natura-biology — lexicon

RENDERED FROM STRUCTURED DATA — this page is not a markdown file. Source: lexicon/terms/natura-biology.json in uni-cookbook at 575fc93d9d31. The sha256 recorded for this page is the digest of that JSON, not of this text. The renderer is generators/derive_docs.cjs; it copies fields and adds no facts. Every value below is a field of the source, verbatim, including the ones that read NOT RUN and PENDING.

authored 2026-07-15 · reviewer: NONE - PENDING EXPERT REVIEW. An adversarial CITATION AUDIT (not an expert review) was run 2026-07-15 -- see adversarial_audit_2026_07_15. It checked the authorities; it did NOT check the Sanskrit or Latin competence of the renderings, and it is not a substitute for the expert review still owed.

Falsifiers for this file

• Any Sanskrit or Latin headword below is shown not to exist in the cited authority, or not to carry the sense in which this file quotes it. That entry is a pseudopropheta and must be struck to PENDING-CITATION.
• Any entry marked CITED is shown to rest on a sense the authority does not carry.
• Any COINED entry is shown to have an attested classical equivalent that was missed. The coinage must then be withdrawn in favour of the attestation.
• Any back_translation_test below is RUN and FAILS. Every test in this file is NOT-RUN; no register here has passed anything.
• Any entry is shown to render a TRUE-store concept with a term that asserts an experiencer, a soul, or a metaphysics the definition withholds.
• A domain value other than 'natura' appears in any entry of this file.
• Any quotation of an authority in this file is shown to be TRIMMED, re-attributed, or otherwise shaped so as to make a register look better-attested than the authority supports. This falsifier was ADDED 2026-07-15 after an adversarial audit found six such defects, all running in that one direction (see adversarial_audit_2026_07_15). The original falsifier list guarded only against INVENTED entries; it did not guard against motivated quotation of real ones, which is the defect this file actually had.

CELL —

Store: TRUE · Domain: natura · Confidence: Sanskrit: headword read, claimed sense REFUTED by the authority -- coinage stands or falls on expert review. Latin: form read and carried, biological sense read and ABSENT. Hindi: read negative in Platts, unsourced otherwise. Spanish: unsourced. Overall: LOW. The most valuable content in this entry is the gap between the modern word and what the 1899 and 1879 authorities actually say. · Reviewer: NONE - PENDING EXPERT REVIEW

The design unit of biology and the canonical Markov blanket, plasma-membrane bounded and dominated by viscosity and diffusion at Re of order 1e-5.

register	form	literal gloss	status
sanskrit	कोशिका (kosika)	little vessel; little container	COINED
latin	cellula	a small store-room; a little chamber	CITED (form only, NOT the biological sense)
english	cell	The design unit of biology and the canonical Markov blanket, plasma-membrane bounded and dominated by viscosity and diffusion at Re of order 1e-5.	—
hindi	कोशिका ()	cell (biological) -- the author's working knowledge, UNSOURCED	PENDING-CITATION
spanish	celula (célula)	cell (biological) -- the author's working knowledge, UNSOURCED	PENDING-CITATION

Ambiguity notes

• kośa/kośikā is a CONTAINER word. Its whole attested field is vessel, cask, box, sheath, scabbard, pod, nut-shell, cocoon. It therefore imports the image of a hollow thing whose interesting property is that it holds something. A cell is not primarily a container: the definition's load-bearing property is the BOUNDARY and its conditional independence, not the volume enclosed.
• THE HEAVIEST FREIGHT IN THIS FILE: MW s.v. कोश explicitly carries 'the three sheaths' as a Vedanta technical sense (authority: Vedāntas.). This is the pancakosa doctrine, in which the kosas are the sheaths ENVELOPING THE ATMAN. Rendering 'cell' as kośa/kośikā to a reader trained in Vedanta says that a cell is a sheath around a self. That asserts exactly the metaphysics the corpus withholds, and it does so invisibly, because the word looks like a neutral technical term.
• MW s.v. कोश also gives 'a dictionary, lexicon or vocabulary'. A word meaning 'lexicon' is a poor choice of headword INSIDE a lexicon file: kośa is the standard Sanskrit word for the genre this very file belongs to (cf. Amarakosa). The collision is real and will confuse readers of this document specifically.
• MW s.v. कोश carries anatomical senses that are near-misses and therefore dangerous. CORRECTED 2026-07-15 BY ADVERSARIAL RE-READ: an earlier version of this note attributed all five to 'L.' That blanket attribution was FALSE and has been struck. MW's actual authorities, re-read at getword.php?key=koSa: 'the membrane covering an egg (in the womb)' -- Suśr.; VarBṛ.; MārkP. (NOT L.); 'a testicle or the scrotum' -- Suśr.; VarBṛS. (NOT L.); 'the penis' -- W. (Wilson, NOT L.); 'the vulva' -- L.; 'an egg' -- L. Only two of the five are L. THE CORRECTION CUTS AGAINST THE AUTHOR AND SHARPENS THE WARNING: kośa's anatomical senses are NOT lexicographers-only curiosities, they are attested in Suśruta -- the classical medical corpus itself. A reader may therefore take kośa as already-anatomical, ON GOOD LITERARY EVIDENCE, and conclude the biological sense is attested. It is attested for THOSE organs, in medicine, and NOT for the cell. The false '(all L.)' had made this trap look smaller than it is.
• Latin cellula imports a SMALL ROOM: the monastic and domestic register. A room is a negative space defined by its walls; the concept's boundary is an active, selectively permeable, physically-constituted surface. The metaphor inverts which part is the object.
• The definition calls the cell 'the canonical Markov blanket'. Read this WITH the MEMBRANE entry: 'canonical' names the standard EXAMPLE, and does not mean I(mu;eta|b) was ever estimated for a cell. It was not. No register may be chosen so as to make the blanket claim sound settled.
• Re of order 1e-5 is a scope, not a decoration: it fixes the cell in the viscous regime. A rendering that ports the word to a macroscopic 'chamber' loses the only property that makes the physics of the referent what it is.

