Home·The theorem·The proof in one reading·Architecture·Units·Manuscript·Glossary·FAQ·Work items·Data·Method·Provenance·Changes
← IV.1 Infinite volume: extraction of a subsequence and the Euclidean axioms of the limitPart V·Detailed notesV.2 Osterwalder-Schrader reconstruction of a Wightman theory →

V.1. Decomposition of the limit into phases

Contents

1. What this unit proves 2

2. What it uses 3

3. Strategy 4

4. Steps 4

5. Where the difficulty lies 4

6. Notation 5

7. What remains unexplained 5

Code dependencies 5

Lean source 5

Additional source locations 6

Key statements 10

Part V — Phases and reconstruction.

Decomposes the mixed Euclidean limit into phases and explains the selection of a phase with a unique vacuum.

Status: edited. Condensed from the unit's detailed notes, which were checked against the linked Lean in two rounds and edited; the local limitations still apply. Status meanings.

Overview · All units · Detailed notes · ← Previous: IV.1 · Next: V.2 →

Results · Inputs · Strategy · Proof steps · Notation · What remains unexplained · Code dependencies · Lean source · Key statements

Reading scope. The phase selection and its reconstruction do not prove a mass gap. See V.2 and V.3.

Source inventory and conventions

Part: V, phases and reconstruction. Lean: YM/OS, YM/Phase, YM/Ground, YM/Heat, YM/SNAG, YM/Paley, YM/Spectral, the work items B01–B18 in YM/Work, and the assembly files Bid/Phase.lean, Bid/Fetch.lean, Bid/Market.lean, Org/Phase.lean. Size: about 5,700 lines in line-counted declarations in the source allocation, including about 2,700 in B01–B18, plus the assembly files.

1. What this unit proves

\(S\) is the limit table; words are separated unless stated otherwise. “Hierarchy” means permutation symmetry, normalization and unit labels, signed-axis symmetries and reality, translation invariance, point and smeared reflection positivity for every coordinate axis, smoothness on separated configurations, and caps on all derivatives (IV.1). A cap bounds each fixed correlation derivative by \(Cr^{- A}\) for \(0 < r \leqslant 1\) and mutual sup-distance \(\geqslant r\), with \(C \geqslant 0\), \(A \in \mathbb{N}\) depending on arity, degrees, labels and derivative order. The OS inner product is \(\langle b, b' \rangle = S (\Theta b \cdot b')\), and \(\Omega = | \emptyset \rangle\).

Proposition V.1.1 (chambers and ground space). For every \(S\) satisfying the hierarchy and coordinate axis \(a\), its OS space carries a strongly continuous self-adjoint contraction semigroup and a commuting strongly continuous spatial unitary group, \(E_t |b \rangle = |b + te_a \rangle\), \(U_y |b \rangle = |b + y \rangle\) (\(t \geqslant 0\), \(y \perp e_a\)). These identities hold for every \(a\)-positive \(b\). This is a chamber. The fixed space \(\mathcal{H}_0\) of all \(E_t\) contains \(\Omega\), and \(E_t \psi \to P_0 \psi\) for every \(\psi\), where \(P_0\) projects onto \(\mathcal{H}_0\). Status: theorem. Lean: OS.implement, OS.Semigroup.emerges.

Proposition V.1.2 (spectral calculus). In every chamber, writing \(E_t = e^{- tH}\), there is a strongly continuous unitary group determined by \(e^{isH} E_1 \psi = E_1^{1 - is} \psi\) through continuous functional calculus. The strongly continuous unitary representation \(V (x) = e^{ix^a H} U_{x - x^a e_a}\) has a projection-valued measure \(P\); for every \(\psi\), \(\mu_{\psi} (B) = \langle \psi, P (B) \psi \rangle\) is supported in \(p^a \geqslant 0\), and, for every \(x\) and \(\text{Im} \hspace{0.17em} z \geqslant 0\),

\(\displaystyle \langle \psi, V (x) e^{i (\text{Re} \hspace{0.17em} z) H} E_{\text{Im} \hspace{0.17em} z} \psi \rangle = \int e^{ip \cdot x} e^{ip^a z} \hspace{0.17em} d \mu_{\psi} (p) .\)

Status: theorem. Lean: Semigroup.unit, Spectral.Orbit.total, Semigroup.heat.

