Beilinson–Bloch heights · the structure of the arguments
Complete logical diagram
Results, exact premises and open steps, arranged in top-down diagrams. Parallel inputs meet at numbered logical connections; each connection links to its full hypotheses, explanation and sources below the graph.
Follow the detailed proof diagrams for every Progress result →
How to read the logic
Read from top to bottom. Parallel premises meet at a numbered connection, then lead to a conclusion. Select a connection to read its hypotheses and deduction; select a node to read the linked statement.
The full scope of each occurrence is recorded below its diagram. A proved conditional implication and an open geometric premise keep their separate statuses.
LOGIC 01
From a constructed pairing to the full global target
A complete construction route + functoriality + normalization + comparison ⇒ the full height target.
Scope. Fix a number field F, a smooth projective geometrically integral X/F and complementary r+s=dim(X)+1. The universal conjecture requires the same properties for every required variety and finite extension.
Scroll inside the frame to follow a large branch. “Fit width” shows the whole layout; “Readable size” restores the text to 100%.
Read the 8 connections
Each numbered card gives the exact inputs, the deduction and its sources.
01IFF · EQUIVALENT lifting criterion
Premises and stated hypotheses
- OpenThe full vertical correction equation is solvable
The finite-matrix test is valid only after all tests are proved to factor through the chosen finite-dimensional spaces.
Scope & sources
How the deduction works
Zero model obstruction is equivalent to solvability of the full vertical correction equation and hence to a numerically admissible lifting.
Conclusion / target
For the chosen closure, solve the correction equation against ALL vertical tests.
Scope & sourcesEvidence and further reading
Back to connection 01 in the diagram ↑02AND · ALL local and descent hypotheses
Premises and stated hypotheses
- OpenLefschetz and Hodge standard conjectures on every required geometric stratum
The strata come from the specified strictly semistable model.
Scope & sources - HypothesisSemistable model, local lifting theorem and cycle descent
Keep the stated base valuation rings and the descent of the model and cycle; use the separately stated alteration argument when necessary.
Scope & sources
How the deduction works
Retain the geometric-stratum hypotheses and descent from the completion of the maximal unramified extension.
Conclusion / target
This is a published conditional lifting theorem, not a proof of its general standard-conjecture premises.
Scope & sourcesEvidence and further reading
Back to connection 02 in the diagram ↑03OR · ANY complete source of the required liftings
Alternative complete routes
- OpenVanishing of the model numerical obstruction
For the chosen closure, solve the correction equation against ALL vertical tests.
Scope & sources - ConditionalThe standard-conjecture route on semistable strata
This is a published conditional lifting theorem, not a proof of its general standard-conjecture premises.
Scope & sources
How the deduction works
Apply the selected method to every relevant class, in both degrees and at every required finite place.
Conclusion / target
On a regular flat projective arithmetic model, lift every class numerically orthogonally to vertical cycles; use zero-curvature Green currents at infinity.
Scope & sourcesEvidence and further reading
Back to connection 03 in the diagram ↑04AND · ALL analytic and global inputs
Premises and stated hypotheses
- OpenStrong zero-curvature Green forms
STRONG-GRequired at every finite completion, in BOTH complementary codimensions, over the original valued field.
Scope & sources - HypothesisRestriction, principal-cycle operations and reciprocity
Use the stated current, intersection and principal-divisor calculus, including the auxiliary varieties and both ordered degrees.
Scope & sources - HypothesisMoving, archimedean identities and good-place vanishing
Moving handles representatives; the archimedean identities and almost-everywhere vanishing make the normalized sum well-defined.
Scope & sources
How the deduction works
Strong G alone supplies neither the principal-divisor calculus nor the finite global sum.
Conclusion / target
The principal transformation law and the product formula give descent through rational equivalence in both variables.
Scope & sourcesEvidence and further reading
Back to connection 04 in the diagram ↑05AND · BOTH the kernel construction and full-domain vanishing
Premises and stated hypotheses
- ConditionalH16 on its fixed global kernel domain
BB-H16-KERNEL-DOMAINRetain the local height identities and the source-version hypotheses. For the finite-detector presentation, retain the FD17 witness comparison.
Scope & sources - OpenZero actual obstruction for all localized global classes in degrees r AND s
Vanishing in only one degree enlarges only one factor.
Scope & sources
How the deduction works
The same kernel-domain construction has full domain exactly when Hʳ=Vʳ and Hˢ=Vˢ. This is not a claim about every possible extension of that pairing. A finite-detector description needs its separate fixed-witness comparison.
Conclusion / target
Vanishing must hold on localizations of GLOBAL classes in BOTH degrees. This target can be weaker than Strong G for every local Chow class.
Scope & sourcesEvidence and further reading
Back to connection 05 in the diagram ↑06AND · ALL lifting and descent inputs
Premises and stated hypotheses
- HypothesisAdmissible liftings in BOTH complementary degrees
On a regular flat projective arithmetic model, lift every class numerically orthogonally to vertical cycles; use zero-curvature Green currents at infinity.
Scope & sources - HypothesisProjection formulas and degree-normalized model descent
Retain native total degrees, compatible field extension and a common dominating alteration for comparisons.
Scope & sources
How the deduction works
The standard conjectures provide a published conditional route to the liftings. The curve–surface theorem is a known case, not a proof for arbitrary X.
Conclusion / target
Zhang’s model-theoretic approach constructs arithmetic-intersection heights. It does not, by itself, construct actual non-Archimedean Green forms.
Scope & sourcesEvidence and further reading
Back to connection 06 in the diagram ↑07OR · ANY complete construction route
Alternative complete routes
- ConditionalDirect actual-form construction
The principal transformation law and the product formula give descent through rational equivalence in both variables.
Scope & sources - ConditionalEnlarge the obstruction-kernel pairing
Vanishing must hold on localizations of GLOBAL classes in BOTH degrees. This target can be weaker than Strong G for every local Chow class.
Scope & sources - ConditionalAdmissible arithmetic liftings
Zhang’s model-theoretic approach constructs arithmetic-intersection heights. It does not, by itself, construct actual non-Archimedean Green forms.
Scope & sources
How the deduction works
One route may construct a pairing for the varieties and classes satisfying all of its premises. A result for one geometric family does not settle the universal target.
Conclusion / target
Fix X/F and complementary codimensions r+s=dim(X)+1. The three methods below have different inputs; their pairings are not identified merely by listing the routes.
Scope & sourcesEvidence and further reading
Back to connection 07 in the diagram ↑08AND · ALL requirements of the full target
Premises and stated hypotheses
- ConditionalA symmetric pairing on both full homologically trivial Chow groups
Fix X/F and complementary codimensions r+s=dim(X)+1. The three methods below have different inputs; their pairings are not identified merely by listing the routes.
Scope & sources - HypothesisH2: normalized extension, isomorphism and Galois compatibility
For the direct construction, finite trace and its degree formula are additional inputs.
Scope & sources - HypothesisH3: full correspondence adjunction
The adjunction theorem retains its proper-intersection hypotheses and the specified height domains.
Scope & sources - OpenH4 and classical agreement on common domains
Match the pointed-curve Néron–Tate sign and Poincaré normalization, and compare with classical heights under the stated model hypotheses.