Forbidden mistranslations

register	form	why it is wrong
		Imports an animating soul and, in Jain and Vedantic usage, a transmigrating self. The concept is a membrane-bounded physical unit. This asserts an experiencer where the definition names none, and it is the single most attractive wrong answer because 'cell' and 'living thing' feel adjacent in English.
		Wrong rung of the scale ladder by roughly five orders of magnitude, and it imports Vaisesika atomism. A cell is not a minimal particle; it is a bounded system with interior dynamics.
		L&S s.v. cella gives, alongside the storeroom sense, 'The part of a temple in which the image of a god stood, the chapel' (II. A.), an apartment in a bathing-house (II. B.), the chamber of servants and slaves, and a room in a brothel (II. D., Petr.; Juv.). Every one of those registers is imported with the word. LAST SENTENCE STRUCK 2026-07-15 AS REFUTED BY THE CITED AUTHORITY: it read 'The diminutive cellula at least narrows to the store-room, which is why the diminutive and not the parent is proposed.' L&S s.v. cellula explicitly carries 'acc. to cella, II. D., Petr. 11, 1' -- the brothel sense passes straight into the diminutive. cellula escapes the temple and bath senses, not the brothel one. That is a smaller argument for the diminutive than this file made, and the smaller one is the true one.
		L&S was read and REFUTES this: it carries only 'a small store-room or apartment' (Ter.; Col.). Presenting the biological sense as classical fabricates an attestation that the authority does not contain.
		MW was read and REFUTES this: 'koSikA f. a drinking-vessel, L.' The headword is real; the sense is modern. This is the exact shape of the pseudopropheta the brief warns against -- a real headword carrying a sense its authority never gave it.

Back-translation test — status: NOT-RUN

MEMBRANE —

Store: TRUE · Domain: natura · Confidence: Latin: the best-attested register in this file -- headword read, COVERING sense carried, marked classical by L&S, Cicero cited; the concept's defining sense (lipid, selectively permeable) read for and ABSENT. 'Sense carried' was struck 2026-07-15 as ambiguous between those two and read as the second. Sanskrit: tvac read and carried in the covering sense, silent on every defining property. Hindi: unsourced, and the obvious candidate was read and refuted in the classical register. Spanish: unsourced. Overall: MEDIUM for Latin topology, LOW for the concept's actual discriminating properties, which NO register here carries. · Reviewer: NONE - PENDING EXPERT REVIEW

A selectively permeable lipid boundary, which is a surface for reasons of lipid physics and history, and whose satisfying I(mu;eta|b) = 0 is a DIFFERENT question whose answer is NOT-MEASURED.

register	form	literal gloss	status
sanskrit	त्वच् (tvac)	skin; hide; bark; a covering	CITED (as 'covering/skin' -- NOT as 'lipid bilayer')
latin	membrana	the skin or membrane covering parts of the body	CITED (as 'the skin or membrane that covers parts of the body' -- NOT as 'a selectively permeable LIPID boundary')
english	membrane	A selectively permeable lipid boundary, which is a surface for reasons of lipid physics and history, and whose satisfying I(mu;eta|b) = 0 is a DIFFERENT question whose answer is NOT-MEASURED.	—
hindi	झिल्ली ()	membrane -- the author's working knowledge, UNSOURCED	PENDING-CITATION
spanish	membrana	membrane -- the author's working knowledge, UNSOURCED	PENDING-CITATION

