Alphabeta Math
Pipeline-generated
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.

Complete retarded point fields

1 · Prerequisites

2 · Summary

The completion of the retarded point-field layer of the classical electromagnetism mathematics track. The single item, Complete retarded point fields: unique retarded root, the delta substitution with denominator 1−n⋅β, Liénard–Wiechert potentials and fields, and a uniform far-zone remainder, fixes a C3 worldline with a uniform subluminal bound ∣z′∣≤v∗<c and assembles the full point-source computation in four clauses. Clause (a) states the event form of the retarded root: every event off the worldline has exactly one root s=s(x,t)<t of t−s=∣x−z(s)∣/c, the root map is C3 with derivative formulas ∂ts=κ−1 and ∇xs=−n/(cκ), the emission-time Jacobian 1−n⋅β=κ is uniformly bounded below by 1−v∗/c, and κ=1−n⋅β is exactly the denominator produced by the delta substitution. Clause (b) records the delta substitution as the exact history-integral pairing identity with weight (κr)−1, under Countable Choice, and identifies the Liénard-Wiechert potentials φ=C/(κr) and A=βφ/c as C2 functions off the worldline whose zero extensions are locally integrable and define regular distributions. Clause (c) gives E=−∇φ−∂tA and B=curl⁡A as C1 fields off the worldline together with the velocity/acceleration split and the uniform bound ∣Evel∣≤∣C∣(1+v∗/c)(1−v∗/c)−3r−2. Clause (d) evaluates the field on the ray family xr=z(s0)+rn, tr=s0+r/c, where the root is exactly the emission time s0, and proves the uniform far-zone remainder with explicit constant and the equivalent O(r−1) limit of rE.

The item is mathematics: it carries domain: mathematics, depends only on mathematical suppliers, and asserts no self-force law, radiation-reaction postulate, physical Maxwell coupling, finite-energy or radiation-flux conclusion. Its physical consumers are the classical electromagnetism pages that use the retarded fields; empirical qualifications of any physical use remain with the consuming physical item. The companion examples page records the three hypothesis/convention boundaries on which the item's hypotheses depend.

3 · Logical flowchart

4 · Definitions, postulates, theorems and proofs

TheoremStatement: AI-adaptedProof: AI-adaptedprecheck passOpen item page →

Complete retarded point fields: unique retarded root, the delta substitution with denominator 1−n⋅β, Liénard–Wiechert potentials and fields, and a uniform far-zone remainder

Statement

Assume the Axiom of Countable Choice (ACω), used through the change-of-variables corollaries inside the proof of the retarded-convolution identity cited in clause (b) (The Axiom of Countable Choice (ACω)). Let c>0, let v∗<c, let z:R→R3 be of class C3 with ∣z′(s)∣≤v∗ for every s∈R (The derivative and the Riemann integral of a vector-valued function: an intrinsic derivative and a componentwise integral, Ck Euclidean maps and diffeomorphisms), let q∈R and ε0>0, put μ0:=(c2ε0)−1 and C:=q/(4πε0), and write Ω:=R3×R and Γ:={(z(s),s):s∈R}.

(a) [Root and emission-time Jacobian] For every (x,t)∈Ω∖Γ (equivalently x≠z(t)) there is exactly one s=s(x,t)<t with t−s=∣x−z(s)∣/c. The map s:Ω∖Γ→R is of class C3, and with R:=x−z(s), r:=∣R∣, n:=R/r, β:=z′(s)/c and κ:=1−n⋅β (The Euclidean inner product ⟨x,y⟩=∑k<nxkyk on Rn) one has ∂ts(x,t)=κ−1,∇xs(x,t)=−nc κ,κ∈[1−v∗/c, 1+v∗/c], so κ≥1−v∗/c>0 uniformly on Ω∖Γ. Moreover, for the fixed x and near the root, the function s′↦s′+∣x−z(s′)∣/c is differentiable at s′=s with derivative κ(x,t): this emission-time Jacobian 1−n⋅β is the denominator produced by the delta substitution of clause (b).