Scope & sources
How the deduction works
This separates the requirements in the target statement. A construction theorem must be supplemented by the additional comparison and functoriality inputs on its own domain.
Conclusion / target
BB-GLOBALThe root is the full target, not merely the existence of a bilinear pairing.
Scope & sourcesEvidence and further reading
Back to connection 08 in the diagram ↑All 21 nodes: scope, status and sources
ConditionalA curve-normalized height system
Defines the complete H1–H4 interface. The construction and comparison gates must establish these properties on the required domains; the definition does not establish them.
Back to this node in the diagram ↑OpenA canonical global Beilinson–Bloch pairing
The root is the full target, not merely the existence of a bilinear pairing.
Back to this node in the diagram ↑ConditionalA symmetric pairing on both full homologically trivial Chow groups
Fix X/F and complementary codimensions r+s=dim(X)+1. The three methods below have different inputs; their pairings are not identified merely by listing the routes.
Back to this node in the diagram ↑ConditionalDirect actual-form construction
The principal transformation law and the product formula give descent through rational equivalence in both variables.
Back to this node in the diagram ↑OpenStrong zero-curvature Green forms
Required at every finite completion, in BOTH complementary codimensions, over the original valued field.
Back to this node in the diagram ↑HypothesisRestriction, principal-cycle operations and reciprocity
Use the stated current, intersection and principal-divisor calculus, including the auxiliary varieties and both ordered degrees.
Back to this node in the diagram ↑HypothesisMoving, archimedean identities and good-place vanishing
Moving handles representatives; the archimedean identities and almost-everywhere vanishing make the normalized sum well-defined.
Back to this node in the diagram ↑ConditionalEnlarge the obstruction-kernel pairing
Vanishing must hold on localizations of GLOBAL classes in BOTH degrees. This target can be weaker than Strong G for every local Chow class.
Back to this node in the diagram ↑ConditionalH16 on its fixed global kernel domain
Retain the local height identities and the source-version hypotheses. For the finite-detector presentation, retain the FD17 witness comparison.
Back to this node in the diagram ↑OpenZero actual obstruction for all localized global classes in degrees r AND s
Vanishing in only one degree enlarges only one factor.
Back to this node in the diagram ↑ConditionalAdmissible arithmetic liftings
Zhang’s model-theoretic approach constructs arithmetic-intersection heights. It does not, by itself, construct actual non-Archimedean Green forms.
Back to this node in the diagram ↑HypothesisAdmissible liftings in BOTH complementary degrees
On a regular flat projective arithmetic model, lift every class numerically orthogonally to vertical cycles; use zero-curvature Green currents at infinity.
Back to this node in the diagram ↑OpenVanishing of the model numerical obstruction
For the chosen closure, solve the correction equation against ALL vertical tests.
Back to this node in the diagram ↑OpenThe full vertical correction equation is solvable
The finite-matrix test is valid only after all tests are proved to factor through the chosen finite-dimensional spaces.
Back to this node in the diagram ↑ConditionalThe standard-conjecture route on semistable strata
This is a published conditional lifting theorem, not a proof of its general standard-conjecture premises.
Back to this node in the diagram ↑OpenLefschetz and Hodge standard conjectures on every required geometric stratum
The strata come from the specified strictly semistable model.
Back to this node in the diagram ↑HypothesisSemistable model, local lifting theorem and cycle descent
Keep the stated base valuation rings and the descent of the model and cycle; use the separately stated alteration argument when necessary.
Back to this node in the diagram ↑HypothesisProjection formulas and degree-normalized model descent
Retain native total degrees, compatible field extension and a common dominating alteration for comparisons.
Back to this node in the diagram ↑HypothesisH2: normalized extension, isomorphism and Galois compatibility
For the direct construction, finite trace and its degree formula are additional inputs.
Back to this node in the diagram ↑HypothesisH3: full correspondence adjunction
The adjunction theorem retains its proper-intersection hypotheses and the specified height domains.
Back to this node in the diagram ↑OpenH4 and classical agreement on common domains
Match the pointed-curve Néron–Tate sign and Poincaré normalization, and compare with classical heights under the stated model hypotheses.
Back to this node in the diagram ↑LOGIC 02
How an exact-support Green form is obtained
A curved Green pair + a global actual primitive ⇒ a strong Green form for the same cycle.
Scope. Use BB-LOCAL-DATA: X/K is smooth projective geometrically integral, K/ℚₚ is the original finite extension, and Z is a rational codimension-r cycle with 1≤r≤dim(X). Every conclusion keeps the prescribed complement and full actual tests.
Scroll inside the frame to follow a large branch. “Fit width” shows the whole layout; “Readable size” restores the text to 100%.
Read the 9 connections
Each numbered card gives the exact inputs, the deduction and its sources.
01AND · BOTH the abelian numerical certificate and finite numerical descent
Premises and stated hypotheses
- HypothesisKünnemann’s semistable input and the abelian numerical certificate
For the prescribed homologically trivial class on A, retain the semistable component exactness, cycle-class injectivity and the numerical vanishing on every vertical testing cycle.
Scope & sources - Draft · audit pendingDescent from a vertical numerical certificate
BB-FINITE-NUMERICAL-DESCENTAssume the proposed analytic descent theorem for the finite map A_F→A. Retain the numerical detector, generic-Chow-zero primitive, trace on the original scalar–curvature generators, and full degree [F:K]. A proper alteration alone is not this finite map.
Scope & sources
How the deduction works
After finite extension, the semistable abelian model produces a certificate for the pullback of the given geometrically homologically trivial class. Finite numerical descent returns its obstruction to zero over the original field.
Conclusion / target
BB-ABELIAN-STAGEThe abelian conclusion retains both the geometric certificate and the draft analytic descent input.
Scope & sourcesEvidence and further reading
Back to connection 01 in the diagram ↑02AND · ALL Theorem A premises
Premises and stated hypotheses
- HypothesisZ is geometrically algebraically trivial
This domain is smaller than the full homologically trivial Chow group.
Scope & sources - Draft · audit pendingA Chow-invariant actual Bott–Chern obstruction
BB-ACTUAL-OBSTRUCTIONAssume the Chow-invariant actual obstruction (A1), together with the curved Green construction (A2).
Scope & sources - Draft · audit pendingUniform constancy along nearby rational fibres
BB-RATIONAL-FIBRE-RIGIDITYThe rational-fibre rigidity input R is still a draft.
Scope & sources - Internal proof availableCompact additive maps into torsion-free groups
BB-COMPACT-RIGIDITYThis is the proved compact-group step, not a substitute for the analytic rigidity input.
Scope & sources
How the deduction works
Rational-fibre rigidity makes the additive Jacobian map locally constant; compactness and torsion-freeness force its image to vanish.
Conclusion / target
BB-ALG-TRANSFERTheorem A applies to geometrically ALGEBRAICALLY trivial classes.
Scope & sourcesEvidence and further reading
Back to connection 02 in the diagram ↑03AND · ALL Theorem B premises
Premises and stated hypotheses
- HypothesisThe analytic packages A1, A2, N and C
These include the stated obstruction, support, numerical descent and correspondence compatibilities; R is not an extra premise of this route.