Ambiguity notes

• THE SCALE TRAP, and it is the sharpest one here BECAUSE the Latin is genuinely CITED. Classical membrana is a TISSUE-SCALE object you could see and peel: Cicero's 'natura oculos membranis tenuissimis vestivit' is about the coats of the eye. The concept is a MOLECULAR lipid bilayer roughly 5 nm thick, whose defining properties are lipid physics and selective permeability. A fully-sourced classical citation can still be the wrong referent, and here it is: L&S carries the word, and the word does not carry the chemistry. An entry can be CITED and still not mean what you need it to mean -- this file's own worked example of that. ('A perfect classical citation' was the original wording; 'perfect' is banned in author voice by the same charter rule 18.14 this file invokes in not_claimed, and the word was doing rhetorical work here that the evidence does not need.)
• membrana's transferred classical sense is 'parchment' (L&S, citing Quintilian and Horace). Parchment is a DEAD, PROCESSED, IMPERMEABLE sheet. That is the near-exact inverse of a selectively permeable living boundary, and it is a sense a classically-trained reader carries as live.
• THE FREIGHT THE DEFINITION ITSELF GUARDS: the concept's own definition says the membrane's satisfying I(mu;eta|b) = 0 is a DIFFERENT question whose answer is NOT-MEASURED. So no register may be chosen because it suggests boundary-hood in the blanket sense. The corpus's note on this concept is blunt -- 'the observation that organisms have surfaces is not evidence that the surface is a Markov blanket' -- and it names the rule: do not let a membrane and a blanket be rendered with one term. Whatever renders MARKOV_BLANKET, it must not be tvac and must not be membrana.
• tvac's ritual register is live and specific: MW carries 'a cow's hide used in Soma pressing' and 'a leather bag' (RV. v, 33, 7). The word arrives carrying sacrificial equipment. It also carries 'darkness (as black cover)' and 'a mystical name for the letter ya', either of which will read as deliberate symbolism to a Sanskritist encountering it in a technical document.
• tvac means bark as readily as skin. Bark is dead secondary tissue. If the rendering is read botanically, the referent becomes a dead structural layer -- again the inverse of the living, selectively permeable object.

Forbidden mistranslations

register	form	why it is wrong
		READ AND REFUTED. MW s.v. झिल्ली gives: 'JillI f. a cricket, BhP. v; the wick of a lamp, L.; = JallikA, q.v., L.; rice burnt by cooking in a saucepan, L.; cymbals, W.; parchment, W.' The FIRST and only literarily-attested sense (BhP.) is A CRICKET -- the insect. jhilli is the standard MODERN word for membrane and a translator will reach for it with total confidence; a reader of classical Sanskrit receives 'a cricket'. This is the most concrete pseudopropheta this file found. Note that MW does give 'parchment, W.' (Wilson) -- which converges on membrana's dead-sheet sense, not on the living bilayer, so even the closest MW sense fails the concept.
		Forbidden by the concept's own definition and by the corpus note attached to it. Sharing a term silently converts 'organisms have surfaces' (observed) into 'that surface is a Markov blanket' (NOT-MEASURED for every real biological system). This is the wing's single sharpest discipline and a shared word would defeat it in one stroke.
		Both name the macroscopic integument as material -- hide to be worked. They lose permeability entirely and import the tanner's register. membrana is the better Latin precisely because L&S ties it to membrum and to the coats of the eye.
		MW was read: it gives skin, hide, bark, covering, leather bag, Earth's surface, darkness. Nothing in the entry touches selectivity, permeability, or lipid. Citing tvac for the concept's defining properties would claim a sense the authority does not carry.

Back-translation test — status: NOT-RUN

MORPHOGENESIS —

Store: TRUE · Domain: natura · Confidence: Sanskrit: both members read and cited; the compound is proposed and is predicted by its own author to fail its own test on the agent question. Latin: nothing proposed, honestly. Hindi: nothing proposed. Spanish: unsourced. Overall: LOW, and the entry's value is the recorded conflict between janana's 'creator' sense and the definition's 'without a blueprint'. · Reviewer: NONE - PENDING EXPERT REVIEW

The generation of form from local rules during growth, in which pattern arises without a blueprint.

