring: a commutative Noetherian ring is regular when all its prime localisations are regular local rings.
Proof
Every finite purely inseparable extension of is trivial. Let be finite purely inseparable. If the characteristic is , then by [F3]. If the characteristic is , fix ; by [F3] there is with , and since the Frobenius map of is surjective by [F1] we may write with . Then by [F4], and is a field, so . Hence .
Applying the field tests. The algebra is regular by hypothesis, and by step 1.1 the only finite purely inseparable extension of is itself, for which is regular. Hence [F5] makes geometrically regular over , and its second clause makes regular for every field extension .
Thus a regular finite-type algebra over a perfect field is geometrically regular, and every scalar extension is regular, whether or not is perfect or finitely generated; in characteristic the perfectness hypothesis is automatic by
Jacobian criterion and openness of the regular locus over a perfect field
Statement
Assume the Axiom of Choice (The Axiom of Choice). Let be a perfect field (Perfect fields: every irreducible polynomial is separable), let with , let be an ideal and put . Fix elements generating and let be the Jacobian matrix (Differentials of a polynomial quotient and the Jacobian cokernel), of size , with entries viewed in .
- Closed-point criterion. Let be a maximal ideal whose residue field is a finite separable extension of , and let be the matrix over obtained by evaluating the entries of . Then is a regular local ring if and only if In particular the rank of does not depend on the chosen generating set of .
- Local charts. If and is regular, then there are and such that is generated by after inverting , some minor of the Jacobian matrix is a unit of , and therefore is a standard smooth -algebra (Standard smooth presentations and locally standard smooth maps). Here (The height of a prime ideal).
- Openness and density. The regular locus is open in . If is a minimal prime of with reduced, then the regular locus contains a dense open subset of ; when is reduced this holds for every irreducible component.
The remaining relations of beyond the chosen ones are killed by a Nakayama argument, so no appeal to a later local-presentation theorem is needed.
Facts & Assumptions
Given: A perfect field , the polynomial ring , an ideal generated by , the algebra , and the Axiom of Choice.
Differentials of a polynomial quotient and the Jacobian cokernel: is free on , and for the module is the cokernel of the Jacobian matrix acting from to .
Separable residue and the cotangent sequence of a local algebra: for a Noetherian local -algebra with maximal ideal and residue field finitely generated and separably generated over , the sequence is exact; if is finite separable, then and the first map is an isomorphism.
Finitely generated extensions of a perfect field are separably generated: every finitely generated field extension of a perfect field is separably generated.
Tensoring is right exact: tensoring is right exact, so the cokernel of the Jacobian matrix base changes to the cokernel of the base-changed matrix.
regular system of parameters equivalent basis: under the Axiom of Choice, in a Noetherian local ring the classes of a regular system of parameters form a basis of , and conversely a lift of any basis generates the maximal ideal as a system of parameters.
regular local regular quotient ideal is parameter generated: under the Axiom of Choice, for a regular local ring of dimension and an ideal , the quotient is regular if and only if is generated by an initial part of a regular system of parameters, if and only if .
localisation and polynomial extension of regular rings: under the Axiom of Choice, localisations and finite polynomial extensions of a regular Noetherian ring are regular; in particular every prime localisation of the polynomial ring is a regular local ring of dimension .
embedding dimension and regular local ring: for a nonzero Noetherian local ring one has , and is regular if and only if .
Fibres of standard smooth algebras are regular of relative dimension: under the Axiom of Choice, for a standard smooth presentation with variables and equations over a commutative ring , every local ring of the base change to any field extension of the residue field of any is regular; over the field and the extension this says that every local ring of a standard smooth -algebra is regular.
Prime ideals of a localization are exactly the primes disjoint from the denominator set: contraction is a bijection from the primes of a localisation onto the primes of the base ring avoiding the multiplicative set, so an element outside a prime stays outside every prime of the localisation at that prime and is a unit after further inverting it.
Minimal primes are exactly the primes of height zero: a minimal prime ideal has height zero, that is, for a minimal prime of .
A Noetherian ring is Artinian exactly when every prime ideal is maximal and An Artinian local ring has nilpotent maximal ideal, and its finite modules have finite length: under the Axiom of Choice, a Noetherian ring is Artinian exactly when all its primes are maximal, and the maximal ideal of an Artinian local ring is nilpotent.
Irreducible topological spaces and irreducible subsets in the subspace topology: a nonempty open subset of an irreducible topological space is dense.
Perfect fields: every irreducible polynomial is separable: a field is perfect when every algebraic extension of it is separable, equivalently when its Frobenius endomorphism is surjective in characteristic .
Proof
The closed-point criterion. Let be maximal with finite separable over . The local ring is a Noetherian local -algebra with residue field , which is finitely generated and separably generated over the perfect field by [F3]; since is finite separable, [F2] gives , and . By [F1] and [F4] the latter is the cokernel of the matrix acting from to , so its dimension is . Hence by [F8], and by [F8] again is regular if and only if , that is, if and only if ; both and are intrinsic, so the rank is independent of the generating set of .
Local charts at regular points. Let with regular, let correspond to , and put , a regular local ring of dimension by [F7], with ideal and quotient of dimension . By [F6] there are forming an initial part of a regular system of parameters of , with , and generating . Lifting the fractions to elements that still generate , the classes of the span modulo with by [F6], and they are linearly independent by [F5]. The map is injective by [F2], the residue field being finitely generated and separably generated over the perfect field by [F3], and the images of the classes of the are the vectors by [F1]; these vectors are therefore linearly independent over , so some minor of the Jacobian matrix of satisfies . Since is finitely generated and , there is with ; put . By [F10] the element is not in and becomes a unit in , and in the ideal is generated by while is a unit, so is a standard smooth -algebra with variables and equations.