Scope & sources - ConditionalThe semistable abelian numerical stage
BB-ABELIAN-STAGEThe abelian conclusion retains both the geometric certificate and the draft analytic descent input.
Scope & sources - HypothesisAn actual abelian Chow retraction after finite extension
The rational Chow identity uses opposite Tate shifts. Z must be geometrically homologically trivial.
Scope & sources
How the deduction works
Use abelian obstruction vanishing and the actual Chow identity for the diagonal, then descend with the full finite degree. A numerical or cohomological factorization is not enough.
Conclusion / target
BB-MOTIVIC-TRANSFERTheorem B applies to potential abelian Chow retracts, with geometrically homologically trivial Z.
Scope & sourcesEvidence and further reading
Back to connection 03 in the diagram ↑04AND · BOTH detection and vanishing
Premises and stated hypotheses
- OpenComparison with the ORIGINAL actual obstruction
First identify [ΩZ]=Ob([Z]) for the chosen Green pair. Fix the period functionals and prove that their joint vanishing detects this Ob on the specified Chow source.
Scope & sources - OpenAll selected periods vanish on the required classes
The general target requires this for EVERY original homologically trivial class; one calculation only treats that class. Finite-dimensionality of a cohomology group alone proves neither assertion.
Scope & sources
How the deduction works
A sufficient detector comparison is intersection of period kernels contained in the obstruction kernel. Equality of kernels is a stronger statement.
Conclusion / target
BB-RESIDUAL-PERIODSOnly the image of the original Chow source needs to vanish; the entire obstruction space need not vanish. The general record concerns that whole source; this local branch applies the criterion to the fixed class [Z].
Scope & sourcesEvidence and further reading
Back to connection 04 in the diagram ↑05OPEN · OPEN realization bridge
Premises and stated hypotheses
- OpenA primitive of the original curvature in the chosen tropical complex
The complex and the relevant Hodge/Aeppli comparison must be specified.
Scope & sources - OpenActual realization, both differentials, integration, gluing and descent
Preserve all actual tests, ordinary refinements, overlaps, the exact support and the original valued-field normalization.
Scope & sources
Remaining step
A primitive in a tropical complex becomes usable only after a comparison into global actual forms with every listed compatibility.
Conclusion / target
BB-TROPICAL-INTERFACEAn alternative construction of Ψ, not a consequence of topological inverse limits alone.
Scope & sourcesEvidence and further reading
Back to connection 05 in the diagram ↑06OR · ANY applicable method for the primitive
Alternative complete routes
- ConditionalAlgebraic equivalence and compact parameters
BB-ALG-TRANSFERTheorem A applies to geometrically ALGEBRAICALLY trivial classes.
Scope & sources - ConditionalTransfer through abelian Chow retracts
BB-MOTIVIC-TRANSFERTheorem B applies to potential abelian Chow retracts, with geometrically homologically trivial Z.
Scope & sources - OpenVanishing on the original Chow source
BB-RESIDUAL-PERIODSOnly the image of the original Chow source needs to vanish; the entire obstruction space need not vanish. The general record concerns that whole source; this local branch applies the criterion to the fixed class [Z].
Scope & sources - OpenTropical realization on the full actual test space
BB-TROPICAL-INTERFACEAn alternative construction of Ψ, not a consequence of topological inverse limits alone.
Scope & sources
How the deduction works
Geometric cases do not exhaust all homologically trivial cycles. The general source-vanishing and tropical interfaces remain open.
Conclusion / target
When the obstruction construction identifies [ΩZ] with Ob([Z]), zero obstruction is equivalent to this primitive. A directly constructed primitive does not require that obstruction construction. Each method below keeps its own premises.
Scope & sourcesEvidence and further reading
Back to connection 06 in the diagram ↑07AND · ALL inputs of curvature correction
Premises and stated hypotheses
- Draft · audit pendingGreen pairs on the prescribed support complement
BB-SUPPORTED-GREEN-PAIRThe exact-support construction remains a draft analytic input.
Scope & sources - OpenA global actual primitive DΨ=Ω
When the obstruction construction identifies [ΩZ] with Ob([Z]), zero obstruction is equivalent to this primitive. A directly constructed primitive does not require that obstruction construction. Each method below keeps its own premises.
Scope & sources - HypothesisActual Stokes against every complementary test
The primitive is global and belongs to the actual form space, not merely to a larger current space.
Scope & sources
Supporting results used in this deduction
- ConditionalRemoving an exact global curvature
BB-GREEN-CORRECTIONConditional curvature-correction rule: the prescribed curved pair and a global actual primitive, with actual Stokes, give the strong identity on the SAME complement and every stated test.
Scope & sources
How the deduction works
Subtract the actual Stokes identity for the global primitive from the curved Green identity. No additional support is removed.
Conclusion / target
The correction is γ=G−Ψ on the original complement. This is a sufficient construction; an arbitrary strong Green form need not arise from the selected obstruction map.
Scope & sourcesEvidence and further reading
Back to connection 07 in the diagram ↑All 32 nodes: scope, status and sources
DefinitionThe local actual-form problem
Fixes the original absolute value, ordinary actual form sheaf, prescribed cycle coefficients and exact complement, full actual test space, integrability and native finite degrees.
Back to this node in the diagram ↑ConditionalA strong zero-curvature Green form for the prescribed Z
The correction is γ=G−Ψ on the original complement. This is a sufficient construction; an arbitrary strong Green form need not arise from the selected obstruction map.
Back to this node in the diagram ↑ConditionalRemoving an exact global curvature
Conditional curvature-correction rule: the prescribed curved pair and a global actual primitive, with actual Stokes, give the strong identity on the SAME complement and every stated test.
Back to this node in the diagram ↑Draft · audit pendingGreen pairs on the prescribed support complement
The exact-support construction remains a draft analytic input.
Back to this node in the diagram ↑OpenA global actual primitive DΨ=Ω
When the obstruction construction identifies [ΩZ] with Ob([Z]), zero obstruction is equivalent to this primitive. A directly constructed primitive does not require that obstruction construction. Each method below keeps its own premises.
Back to this node in the diagram ↑ConditionalAlgebraic equivalence and compact parameters
Theorem A applies to geometrically ALGEBRAICALLY trivial classes.
Back to this node in the diagram ↑HypothesisZ is geometrically algebraically trivial
This domain is smaller than the full homologically trivial Chow group.
Back to this node in the diagram ↑Draft · audit pendingA Chow-invariant actual Bott–Chern obstruction
Assume the Chow-invariant actual obstruction (A1), together with the curved Green construction (A2).
Back to this node in the diagram ↑Draft · audit pendingUniform constancy along nearby rational fibres
The rational-fibre rigidity input R is still a draft.
Back to this node in the diagram ↑Internal proof availableCompact additive maps into torsion-free groups
This is the proved compact-group step, not a substitute for the analytic rigidity input.
Back to this node in the diagram ↑ConditionalTransfer through abelian Chow retracts
Theorem B applies to potential abelian Chow retracts, with geometrically homologically trivial Z.
Back to this node in the diagram ↑HypothesisThe analytic packages A1, A2, N and C
These include the stated obstruction, support, numerical descent and correspondence compatibilities; R is not an extra premise of this route.