register	form	literal gloss	status
sanskrit	रूपजनन (rupajanana (rūpajanana))	the production of form -- CLAIMED, NOT SETTLED. The competing parse, which MW's own 'ifc.' marking points at, is the adjectival 'form-producing'. This file cannot adjudicate between them and does not.	COINED
latin	NOT-ATTEMPTED	—	NOT-ATTEMPTED
english	morphogenesis	The generation of form from local rules during growth, in which pattern arises without a blueprint.	—
hindi	—	—	NOT-ATTEMPTED
spanish	morfogenesis (morfogénesis)	morphogenesis -- the author's working knowledge, UNSOURCED	PENDING-CITATION

Ambiguity notes

• THE DEFINITION'S LOAD-BEARING CLAUSE IS 'WITHOUT A BLUEPRINT', AND THE PROPOSED SANSKRIT MEMBER ATTACKS IT DIRECTLY. MW s.v. जनन carries, as a masculine, 'a progenitor, creator' with a Rgveda citation (RV. ii, 40, 1). The whole point of morphogenesis is that form arises from LOCAL RULES with NO designer and NO plan: the corpus's own exemplar is phyllotaxis, where 'nothing golden was put in -- repulsion, advection and periodic deposition were put in, and the golden angle came out'. A member word that can read as 'creator' imports the exact agent the concept exists to eliminate. This is recorded, not smoothed: it may be sufficient reason to reject the coinage outright.
• rūpa carries heavy Buddhist doctrinal freight: MW gives '(with Buddhists) material form i.e. the organized body ... one of the 5 constituent elements or Skandhas'. To a Buddhist-trained reader, rūpa is the rūpa-skandha, the first aggregate, and arrives inside a whole analysis of person-constituents. The concept means geometric form arising in growth and asserts nothing about aggregates, persons, or the constituents of experience.
• rūpa also means 'beauty, loveliness, grace, splendour' (MW, RV. &c.). A term for form-that-is-beautiful will quietly re-import the aesthetic teleology this wing spends its effort removing -- the golden-ratio-hunting the corpus explicitly warns against under RATIO. The word carries the very bias the chapter guards.
• MW also gives rūpa as '(in alg.) the arithmetical unit' and as a term for the number 'one' (VarBrS.; Ganit.). In a document that is partly mathematical, a headword meaning 'the arithmetical unit' is a live collision.
• janana's neuter 'birth, coming into existence' is closer to the concept than the masculine 'progenitor' -- so the coinage's viability turns on the compound forcing the neuter reading. That is a real grammatical question and it is NOT settled here. It is exactly what an expert reviewer is for.
• ADDED 2026-07-15 BY ADVERSARIAL RE-READ, AND IT WORSENS THE COINAGE'S POSITION: MW marks the adjectival sense of janana 'IFC.' -- in fine compositi, i.e. THAT is the sense the word carries WHEN IT STANDS AT THE END OF A COMPOUND. In रूपजनन, janana stands at the end of a compound. So the authority's own marking predicts the reading 'form-producing' / 'form-generating' -- an ADJECTIVE, describing whatever produces form -- rather than the neuter action-noun 'the production of form' that this file's literal_gloss claims. The note above framed the neuter-vs-masculine question as open; the sharper and previously-unstated problem is that MW's ifc. marking points AWAY from the noun reading altogether, toward an adjective that presupposes a producer. That compounds the 'creator' freight already recorded, rather than relieving it. This does NOT settle the grammar -- a tatpuruṣa in -ana can be a legitimate action-noun and the author is not competent to adjudicate it -- but it means the coinage now faces TWO independent routes to importing an agent, both flagged by the authority itself, and neither yet answered.
• ALSO ADDED 2026-07-15: MW carries a further sense of janana omitted from this entry's original citation -- 'race, lineage, L.' A term for morphogenesis whose final member can read as 'lineage' will collide directly with CONVERGENT_EVOLUTION, whose whole evidential content is INDEPENDENT LINEAGES. Two entries in this same file would then share a member word across concepts the corpus requires be kept apart. Attestation is L. (lexicographers only), so the collision is weak -- but it was not recorded at all, and an unrecorded weak collision is still an unrecorded one.

Forbidden mistranslations

register	form	why it is wrong
		srsti is the cosmological creation-word and arrives with a creator and an act of creation. The concept is pattern WITHOUT a blueprint. Using a creation word for a blueprint-free process asserts the opposite of the definition while sounding elevated and correct.
		This is preformationism -- the discredited view that the form is already present and merely made visible. The definition says pattern ARISES; local rules generate what was not there. Cf. the CONVERGENT_EVOLUTION entry, where Latin evolutio has this same defect built into its root sense.
		MW carries no such headword. It is coined here. Presenting it as attested would fabricate an entry, which is the worst defect available in this repo.