Openness of the regular locus. In the situation of step 1.2, [F9] applied over the field shows that every local ring of the standard smooth -algebra is regular; hence the whole distinguished open consists of regular points and is a neighbourhood of . As was an arbitrary regular point, the regular locus is open in .
Density along a generically reduced component. Let be a minimal prime of with reduced. By [F11] the local ring is Noetherian local of dimension , so all its primes are maximal and it is Artinian by [F12]; its maximal ideal is therefore nilpotent by [F12], and reducedness forces it to be zero, so is a field, in particular regular. Step 2.1 then provides with contained in the regular locus, and is a nonempty open subset of the irreducible space , hence dense in by [F13]. When is reduced, every localisation at a minimal prime is reduced, so this applies to every irreducible component.
The three assertions are steps 1.1, 1.2 and 2.1 with the density statement of step 3.1; all of them use only the perfectness of through [F3] and [F2], and no later local-presentation theorem. ∎
Flat maps with geometrically regular fibres have standard smooth local presentations
Statement
Assume the Axiom of Choice (The Axiom of Choice). Let be a ring map of finite presentation. Choose a presentation with and , and let be the preimage of ; put , and Assume
- the local ring homomorphism is flat, and
- the fibre is geometrically regular at (Geometrically regular algebras and geometrically regular fibres): for every field extension and every prime of lying over the prime of corresponding to , the local ring there is regular.
Then there are an integer , elements selected from the chosen generating list for , a polynomial , and a Jacobian minor of these whose image is a unit in , such that, writing for the image of in , there is an -algebra isomorphism Thus is standard smooth over and witnesses that is standard smooth at (Standard smooth presentations and locally standard smooth maps). This is the converse direction of the equivalence between local standard smoothness and flatness with geometrically regular fibres.
Facts & Assumptions
Given: A ring map of finite presentation, a presentation with , a prime with , the primes and lying over , the fibre , flatness of , geometric regularity of at , and the Axiom of Choice.
Geometrically regular algebras and geometrically regular fibres: for a finitely presented -algebra and over , the fibre is geometrically regular at when for every field extension every local ring of at a prime lying over the prime corresponding to is a regular local ring; the fibre is , and .
Standard smooth presentations and locally standard smooth maps: a standard smooth -presentation is a presentation with in which some minor of the Jacobian matrix has image a unit of ; the invertible minor may be taken to be the leading one, in the first columns, and is the relative dimension.
Differentials of a polynomial quotient and the Jacobian cokernel: for the module is free on , the partial derivatives are computed on the monomial basis and extended -linearly, , and the Jacobian matrix governs .
Jacobian criterion and openness of the regular locus over a perfect field: under the Axiom of Choice, for a perfect field , , and a maximal ideal whose residue field is a finite separable extension of , the local ring is regular if and only if , where is the Jacobian matrix of a generating set of evaluated at .
regular local regular quotient ideal is parameter generated: under the Axiom of Choice, for a regular local ring of dimension and an ideal , the following are equivalent: is regular; is generated by an initial part of a regular system of parameters; and .
Assuming the Axiom of Choice, Nakayama's lemma: under the Axiom of Choice, if is a commutative ring, and is a finitely generated -module with , then .
Assuming Choice, every field has an algebraic closure, An algebraic closure of a field and Fields of characteristic zero, finite fields, and algebraically closed fields are perfect: under the Axiom of Choice every field has an algebraic closure; an algebraic closure of a field is an algebraically closed algebraic extension; and every algebraically closed field is perfect.
Height plus quotient dimension equals ambient dimension in an affine domain and A polynomial ring in n variables over a field has dimension n: under the Axiom of Choice, for a field , a finite-type -domain and one has , and .
Flatness is equivalent to preserving injections and to the ideal and finitely generated ideal tests and The long exact Tor sequence in the left-module variable: is flat over exactly when is injective for every ideal ; and under the Axiom of Dependent Choice, a short exact sequence of left -modules and a right module give the long exact sequence ; in particular for flat .
Localisation of modules is extension of scalars and Tensoring is right exact: localisation of modules is given by tensoring with the localised ring, so localising commutes with base change of scalars, and is right exact, so a surjection stays surjective and .
Equality, vanishing, and the kernel of the localisation map and Universal property of localisation: maps that invert factor uniquely through : in a fraction is zero exactly when for some . If a module is finitely generated, then exactly when some annihilates : choose an annihilator outside for each of its finitely many generators and take their product. For a ring map carrying a multiplicative set into the units there is a unique extension to the localisation, so elements of the multiplicative set become units.
The height of a prime ideal and is local with unique maximal ideal : , and for a prime the localisation is a local ring with maximal ideal .
The axiom of dependent choice: a relation in which every element is related to something admits an -indexed chain: the Axiom of Dependent Choice, used only through the Tor long exact sequence of [F9]; it is a consequence of the Axiom of Choice assumed in the statement.
localisation and polynomial extension of regular rings: under the Axiom of Choice, finite polynomial extensions and localizations of a commutative regular Noetherian ring are regular. In particular, a finite polynomial ring over a field and its localization at any prime are regular local at that prime.