Back to this node in the diagram ↑ConditionalThe semistable abelian numerical stage
The abelian conclusion retains both the geometric certificate and the draft analytic descent input.
Back to this node in the diagram ↑HypothesisKünnemann’s semistable input and the abelian numerical certificate
For the prescribed homologically trivial class on A, retain the semistable component exactness, cycle-class injectivity and the numerical vanishing on every vertical testing cycle.
Back to this node in the diagram ↑Draft · audit pendingDescent from a vertical numerical certificate
Assume the proposed analytic descent theorem for the finite map A_F→A. Retain the numerical detector, generic-Chow-zero primitive, trace on the original scalar–curvature generators, and full degree [F:K]. A proper alteration alone is not this finite map.
Back to this node in the diagram ↑HypothesisAn actual abelian Chow retraction after finite extension
The rational Chow identity uses opposite Tate shifts. Z must be geometrically homologically trivial.
Back to this node in the diagram ↑OpenVanishing on the original Chow source
Only the image of the original Chow source needs to vanish; the entire obstruction space need not vanish. The general record concerns that whole source; this local branch applies the criterion to the fixed class [Z].
Back to this node in the diagram ↑OpenComparison with the ORIGINAL actual obstruction
First identify [ΩZ]=Ob([Z]) for the chosen Green pair. Fix the period functionals and prove that their joint vanishing detects this Ob on the specified Chow source.
Back to this node in the diagram ↑OpenAll selected periods vanish on the required classes
The general target requires this for EVERY original homologically trivial class; one calculation only treats that class. Finite-dimensionality of a cohomology group alone proves neither assertion.
Back to this node in the diagram ↑OpenTropical realization on the full actual test space
An alternative construction of Ψ, not a consequence of topological inverse limits alone.
Back to this node in the diagram ↑OpenA primitive of the original curvature in the chosen tropical complex
The complex and the relevant Hodge/Aeppli comparison must be specified.
Back to this node in the diagram ↑OpenActual realization, both differentials, integration, gluing and descent
Preserve all actual tests, ordinary refinements, overlaps, the exact support and the original valued-field normalization.
Back to this node in the diagram ↑HypothesisActual Stokes against every complementary test
The primitive is global and belongs to the actual form space, not merely to a larger current space.
Back to this node in the diagram ↑OpenFinite traces and exact-support Green pairs
The open verification problem for the actual-form construction. These draft ingredients are obligations to establish, not a completed proof package.
Back to this node in the diagram ↑LOGIC 03
The exact arithmetic reduction for products of curves
Under H1–H4: off-central vanishing + AJ injectivity + residual Northcott ⇔ the full product package.
Scope. Fix the pointed products and product polarization. H1–H4 must hold on all required subproducts, source curves of correspondences and finite extensions.
Scroll inside the frame to follow a large branch. “Fit width” shows the whole layout; “Readable size” restores the text to 100%.
Read the 3 connections
Each numbered card gives the exact inputs, the deduction and its sources.
01IFF · EQUIVALENT to ALL three assertions, under H1–H4
Premises and stated hypotheses
- OpenOff-central Chow vanishing
OPEN-OFFCENTRALFor all required subproducts. R2-L1 identifies this with arithmetic hard Lefschetz; it does not prove the vanishing.
Scope & sources - OpenCentral algebraic Abel–Jacobi injectivity
OPEN-AJ-KERNELEliminate the algebraic contribution to the central height radical.
Scope & sources - OpenResidual integral Northcott
OPEN-RESIDUAL-NCBounded-height, bounded-degree finiteness for the residual INTEGRAL image.
Scope & sources
Supporting results used in this deduction
- ConditionalAn equivalent package for products of curves
R2-L3Main equivalence rule under H1–H4 on every required subproduct, correspondence source and finite extension. Its conditional theorem status does not resolve any of the three open inputs.
Scope & sources - Internal proof availableArithmetic Lefschetz from off-central vanishing
R2-L1Proof ingredient: for the fixed pointed product and fixed E, off-central vanishing on ALL subproducts is equivalent to arithmetic hard Lefschetz. No height axiom is needed for this equivalence.
Scope & sources - ConditionalThe primitive sum-of-squares identity
R2-L2Proof ingredient: H1–H4 and the central Chow-block construction give the primitive sum-of-squares identity with one common denominator ∏aᵢ. Off-central vanishing is NOT a premise of that identity; it is used separately when passing to the whole product.
Scope & sources
How the deduction works
The same package then implies real primitive positivity and complementary nondegeneracy. The sum-of-squares identity R2-L2 is the propagation rule, with its common coordinate denominator.
Conclusion / target
R2-L3 proves an equivalence under the height axioms. It does not establish the three open inputs.
Scope & sourcesEvidence and further reading
Back to connection 01 in the diagram ↑All 19 nodes: scope, status and sources
DefinitionPointed products and central residual groups
Defines the pointed product, anchored positive integral polarization, Chow cores U, algebraic part A and quotient G. The primitive and central ranges stay fixed throughout each use.
Back to this node in the diagram ↑ConditionalA curve-normalized height system
Assume H1–H4 on all involved varieties and finite extensions for conclusions involving heights. The Chow equivalence R2-L1 and the geometric projector construction do not themselves need these height axioms.
Back to this node in the diagram ↑DefinitionIntegral images and bounded-degree Northcott
Defines the projected images of integral cycles and bounded-degree Northcott. Counting takes place after rational equivalence and torsion have been removed, in one geometric rational group.
Back to this node in the diagram ↑OpenArithmetic hard Lefschetz AND primitive integral bounded-degree Northcott
R2-L3 proves an equivalence under the height axioms. It does not establish the three open inputs.
Back to this node in the diagram ↑ConditionalAn equivalent package for products of curves
Main equivalence rule under H1–H4 on every required subproduct, correspondence source and finite extension. Its conditional theorem status does not resolve any of the three open inputs.
Back to this node in the diagram ↑Internal proof availableArithmetic Lefschetz from off-central vanishing
Proof ingredient: for the fixed pointed product and fixed E, off-central vanishing on ALL subproducts is equivalent to arithmetic hard Lefschetz. No height axiom is needed for this equivalence.
Back to this node in the diagram ↑ConditionalThe primitive sum-of-squares identity
Proof ingredient: H1–H4 and the central Chow-block construction give the primitive sum-of-squares identity with one common denominator ∏aᵢ. Off-central vanishing is NOT a premise of that identity; it is used separately when passing to the whole product.
Back to this node in the diagram ↑OpenOff-central Chow vanishing
For all required subproducts. R2-L1 identifies this with arithmetic hard Lefschetz; it does not prove the vanishing.
Back to this node in the diagram ↑OpenCentral algebraic Abel–Jacobi injectivity
Eliminate the algebraic contribution to the central height radical.
Back to this node in the diagram ↑OpenResidual integral Northcott
Bounded-height, bounded-degree finiteness for the residual INTEGRAL image.
Back to this node in the diagram ↑LOGIC 04
Two ways to establish fixed-field Northcott
Positive height + common lattice, OR positive height + genuine gap + finite Kummer quotient.
Scope. Assume H1–H4 and fix one odd central block, one finite extension E/F₀ of number fields and the projected integral image ΛG(E). Do not replace it by all rational classes.