Back-translation test — status: NOT-RUN

ALLOMETRY —

Store: TRUE · Domain: natura · Confidence: Sanskrit: nothing proposed. Latin: nothing proposed. Hindi: nothing proposed. Spanish: unsourced. Overall: this entry carries NO rendering in four of five registers, and that emptiness is its honest content. The brief predicted that Sanskrit technical vocabulary for modern quantitative science largely does not exist; this entry is that prediction coming true rather than being worked around. · Reviewer: NONE - PENDING EXPERT REVIEW

The scaling of a biological quantity with body mass, canonically BMR proportional to M^b, whose exponent moves with analysis choices and changes at the major evolutionary transitions.

register	form	literal gloss	status
sanskrit	—	—	NOT-ATTEMPTED
latin	—	—	NOT-ATTEMPTED
english	allometry	The scaling of a biological quantity with body mass, canonically BMR proportional to M^b, whose exponent moves with analysis choices and changes at the major evolutionary transitions.	—
hindi	—	—	NOT-ATTEMPTED
spanish	alometria (alometría)	allometry -- the author's working knowledge, UNSOURCED	PENDING-CITATION

Ambiguity notes

• THE CONCEPT IS NOT A CONSTANT AND NO REGISTER MAY MAKE IT SOUND LIKE ONE. The definition says the exponent MOVES -- with which taxa are excluded, with whether body temperature is normalised, with whether the regression unit is species or order. CONCEPTS.json records that this mobility IS the dispute and classes it OBSERVED-CONTESTED. Any rendering that reads as 'the law of body-size scaling' asserts a settled exponent and takes a side in a live dispute by wording alone. The safest renderings will sound tentative; that is correct, not weak.
• 'Allometry' in English is a term of art with a specific antonym (isometry) and a specific home (relative growth). A rendering that back-translates to the ordinary sense 'measurement of bodies' loses the entire content -- the point is not that bodies are measured but that a quantity scales as a POWER of mass with an exponent that is not 1.
• The definition ties the exponent's change to 'the major evolutionary transitions'. Read with the TRANSITION entry in CONCEPTS.json, which registers a three-way homonym: the POMDP B-tensor transition, the evolutionary MAJOR_TRANSITION (Maynard Smith & Szathmary), and the DD-TDD state transition. A register that renders 'transition' with one word across this corpus will merge three unrelated objects. That trap is inherited by this entry through its definition.
• CONCEPTS.json records the curvature finding: superlinear in prokaryotes, linear in protists, sublinear in metazoans -- 'a rung boundary showing up in the residuals'. So 'the' exponent is not one number even in principle across the ladder. A singular-definite rendering in any register is false to the concept.
• BMR is a scoped measurand (basal metabolic rate, under stated conditions), not 'energy use'. A register that glosses it as 'the body's energy' drops the scope, and per the honesty rail a number quoted outside its scope is a defect even when the number is right.

Forbidden mistranslations

register	form	why it is wrong
		Imports the artist's canon of proportion (Vitruvian, Polykleitan) -- a normative aesthetic ideal of how a body OUGHT to be proportioned. Allometry is a fitted, contested, regime-dependent empirical exponent describing how rates actually scale. The two are opposite in kind: one prescribes, one measures.
		Pins a contested exponent into the term itself. CONCEPTS.json classes the exponent OBSERVED-CONTESTED and records under MODELED that West-Brown-Enquist derive b = 3/4 from three assumptions, any one of which dropped means 3/4 does not follow. Naming the concept after one contested value would let the word decide the dispute.
		This is the specific failure this entry refuses. A confident compound built from half-remembered headwords is indistinguishable, to every downstream reader, from a compound built from read ones -- and that is what makes it the worst defect available here. NOT-ATTEMPTED is the honest status and it costs nothing but the appearance of completeness.

Back-translation test — status: NOT-RUN (no candidate)

CONVERGENT_EVOLUTION —

Store: TRUE · Domain: natura · Confidence: Sanskrit: nothing proposed. Latin: nothing proposed, and the obvious candidate was read and refuted. Hindi: nothing proposed. Spanish: unsourced, and etymologically inherits the same defect. Overall: LOW as a lexicon entry, and the entry's real value is the read refutation of evolutio -- a term a translator would have used without hesitation, which carries the discredited theory in its root sense. · Reviewer: NONE - PENDING EXPERT REVIEW

Independent lineages arriving at the same solution, which is EVIDENCE of a constraint-optimum and never proof of one.