Proof
Set-up. Write , , and , so that and the primes , correspond to one another and contract to ; in the fibre, corresponds to the prime and has fraction field because is a domain.
The fibre is regular at its own residue field, so the fibre ideal has generators. Taking for the identity extension in [F1], the local ring is regular local by [F14], since the field is regular Noetherian, and has dimension by [F12], and its quotient is a regular local ring by the hypothesis. Since and every lies in , [F5] applies and shows that is generated by an initial part of a regular system of parameters of ; put so that is generated by elements whose classes in are -independent. The classes of the images of span that -vector space of dimension , so after renumbering, the images of form a basis and generate by [F6] applied to the finitely generated -module .
The -rational point of the fibre and the rank computation. By [F7] choose an algebraic closure ; it is perfect by [F7]. The composite , , obtained from the algebraic closure , is a surjective -algebra homomorphism (it is -linear and hits ), and restricting it to recovers the quotient map ; hence its kernel is a maximal ideal of with , so lies over and its preimage is maximal with . By [F1] applied to the local ring is regular local; put .
The chosen equations also generate the ideal of the -fibre. Put and , so . Since the images of generate by step 2.1, the cokernel is zero, so by [F10] (right exactness of base change of scalars, and localisation commuting with it). Hence satisfies , which is regular local of dimension . Now is a maximal ideal of the finite-type -domain , so by [F8], that is, .
The remaining relations die after inverting. Put and ; the ideal is finitely generated because is. Since generate by step 2.1, the map is an isomorphism. The ring is flat over by hypothesis, so [F9] gives ; the Tor long exact sequence of [F9] applied to therefore makes injective, while its composite with the isomorphism above is zero; hence , that is, . The Axiom of Dependent Choice assumed through [F9] is a consequence of the Axiom of Choice assumed in the statement, and is used only here.
The rank and the minor. Apply [F4] with (perfect), , — a generating set of that ideal, with the now viewed in — and : this maximal ideal has residue field , a finite separable extension of , and is regular local of dimension by step 3.1. The criterion gives where is the Jacobian matrix of . Moreover where the middle identification is base change of scalars by [F10] applied to the quotient of step 2.1 and the last inequality is that the dimension of a vector space cannot drop under a field extension. Therefore , so some minor of the Jacobian matrix has nonzero image in .
The minor descends to . By the monomial formula of [F3], partial differentiation is linear over the coefficient ring, so the image in of the minor is the corresponding minor of the images of ; since its image in is nonzero we get , hence because lies over , and hence .
Conclusion. The ring is local with maximal ideal , which contains because , and is a finitely generated -module; so forces by [F6]. By [F11] there is , with , that is as -algebras, where is the image of in . Since by step 5.1, put and write for its image in . The image of is a unit in because is inverted, and because ; hence with the Jacobian minor a unit, a standard smooth presentation of relative dimension by [F2]. Since , its image , so this chart witnesses that is standard smooth at .
Locally standard smooth iff flat with geometrically regular fibres
Statement
Assume the Axiom of Choice (The Axiom of Choice). Let be a ring map of finite presentation (Finitely presented modules and finitely presented algebras).
- Pointwise criterion. Let , put and . Then is standard smooth at (Standard smooth presentations and locally standard smooth maps) if and only if the local ring homomorphism is flat and the fibre is geometrically regular at (Geometrically regular algebras and geometrically regular fibres).
- Global form. The map is locally standard smooth if and only if is flat and every fibre , , is geometrically regular.
- Field case. Let be a field and a finite-type -algebra. Then is geometrically regular over if and only if the structure map is locally standard smooth; equivalently, if and only if admits a standard smooth presentation over at every prime. In that case the relative dimension of a standard smooth chart at a prime is the dimension of the regular local ring when is a -rational point.
Clause 1 is the pointwise form of the classical equivalence between smoothness and flatness with geometrically regular fibres; clause 2 is its global form, and finite presentation is needed in both directions (locally standard smooth maps are finitely presented by definition, and the fibre condition is only defined for a finitely presented -algebra). No hypothesis is placed on .
Facts & Assumptions
Given: A ring map of finite presentation, a prime with and , the fibre , a finite-type -algebra in clause 3, and the Axiom of Choice.
Standard smooth presentations and locally standard smooth maps: a standard smooth presentation of an -algebra consists of integers , elements and of with such that some minor of the Jacobian matrix has image a unit of ; is the relative dimension, the invertible minor may be assumed leading, and a further principal localisation may be absorbed. The map is standard smooth at when has a standard smooth presentation over for some , and locally standard smooth when this holds at every prime; finite presentation of over is part of the definition of standard smoothness at a prime, as well as of the fibre condition.
Standard smooth algebras are finitely presented and flat: under the Axiom of Choice, a standard smooth -algebra is a finitely presented -algebra and is flat over , for every commutative ring .
Fibres of standard smooth algebras are regular of relative dimension: under the Axiom of Choice, for a standard smooth -algebra with leading minor a unit, a prime and a field extension , every local ring of the fibre is a regular local ring with , where is the prime corresponding to , and every irreducible component of has dimension .
Flat maps with geometrically regular fibres have standard smooth local presentations: under the Axiom of Choice, if is of finite presentation, , , the local homomorphism is flat and the fibre is geometrically regular at , then there is such that admits a standard smooth presentation over ; that is, is standard smooth at .