Scroll inside the frame to follow a large branch. “Fit width” shows the whole layout; “Readable size” restores the text to 100%.
Read the 7 connections
Each numbered card gives the exact inputs, the deduction and its sources.
01AND · One sufficient route: ALL-prime saturation
Premises and stated hypotheses
- OpenResidual rational detection
OPEN-SCALAR-DETECTIONRequire scalar-extended injective detectors at EVERY rational prime p.
Scope & sources - Internal proof availableLocal integral images of residual regulators
R2-R3Choose full ℤₚ-lattices Mₚ in the finite-dimensional targets containing the actual integral regulator images. These are local bounds.
Scope & sources - OpenAlmost-everywhere saturation relative to ONE full rational lattice
For almost all p, Dₚ⁻¹(Mₚ)=Λ₀⊗ℤₚ for that same Λ₀. This is additional to local finite generation.
Scope & sources
Supporting results used in this deduction
- Internal proof availableAn all-prime saturation criterion
R2-N4Inference rule under finite rational dimension, injectivity at every prime, full local lattices and the stated almost-everywhere equality with ONE rational lattice. It does not provide that equality.
Scope & sources
How the deduction works
R2-N4 needs agreement with the same rational lattice at almost every prime. Separate local integral bounds do not supply this agreement.
Conclusion / target
OPEN-INTEGRAL-LATTICEA single finite-rank rational lattice contains the entire integral image up to one common denominator.
Scope & sourcesEvidence and further reading
Back to connection 01 in the diagram ↑02AND · BOTH local integrality and scalar detection
Premises and stated hypotheses
- OpenResidual rational detection
OPEN-SCALAR-DETECTIONInjectivity after extension of scalars to ℚₚ, not merely injectivity over ℚ.
Scope & sources - Internal proof availableLocal integral images of residual regulators
R2-R3Use the actual integral regulator bound at that prime.
Scope & sources
How the deduction works
The scalar-injective image embeds Λ⊗ℤₚ into a finite ℤₚ-module. Taking its own quotient modulo p gives a finite Kummer quotient.
Conclusion / target
One coefficient prime may supply m=p under scalar detection.
Scope & sourcesEvidence and further reading
Back to connection 02 in the diagram ↑03AND · BOTH positivity and a common lattice
Premises and stated hypotheses
- OpenReal positivity on the residual Griffiths quotient
OPEN-REAL-POSITIVITYReal positive definiteness of the fixed-field residual form.
Scope & sources - OpenResidual integral discreteness
OPEN-INTEGRAL-LATTICEA single finite-rank rational lattice contains the entire integral image up to one common denominator.
Scope & sources
Supporting results used in this deduction
- Internal proof availableFive equivalent lattice conditions
R2-N3The finite-rank lattice equivalence justifies this sufficient criterion. Its proved status belongs to the equivalence, not to real positivity or the existence of the required geometric lattice.
Scope & sources
How the deduction works
A common finite-rank lattice supplies the required discreteness. Positive definiteness by itself does not.
Conclusion / target
In finite rational rank with a real positive form, Northcott, a positive gap, finite generation, a common denominator lattice and finite-adelic compactness are equivalent.
Scope & sourcesEvidence and further reading
Back to connection 03 in the diagram ↑04AND · ALL three weak Mordell–Weil inputs
Premises and stated hypotheses
- OpenReal positivity on the residual Griffiths quotient
OPEN-REAL-POSITIVITYUse the real positive quadratic form.
Scope & sources - OpenA uniform positive gap on nonzero integral classes
Find ε>0 with q(λ)≥ε for every nonzero λ in the fixed-field integral image.
Scope & sources - ConditionalA finite quotient Λ/mΛ for some m≥2
One coefficient prime may supply m=p under scalar detection.
Scope & sources
Supporting results used in this deduction
- Internal proof availableWeak Mordell–Weil and a positive height gap
R2-N5The descent theorem takes real positivity, a finite Kummer quotient and a genuine uniform positive gap as premises. Finite generation is its conclusion; finite rank is not assumed in advance.
Scope & sources
How the deduction works
The uniform positive height gap is an independent requirement. A finite quotient alone does not establish it.
Conclusion / target
This descent criterion does not assume finite rank in advance; finite generation and Northcott are conclusions.
Scope & sourcesEvidence and further reading
Back to connection 04 in the diagram ↑05OR · ANY complete fixed-field criterion
Alternative complete routes
- ConditionalReal positivity AND one common rational lattice
In finite rational rank with a real positive form, Northcott, a positive gap, finite generation, a common denominator lattice and finite-adelic compactness are equivalent.
Scope & sources - ConditionalReal positivity AND a genuine height gap AND a finite Kummer quotient
This descent criterion does not assume finite rank in advance; finite generation and Northcott are conclusions.
Scope & sources
How the deduction works
These are sufficient routes, not completed geometric hypotheses. Real positivity is required in each route.
Conclusion / target
Only the projected images of integral cycles are counted. Bounded height among arbitrary rational multiples is not the target.
Scope & sourcesEvidence and further reading
Back to connection 05 in the diagram ↑All 24 nodes: scope, status and sources
DefinitionIntegral images and bounded-degree Northcott
The group here is the image of integral cycles in the fixed residual rational quotient. Fixed-field Northcott fixes E; rational multiples with unbounded denominators are outside this target.
Back to this node in the diagram ↑OpenNorthcott over one fixed number field E
Only the projected images of integral cycles are counted. Bounded height among arbitrary rational multiples is not the target.
Back to this node in the diagram ↑ConditionalReal positivity AND one common rational lattice
In finite rational rank with a real positive form, Northcott, a positive gap, finite generation, a common denominator lattice and finite-adelic compactness are equivalent.
Back to this node in the diagram ↑Internal proof availableFive equivalent lattice conditions
The finite-rank lattice equivalence justifies this sufficient criterion. Its proved status belongs to the equivalence, not to real positivity or the existence of the required geometric lattice.
Back to this node in the diagram ↑OpenReal positivity on the residual Griffiths quotient
Real positive definiteness of the fixed-field residual form.
Back to this node in the diagram ↑OpenResidual integral discreteness
A single finite-rank rational lattice contains the entire integral image up to one common denominator.
Back to this node in the diagram ↑Internal proof availableAn all-prime saturation criterion
Inference rule under finite rational dimension, injectivity at every prime, full local lattices and the stated almost-everywhere equality with ONE rational lattice. It does not provide that equality.
Back to this node in the diagram ↑OpenResidual rational detection
Require scalar-extended injective detectors at EVERY rational prime p.
Back to this node in the diagram ↑Internal proof availableLocal integral images of residual regulators
Choose full ℤₚ-lattices Mₚ in the finite-dimensional targets containing the actual integral regulator images. These are local bounds.
Back to this node in the diagram ↑OpenAlmost-everywhere saturation relative to ONE full rational lattice
For almost all p, Dₚ⁻¹(Mₚ)=Λ₀⊗ℤₚ for that same Λ₀. This is additional to local finite generation.
Back to this node in the diagram ↑ConditionalReal positivity AND a genuine height gap AND a finite Kummer quotient
This descent criterion does not assume finite rank in advance; finite generation and Northcott are conclusions.