(b) [Delta substitution and Liénard-Wiechert potentials] Let ρ,J∈D′(Ω) be the point-source distributions of Point-support spacetime densities: distributions, continuity equation and endpoint fluxes attached to q and the worldline z, let K be the retarded wave kernel of The retarded wave kernel is a distribution supported on the cone, solves the point-wave equation, and convolves to the retarded solution, and for ψ∈D(Ω) put (K⊛ψ)(y,τ):=⟨K,ψ(y+⋅,τ+⋅)⟩ and define the retarded-history pairings by the explicit integrals ⟨Φ,ψ⟩:=(q/ε0)∫R(K⊛ψ)(z(s),s) ds, ⟨Υ,ψ⟩:=μ0q∫Rz′(s)(K⊛ψ)(z(s),s) ds, componentwise in the second case. These are the history-integral interpretation supplied by clause (b) of Retarded potentials of a point history as the retarded convolution of the point sources, and the off-worldline Lienard-Wiechert fields, not ordinary distribution action on a noncompact test. Then these pairings are finite for every ψ and their exact retarded-convolution values are ⟨Φ,ψ⟩=∫Ω∖ΓC ψκ r dx dt,⟨Υ,ψ⟩=∫Ω∖Γβc C ψκ r dx dt (the second identity componentwise); equivalently, Φ and Υ are the regular distributions of the locally integrable zero extensions to Γ of the C2 functions φ:=C/(κr) and A:=βφ/c on Ω∖Γ. The weight (κr)−1 with κ=1−n⋅β is the exact form of the substitution with denominator 1−n⋅β (the rigorous statement is the displayed pairing identity; composition of δ with a function is not part of the distribution calculus used here).

(c) [Fields and the velocity/acceleration split] With E:=−∇φ−∂tA and B:=curl⁡A on Ω∖Γ (Scalar-vector potential pairs, gauge transformation, Lorenz divergence and the wave expression), and writing β˙:=z′′(s)/c at the root, E=Evel+Eacc,Evel=C(1−∣β∣2)(n−β)κ3r2,Eacc=C n×((n−β)×β˙)c κ3r,B=n×Ec at every event of Ω∖Γ (The cross product in R3, The cross product is bilinear, alternating, and orthogonal to both factors). Both E and B are C1 on Ω∖Γ, and the velocity field obeys the bound ∣Evel(x,t)∣≤∣C∣ 1+v∗/c(1−v∗/c)3 r(x,t)−2 at every event.

(d) [Uniform far-zone remainder] For every s0∈R and every n∈R3 with ∣n∣=1, put xr:=z(s0)+rn and tr:=s0+r/c for r>0, and set β0:=z′(s0)/c, β˙0:=z′′(s0)/c, κ0:=1−n⋅β0. Then (xr,tr)∈Ω∖Γ and s(xr,tr)=s0 for every r>0, and E(xr,tr)=C n×((n−β0)×β˙0)c κ03 r+ρE(s0,n,r),B(xr,tr)=n×E(xr,tr)c, where the remainder is bounded uniformly in the direction, ∣ρE(s0,n,r)∣≤∣C∣ 1+v∗/c(1−v∗/c)3 r−2for every n∈S2 and every r>0. Equivalently, rE(xr,tr)→C n×((n−β0)×β˙0)/(c κ03) as r→∞, uniformly over n∈S2 at the explicit rate O(r−1) given by the same bound multiplied by r. The uniformity is in the direction n at a fixed emission event s0; no uniformity in s0 is asserted (z′′ is not assumed bounded), and no finite-energy, decay or radiation-flux conclusion is drawn.

This is a mathematical model computation for the stipulated C3 uniformly subluminal worldline and the retarded choice; no self-force law, no radiation-reaction postulate and no physical Maxwell coupling is asserted or proved, and the uniform speed bound and the exclusion of Γ are essential hypotheses (see the companion example page).

Facts & Assumptions

Given: A speed c>0, a real bound v∗<c, a worldline z:R→R3 of class C3 with ∣z′(s)∣≤v∗ for every s∈R, a real q and a positive ε0, together with the constants μ0=(c2ε0)−1, C=q/(4πε0) and the sets Ω=R3×R, Γ={(z(s),s):s∈R}.

[F1]

Let c>0, 0≤v∗<c and z:R→R3 of class C1 with ∣z′(s)∣≤v∗ for every s. Then for every (x,t)∈R3×R there is exactly one real root s=s(x,t) of t−s=∣x−z(s)∣/c, and s≤t with equality exactly when x=z(t); on the open off-instantaneous-worldline domain D={(x,t):x≠z(t)} one has s<t, r=∣x−z(s)∣=c(t−s)>0, the root is C1, and with n=(x−z(s))/r, β=z′(s)/c, κ=1−n⋅β its derivatives are ∂ts=1/κ and ∇xs=−n/(cκ), where 1−v∗/c≤κ≤1+v∗/c. If z is Ck for an integer k≥1, the root is Ck on D; in particular a C3 trajectory gives a C3 root (Global existence, uniqueness and differentiability of the retarded root).