Geometrically regular algebras and geometrically regular fibres: a finite-type -algebra is geometrically regular over when is a regular Noetherian ring for every finitely generated field extension ; for a finitely presented -algebra , a prime with , the fibre is and it is geometrically regular at when for every field extension and every prime of lying over the image of the local ring there is regular; a fibre is geometrically regular when it is geometrically regular at each of its points.
Field tests for geometric regularity: under the Axiom of Choice, for a finite-type -algebra : is geometrically regular over if and only if is regular and is regular for every finite purely inseparable ; if is geometrically regular over then is regular for every field extension ; and if is geometrically regular over for one field extension then is geometrically regular over .
Modules over a field are projective, flat, and injective: under the Axiom of Choice every module over a field is free, hence projective and flat.
Flatness is equivalent to preserving injections and to the ideal and finitely generated ideal tests, Assuming the Axiom of Choice, local criteria for zero modules and for injective, surjective, and bijective maps: an -module is flat if and only if is injective for every ideal ; and under the Axiom of Choice an -module is zero if and only if for every maximal ideal , equivalently for every prime.
Every localization is flat, and localizing a flat module preserves flatness: for a commutative ring and multiplicative set , the localisation is a flat -algebra, and a -module is flat over if and only if it is flat over .
Localisation of modules is extension of scalars, Localisation commutes with kernels images and cokernels, Injective module maps remain injective after localisation, Localising twice is localising once at the multiplicative set generated by both denominator sets: localisation of modules is given by tensoring with the localised ring and commutes with kernels, images and cokernels, so localising preserves injectivity and commutes with base change of scalars; and for multiplicative sets the iterated localisation is the localisation at the multiplicative set generated by and .
regular noetherian ring: a commutative Noetherian ring is regular when its localisation at every prime is a regular local ring; this holds vacuously for the zero ring.
Finitely presented modules and finitely presented algebras: a commutative -algebra is finitely presented when for some and a finitely generated ideal .
The Axiom of Choice: the Axiom of Choice, assumed in the statement and used through [F2], [F3], [F4], [F6], [F7] and [F8].
Every affine scheme is quasi-compact: every affine scheme is quasi-compact, so and are quasi-compact, and a family of principal opens covering either of them has a finite subcover whose elements generate the unit ideal.
A polynomial ring in n variables over a field has dimension n, Maximal ideals of an affine domain have full height: for a field one has , and a maximal ideal of a finite-type -domain has height equal to the dimension of that domain; in particular a maximal ideal of satisfies .
Every algebra of finite type over a Noetherian ring is finitely presented: a finite-type algebra over a Noetherian ring is finitely presented; in particular this holds over a field.
Proof
Set-up and conventions. Write for the fibre over ; by [F5] the fibre is defined because is finitely presented, and "geometrically regular at " means that for every field extension and every prime of lying over the image of in , the local ring is regular. Since a standard smooth chart at is by definition a standard smooth presentation of some , [F1], and since is finitely presented over when is [F2, F12], both sides of clause 1 only concern finitely presented -algebras.
Flatness from a principal cover. Suppose generate the unit ideal of and each is flat over . Then is flat over . Indeed, let be an ideal and let ; localising the map at gives the map by [F10], which is injective because is flat over , so for every by [F10]. If , then for some maximal ideal by [F8], and since the generate the unit ideal some ; then by [F10], a contradiction. Hence , so every such multiplication map is injective and is flat over by [F8].
Clause 1, only-if: flatness at . Assume is standard smooth at , and choose such that has a standard smooth presentation over [F1]. Then is flat over by [F2], so for every ideal the map is injective by [F8]; the localisation at is a localisation of the -module , so is the localisation of that injective map and is injective by [F10]; hence is flat over by [F8]. Because maps into , the ring is an -algebra, so [F9] upgrades flatness over to flatness over : the local homomorphism is flat.
Clause 3, if direction. Conversely let be locally standard smooth, and let ; choose with standard smooth over [F1]. For every field extension , [F3] applied to the standard smooth -algebra with and fibre shows that every local ring of is regular; a prime lying over does not contain , and [F10] identifies with the local ring of at the corresponding prime, so it is regular. As was arbitrary, is geometrically regular at in the sense of [F5]; in particular, taking finitely generated over , the finite-type -algebra has all its prime localisations regular, so it is a regular Noetherian ring by [F11] and is geometrically regular over .
Clause 1, only-if: geometric regularity of the fibre at . Keep the chart of step 1.3. Since is finitely presented, so is the coefficient extension , and by [F10]; write for the image of in , so , where is the image of . Let be a field extension and let be a prime lying over ; then , and [F10] identifies with the local ring of at the corresponding prime. That local ring is regular by [F3] applied to the standard smooth -algebra and the extension . Since and were arbitrary, is geometrically regular at by [F5].
Clause 1, if direction. If is flat and the fibre is geometrically regular at , then [F4] produces with standard smooth over , that is, is standard smooth at . Steps 1.3 and 2.1 give the converse, so clause 1 holds.
Clause 2, only-if. Assume is locally standard smooth. For each choose with standard smooth over [F1]; the open sets cover the affine, hence quasi-compact, scheme [F14], so finitely many of them, say for , already cover, and their elements generate the unit ideal of . Each is flat over by [F2], so is flat over by step 1.2. For the fibres, let and let be a point of the fibre, with image ; choosing the chart at that and applying step 2.1 shows that the local ring of the fibre at the prime corresponding to — after any field extension of — is regular, so is geometrically regular at and hence the whole fibre over is geometrically regular by [F5]. As was arbitrary, is flat with geometrically regular fibres.