Back to this node in the diagram ↑Internal proof availableWeak Mordell–Weil and a positive height gap
The descent theorem takes real positivity, a finite Kummer quotient and a genuine uniform positive gap as premises. Finite generation is its conclusion; finite rank is not assumed in advance.
Back to this node in the diagram ↑OpenReal positivity on the residual Griffiths quotient
OpenA uniform positive gap on nonzero integral classes
Find ε>0 with q(λ)≥ε for every nonzero λ in the fixed-field integral image.
Back to this node in the diagram ↑ConditionalA finite quotient Λ/mΛ for some m≥2
One coefficient prime may supply m=p under scalar detection.
Back to this node in the diagram ↑OpenResidual rational detection
Injectivity after extension of scalars to ℚₚ, not merely injectivity over ℚ.
Back to this node in the diagram ↑Internal proof availableLocal integral images of residual regulators
Use the actual integral regulator bound at that prime.
Back to this node in the diagram ↑LOGIC 05
The additional step from fixed field to bounded degree
Fixed-field Northcott over every finite extension + uniform class-field ramification ⇔ bounded-degree Northcott.
Scope. Assume H1–H4 and apply the criterion separately to EVERY odd subproduct, with the field- and Galois-invariant residual form and restriction/trace-compatible integral images. Use the minimal field of the cycle CLASS, not cohomological ramification over a fixed field.
Scroll inside the frame to follow a large branch. “Fit width” shows the whole layout; “Readable size” restores the text to 100%.
Read the 3 connections
Each numbered card gives the exact inputs, the deduction and its sources.
01AND · A STRONGER sufficient condition
Premises and stated hypotheses
- OpenA quantitative discriminant bound
OPEN-DISCRIMINANTA uniform bound in terms of height and degree gives finitely many possible discriminants at fixed bounds.
Scope & sources
How the deduction works
The discriminant estimate is one possible route to ramification control; the converse is not asserted.
Conclusion / target
OPEN-CLASS-RAMIFICATIONFor bounded height and degree, bound a finite set of primes ramifying in the minimal class fields.
Scope & sourcesEvidence and further reading
Back to connection 01 in the diagram ↑02IFF · EQUIVALENT to BOTH requirements
Premises and stated hypotheses
- OpenFixed-field Northcott for EVERY finite extension E/F₀
Prove one of the fixed-field criteria for each field. This alone does not control a union of varying fields.
Scope & sources - OpenRamification of residual class fields
OPEN-CLASS-RAMIFICATIONFor bounded height and degree, bound a finite set of primes ramifying in the minimal class fields.
Scope & sources
Supporting results used in this deduction
- Internal proof availableRamification is the remaining variable-field condition
R2-F2Ramification equivalence rule for a field- and Galois-invariant height and trace-compatible integral images. Its theorem does not establish the open ramification assertion for geometric residual classes.
Scope & sources - Internal proof availableTrace to the field of a cycle class
R2-F1Descent lemma used in the proof: for ξ represented integrally over E and class field K, [E:K]ξ lies in Λ(K), with multiplier at most the bounded field degree. It does NOT assert ξ∈Λ(K).
Scope & sources
How the deduction works
R2-F2 uses Hermite–Minkowski for finitely many candidate fields and R2-F1 for a bounded descent multiplier, then applies fixed-field Northcott.
Conclusion / target
OPEN-RESIDUAL-NCThe bounded-degree target needs both arithmetic finiteness and field control.
Scope & sourcesEvidence and further reading
Back to connection 02 in the diagram ↑All 9 nodes: scope, status and sources
DefinitionIntegral images and bounded-degree Northcott
The same compatible projected integral images are used over all finite E. The minimal field of a rational Chow class need not already carry an integral representative of that class.
Back to this node in the diagram ↑OpenResidual integral Northcott
The bounded-degree target needs both arithmetic finiteness and field control.
Back to this node in the diagram ↑Internal proof availableRamification is the remaining variable-field condition
Ramification equivalence rule for a field- and Galois-invariant height and trace-compatible integral images. Its theorem does not establish the open ramification assertion for geometric residual classes.
Back to this node in the diagram ↑Internal proof availableTrace to the field of a cycle class
Descent lemma used in the proof: for ξ represented integrally over E and class field K, [E:K]ξ lies in Λ(K), with multiplier at most the bounded field degree. It does NOT assert ξ∈Λ(K).
Back to this node in the diagram ↑OpenFixed-field Northcott for EVERY finite extension E/F₀
Prove one of the fixed-field criteria for each field. This alone does not control a union of varying fields.
Back to this node in the diagram ↑OpenRamification of residual class fields
For bounded height and degree, bound a finite set of primes ramifying in the minimal class fields.
Back to this node in the diagram ↑OpenA quantitative discriminant bound
A uniform bound in terms of height and degree gives finitely many possible discriminants at fixed bounds.
Back to this node in the diagram ↑LOGIC 06
Metric comparison: established pieces and the open bridges
Complex comparison and good-reduction comparison are parallel inputs; bad reduction and families need further arguments.
Scope. Respect rank, degree, source-model, boundary, sign and normalization hypotheses. None of these comparisons alone establishes a global family-height theorem.
Scroll inside the frame to follow a large branch. “Fit width” shows the whole layout; “Readable size” restores the text to 100%.
Read the 4 connections
Each numbered card gives the exact inputs, the deduction and its sources.
01AND · ALL hypotheses for the cycle comparison
Premises and stated hypotheses
- Internal proof availableAn exact normalized Quillen isometry
R2-M1Retain the rigidification, mixed Poincaré curvature and product Kähler hypotheses.
Scope & sources - HypothesisVirtual rank zero and the filtered GRR cycle hypotheses
These identify the determinant with the flat Poincaré metric of the transpose divisor.
Scope & sources
How the deduction works
The normalized Quillen theorem is combined with rank-zero Deligne comparison and the filtered Riemann–Roch cycle calculation.
Conclusion / target
R2-M2Rank-zero and cycle determinant comparisons at the complex places.
Scope & sourcesEvidence and further reading
Back to connection 01 in the diagram ↑02OPEN · OPEN global comparison bridge
Premises and stated hypotheses
- ConditionalA constructed global pairing on the desired domain
Use one complete construction route from the first tree.
Scope & sources - Internal proof availableRank zero, Deligne pairings, and filtered Riemann–Roch
R2-M2Rank-zero and cycle determinant comparisons at the complex places.
Scope & sources - Internal proof availableGood-reduction determinant metrics
R2-M3Independent good-reduction input: a smooth proper source-curve model, section, rank zero, determinant degree zero and perfect pushforward.
Scope & sources - OpenBad-reduction metric compatibility
OPEN-BAD-REDUCTION-METRICThe semistable/Néron extension needs admissible vertical correction and matching actual-form trace, length and normalization.
Scope & sources - HypothesisCompatible projection, biextension and sign conventions
Keep the extra hypotheses of the chosen classical comparison, including singular-fibre conditions where relevant.
Scope & sources
Remaining step
The work package combines a constructed pairing with placewise comparisons and compatible projection, biextension and sign conventions.
Conclusion / target
This supplies part of the comparison work in the first tree; it is not an automatic consequence of height existence.