[F2]

For f:A→Rm with A⊆R and a limit point c∈A, f is differentiable at c when the difference quotients qf,c(x)=1x−c(f(x)−f(c)) converge as x→c, and then the derivative is the limit f′(c):=lim⁡x→cqf,c(x)∈Rm; the notion is intrinsic and is computed componentwise (The derivative and the Riemann integral of a vector-valued function: an intrinsic derivative and a componentwise integral).

[F3]

A map f:U→Rq on an open U⊆Rm is of class Ck when every word of at most k coordinate partial derivatives of each component exists and is continuous, including the word of length zero; it is smooth when it is Ck for every k (Ck Euclidean maps and diffeomorphisms).

[F4]

On Rn the Euclidean inner product is ⟨x,y⟩=∑k<nxkyk, it is bilinear and symmetric, ⟨z,z⟩≥0 with ⟨z,z⟩=0 only for z=0, and the Euclidean norm is ∥x∥2=⟨x,x⟩ (The Euclidean inner product ⟨x,y⟩=∑k<nxkyk on Rn).

[F5]

For all x,y∈Rn one has ∣⟨x,y⟩∣≤∥x∥2∥y∥2; ∥⋅∥2 is a norm, so ∥u+v∥2≤∥u∥2+∥v∥2 and ∥λu∥2=∣λ∣∥u∥2, and consequently the reverse triangle inequality ∣∥u∥2−∥v∥2∣≤∥u−v∥2 holds (Cauchy-Schwarz ∣⟨x,y⟩∣≤∥x∥2∥y∥2 with its equality case, the triangle inequality for ∥⋅∥2, the parallelogram law and polarisation).

[F6]

Let I⊆R be an open interval, z:I→R3 a C1 trajectory, q∈R, and Ω⊇Γ open on which the graph map is proper. Then ⟨ρ,ψ⟩=q∫Iψ(z(t),t)dt and ⟨J,ψ⟩=q∫Iz′(t)ψ(z(t),t)dt define distributions with ∂tρ+div⁡J=0; properness is automatic for Ω=R3×I, in particular for the all-time history I=R (Point-support spacetime densities: distributions, continuity equation and endpoint fluxes).

[F7]

The retarded wave kernel is the distribution K on R3×R given by ⟨K,ψ⟩=14π∫R3ψ(y,∣y∣/c)∣y∣dy, and for (y,τ)∈R3×R the shifted pairing (K⊛ψ)(y,τ):=⟨K,ψ(y+⋅,τ+⋅)⟩ equals the absolutely convergent integral 14π∫ψ(y+z,τ+∣z∣/c)∣z∣dz (The retarded wave kernel is a distribution supported on the cone, solves the point-wave equation, and convolves to the retarded solution).

[F8]

Assume ACω. Let c>0, q∈R, ε0>0 and let z:R→R3 be C3 with ∣z′(s)∣≤v∗<c for all s. With K the retarded kernel of [F7] and, for tests ψ, ⟨Φ,ψ⟩=(q/ε0)∫R(K⊛ψ)(z(s),s)ds, ⟨Υ,ψ⟩=μ0q∫Rz′(s)(K⊛ψ)(z(s),s)ds componentwise, these pairings are finite for every ψ and satisfy the exact retarded-convolution identities ⟨Φ,ψ⟩=∫Ω∖ΓCψκrdxdt and ⟨Υ,ψ⟩=∫Ω∖ΓβcCψκrdxdt; consequently Φ and Υ are the regular distributions of the locally integrable zero extensions of the C2 functions φ=C/(κr) and A=βφ/c on Ω∖Γ (Retarded potentials of a point history as the retarded convolution of the point sources, and the off-worldline Lienard-Wiechert fields).

[F9]

In the situation of [F8], the fields E=−∇φ−∂tA and B=curl⁡A on Ω∖Γ satisfy E=C[(1−∣β∣2)(n−β)κ3r2+n×((n−β)×β˙)cκ3r],B=n×Ec and are of class C1 there, where β˙=z′′(s)/c is evaluated at the retarded root (Retarded potentials of a point history as the retarded convolution of the point sources, and the off-worldline Lienard-Wiechert fields).

[F10]

For c>0 and an open Ω⊆R3×R, a Ck scalar-vector potential pair (φ,A) on Ω with k≥1 has fields E=−∇φ−∂tA and B=curl⁡A, both of class Ck−1 (Scalar-vector potential pairs, gauge transformation, Lorenz divergence and the wave expression).

[F11]

The right-handed cross product on R3 is u×v=(uyvz−uzvy, uzvx−uxvz, uxvy−uyvx) (The cross product in R3); it is bilinear and alternating, ⟨u×v,w⟩=det⁡[u v w], and u×v is orthogonal to both u and v (The cross product is bilinear, alternating, and orthogonal to both factors).

