PKL Rev1
DE · EN · Repository

The Mathematical Shape of the Architecture

Polycontextural Logic in Lean 4 and Mathlib — Edition PKL Rev1

Preliminary note: what distinguishes this edition from the previous one

This is not a contribution to definability theory. It is the account of a formalisation that labels its own claims, and of a reviewing practice meant to be transferable. Its subject is the reformulation of Gotthard Günther's polycontextural logic in a type-theoretic proof environment (Lean 4 with Mathlib).

Every claim carries its label where it stands — proved (machine-verified theorem), measured (machine-counted figure), computed (exact calculation outside the corpus), posited (explicit premise), source-verified (verbatim in Günther), interpretation (assignment of formal to conceptual), programme (named, not built).

The retraction that makes this edition necessary

The previous edition opened with a sentence that did not hold:

The first posit is named first: discontexturality — that between two logics no translation of truth exists — is not derived but stated as a premise. … in Günther too it is a posit.

Both halves are refuted by the text. Günther's most elementary case of discontexturality is the opposition of Being and Nothing — and in his account these two are isomorphic; he cites Baer (1931) for it. A translation between them does exist, and they are discontextural nonetheless. That disposes of the first half. And since Günther argues from the text rather than positing, the second half falls with it.

What takes its place is stronger, not weaker: in this edition discontexturality is no longer a premise but a theorem — proved in four stages, finally as a characterisation for every number of values from four upwards. The posit has not vanished; it has been displaced and named: what remains posited is the identification of a two-element value set with Günther's elementary contexture, not the separation itself.

What else has been added

previous edition (20 July)this edition (24 July)
discontexturalitypositedproved, limit of interpretation named
clone boundone witnessfour stages, finally for all m ≥ 4
second discontexturalitynot treatedown section (symmetric / asymmetric)
limit of applicationundeterminedconceptually grounded (M3)
open proof gapsfour declarednone — whitelist empty
axiom guards174 across 23 files299 across 35 files
constants checked2,5812,984

I · The ground: a combinatorics prior to logic

At the very bottom lies a structure that is not yet a logic: the kenogrammatic layer, a combinatorics of empty-place patterns. Finite sequences of places are classified solely by which places repeat which. No values, no negation, no truth; only distinguishability and repetition. Everything later arises as an assignment on these patterns: a logic is a choice of values on the ground.

Three granularities

Günther distinguishes three increasingly fine granularities. The proto-structure counts only how many distinct marks occur. The deutero-structure records the repetition profile — which block sizes occur with which multiplicity, irrespective of position. The trito-structure retains the position-exact pattern.

sequenceprotodeuterotrito
a b b a cone assignment3 marks2, 2, 10, 1, 1, 0, 2
x y y x zanother3 marks2, 2, 10, 1, 1, 0, 2

Only the third is fine enough to serve as the ground of a logic; the two coarser ones are its quotients. The reason is structural, not aesthetic: an assignment must be able to say which places carry the same value. Whoever knows only the block sizes does not know the assignment. The coarser granularities are therefore readable invariants of the ground, not possible grounds.

sequence 1 sequence 2 normal form abb ac xyy xz 011 02 proto3 marks deutero2, 2, 1 trito0, 1, 1, 0, 2 coarser → finer
Figure 1 — Two assignments with the same pattern; the normal form makes them identical. Below, the three granularities as increasingly fine classifications of the same sequence.

The normal form

Formally the trito-structure is represented by a normal form: a sequence of natural numbers beginning with 0 in which each new mark exceeds the previous maximum by exactly one — known in combinatorics as a restricted growth string. The predicate is decidable, and the canonisation of an arbitrary sequence is built as a procedure, not merely asserted (proved).

The built-in directional asymmetry

The maps between patterns are not symmetric. Extending a shorter sequence to a longer one is something other than retracting a longer one to a shorter.