Scope & sourcesEvidence and further reading
Back to connection 02 in the diagram ↑All 14 nodes: scope, status and sources
OpenClassical comparison of a constructed global pairing
This supplies part of the comparison work in the first tree; it is not an automatic consequence of height existence.
Back to this node in the diagram ↑ConditionalA constructed global pairing on the desired domain
Use one complete construction route from the first tree.
Back to this node in the diagram ↑Internal proof availableRank zero, Deligne pairings, and filtered Riemann–Roch
Rank-zero and cycle determinant comparisons at the complex places.
Back to this node in the diagram ↑Internal proof availableAn exact normalized Quillen isometry
Retain the rigidification, mixed Poincaré curvature and product Kähler hypotheses.
Back to this node in the diagram ↑HypothesisVirtual rank zero and the filtered GRR cycle hypotheses
These identify the determinant with the flat Poincaré metric of the transpose divisor.
Back to this node in the diagram ↑Internal proof availableGood-reduction determinant metrics
Independent good-reduction input: a smooth proper source-curve model, section, rank zero, determinant degree zero and perfect pushforward.
Back to this node in the diagram ↑OpenBad-reduction metric compatibility
The semistable/Néron extension needs admissible vertical correction and matching actual-form trace, length and normalization.
Back to this node in the diagram ↑HypothesisCompatible projection, biextension and sign conventions
Keep the extra hypotheses of the chosen classical comparison, including singular-fibre conditions where relevant.
Back to this node in the diagram ↑Complete record index 54 records · supporting topic index
This supplementary index preserves all source records. Branches group related topics for reading; the logical relations are in the proof trees above. “Internal proof available” indicates a supplied argument, not independent certification.
54 records in 5 branches
Local Green formsConstruct actual Green forms on the prescribed support complement.20 records
The local problem uses the original valued field, ordinary actual forms and every compact complementary actual test. Finite descent retains the full field degree.
The local setting and conjecture2
- The local actual-form problemDefinition
The original valued field, the precise support complement, and the full actual test class.
BB-LOCAL-DATA Construct an actual Green form for every geometrically homologically trivial cycle.
STRONG-G
Green pairs, trace and curvature5
- Numerical model curvature detectionDraft · audit pending
A special-fibre numerical condition is proposed to detect equality of actual model Chern forms.
BB-MODEL-DETECTOR - Finite trace on scalar–curvature formsDraft · audit pending
The proposed trace has a specified actual-form domain and adjunction against all compact actual tests.
BB-RESTRICTED-TRACE - Green pairs on the prescribed support complementDraft · audit pending
A supported model complex supplies the proposed curved Green identity before curvature is removed.
BB-SUPPORTED-GREEN-PAIR - A Chow-invariant actual Bott–Chern obstructionDraft · audit pending
The curvature of a model Green pair is proposed to depend only on its generic Chow class.
BB-ACTUAL-OBSTRUCTION - Removing an exact global curvatureConditional
A formal correction preserves the original support complement.
BB-GREEN-CORRECTION
Algebraically trivial cycles3
- Compact additive maps into torsion-free groupsInternal proof available
An elementary compactness lemma used in the algebraic-equivalence argument.
BB-COMPACT-RIGIDITY - Uniform constancy along nearby rational fibresDraft · audit pending
The proposed actual-form rigidity supplies local constancy of the obstruction map.
BB-RATIONAL-FIBRE-RIGIDITY - Algebraic equivalence and compact parametersConditional
A geometric deduction from a Chow-invariant actual obstruction and its support theorem.
BB-ALG-TRANSFER
Abelian varieties and Chow transfer3
- Descent from a vertical numerical certificateDraft · audit pending
A certificate controls special-fibre intersections before a finite trace returns to the original field.
BB-FINITE-NUMERICAL-DESCENT - The semistable abelian numerical stageConditional
Vertical correction on a Künnemann model supplies the certificate needed by finite numerical descent.
BB-ABELIAN-STAGE - Transfer through abelian Chow retractsConditional
An actual diagonal factorization transports the abelian input.
BB-MOTIVIC-TRANSFER
Products of curves4
- The Chow diagonal of a pointed productInternal proof available
A purely algebraic factorization through one abelian variety, with degree shifts.
BB-CURVE-CHOW-DIAGONAL - Simultaneous regular models for the product routeDraft · audit pending
The geometric reduction needs regular models as well as a Chow diagonal.
BB-REGULAR-PRODUCT-MODELS The expanded draft has a more explicit model-sensitive transfer criterion.
BB-REGULAR-CHOW-CRITERION- Actual Green forms on products of curvesDraft · audit pending
The standalone draft combines finite traces, numerical liftings, and curve-motive decompositions.
BB-CURVE-PRODUCT-DRAFT
The remaining general case3
The analytic interfaces that govern promotion of the current standalone draft.
BB-ANALYTIC-INTERFACEKill the source image in a residual dual, not necessarily the entire obstruction space.
BB-RESIDUAL-PERIODSPass from chartwise primitives to an ordinary actual form on the exact complement.
BB-TROPICAL-INTERFACE
Research planNext problems
- Green forms and finite trace
Construct a Green pair for every prescribed cycle over the original non-Archimedean field.
Precise problem and proposed argument - Vanishing of the Bott–Chern class
Show that homologically trivial cycles have zero curvature class in the actual Bott–Chern quotient.
Precise problem and proposed argument
- Green forms and finite trace
Global height pairingsAssemble the local terms and determine the domain of the pairing.2 records
The kernel-domain construction is conditional on its Green-form and reciprocity hypotheses. Extending it to the full homologically trivial Chow groups requires local vanishing in both complementary codimensions.
The pairing and its domain2
- H16 on its fixed global kernel domainConditional
The recovered assembly proof constructs a pairing on admitted kernels; the fixed finite-detector presentation requires its separate witness comparison.
BB-H16-KERNEL-DOMAIN Assemble local contributions on the complete homologically trivial Chow groups.
BB-GLOBAL
Three construction routes3 routes
Alternative approaches to the global pairing, each with its own hypotheses.
- Vanishing of local curvature obstructions
Conditional constructions; general vanishing open
Start with the pairing on the common kernels of the local curvature maps. Prove that both kernels contain all homologically trivial classes.
Formal article and PDF - Direct construction of zero-curvature Green forms
Strong Green existence and analytic compatibility open in general
Construct strong Green forms directly, then prove the local identities needed to sum their pairings over all places.
Formal article and PDF - Admissible arithmetic liftings: Zhang’s approach
Published conditional theorem; unconditional curve–surface case
Correct cycle closures by vertical cycles on arithmetic models, then use arithmetic intersection. The general construction follows conditionally from the standard conjectures.
Formal article and PDF
- Vanishing of local curvature obstructions
Pairing domains and geometric cases13 cases
These constructions retain the Green-form and reciprocity hypotheses. Each case states its additional geometric or model assumptions.