Proposition V.1.3 (slab operators). Under the hierarchy and bench V.1.7, every time-axis chamber has unique bounded \(T_d (w)\) for \(d > 0\), \(w\) in \((0, d)\), satisfying \(T_d (w) |b \rangle = |w \cdot \tau_d b \rangle\) for every positive \(b\). For every \(D \in \mathbb{N}\), \(\rho > 0\), some \(C > 0\), \(\kappa \in \mathbb{N}\) give

\(\displaystyle \|T_d (w)\| \leqslant C^{|w|} (|w| / r)^{\kappa |w|}\)

for all \(d > 0\), \(0 < r \leqslant 1\), and \(r\)-buffered \(w\) with degrees \(\leqslant D\), label norms \(\leqslant \rho\); the empty-word bound is \(1\). With the plan's derivative identities, for fixed \(d\) and labels, \(T_d (w (x))\) is norm-\(C^{\infty}\) on the open slab configuration set, and coordinate differentiation replaces the corresponding label by \(\partial_j v\). More generally this smoothness holds for any slab family satisfying (B) on a time-axis chamber of \(S'\), provided \(S, S'\) satisfy the hierarchy, \(S'\) respects \(S\), and \(S'\) satisfies the derivative identities of the plan on \(S\). Status: theorem conditional on the bench. Lean: Work.B01.answer, Work.B02.answer.

Proposition V.1.4 (compressed algebra). Assume the hierarchy, \(O (4)\)-covariance, a time-axis chamber and a slab family satisfying V.1.3's insertion identity. Then spatial translations fix \(\mathcal{H}_0\) pointwise. For every separated \(w\), \(M (w) = P_0 T_d (w + s) |_{\mathcal{H}_0}\) is independent of admissible \(s, d\); \(M (b) \Omega = P_0 |b \rangle\) for positive \(b\), and \(\langle \Omega, M (w) \Omega \rangle = S (w)\). For any finite collection of separated words \(q_i\) and shifts \(a_i (n)\) with \(t (a_j (n) - a_i (n)) \to + \infty\) for \(i < j\), the concatenations are eventually separated and their correlations tend to \(\langle \Omega, \prod_i M (q_i) \Omega \rangle\), regardless of spatial shifts. The generated norm-closed \(\ast\)-algebra \(\mathcal{A}\) is commutative; \(\Omega\) is cyclic and separating. With smooth slab laws, \(M\) is permutation/translation invariant, satisfies \(M (\emptyset) = 1\), permits dropping unit labels together with their points, satisfies \(M (w)^{\ast} = M (\Theta w)\), is multilinear through products of \(F_d (S)\) and norm-smooth in positions; \(\mathcal{A}\) is separable. Status: theorem. Lean: Ground.adjunction, Work.B03.answer.

Proposition V.1.5 (decomposition). Under V.1.4 including smoothness, \(X = \text{Spec} (\mathcal{A})\) is compact metrizable, with full-support Borel probability \(\nu\). Its points bijectively parametrize characters \(\chi_{\xi}\); every evaluation is continuous and \(\int_X \chi_{\xi} (p) d \nu = \langle \Omega, p \Omega \rangle\) for all \(p \in \mathcal{A}\). Each \(S^{\xi} (w) = \chi_{\xi} (M (w))\), extended by zero off separated configurations, is a table; evaluations are continuous in \(\xi\), and \(S (w) = \int_X S^{\xi} (w) d \nu\) for every separated \(w\). Status: theorem. Lean: Ground.resolution, Resolution.barycenter.