Clause 2, if direction. Assume is flat and every fibre is geometrically regular. Fix , . Flatness of over localises: is flat over by [F9, F10] applied to the localisation of the flat -module , hence flat over by [F9] since is an -module. The fibre condition is exactly hypothesis 2 of [F4] at , because a fibre that is geometrically regular at each of its points is geometrically regular at [F5]. So [F4] gives with standard smooth over . As was arbitrary, is locally standard smooth.
Clause 3, only-if. Let be a field and a finite-type, hence finitely presented [F16], -algebra that is geometrically regular over ; then the local homomorphism is flat for every prime because every -module is flat [F7, F9], and the only prime of is with residue field , so the fibre is . By [F6] the geometric regularity of over makes a regular ring for every field extension , not only the finitely generated ones; by [F11] this says precisely that every local ring of at a prime lying over a given prime of is regular, so is geometrically regular at every prime in the sense of [F5]. Clause 1 (step 3.1) then gives a standard smooth chart of over at every prime, that is, is locally standard smooth.
The relative-dimension clause of clause 3. Let be a -rational point of the finite-type -algebra , that is as -algebras, and let , , be a standard smooth chart of over at with leading minor a unit of the localisation [F1]. Write for the prime corresponding to . The composite is a -algebra map whose kernel is , so is a -subalgebra of the field containing the image of , hence equal to ; thus is maximal and by [F15]. Applying [F3] to the standard smooth -algebra with , and the local ring gives , which is the relative dimension of the chart, in the situation of step 1.4.
Base change and composition of standard smooth presentations
Statement
Let be a homomorphism of commutative rings and let be an arbitrary ring homomorphism. Write standard smooth presentations (Standard smooth presentations and locally standard smooth maps) as of relative dimensions and , with leading Jacobian minors and mapping to units.
- Base change. is a standard smooth -algebra with the same parameters , relative dimension , and with the image of a unit. If moreover is standard smooth at a prime , then is standard smooth at every prime of lying over ; consequently locally standard smooth maps are stable under arbitrary base change of the base ring.
- Composition. carries a standard smooth -presentation with variables, equations and relative dimension ; thus the relative dimensions of these displayed presentations add. If is standard smooth at and is standard smooth at with , then is standard smooth at ; consequently a composite of locally standard smooth maps is locally standard smooth.
No hypothesis is placed on or on , no regularity theorem is used, and no form of the Axiom of Choice is used: all statements are formal consequences of the displayed polynomial presentations. The relative dimension of a presentation is the integer ; its identification with the dimension of a nonempty fibre is a separate matter, proved under the Axiom of Choice elsewhere on this page and used nowhere below.
Facts & Assumptions
Given: A homomorphism with a standard smooth presentation of relative dimension and leading minor a unit, an arbitrary ring homomorphism , and an -algebra with a standard smooth -presentation of relative dimension and leading minor a unit (with localisation denominator ).
Standard smooth presentations and locally standard smooth maps: a standard smooth presentation of an -algebra consists of , and with such that some minor of the Jacobian matrix has image a unit of ; is the relative dimension and the invertible minor may be assumed to be the leading one in the first columns. For a finitely presented -algebra , the map is standard smooth at when has a standard smooth presentation over for some , and locally standard smooth when this holds at every prime.
Base change of standard smooth presentations: for any ring map and a standard smooth presentation with minor a unit, there is a unique -algebra isomorphism sending to , and the target is standard smooth over with the same and relative dimension, the image of again a unit.
Differentials of a polynomial quotient and the Jacobian cokernel: for the partial derivatives are computed on the monomial basis by and extended -linearly, so that is -linear and is zero on polynomials not involving ; , and the Jacobian matrix governs the cokernel presentation of .
Universal mapping property of the tensor product of commutative algebras, Localisation of modules is extension of scalars, A polynomial ring on a finite ordered family agrees canonically with the iterated polynomial-ring construction: for a ring homomorphism there is an -algebra isomorphism ; for a multiplicative set of an -algebra the localisation of an -module is , so ; and the iterated polynomial ring is canonically .
Tensoring is right exact: tensoring an exact sequence with a module preserves exactness; in particular for an ideal one has , giving .
Universal property of localisation: maps that invert factor uniquely through , Multiplicative subsets and the localisation as equivalence classes of fractions: a unital homomorphism carrying a multiplicative set into the units factors uniquely through the localisation, and in the element is a unit; localisation is functorial for ring maps.
Localising twice is localising once at the multiplicative set generated by both denominator sets: for multiplicative sets of a commutative ring , the iterated localisation is the localisation of at the multiplicative set generated by ; in particular localising successively at and at is localising at , and an element which is a unit remains a unit.
Proof
Notation. Fix a standard smooth presentation with leading minor a unit of [F1], put and , so that . Fix also a standard smooth -presentation with leading minor a unit of [F1]. Finally fix a ring map .
Base change of presentations. By [F2] applied to the presentation of step 1.1 and the ring map there is an -algebra isomorphism , where are the images of ; the target is a standard smooth -presentation with the same and relative dimension , and the image of is a unit. This is the first assertion of clause 1.
The polynomial presentation of . The coefficient extension holds by [F5], since and by [F4]; combining it with from [F4] and with gives We use this isomorphism to read the presentation of in the polynomial ring over .