[F12]

ACω: every family (Xn)n∈N of nonempty sets has a choice function (The Axiom of Countable Choice (ACω)).

Proof

Given: The hypotheses of the Statement: c>0, v∗<c, a C3 worldline z with ∣z′(s)∣≤v∗ for every s, q∈R, ε0>0, and the derived data μ0, C, Ω, Γ.

Proof technique: direct: clause (a) is the root lemma together with the one-variable computation of the emission-time Jacobian; clause (b) is the supplier's pairing identity read in the stated history-integral interpretation; clause (c) is the supplier's field formula with an elementary uniform bound; clause (d) evaluates the field on the ray family and reads off the uniform limit.

1.1F1F2F3given

Hypothesis audit and the root. The hypothesis gives 0≤∣z′(0)∣≤v∗, so 0≤v∗<c, and z is of class C3, hence C1; the root lemma [F1] therefore applies with these c, v∗, z, and since z is C3 its Ck branch with k=3 applies. Hence for every (x,t)∈Ω∖Γ there is exactly one s=s(x,t)<t with t−s=∣x−z(s)∣/c, the root map is of class C3 on Ω∖Γ, and with R=x−z(s), r=∣R∣>0, n=R/r, β=z′(s)/c, κ=1−n⋅β one has ∂ts=κ−1, ∇xs=−n/(cκ) and 1−v∗/c≤κ≤1+v∗/c, so κ≥1−v∗/c>0. The differentiability vocabulary and the Ck vocabulary used throughout are [F2] and [F3].

1.2F2F4F5algebra

The norm term. Fix (x,t)∈Ω∖Γ with root s and put R=x−z(s), r=∥R∥2>0, n=R/r; I show that ψ(s′):=∥x−z(s′)∥2 is differentiable at s′=s with ψ′(s)=−⟨z′(s),R⟩/r. For h≠0 put Vh:=z(s+h)−z(s)h; by [F2] one has Vh→z′(s) as h→0, so with M:=∥z′(s)∥2+1 there is δ>0 with ∥Vh∥2≤M for 0<∣h∣<δ; shrinking δ if necessary so that Mδ<r/2, the vector w2:=x−z(s+h) satisfies ∥w2−R∥2=∥z(s+h)−z(s)∥2=∣h∣∥Vh∥2≤M∣h∣ and ∥w2∥2≥r−M∣h∣>r/2. Writing w1:=R, the identity ∥w2∥2−∥w1∥2=⟨w2−w1, w2+w1⟩∥w2∥2+∥w1∥2 holds, because ∥w2∥22−∥w1∥22=⟨w2+w1,w2−w1⟩ by bilinearity and symmetry and the denominator is positive [F4]; since w2−w1=−hVh, dividing by h gives ψ(s+h)−ψ(s)h=−⟨Vh,Uh⟩ with Uh:=w1+w2∥w1∥2+∥w2∥2. The triangle inequality gives ∥Uh∥2≤1, and ∥Uh−n∥2=∥rw2−∥w2∥2w1∥2r(r+∥w2∥2)≤2∥w2−w1∥2r+∥w2∥2≤2M∣h∣r, where the middle step is the reverse triangle inequality [F5]. Cauchy-Schwarz then gives ∣⟨Vh,Uh⟩−⟨z′(s),n⟩∣≤∥Vh−z′(s)∥2∥Uh∥2+∥z′(s)∥2∥Uh−n∥2≤∥Vh−z′(s)∥2+∥z′(s)∥22M∣h∣r, which tends to 0 as h→0; hence the difference quotients converge and ψ′(s)=−⟨z′(s),n⟩=−n⋅z′(s) [F5].

1.3F1givenalgebra

The ray family. Fix s0∈R and n∈R3 with ∥n∥2=1, and let r>0; put xr:=z(s0)+rn and tr:=s0+r/c. Then tr−s0=r/c=∥xr−z(s0)∥2/c, so s0 solves the retarded equation at the event (xr,tr); by the uniqueness in [F1] applied at that event the root is s(xr,tr)=s0, and since s0<tr, the equality clause of [F1], which says that the root equals t exactly when the observation point is z(t), gives xr≠z(tr), that is (xr,tr)∈Ω∖Γ.

2.1step 1.2F1F2algebra

Emission-time Jacobian. With the fixed x of step 1.2, the function gx(s′):=s′+1c∥x−z(s′)∥2 is differentiable at s′=s, as the sum of the identity map, whose derivative is 1, and the constant multiple 1cψ of the differentiable map ψ of step 1.2, and its derivative is gx′(s)=1−1cn⋅z′(s)=1−n⋅β=κ(x,t), using β=z′(s)/c and κ=1−n⋅β from step 1.1; this is exactly the emission-time Jacobian 1−n⋅β asserted in clause (a).