The asymmetry can be counted. Take the pattern 0, 1, 1, 0 — four places, two marks. For the fifth place there are three possibilities: it repeats the first mark, it repeats the second, or it introduces a new one. In general a pattern whose largest mark is k has exactly k + 2 extensions. Retraction, by contrast, is unique: delete the last place and precisely one pattern remains.

This is no notational nicety but the structure itself: ascent chooses, descent computes. Whoever lengthens a sequence makes a decision that does not follow from what is given; whoever shortens one performs an operation whose result is fixed. Between the two directions lies not a change of sign but a difference in the kind of determinacy.

011 emanative unique pattern 01 10 011 00 011 01 011 02 evolutive branches
Figure 2 — The asymmetry in its sharpest form: retraction is unique, extension chooses. A pattern with largest mark k has k + 2 extensions — any existing mark or a new one — and exactly one retraction (computed).

Günther calls one direction evolutive, the other emanative (source-verified). This is the germ of what will later be called time — long before any value enters. And it is the same finding that recurs at the other end of the architecture: that ascent through the contextural stages is not determined stands there as a theorem (Section II.3) — here it stands as a count.

A caveat belongs here and stands in the corpus as well: the finer three-way division of relation types that the architecture employs is not Günther's terminology; he carries a two-way division. The refinement is named here and does not claim him as its authority (interpretation).

II · Contextures: two discontexturalities, one of them proved

A contexture is an assignment on the ground with the rank of a logic of its own: a logical locus with its own truth-value classifier — not a region within an encompassing logic, but a logic. The overall picture is therefore not a topos but a web of topoi over a shared pre-logical ground.

This section is entirely recast relative to the previous edition.

1 · What discontexturality is not

The previous edition held that between two contextures no translation of truth exists. This is false at Günther's own basic case. Being and Nothing are for him the most elementary discontexturality — and at the same time isomorphic; he appeals to Baer (1931). Between them there is a complete, structure-preserving map.

The isomorphism exists. It simply is not an intra-contextural process: it is stated from outside, not executed from within. That is Günther's own distinction, and it is the one with which he handles the tertium non datur — valid for both sides, applicable only to one.

From this follows the formalisation decision that carries this edition: discontexturality appears not as the non-existence of a map, but as non-generability in the term clone — as the impossibility of assembling a mediating operation from intra-contextural means. What is missing is not the map, but the path to it from within.

no assembly from intra-contextural means for all m ≥ 4: in the clone exactly min and max (proved) contexture K₁ its own logic contexture K₂ its own logic isomorphism exists stated from outside kenogrammatic ground no values, no truth
Figure 3 — What is missing is not the map but the path to it from within. The isomorphism exists (Günther's basic case); what cannot be assembled is the mediating operation.

2 · The first relation: symmetric, and proved

Günther names two fundamental relations. The first is the exchange relation between elementary contextures: an unordered pair, symmetric, without directional sense (source-verified, Definitions §4).

For it, this edition offers a theorem where a premise formerly stood:

Theorem (E3). For every m ≥ 4, a contexture-preserving, locally classical operation on m values lies in the clone of the intra-contextural connectives if and only if it is the minimum or the maximum. (proved, without the axiom of choice, without a fixed number of values)

This is a characterisation, not an existence claim: the set of non-assemblable operations is not merely non-empty but completely determined. Any policy that decides differently on different elementary contextures is unreachable from the local means — in whatever wiring and to whatever depth.

3 · The second relation: asymmetric, and open

The second relation holds between contexture and trans-contexturality. Günther calls it the one which, by virtue of its asymmetry, gives us the possibility of the new (source-verified). It is an ordered pair with directional sense, stepwise progression and loss of rank.

Its standing in this edition is a result, but a divided one:

And the hardness of the three features is unequal: one is inherited from the place count, one holds by definition, and only the loss of determination carries content of its own. To infer three equally strong findings from the occupied row would be to misread the theorem.