Pairings on cycle groups4 cases
- A height pairing when the local obstruction vanishes
Conditional construction
Statement and analytic hypotheses - Height pairings on algebraically trivial cycles
Conditional consequence
Proof for algebraically trivial cycles - Divisors, zero-cycles and one-cycles
Conditional consequences
Proofs for divisors, zero-cycles and one-cyclesGeometric applications (1)
- A period criterion for individual cycle classes
Conditional criterion
The period criterion and Green formsGeometric applications (1)
- A height pairing when the local obstruction vanishes
Chow decompositions and geometric applications4 cases
- Varieties with a Chow motive of abelian type
Conditional theorem
Proof by Chow correspondences from abelian varieties - Products of curves and surfaces
Conditional theorem with an additional model hypothesis
Proof for products of curves and surfaces - Varieties with a decomposition of the diagonal
Conditional theorem
Proof using the decomposition of the diagonal - Chow niveau, complete intersections and fibrations
Conditional theorem and geometric applications
Proofs using Chow niveau and fibration theorems
- Varieties with a Chow motive of abelian type
Model criteria and arithmetic families3 cases
- Numerical corrections on regular models
Conditional model criteria
Numerical correction and its applications - Good reduction and semistable reduction
Conditional consequences of model theorems
Proofs for ordinary quadratic and semistable models - Arithmetic families from Lefschetz pencils
Conditional constructions of arithmetic families
Construction by Lefschetz pencils and congruences
- Numerical corrections on regular models
Comparison and arithmetic properties2 cases
- Comparison with classical height pairings
Conditional comparison formulas
Proofs of the height comparison formulasGeometric applications (2)
- Arithmetic hard Lefschetz, positivity and finiteness
Conditional reductions
Proofs of the Lefschetz and Northcott reductions
- Comparison with classical height pairings
Research planNext problem
- Construction of the global height pairing
Establish a local sum which is independent of Green representatives and descends through rational equivalence in both variables.
Precise problem and proposed argument
- Construction of the global height pairing
Chow groups and positivityDecompose the Chow groups and isolate the arithmetic conjectures.9 records
The height identities use the curve-normalized axioms H1–H4 on the stated pointed products, subproducts, source curves of correspondences and finite extensions. The Lefschetz result is an equivalence with off-central vanishing.
Central Chow components and height axioms2
The all-h¹ core and its algebraic and Griffiths parts.
BB-CENTRAL-DATA- A curve-normalized height systemConditional
The four explicit assumptions used by the height-properties theorems.
BB-HEIGHT-DATA
Chow decompositions and Lefschetz theory3
- The maximal curve-visible Chow projectorInternal proof available
An actual idempotent separates a fixed abelian summand from the residual motive.
BB-CHOW-PROJECTOR - Arithmetic Lefschetz from off-central vanishingInternal proof available
An unconditional equivalence in the actual Chow group of a pointed product.
R2-L1 Only the arithmetic middle degree should survive in the all-h¹ homological kernel.
OPEN-OFFCENTRAL
Central heights and the residual pairing4
Eliminate the algebraic contribution to the height radical.
OPEN-AJ-KERNEL- The algebraic radical and residual pairingConditional
Orthogonalize before passing to the Griffiths quotient.
BB-RESIDUAL-PAIRING - The primitive sum-of-squares identityConditional
An exact factorial and sign identify primitive heights with central heights.
R2-L2 - An equivalent package for products of curvesConditional
The geometric, radical, and integral conjectures assemble into one exact equivalence.
R2-L3
Research planNext problem
- Arithmetic Lefschetz and positivity
For pointed products of curves, prove the Chow vanishing and central positivity assertions used by the arithmetic reduction theorem.
Precise problem and proposed argument
- Arithmetic Lefschetz and positivity
Integral NorthcottControl integral cycle classes, their heights and their fields of definition.18 records
The objects being counted are projected integral cycle classes. Real positivity, a common integral lattice and control of the fields of definition are separate parts of the finiteness problem.
Finiteness and real positivity5
Count geometric classes with integral representatives, not all rational multiples.
BB-INTEGRAL-IMAGEBounded height and bounded field degree should leave finitely many residual cycle classes.
OPEN-RESIDUAL-NC- The exact central Northcott criterionConditional
Central finiteness is equivalent to algebraic AJ injectivity plus residual finiteness.
R2-N1 A weaker target than residual integral Northcott.
OPEN-REAL-POSITIVITY- Northcott forces real positivityInternal proof available
Finiteness excludes real null directions, even without a finite-rank assumption.
R2-N2
Arithmetic regulators and detection4
- The residual arithmetic regulatorInternal proof available
An actual Chow projector yields a regulator on the Griffiths quotient.
R2-R1 - Weight −1 and finite-dimensional arithmetic targetsInternal proof available
The full rational cohomological target is controlled over each fixed field.
R2-R2 - Local integral images of residual regulatorsInternal proof available
For each coefficient prime, integral residual cycle images lie in a finite module.
R2-R3 Injectivity must survive extension of scalars to the coefficient field.
OPEN-SCALAR-DETECTION
Integral control over a fixed field5
A single rational lattice should control all integral residual classes over a fixed field.
OPEN-INTEGRAL-LATTICE- Five equivalent lattice conditionsInternal proof available
Finite-rank Northcott is exactly a discreteness condition.
R2-N3 - An all-prime saturation criterionInternal proof available
Local integral detectors yield a global denominator when they agree almost everywhere.
R2-N4 - Weak Mordell–Weil and a positive height gapInternal proof available
A quantitative descent criterion without a prior finite-rank assumption.
R2-N5 - Why p-adic finiteness alone is not NorthcottInternal proof available
An abstract denominator example isolates a missing global condition.
BB-DENOMINATOR-EXAMPLE
Varying the field4
- Trace to the field of a cycle classInternal proof available
The descent multiplier is the relative field degree.
R2-F1 - Ramification is the remaining variable-field conditionInternal proof available
Fixed-field Northcott plus finite ramification sets is equivalent to bounded-degree Northcott.
R2-F2 Bound the places of ramification as both the cycle class and its field vary.
OPEN-CLASS-RAMIFICATIONA stronger route from height to control of the class field.
OPEN-DISCRIMINANT
Research planNext problem
- Integral Northcott finiteness
Prove bounded-height finiteness for the projected images of integral cycles, first over a fixed field and then in bounded degree.
Precise problem and proposed argument
- Integral Northcott finiteness
Metrics and familiesCompare normalized metrics and study degeneration and height variation.5 records
The determinant comparisons retain the stated normalization and model hypotheses. Bad reduction and height estimates in families are further research problems.
Normalized determinant metrics3
- An exact normalized Quillen isometryInternal proof available
Origin normalization removes the constant in a determinant comparison.
R2-M1 - Rank zero, Deligne pairings, and filtered Riemann–RochInternal proof available
The determinant metric becomes the flat Poincaré metric of a divisor class.
R2-M2 - Good-reduction determinant metricsInternal proof available
A smooth source model fixes the non-Archimedean metric exactly.
R2-M3
Degeneration and variation2
Determine the correction that connects semistable model determinants to actual Green heights.
OPEN-BAD-REDUCTION-METRICConnect the local theory to adelic positivity and Diophantine finiteness.
BB-FAMILY-HEIGHTS
Research planNext problem
- Correspondence adjunction and classical heights
Prove that the constructed pairing has the required correspondence identity and agrees with the classical height on common domains.
Precise problem and proposed argument
- Correspondence adjunction and classical heights
No matching item. Try a shorter title or a permanent ID.