Base change at a prime. Finite presentation is preserved by base change: tensoring with gives by [F4, F5], where the primes denote coefficient images. Suppose is standard smooth at , witnessed by an element with standard smooth over [F1]; by [F2] the base change is standard smooth over . Let be a prime with , i.e. lying over ; then , since would give . Hence lies in the principal open of , and the localisation — which is by [F4] — is standard smooth over [F6]. Therefore is standard smooth at ; as was an arbitrary prime over , this gives the pointwise form of clause 1, and taking the witnessing chart at every prime of gives stability of local standard smoothness under base change.
Clearing denominators and the composite presentation. By step 2.2 the -presentation of is a presentation in the ring , with ; write and with coefficients in , and choose representatives and with and . Put , (so empty families give ) and define Multiplying the displayed identities by and shows and in . Since is a unit of , the ideals and coincide there, and is a unit multiple of ; by [F7] localising at and then at is localising at . Hence the composite presentation of over .
The Jacobian minor of the composite. In the ring , the sum formula and coefficient linearity of [F3] give , because the coefficients represent and is -linear on . Hence the leading block has determinant , a unit of because and are units. Moreover for all , since does not involve the 's [F3]. Therefore the minor of the Jacobian matrix of on the columns and is block triangular with diagonal blocks and , so its determinant is which is a unit of because maps to a unit of and hence of , and and are units of .
The composite is standard smooth. By step 3.2 the algebra is presented over as with variables and equations, and by step 4.1 the displayed minor of the Jacobian matrix is a unit of ; moreover the invertible minor may be assumed leading after permuting variables, so this is a standard smooth -presentation [F1]. Its relative dimension is , the sum of the relative dimensions of the two given presentations. This proves the first assertion of clause 2.
Composition at a point. Finite presentation is preserved by composition: from and , lift the finitely many coefficients of the to ; then by [F4, F5]. Thus the finite-presentation prerequisite in [F1] holds for the composite. Suppose is standard smooth at and is standard smooth at with . Choose with standard smooth over and with standard smooth over [F1]. Since , both and lie outside , so . Base change of the standard smooth -presentation of along gives the standard smooth -algebra by step 2.1 and [F4, F7]. Applying step 5.1 to over and over exhibits as standard smooth over ; since , this witnesses that is standard smooth at [F1]. As was arbitrary, a composite of locally standard smooth maps is locally standard smooth, which completes clause 2.
Submersion criterion for locally standard smooth morphisms
Statement
Assume the Axiom of Choice (The Axiom of Choice). Let be a field, let and be -schemes that are locally standard smooth over at the points considered below (Standard smooth presentations and locally standard smooth maps), and let be a morphism of -schemes of finite type (Locally finite type and finite type morphisms). Let be a -rational point and put , assumed -rational as well, so that (The residue field at a point of an affine scheme). Write , , with maximal ideals , , and let be the induced local homomorphism. Let be the scheme-theoretic fibre (Scheme-theoretic fibre), whose local ring at is . Then:
- Submersion criterion. is locally standard smooth at — that is, there are affine opens and with , and standard smooth at the prime of corresponding to — if and only if the -linear map induced by is injective.
- Flatness and fibres. If the equivalent conditions of clause 1 hold and , , then is flat over , that is is flat at , and is a regular local ring of dimension ; in other words the fibre is regular at of dimension . Any standard smooth chart of at has relative dimension .
The two schemes are only required to be locally standard smooth at and , not globally; is of finite type as assumed above, and no hypothesis is imposed on the base field. The Axiom of Choice is used through the local-flatness, regular-parameter and geometric-regularity suppliers cited below.
Facts & Assumptions
Given: A field , -schemes locally standard smooth over at a -rational point and at its image , a finite-type morphism of -schemes , the local rings , with maximal ideals and residue fields , the induced local homomorphism , and the Axiom of Choice.
Standard smooth presentations and locally standard smooth maps: a standard smooth presentation of an -algebra consists of , and with such that some Jacobian minor has image a unit of ; is the relative dimension, the invertible minor may be assumed leading, and a further principal localisation may be absorbed. For a finitely presented -algebra map and a prime , standard smooth at means that has a standard smooth presentation over for some ; locally standard smooth means this holds at every prime.
Standard smooth algebras are finitely presented and flat: under the Axiom of Choice, a standard smooth -algebra is a finitely presented -algebra and is flat over , for every commutative ring .
Fibres of standard smooth algebras are regular of relative dimension: under the Axiom of Choice, for a standard smooth -algebra with leading minor a unit, a prime and a field extension , every local ring of is regular local of dimension , where corresponds to , and every irreducible component of has dimension ; for this says that a localisation of a polynomial ring over a field is regular local of dimension .
Separable residue and the cotangent sequence of a local algebra: let be a Noetherian local -algebra with maximal ideal and residue field , finitely generated and separably generated over . Then is short exact, the first map sending the class of to ; if is finite separable then and that map is an isomorphism .
Transitivity sequence for differentials: for homomorphisms of commutative rings the sequence of -modules is exact, the first map being the extension of scalars of .
Localization, base change and functoriality of differentials: for ring maps there is a natural isomorphism , and for multiplicative sets , with the image of in contained in there is a -module isomorphism .