register	form	literal gloss	status
sanskrit	—	—	NOT-ATTEMPTED
latin	—	—	NOT-ATTEMPTED
english	convergent evolution	Independent lineages arriving at the same solution, which is EVIDENCE of a constraint-optimum and never proof of one.	—
hindi	—	—	NOT-ATTEMPTED
spanish	evolucion convergente (evolución convergente)	convergent evolution -- the author's working knowledge, UNSOURCED	PENDING-CITATION

Ambiguity notes

• THE ETYMOLOGY IS NOT NEUTRAL AND IT IS NOT DEAD -- IT IS THE DISCREDITED THEORY. Lewis & Short, s.v. evolutio, was read and gives exactly one sense: 'an unrolling or opening of a book; hence, a reading' (Cic. Fin. 1.7.25). The image is a scroll being unrolled: EVERYTHING WAS ALREADY WRITTEN, and unrolling merely makes it visible. That is preformationism -- the view that the adult form is already present and only becomes manifest. Darwinian evolution is the REFUTATION of that view. The English word 'evolution' carries this defect silently because the metaphor is dead in English; in LATIN THE METAPHOR IS ALIVE AND IS THE PRIMARY SENSE. A Latin-register reader receives the wrong theory, stated confidently, from the word itself. This is the strongest single argument in this file for not reaching for the cognate.
• 'Convergent' in English is itself a metaphor from geometry: lines approaching a point. The concept is not about approach but about ARRIVAL AT THE SAME SOLUTION from independent starts. A register that renders convergence as 'moving toward one another' imports an interaction between lineages -- which is exactly what did NOT happen, and independence is the whole evidential content.
• THE DEFINITION'S FENCE IS THE HARD PART TO TRANSLATE AND THE EASIEST TO DROP: convergence is EVIDENCE of a constraint-optimum and NEVER proof of one. The concept and CONSTRAINT_OPTIMUM are deliberately separate entries in CONCEPTS.json, whose note reads: 'The evidence; CONSTRAINT_OPTIMUM is what it is evidence OF. Do not render as one term.' A register that fuses them turns a defeasible inference into a conclusion, in one word, invisibly.
• The counterweight travels with this concept EVERYWHERE (see ADAPTATIONIST_COUNTERWEIGHT). A rendering that reads as 'lineages discovering the best design' has already violated it: not every trait is an adaptation, and convergence is a hypothesis generator, never a proof of optimality.
• 'Solution' in the definition is engineering register, not biology's. Nothing was solving anything. A register that renders it with a word for answering a question or overcoming a difficulty imports a problem-solver and hence an agent -- the same freight the AGENT entry in CONCEPTS.json guards against.

Forbidden mistranslations

register	form	why it is wrong
		READ AND REFUTED. L&S s.v. evolutio carries ONE sense: 'an unrolling or opening of a book; hence, a reading', citing Cicero, de Finibus 1.7.25. It does not mean biological evolution and it never did. Worse than merely absent: the root image (a scroll whose text is already written, merely being unrolled) asserts preformationism, the theory Darwinian evolution overturned. The word does not fail to carry the meaning -- it carries the opposite one.
		parinama is Samkhya's technical term for the transformation of prakrti, and arrives inside satkaryavada -- the doctrine that the effect PRE-EXISTS in its cause. That is preformationism again, arriving by a second route and in a second language. Any register reaching for a transformation-word must be checked against this specific failure, because the two most natural candidates in two different languages share it.
		Forbidden by CONCEPTS.json's own note on both entries. It converts evidence into proof by vocabulary, which is the cheapest possible way to make an unearned claim and the hardest to spot after the fact.
		Asserts contact or fusion between the lineages. The lineages are INDEPENDENT -- that independence is the entire reason the resemblance is evidence of anything at all. A rendering implying contact destroys the inference while looking like a translation of 'converge'.

Back-translation test — status: NOT-RUN (no candidate)

CONSTRAINT_OPTIMUM —

Store: TRUE · Domain: natura · Confidence: No rendering is proposed in any of the four non-English registers. The entry's content is its ambiguity_notes and its forbidden_mistranslations, both of which are usable now by a translator who attempts what this pass did not. · Reviewer: NONE - PENDING EXPERT REVIEW

A solution repeatedly rediscovered by independent lineages under real physical constraints, which is what convergent evolution is evidence of.

register	form	literal gloss	status
sanskrit	—	—	NOT-ATTEMPTED
latin	—	—	NOT-ATTEMPTED
english	constraint-optimum	A solution repeatedly rediscovered by independent lineages under real physical constraints, which is what convergent evolution is evidence of.	—
hindi	—	—	NOT-ATTEMPTED
spanish	—	—	NOT-ATTEMPTED