4 · Where the posit now sits

The posit has not disappeared; it has moved to a place where it can be examined. It no longer reads there is no translation, but:

A two-element set of neighbouring values is an elementary contexture in Günther's sense. (posited — the load-bearing choice of the entire bound series)

There is a conceptual reason for this choice, and it is not cut to fit after the fact: a two-element set {x, y} is closed under the local operations if and only if x and y are comparable. On incomparable pairs the minimum leaves the set — and then it is no longer a self-contained connection, hence no contexture in the sense of Definitions §1. The limit of the formalisation lies in the concept, not in the proof.

5 · Standing with respect to classical mathematics

Every proved theorem of this architecture is ordinary mathematics. What was built, inside a classical proof environment, is a family of theorems whose content is precisely the failure of a classical assembly; reading that failure as contexturality remains a posit.

Whether set theory decides between the readings is not an open question but a misposed one: it is silent — not because independence had been proved, but because the question is not stated in its language. This blocks the convenient exit of adding a decisive axiom: there is no formula to add one to. Whoever rejects the contextural reading thereby occupies a position outside the formal language as well; neither of the two is the neutral default.

III · The mode of proof: the differential

How does one prove what is inalienable in a structure, without the self-contradiction of a system certifying its own limit from within? Not positively within one system, but differentially between two. For each structural feature a poor class is named and a pair is exhibited: over all instances of the poor class the feature is impossible — as a theorem, universally quantified — and a rich witness carries it.

the hard halfthe easy half
X is impossible — universally quantifiedX is exhibited — a construction
all strict orders, the generated cloneone concrete structure, built internally

The impossibility is the hard half; the construction alone would be the easy one, and passing it off for the whole would be the differential form of immunisation.

poor class all strict orders, the generated clone X is impossible — universally the hard half ⊢ rich witness one concrete structure, built internally X is exhibited — a construction the easy half
Figure 4 — The differential and its two unequally heavy halves. It is delivered only with the negative one.

Three marks discipline the mode. First: the poor model is built internally, not asserted in the metalanguage. Second: the result is conditional relative to the posited premises; it symmetrises the sceptic, it does not compel him. Third: what is proved is the form, never the naming — that a contexture-relative choice of operation is unreachable from local means is a theorem; that this choice is called mediation remains interpretation. The file in the corpus is therefore named NonUniformCloneBound and not Mediation.

And the provenance of the poor classes is itself a rubric carried along — for here, not with the theorems, sits the burden of proof of the construction. Canonical is the procedure: the clone concept and generation are standard clone theory. Chosen is the basis: these operations, this embedding. The objection that these choices are cut to fit is admissible and is not answered but declared.

IV · The witnesses: the series of bounds

The argument of the architecture is a narrow series of impossibility theorems; the rest of the building is substrate, location and connection. Relative to the previous edition, one witness has become a complete series.

The series at a glance

theoremkind of non-internalityreach
Drejection does not lie in the clonetranscendence — it leaves the contexturem = 3
E1complete classificationnon-uniformity — it stays and chooses differentlym = 3
E2characterisationexactly minimum and maximumm = 4
E3characterisationexactly minimum and maximumall m ≥ 4
E4stage aggregationface of applicationm = 4
D m = 3 transcendence leaves the contexture E1 m = 3 non-uniformity stays and chooses differently E2 m = 4 characterisation exactly min, max E3 all m ≥ 4 characterisation no fixed value count robustness under extension by constants: D falls · E2, E3, E4 hold
Figure 5 — The series of bounds: from two reasons for non-internality through classification to a characterisation without a fixed number of values.

1 · D — locating the first bound

The rejection operation refuses an offered pair of values as a whole and does not lie in the clone generated by the basis operations over three values (proved; the proof is arity-blind and runs via contexture preservation).