Proposition V.1.6 (phase laws). Assume V.1.3–V.1.5 with smooth laws for the field-algebra plan, the Seed, grid hypothesis and field-algebra laws on \(S\). For every separated word \(w\), \(M (w) = M (\vartheta w)\) with arbitrary complex labels, and \(M (w) = M (Kw)\) for every \(K \in O (4)\). For every \(\xi \in X\), \(S^{\xi}\) satisfies the hierarchy, respects \(S\), is \(O (4)\)-covariant for the same action, inclusion/derivative identities and stress-tensor Ward identities, and \(N_d (S^{\xi}) = N_d (S)\) for every degree \(d\). For every \(D \in \mathbb{N}\), \(\rho > 0\), some \(K > 0\), \(\kappa \in \mathbb{N}\) give \(|S^{\xi} (w) | \leqslant K^{|w|} (|w| / r)^{\kappa |w|}\) uniformly in \(\xi\), \(0 < r \leqslant 1\), and words with degrees \(\leqslant D\), label norms \(\leqslant \rho\), mutual sup-distance \(\geqslant r\). Status: theorem. Lean: Work.B05.answer, Retain.lift_detect.

Proposition V.1.7 (bench). For every \(D \in \mathbb{N}\), \(\rho > 0\), some \(C_0 > 0\), \(\kappa \in \mathbb{N}\) work for every \(d > 0\), \(0 < r \leqslant 1\), nonempty \(r\)-buffered \(w\) with degrees \(\leqslant D\), norms \(\leqslant \rho\): there are a lattice level \(b\) and \(s \leqslant d\), \(s \in a_b \mathbb{N}\), with \(w\) in \((0, s)\), such that every positive \(v\) has \(B_v\) satisfying

\(\displaystyle |S (\Theta v_n \cdot v_n) | \leqslant B_v A^{4 n} \quad (n \geqslant 0), \qquad A = C_0^{|w|} (|w| / r)^{\kappa |w|},\)

where \(v_0 = v\), \(v_{n + 1} = \tau_s (\Theta w) \cdot \tau_s (w \cdot \tau_s v_n)\). Status: hypothesis here, proved in VII.4 for every Seed table. Lean: Next.Phase.seats.

Proposition V.1.8 (vacuum uniqueness). Given V.1.5 and the hierarchy for every phase, fix a time-axis chamber for each \(\xi\). Then \(\mathcal{H}_0^{\xi} =\mathbb{C} \Omega_{\xi}\) for \(\nu\)-almost every \(\xi\), so such phases exist. Every time-axis phase chamber has slab operators with \(\|T_d^{\xi} (w)\| \leqslant \|T_d (w)\|\), inheriting (B) whenever the original family has it, and, under V.1.6, smoothness. Status: theorem. Lean: Cluster.lifts, Work.B13.answer.

Proposition V.1.9 (selected-phase certificate). For every Wilson family with extraction order and limit \(S\), assuming its Seed, grid hypothesis and field algebra with laws, there exist a time-axis chamber of \(S\) with spatially fixed ground space, a smooth slab family satisfying (B), decomposition as V.1.5 with every phase satisfying V.1.6's hierarchy, field laws and null-space equality, and a selected phase with a time-axis chamber having ground space \(\mathbb{C} \Omega\) and smooth slab family. Status: theorem. Lean: Phase.Loom.cleanse.

2. What it uses

3. Strategy

Compress bounded slab insertions to the ground space. Their commutative algebra describes infinite-time-separated observables; characters yield phase tables and the vacuum state supplies their mixing measure. Operator identities give laws in every phase; an integrated variance identity selects a unique-vacuum phase almost everywhere.

4. Steps

  1. Construct contractions by OS doubling; functional calculus and SNAG give spectral measures, with positive energy from half-plane analyticity. OS.contract, Semigroup.nonneg.

  2. Turn bench train bounds into slab norms by the power argument; absorb leftover width with \(E_{d - s}\). Drive.against.