Ambiguity notes

• 'OPTIMUM' IS BOUNDED AND THE BOUND IS THE CONCEPT. CONCEPTS.json's note is explicit: it is optimal WITHIN the constraints that obtained, not optimal simpliciter, and the inference from convergence to optimality is DEFEASIBLE. Every register's superlative vocabulary is unbounded by default -- 'the best' does not natively carry 'given these constraints, and revisably'. The rendering must carry the bound or it asserts something the corpus refuses.
• 'Rediscovered' does NOT imply a discoverer. Read with the AGENT entry in CONCEPTS.json: no experiencer, no selfhood, no intention. A register whose word for discovery implies a mind that searched has imported an agent into a statement about lineages. English hides this because 'evolution discovered X' is a dead idiom; in a register where the metaphor is live, it will read as a claim.
• 'Real physical constraints' is doing load-bearing work and is not decoration. The constraints are physics -- viscosity, diffusion, gravity, the square-cube relation -- not preferences, not costs, not trade-offs in the economic sense. A rendering via an economics or optimisation-theory register will import a cost function and an optimiser, neither of which the definition contains.
• This entry and CONVERGENT_EVOLUTION are deliberately separate and MUST NOT share a term: one is the evidence, the other is what the evidence is OF. CONCEPTS.json states this twice, once in each note. It is the same discipline as MEMBRANE and MARKOV_BLANKET not sharing a term, and it fails the same way: silently, by vocabulary, converting a defeasible inference into a conclusion.
• The counterweight applies here with full force: phylogenetic inertia, drift, developmental constraint and contingency mean a repeatedly-rediscovered solution may still not be optimal. The vertebrate retina is wired backwards and is nonetheless near-universal among vertebrates. Repetition is not optimality, and no register may let the word suggest it is.

Forbidden mistranslations

register	form	why it is wrong
		Drops the constraint-bound and the defeasibility in one move, and lands exactly on the adaptationist error Gould & Lewontin named. Nature is full of frozen accidents; a near-universal solution can be a constraint of history rather than an optimum of physics.
		These assert unqualified superiority in a register (social, ritual, moral) that has nothing to do with a maximum over a feasible set. optimus alone additionally carries 'best' as a moral and social judgement throughout classical usage -- Iuppiter Optimus Maximus is the god's cult title. The word arrives with a deity attached.
		Imports a designer. The whole apparatus of NATURE_AS_AUTHORITY plus its counterweight exists to permit nature to be used as a hypothesis generator WITHOUT asserting that anything was designed or intended.
		Explicitly forbidden by CONCEPTS.json's notes on both entries: 'The evidence; CONSTRAINT_OPTIMUM is what it is evidence OF. Do not render as one term.'

Back-translation test — status: NOT-RUN (no candidate)

NATURE_AS_AUTHORITY —

Store: META · Domain: natura · Confidence: Sanskrit: headword read and CITED in the 'nature' sense, and simultaneously carrying a metaphysics that may disqualify it -- the citation is solid and the choice is not. Latin: both members read and CITED; the phrase is composed and is at risk from auctoritas's frequent sense. Hindi: unsourced, and must not inherit the Sanskrit's citation. Spanish: unsourced. Overall: LOW-to-MEDIUM on citation, LOW on suitability. This entry is the file's clearest demonstration that a SOLID citation and a correct rendering are two different achievements. ('A perfect citation' was the original wording; struck 2026-07-15, 'perfect' being banned in author voice.) · Reviewer: NONE - PENDING EXPERT REVIEW

The doctrine, defensible in exactly one form, that nature is authoritative because it is the only system that has already run the experiment -- a very long parallel search under real physical constraints in which the failures were deleted.

register	form	literal gloss	status
sanskrit	प्रकृति (prakrti (prakṛti))	the original or natural form or condition of anything; primary substance	CITED (headword and the sense 'nature') -- BUT SEE THE FREIGHT, WHICH MAY DISQUALIFY IT
latin	natura ut auctoritas	nature as authority / nature as warrant	COINED (phrase) -- both members CITED
english	nature as the authority	The doctrine, defensible in exactly one form, that nature is authoritative because it is the only system that has already run the experiment -- a very long parallel search under real physical constraints in which the failures were deleted.	—
hindi	प्रकृति ()	nature -- the author's working knowledge, UNSOURCED for the Hindi register specifically	PENDING-CITATION
spanish	la naturaleza como autoridad	nature as the authority -- the author's working knowledge, UNSOURCED	PENDING-CITATION