Lying outside a clone is cheap in itself: in the binary fragment the clone comprises 82 of 19,683 operations, less than half a percent (computed, outside the corpus). The claim does not rest there. The basis preserves three invariants, and every term over it preserves all three — not merely the generators (proved, by term induction). Rejection violates exactly one of them and preserves the other two. It therefore lies not somewhere outside, but at a determinate place.

all binary operations: 19,683 generated clone 82 rejection three invariants are preserved by every term (proved): two-valued embedding · fixed point · tolerance relation rejection violates exactly one and preserves the other two
Figure 6 — The location: not somewhere outside, but beyond exactly one of three preserved invariants. Clone counts computed, outside the corpus.

2 · E1 — the second reason, and the complete division

Here lies the break with the previous edition. An operation can fail to be assemblable for two distinct reasons:

The second reason had no witness before. It has one now, and completely so: of the eight choice patterns over three values exactly four are assemblable and exactly four are not (proved). The division is closed under negation conjugation — a transport lemma that replaces three copies of a proof by one theorem.

3 · E2 and E3 — from existence to characterisation

At four values the picture tips: only two of the 64 locally classical operations remain assemblable, namely minimum and maximum themselves (proved). At five values it is 2 of 1,024, at six 2 of 32,768 (computed).

E3:   m ≥ 4  ⟹  (f in the clone  ⟺  f = min or f = max)

E3 is the first theorem of the series of bounds that hangs on no fixed number of values — D and E1 stand at three values, E2 at four. Its proof uses no decide and no case distinction on m; it runs over six families of edges with closed-form breaking points and an induction on path length.

The threshold lies at four, and this is not an artefact. At three values the statement is false — there four of eight patterns are assemblable. At two values there is no witness at all, because there is only one contexture. This meets Günther's own threshold: the classical two-valued system does not yet represent a compound contexture … the second value yields nothing new and stands unmediated over against the first (source-verified). Mediation begins at three — here as a counted zero, not as a thesis.

4 · Robustness: how much each bound claims

A non-generability bound hangs on its invariant. If the basis is extended by constant operations, the bound survives precisely when the invariant is reflexive.

witnessinvariantsurvives constants?
Done elementary contextureno
E2, E3, E4neighbourhood with broken edgesyes

This is no side issue: a bound that falls to an added constant claims less than it appears to. The distinction is written into the corpus as a duty of inspection before every new witness.

5 · E4 — the face of application, and its limit

Four authorisation levels, an aggregation policy conservative everywhere except between the two highest: it is not assemblable from local, contexture-blind checkers — not even when arbitrary constant checkers are added (proved).

The theorem is an instance of E3. Its yield class is therefore naming, not yield, and the file header says so. What is new is the reading, not a theorem.

The limit belongs in the same breath. The bound bites on linearly graded carriers. On a lattice of equal-ranking roles it falls: already on the smallest flat lattice M3 there exists a locally classical operation that is neither minimum nor maximum (computed). A transition graph of named roles without a grading is not carried by this bound; what remains provable there is reachability, and that achieves nothing a type system does not achieve too.

V · The eightfold series

Günther gives the ontological layer an eightfold inner structure of the form 1 + 3 + 3 + 1: a reflectionless ground, three reflections of nature, three subjective reflections, an exhaustion transition (source-verified). In the architecture all eight positions have their term-verified correlate; the assignment of position to correlate remains interpretation.

The edges carry as differentials. The eighth position: exhaustion as a fourfold predicate, the transition as a well-foundedness theorem (proved). The first: the ground as a fixed point recurring under all iteration (proved, axiom-free). And the arc closing both: the self-exhausting gradation provably returns into its ground — the series does not end beside its beginning (proved).

1ground 2nature 3nature 4nature 5spirit 6spirit 7spirit 8transition differential locations on two carriers differential exhaustion provably returns into the ground
Figure 7 — The eightfold series 1 + 3 + 3 + 1: the edges as differentials, the middle as locations; the arc closes the series into its ground.

The six middle positions are locations on two carriers. The three classes of nature — exchange, return-freedom, mediation — are pairwise separated and each purely witnessed (proved). The positions of spirit run as liftings: the same process, repeated on its own contents, inherits exchange and breaks irreversibility.