2.2F6F7F8F12step 1.1given

Clause (b). The definitions of the retarded-history pairings in clause (b) coincide with [F8], with the same signed constant C=q/(4πε0), the same β=z′(s)/c and κ=1−n⋅β of step 1.1, and the same shifted-kernel pairing (K⊛ψ) of [F7]; the point-source distributions and their continuity equation are [F6], whose proper-graph hypothesis is automatic for the all-time history I=R. The supplier [F8] proves, under ACω [F12], that the pairings are finite for every test, that the displayed exact identities hold, and that Φ and Υ are the regular distributions of the locally integrable zero extensions of φ=C/(κr) and A=βφ/c, which are C2 on Ω∖Γ by the supplier's regularity branch fed by the C3 root of step 1.1; this is the whole content of clause (b).

2.3step 1.1F5F8F9F10F11algebra

Clause (c). By step 1.1 and [F8] the pair (φ,A) is a C2 potential pair on Ω∖Γ, so [F10] identifies its fields as E=−∇φ−∂tA and B=curl⁡A, both C1; the supplier's formula [F9] gives the displayed E and B, whose acceleration numerator is the cross product of [F11], and splitting the bracket defines Evel and Eacc as in the Statement. For the bound, at every event ∥n∥2=1 and ∥β∥2=∥z′(s)∥2/c≤v∗/c, so ∣1−∥β∥22∣≤1 and the triangle inequality [F5] gives ∥n−β∥2≤1+∥β∥2≤1+v∗/c; with κ≥1−v∗/c>0 from step 1.1 this yields κ3≥(1−v∗/c)3 and hence ∥Evel(x,t)∥2≤∣C∣1+v∗/c(1−v∗/c)3r−2, the signed constant C being kept in the formula and ∣C∣ used in the bound.

3.1F5F11step 1.3step 2.3algebra

The far-zone expansion. At the event (xr,tr) of step 1.3 the retarded data are the root s0, the separation xr−z(s0)=rn, the retarded distance ∥xr−z(s0)∥2=r, which is the ray parameter, the unit vector n, and β=β0=z′(s0)/c, β˙=β˙0=z′′(s0)/c, κ=κ0=1−n⋅β0; hence step 2.3 gives E(xr,tr)=Cn×((n−β0)×β˙0)cκ03r+ρE(s0,n,r) with ρE(s0,n,r):=Evel(xr,tr) and B(xr,tr)=n×E(xr,tr)c, while the uniform bound of step 2.3 gives ∣ρE(s0,n,r)∣≤∣C∣1+v∗/c(1−v∗/c)3r−2 for every unit n and every r>0. Multiplying by r yields ∥rE(xr,tr)−Cn×((n−β0)×β˙0)cκ03∥2≤∣C∣1+v∗/c(1−v∗/c)31r, which tends to 0 as r→∞ with a constant depending only on C, v∗ and c; hence rE(xr,tr)→Cn×((n−β0)×β˙0)cκ03 uniformly over the unit directions n at the fixed emission event s0, at the explicit rate O(r−1), and no bound on z′′ enters the remainder.

4.1givenstep 1.1step 1.3step 2.1step 2.2step 2.3step 3.1∎

Assembly. Steps 1.1 and 2.1 give clause (a): at every event off the worldline there is a unique retarded root strictly below the observation time, the root map is C3, ∂ts=κ−1, ∇xs=−n/(cκ) and 1−v∗/c≤κ≤1+v∗/c, and the emission-time Jacobian is 1−n⋅β=κ; step 2.2 gives clause (b) with the finiteness of the history pairings, the exact pairing identities and the identification of Φ,Υ as the regular distributions of the zero extensions of the C2 potentials φ=C/(κr) and A=βφ/c; step 2.3 gives clause (c) with the velocity/acceleration split, the C1 regularity and the uniform velocity-field bound; and steps 1.3 and 3.1 give clause (d) with s(xr,tr)=s0, the expansion, the direction-uniform remainder bound and the equivalent O(r−1) limit of rE. All hypotheses are those of the Statement, ACω is assumed and used only through clause (b), the signed constant C is retained in the field formulas while the bounds use ∣C∣, and no self-force, radiation-reaction, Maxwell-coupling, finite-energy, decay or flux conclusion is asserted.

5 · Examples, experiments and counterexamples

None yet.

Sources