Differentials of a polynomial quotient and the Jacobian cokernel: for , is free on ; if then is exact; and if then is the cokernel of the Jacobian matrix, so that with an invertible minor is free of rank .
regular system of parameters equivalent basis: under the Axiom of Choice, for a nonzero Noetherian local ring of dimension and , the tuple is a regular system of parameters if and only if its classes form a -basis of ; in particular every lift of a cotangent basis generates and is a system of parameters.
regular local rings are domains and cohen macaulay: under the Axiom of Choice, a regular local ring of dimension is a domain and Cohen–Macaulay, and for every regular system of parameters the tuple is -regular and is regular local of dimension for all .
regular local regular quotient ideal is parameter generated: under the Axiom of Choice, for a regular local ring of dimension and an ideal , the quotient is regular if and only if , equivalently if and only if is generated by an initial part of a regular system of parameters.
Local flatness criterion by regular parameters: under the Axiom of Choice, for a local homomorphism of Noetherian local rings and a finite -module with , the module is flat over ( need not be finite over ); consequently, if and are regular local and the images in of a regular system of parameters of extend to a regular system of parameters of , then is flat over .
Locally standard smooth iff flat with geometrically regular fibres: under the Axiom of Choice, for a ring map of finite presentation and with , the map is standard smooth at if and only if is flat and the fibre is geometrically regular at ; and for a finite-type -algebra that is locally standard smooth over , the relative dimension of a standard smooth chart at a -rational prime equals .
Base change and composition of standard smooth presentations: base change of a standard smooth presentation along any ring map yields a standard smooth -presentation with the same parameters and relative dimension, standard smoothness at a prime is stable under such base change, and composing standard smooth presentations over yields a standard smooth -presentation of with relative dimension the sum of the two relative dimensions; composition is likewise standard smooth at a prime.
Maximal ideals of an affine domain have full height, A polynomial ring in n variables over a field has dimension n: for a field , and every maximal ideal of a finite-type -domain has height equal to the dimension of that domain; in particular a maximal ideal satisfies .
Geometrically regular algebras and geometrically regular fibres: for a finitely presented -algebra , with , the fibre is geometrically regular at when for every field extension and every prime of lying over the image of the local ring there is regular.
Scheme-theoretic fibre, Base change of objects, morphisms and properties, Universal mapping property of the tensor product of commutative algebras, Existence of all scheme fibre products, Prime ideals of a localization are exactly the primes disjoint from the denominator set, Localising twice is localising once at the multiplicative set generated by both denominator sets, Localisation commutes with kernels images and cokernels: the fibre is ; over affine charts , it is computed by the coproduct with , so that its local ring at the point induced by is ; primes of a localisation are the primes of not containing , localisation commutes with quotients and cokernels, and iterated localisation is localisation at the product of the inverted elements.
Flatness is transitive under a flat change of rings, Every localization is flat, and localizing a flat module preserves flatness: a localisation is flat, a composite of flat ring homomorphisms is flat, and a base change along a flat map is flat.
Tensoring is right exact, Localisation of modules is extension of scalars: tensoring an exact sequence preserves right exactness, so a right-exact sequence stays right exact after tensoring with a module; for an ideal one has .
Every algebra of finite type over a Noetherian ring is a Noetherian ring, Every quotient and every localisation of a Noetherian ring is Noetherian, Every algebra of finite type over a Noetherian ring is finitely presented: a finite-type algebra over the field is Noetherian, as are its quotients and localisations; and a finite-type algebra over a Noetherian ring is finitely presented.
The Axiom of Choice: every family of nonempty sets has a choice function; it is assumed in the statement and used through [F3], [F8], [F9], [F10], [F11] and [F12].
Proof
Setup. Since is of finite type, the point has an affine open neighbourhood and has an affine open neighbourhood with and of finite type; shrinking we may suppose that has a standard smooth presentation with leading minor a unit [F1], and shrinking that has a standard smooth presentation with leading minor a unit. Let be the prime corresponding to and the prime corresponding to ; both are maximal with , since and are -rational points and the -algebra maps and have finite-type domains, hence are isomorphisms. Put and , so that is a local homomorphism of Noetherian local rings with residue field [F19], and put .
Converse direction: extending regular parameters. By [F3] applied over to the two charts of step 1.1, and are regular local rings; write and . Assume now that the map induced by is injective. Choose a -basis of and lift it to ; by [F8] the tuple is a regular system of parameters of . Its images form an independent tuple of elements of the -vector space of dimension , which therefore extends to a -basis; lifting that basis so that the first lifts are and the remaining lifts are new elements gives with for , and [F8] again makes it a regular system of parameters of .
The local rings and the cotangent identifications. Applying [F3] to the two standard smooth presentations of step 1.1 over the base field with and , where the corresponding primes and are maximal and hence of heights and by [F14], shows that is a regular local ring of dimension and that is a regular local ring of dimension ; by [F12] these integers are and , so and . Since the residue fields of and are the field , a finite separable extension of , [F4] gives isomorphisms and carrying the class of an element of the maximal ideal to .
Forward direction: charts give freeness, flatness and the fibre dimension. Assume is locally standard smooth at ; after shrinking the charts of step 1.1 we may suppose that carries a standard smooth presentation over of relative dimension , say with an invertible Jacobian minor [F1]. By [F7] the -module [F6] is the cokernel of the Jacobian matrix , hence is free of rank because the minor is a unit of . Base changing this presentation along exhibits as a localisation of the standard smooth -algebra [F13], which is flat over by [F2]; localisation is flat and flatness is transitive [F17], so is flat over . Finally, is the local ring of the fibre algebra at the prime corresponding to , which is maximal because its residue field is ; so [F3] and [F14] give .
Converse direction: flatness and a regular local fibre. The images in of the regular system of parameters of are the initial segment of the regular system of parameters of from step 2.1, so the second assertion of [F11] shows that is flat over . By [F9] the tuple is -regular and is a regular local ring of dimension ; since , this quotient is , the local ring of the fibre at [F16].