  3. Rotate spatial displacements into time to fix ground vectors; exchange middle clusters in a four-cluster limit to prove commutativity. Ground.inert, Ground.exchanges.

  4. Gelfand–Riesz gives a compact metrizable spectrum and the vacuum measure; separation of \(\Omega\) gives full support. Ground.resolution.

  5. For real labels, \(D = M (w) - M (\vartheta w)\) is skew-adjoint. Positivity of variances of many bounded lattice approximants forces \(- \|D \Omega \|^2 \geqslant 0\); separation gives \(D = 0\). Work.B05.answer.

  6. Transfer positivity and Ward identities before applying characters. Corrected-field detector moments, full support and finite-dimensional injectivity preserve null spaces. Work.B08.answer, Work.B09.answer, Retain.lift_detect.

  7. Dominated convergence makes \(\|P_0^{\xi} |b \rangle \|^2 - |S^{\xi} (b) |^2\) have integral zero; one common countable total family gives uniqueness almost everywhere. Cluster.exp_real, Work.B14.universal.

5. Where the difficulty lies

Phase reflection needs lattice variance positivity; null-space retention needs quantitative corrected-field estimates. Neither follows merely from the abstract character decomposition. No clustering, unique vacuum or spectral gap is assumed.

6. Notation

\(t (x)\): last coordinate; \(e_a\): coordinate unit vector; \(v \in V_d\): degree-\(d\) label. Separated means distinct points. \(|w|\): word length; \(w \cdot w'\): concatenation; \(\tau_s\): time translation; \(\Theta\): antilinear OS reflection with reversal; \(\vartheta\): linear time reflection. Positive words have positive times; \(r\)-buffered means mutual sup-distance \(\geqslant r\) and times in \([r, d - r]\). \(N_d (S)\) comprises labels vanishing against every separated spectator family; \(F_d (S) = V_d / N_d (S)\). A table \(S'\) respects \(S\) when \(N_d (S) \subset N_d (S')\) for every \(d\).

7. What remains unexplained

Ultraviolet packing, bounded approximants and corrected-field estimates remain external inputs. With copy number and approximation index fixed, the variance argument takes cutoff limits before large translations, then sends copy number to infinity and approximation error to zero. Vacuum uniqueness for every phase belongs to V.3; selection here is noncanonical.

Detailed notes: the full account of this unit, with proof sketches for every step, is on the detailed notes page.

Code dependencies

Code dependencies (from the compiled Lean). These lists follow references between declarations allocated to units; they can include shared definitions, record fields and alternative constructions. They are not a mathematical reading order. Only unit pairs with at least 25 references are shown; see the counting rules.

References to: IV.1, I.1, V.2, III.10, II.1, III.16, VIII.1, VI.1.

Referenced by: VIII.1, V.3, V.2, VII.4, VI.1, III.10, III.6, VII.2.

Lean source

Source allocation: 5,713 attributed lines; 635 line-counted declarations; 784 assigned declaration names. Of the assigned names, 776 are in the compiled closure of the seventeen headline results. These quantities have different scopes.

The folders and result chains below are source inventories. Their line counts cover each whole group, including any share allocated to another unit; percentages use attributed lines plus one per assigned name. They need not sum to the unit total.

Additional source locations

Further pinned locations for the definitions and comparisons discussed in this unit.

Key statements

In the selected statement order. K- identifiers name this generated list; proposition and lemma numbers in the outline are separate. Each sentence is an automatically drafted summary of a Lean statement; follow its link for the exact hypotheses and conclusion. The outline status does not certify these summaries. Main result and key result describe importance, not correctness. Full names are in the statement table.

All units · Overview of the proof · Repository

← IV.1 Infinite volume: extraction of a subsequence and the Euclidean axioms of the limitPart V·Detailed notesV.2 Osterwalder-Schrader reconstruction of a Wightman theory →