The difference between the two carriers is the mathematical core of this section. On the first the three properties are found and lie side by side; on the second they arise and are inherited or broken by provable rules. What on the first level is the coexistence of three independent features is on the second a relation of derivation.

The backbone of the location is the triangular numbers as interval openings — the same sequence Günther carries for compound contextures and for the hierarchy of ontologies.

A caveat that lies in the matter itself. The Definitions text gives for compound contextures the formula n(n+1)/2 and the sequence 1, 3, 6, 10, …, and states at the same time that a compound contexture requires at least three values. The two do not agree: the first compound contexture is the three-valued one; the 1 belongs to the ontology sequence. Three formulas are in circulation — n(n+1)/2, m(m−1)/2, m(m+1)/2 — and generate the same sequence at shifted index. The architecture counts the elementary contextures of an m-valued compound contexture as m(m−1)/2; this is compatible with Günther's series, but it is a stipulation, not a derivation.

VI · Movement and time

Over the web lies a modal triad as a stock of movement — three ways in which something can change, without its yet being said what changes:

Their compositions are smooth in exactly one direction; there is no canonical interchange of directions — this is not imagery but theorem (proved, five theorems, all five with the axiom of choice, and across the whole series confined to two files).

The cycle that fits into no order

Before the architecture speaks of time, it is worth looking at the place where Günther first came upon it. In 1937 he sets three propositions about potency side by side — each of them immediately plausible on its own:

  1. Time is more powerful than will. Every decision falls within time; no one can will that it had fallen earlier.
  2. Thought is more powerful than time. Memory runs counter to time and reaches back to the beginning — it is not bound to the direction of the flow.
  3. Will is more powerful than thought. With Fichte: that I think rests on the fact that I will to think.

Take the first two together and it follows that thought is more powerful than will. The third says the opposite. The three principles cannot be brought into a ranking — they close into a circle. Günther draws from this no contradiction but a finding: from them never three groups of statements of transitive character can be formed, and therefore the course of time in concrete consciousness remains always ambiguous (source-verified, Wahrheit, Wirklichkeit und Zeit, Paris 1937).

What the architecture makes of this is a differential and not a thesis. What is proved is the poor side: in every transitive, irreflexive relation — on every carrier, without exception — a three-cycle is impossible, and there is likewise no relation-preserving image of such a cycle into any strict order (proved; the core requires none of the classical axioms). Whatever is intransitive has no place in the poor class.

thoughttimewill memory reaches back to the beginning decisions fall within time “that I think” rests on “that I will to think” ≻ means more powerful than Chain the first two and thought ≻ will follows — the third says the opposite. The cycle closes.
Figure 8 — Günther's three principles of potency of 1937 and their intransitive cycle. That no such cycle fits into a strict order is proved; that this cycle is called thought, will and time remains source-supported interpretation.

It is not claimed that thought, will and time behave this way. What is claimed is: if they do, then no linear order is their measure — and not because none has been found so far, but because none can exist. The assignment of the three names to the three corners remains interpretation; the impossibility is theorem.

The two differentials of time

Time falls into two differentials, separated by a single hard invariant: the contexture count.

τδω displacement · reorganisation · irreversible positing intra — one contexture the dominance cycle lies in no strict order inter — two contextures the cumulative lifting does not close separated by the contexture count
Figure 9 — The modal triad and the two differentials of time; the separation is the contexture count.

VII · The safeguards: guards, profiles, measurements

An architecture that asserts impossibilities must be able to say what its proofs rest on — and must prevent that from changing silently.

The standing

figureprevious editionthis edition
axiom guards174 across 23 files299 across 35 files
of these, axiom-free profiles4147
statement pins7 (pilot)36 across 6 files
constants checked (gate)2,5812,984
open proof gaps4 declared0 — whitelist empty
theorems corpus-wide—698
build jobs—1,267
axiom guards of these axiom-free statement pins constants checked (hundreds) 174 → 299 41 → 47 7 → 36 25.8 → 29.8 open proof gaps: 4 declared → 0 the gate checks a statement without a footnote
Figure 10 — The safeguards in figures: previous edition → this edition (as of commit e89ab47; counting routes in the appendix).