Forward direction: relative dimension and injectivity of the cotangent map. Composing the standard smooth presentation of over from step 2.3 with the standard smooth presentation of over from step 1.1 presents the finite-type -algebra as standard smooth over with relative dimension [F13]; its localisation at the -rational prime is , so [F12] identifies that relative dimension with , whence and, by step 2.3, . The transitivity sequence of [F5] is right exact, and tensoring it with yields, using [F4] and [F18], the exact sequence , in which the first arrow is the map induced by ; since the last term is a -vector space of dimension , so the image has dimension , which equals by step 2.2 and forces the map to be injective.
Converse direction: the fibre is geometrically regular. Write for the maximal ideal corresponding to in the presentation of step 1.1, and put . The parameters generate . Since is Noetherian, after shrinking around and its inverse-image chart around , we may represent every by an element of and arrange that on these charts: first clear their denominators outside , then invert an element outside annihilating the finite module . Absorb the corresponding principal localisations into the polynomial chart of . Write each image of in as with and . Since is a unit of , the images of the numerators generate the same ideal as those of the , and each vanishes at . The finite-type fibre algebra is then presented on this chart by , and its local ring at is for [F16]. The ring is regular local of dimension by [F3] with and [F14], and by step 3.1, so [F10] gives . The classes of the generators span that space, hence form a basis. By [F4] and [F7], their classes in are the rows of the Jacobian matrix evaluated at , so some minor does not lie in . Localising the finite-type algebra at the image of gives a standard smooth -presentation with the displayed equations [F1]; this open chart contains . By [F3], after every field extension every local ring of is regular. Every prime of the extended fibre lying over belongs to this chart because , so the fibre is geometrically regular at in the sense of [F15].
Converse direction: concluding local standard smoothness. The -algebra map of step 1.1 is of finite type, hence finitely presented because is a localisation of a finite-type -algebra and therefore Noetherian [F19]. Its localisation is flat by step 3.1, and step 4.1 proves that the finite-type fibre is geometrically regular at the point induced by (its local ring there is ), so clause 1 of [F12] shows that is standard smooth at ; that is exactly the assertion that is locally standard smooth at . Together with step 3.2 this proves the equivalence of clause 1, step 2.3 and step 3.1 give flatness and the regularity and dimension of the fibre local ring in both directions, and steps 3.2 and 2.3 show that a witnessing chart has relative dimension .
5 · Examples, counterexamples and false statements
None yet.
Sources
- Stacks Algebra 10.131.1–2
- Vakil §22.2.2, p.575
- Stacks Algebra 10.131.3
- Vakil §22.2.17, pp.582–583
- Stacks Algebra 10.131.9, 14–15
- Vakil §§22.2.3, 22.2.12, pp.575, 579
- Stacks Algebra 10.131.8, 12
- Vakil §§22.2.K–L, pp.583–584
- Stacks Algebra 10.131.7
- Vakil §§22.2.9–11, pp.577–579
- Stacks Algebra 10.42.1, 3
- Stacks Algebra 10.44.1–2 and 10.45.2
- Stacks Algebra 10.131.7–8 and 10.44.1–2
- Vakil §22.2.M, p.584
- Stacks Algebra 10.131.9–10 and 10.140.4–5
- Vakil §§22.2.18 and 22.3.9, pp.582–583, 590
- Stacks common principal neighbourhoods, Lemma 26.11.5
- Stacks Algebra 10.131.12 and standard affine tensor base change
- Stacks Algebra 10.166.1–2
- Stacks Algebra 10.137.5–7
- Stacks Algebra 10.137.5–6
- Stacks Algebra 10.137.4 and 10.106.3
- Stacks Algebra 10.99.6–7
- Vakil §25.6.2–3, pp.678–679
- Stacks Algebra 10.99.7, 10.106.3, 10.136.12 and 10.137.5–7
- Vakil §25.6.2–3 and §26.2, pp.678–679, 689–690
- Stacks Algebra 10.137.5–6 (tags 00T6, 00T7)
- Stacks Algebra 10.110.3, 10.133.4–5 (tags 00OF, 00OD, 00OE)
- Vakil §26.2, pp.689–690
- Stacks Algebra 10.42.4 (tag 04KM)
- Stacks Algebra 10.43.3, 10.45.3 and 10.45.4
- Stacks Algebra 10.45.3–4
- Stacks Algebra 10.131.9 and 10.140.3 (Jacobian criterion, smooth locus)
- Vakil §21.2.9 and §25.6, pp.600–601, 678–679
- Stacks Algebra 10.137.15–16 (tags 00TE, 00TF) and 10.136.15 (tag 00SY)
- Vakil §26.2.4, pp.691–693
- Stacks Algebra 10.137.16 (tag 00TF), 10.137.5-6 (tags 00T6, 00T7) and 10.137.12 (tag 00TC)
- Vakil §26.2.2 and proof of §26.2.4, pp.690–693
- Stacks Algebra 10.137.6 (tag 00T7) and 10.137.8 (tag 00T9)
- Vakil §26.2.2 and the Jacobian-block argument of §26.2.4, pp.690–693
- Vakil §26.2.F and the proof of §26.2.4, pp.690–693
- Stacks Algebra 10.140.5 (tag 00TV), 10.137.16 (tag 00TF) and 10.137.5–6 (tags 00T6, 00T7)