The four mechanisms

First, the axiom guards. For every theorem of the load-bearing series its axiom profile is measured and frozen as an expectation; if the proof basis shifts silently, the build breaks. The guard secures the basis, not the wording — a weakened theorem with the same profile passes it.

Second, the statement pins. Precisely that finding forced a second type of guard: pins nail down the full wording and break the build on definitional drift. What was a pilot of seven has become the rule.

Third, the global gate. In this edition it reports a state it never reported before:

AXIOM GATE GREEN — aggregate Reformulation (2984 constants checked)
draws no sorryAx. Whitelist empty.

The four formerly declared gaps have been dissolved. The finding was not four missing proofs but data appearing as theorems: two were not derivable from the given data and became structure fields; two were simply false as stated — a uniqueness claim against empty axioms, and an existence claim over a possibly empty hom-set — and were struck, with a note in place.

What this is and what it is not. Four declarations have disappeared, not four proofs appeared. The Beck–Chevalley condition is as unproved as before; it is now visible as a datum of the structure instead of being presupposed as a consequence. The gain is that the gate checks a statement without a footnote.

Fourth, the consumption measurement. For every asserted dependency between layers it is measured at the proof term whether a theorem of the predecessor enters or whether a concept carries the statement. Consumed in this architecture normally means: the concept carries the statement — stated here so that no reader has to measure it first.

Calibrating the measurement routes

Every measurement route is calibrated before use against cases with a known target — one that must match and one that must not. The reason is experience, not caution: the most dangerous failure mode of a counted architecture is the plausible, clean, too-small number, because it looks like a good result.

This edition has its own instance. Two counting routes for the same figure gave different values — each correct for its route. The cause was that the category hung on a heading in the running text, written in English in one file and in German in the others. Since then every site carries a machine-readable marker, and the rule reads: a counting route that takes prose as its criterion is a time bomb.

VIII · The instruments as a subject of their own

Alongside the theorems something has arisen that knows nothing of Günther and could travel further: a reviewing practice for the formalisation of philosophical texts. Whoever formalises a thinker has a problem foreign to pure mathematics — the interpretation outruns the term. A theorem is built, its name is chosen, and the name claims more than the theorem. Against this gradient no resolutions help, only mechanisms.

Seven rules, each from a misstep

  1. Look for the theorem first, then build only as much carrier as the theorem demands. A base layer that no theorem requests is a renaming of the library.
  2. A trial build before every specification — in the target environment. A trial in a simplified environment measures the wrong system and produces warnings that create work instead of saving it.
  3. The specification names goal, constraints and pitfalls — not the wiring. Where it prescribes the proof path, it obstructs better ones.
  4. Specifications are reviewed too, not only builds. Review before the build found a gap in the requirements, sharpened a demarcation and settled a question passed on as open.
  5. Counting routes must not hang on prose.
  6. Aggregate means free of posits. A file enters the certified body if it carries no posit — not if the effort is small. It is a question of the epistemic signal, not of the work. Addendum (26 July 2026): this rule no longer holds in this form. A measurement of the aggregate found 32 structure fields of type True — the rule was already not covered by the corpus when it was written. It now reads: no theorem of the aggregate depends on a posit. See CLAUDE.md §10, as of 284995b.
  7. The title outruns the theorem. Where building precedes clarification, the description claims more than the build. Three times in succession a claim in a title had to be healed — each time pure documentation, never a line of proof.

The mechanisms

Frozen axiom profiles make the proof basis a checkable quantity: on one occasion a failing expectation guard exposed the hidden classical content of a projected proof route before it was built — the guard was an instrument of discovery, not bookkeeping.

New in this edition is a corpus linter that checks rule 7 mechanically. It searches for claims of rank — first, only, the first to — across Markdown and Lean sources. Its first run confirmed three of four manual healings and named one remaining site, whose assessment it expressly left open. Hence the distinction it cannot make itself: a claim of rank is admissible if its reference set is named in the same paragraph and finite; it needs healing if that set is the corpus, the field or the literature.

The common denominator is a reversal of the burden of proof in one's own house: it is not for the critic to show that a formulation overreaches, but for the text to show that it does not — and mechanically, wherever possible.

The evidence that the review layers bite is this phase's list of errors. A retracted tenet; a false justifying sentence in two probes; an inference from ¬∀ to ∀¬; an outdated calibration line in the very commit that repaired a counting route; an effort estimate off by a factor of 35; three overreaching titles. None of these errors was found by the instance that made it.

IX · Reach and limit

What the architecture claims is conditional and is carried as such. It does not promise to compel the sceptic to acknowledge an irreducibility; it promises to symmetrise him — from the comfortable position this reduces to the known to the explicit position I posit the one encompassing logic, which is itself a posit and as such no longer has precedence.

The three limits

First, mathematical: M3. The characterisation lives on linearly graded carriers and falls on lattices of equal-ranking elements. This is not a proof gap but a counterexample — and the breaking point is exactly the lattice shape an architecture of equal-ranking roles would have. A generalisation leads precisely where it fails.

Second, hermeneutic: the identification. That the two-element value set is Günther's elementary contexture is settled by no theorem. This posit stands on grounded soil — closure under the local operations is a substantive reason, not a retrofit — but it remains a choice.

Third, in principle: the spontaneity of enactment. Not the will as a function, but solely the that of enactment is no structural feature; there is no poorer model in which it is missing as a position, because it is not a position. Here the procedure declares its reach, as a statement about itself.

What is not built

A conclusion that concealed this would be none. The proemial relation itself — the core of Günther's theory — is not formalised. The compound contexture is present in the corpus as a number sequence but not built as a concept. And the asymmetric side has been approached twice, each of the two lines leaving one column open: one binds all features to a single carrier and pays with a posit, the other trades the posit for the established concept and loses substructure preservation.

The picture

That is the shape: a ground that counts repetition and whose normalisation is built as a procedure; logics whose separation is no longer posited but proved — in a form that withstands Günther's basic case; a mode of proof that knows and carries its hard half; a series of witnesses that has grown from an existence claim to a characterisation without a fixed number of values; an eightfold series that provably returns into its ground; a gate with no sorryAx left to administer; and three limits, none of which is a defeat but each an indication of place.

What is posited stands at the beginning; what is proved is counted; what is interpretation is called so.

Appendix · Counting routes

So that the figures are recomputable and not merely credible, each carries a route. The statements hold as of commit e89ab47; any build move renders every exhibited number silently out of date, which is why they carry their standing here.

quantityroute
axiom guards (299)grep -rc '#guard_msgs in #print axioms'
files with guards (35)grep -rl of the same string
axiom-free profiles (47)grep -rc 'does not depend on any axioms'
theorems corpus-wide (698)grep -rhcE '^(theorem|lemma) '
statement pins (36 in 6)grep -rc '^-- STATEMENT-PIN'
constants checked (2,984)build output of the gate
build jobs (1,267)build output

Two rules on counting. First: counting is done with grep over the shell, not with a library function that skips blank lines — one measurement error of this phase traced back to that. Second: aggregate and corpus figures are kept apart. The aggregate is the certified import tree; the corpus additionally comprises the probes that carry a marked posit and therefore stand outside. Conflating them was the cause of a false calibration line.

Sources of the passages cited: Günther, G. (1970), Die historische Kategorie des Neuen, Hegel-Jahrbuch 34–61; Günther, G. (1937), Wahrheit, Wirklichkeit und Zeit, Paris; Günther, G. (1968), Many-Valued Designations and a Hierarchy of First Order Ontologies. The conceptual determinations follow the project's Definitions register.