Ambiguity notes

• THE HEAVIEST DOCTRINAL FREIGHT IN THIS FILE, AND MONIER-WILLIAMS NAMES IT HIMSELF. The entry states that in Samkhya, prakrti is 'the original producer of (or rather passive power of creating) the material world (consisting of 3 constituent essences or Gunas called sattva, rajas and tamas), Nature (distinguished from puruza, Spirit)'. Rendering 'nature' as prakrti therefore imports, in one word: (a) Samkhya's dualism of prakrti and purusa -- matter against spirit; (b) the three gunas as the constituents of matter; (c) a cosmology in which nature is a productive principle. The corpus's doctrine means nothing of the kind: it means the observed biosphere, whose authority is the empirical fact that it has already run a long parallel search under physical constraints. This is the exact failure the operator's own charter names -- Sanskrit does not automatically prevent ambiguity -- and it is worse here than mere ambiguity, because the word does not merely admit a metaphysics, it IS a metaphysics' technical term.
• The freight cuts a second way, and this one is subtle enough to miss: Samkhya's prakrti is the ACTIVE PRODUCER of the world, opposed to purusa the passive witness. The corpus's doctrine is that nature is an authority to LEARN FROM -- an evidential relation, not a productive one. prakrti says nature MAKES; the doctrine says nature SHOWS. Even setting the dualism aside, the term points at the wrong relation.
• MW gives prakrti in politics as 'a king's ministers' and 'the constituent elements or powers of the state'. In a corpus that also carries governance rules, an approval plane, and a sovereignty vocabulary, a headword meaning 'the powers of the state' is a live collision -- the same class of trap CONCEPTS.json registers for POLICY and for HIERARCHY.
• ADDED 2026-07-15 BY ADVERSARIAL RE-READ -- A FREIGHT THIS ENTRY MISSED IN ITS OWN LATIN WHILE CATALOGUING THE SANSKRIT'S: L&S s.v. natura carries, at II. D., 'the natural organs of generation' -- the genitals. This is precisely the near-miss class this file flags at kośa ('the vulva', 'a testicle or the scrotum'), and the author caught it in the Sanskrit and missed it in the Latin. The asymmetry is recorded rather than quietly repaired: a reviewer should assume the Latin registers in this file received LESS freight-scrutiny than the Sanskrit ones, because the author was hunting Sanskrit metaphysics and treating Latin as the safe register. It is not the safe register. Note also that prakṛti carries the SAME sense -- MW gives 'the male or female organ of generation, L.' -- so the collision is bilingual, and a reader of either register may take a doctrine about evidence as a statement about genitals. This is low-probability in context and non-zero, and it is exactly what a bare, unglossed back-translation test is built to surface.
• auctoritas is at risk from the OTHER direction and L&S's own frequency marking is the evidence. The sense wanted is 'a warrant, security for establishing a fact, assertion, credibility' -- nature as EVIDENCE. The sense L&S marks 'very frequent' is 'Might, power, authority, reputation, dignity, influence, weight' -- nature as a POWER whose pronouncements are to be deferred to. The doctrine is 'defensible in exactly one form' and that form is evidential; the frequent sense converts it into an argument from authority, which is a fallacy and is the precise misreading the counterweight exists to prevent. A word that is right in its rare sense and wrong in its frequent one is a bad word, and this is a real argument for rejecting the phrase.
• 'The failures were deleted' is doing exact work and is easy to soften. It means extinction: the search's negative results are not recorded anywhere, they are ABSENT. This is a survivorship-bias statement, and it is the reason nature-as-authority is only a hypothesis GENERATOR. A register that renders it as 'nature perfected its designs' inverts it: the deletion of failures is why we cannot read the search's results as optimal.
• This concept is META store and its rendering must never be admissible as a TRUE signal. Rendering it in a register whose word for 'authority' also means 'valid means of knowledge' -- Sanskrit pramana is the obvious such word, and it is Nyaya's technical term for exactly that -- would silently convert a doctrine about how to use evidence into a claim that nature IS a pramana, i.e. a valid means of knowledge in its own right. That is a metaphysical claim, in the META store, asserted by vocabulary. It is registered here as forbidden.

Forbidden mistranslations

register	form	why it is wrong
		pramana is Nyaya's technical term for a VALID MEANS OF KNOWLEDGE (perception, inference, comparison, testimony). Rendering nature-as-authority via pramana asserts that nature is a pramana -- a whole epistemological claim, made by word choice, in a META-store entry that is never True. It also imports the entire Nyaya-Mimamsa apparatus of which pramanas are valid, a live doctrinal dispute the corpus takes no side in. NOT read against MW in this pass and NOT proposed; recorded here as forbidden so a later pass does not reach for it.
		Refuted by the concept's own mandatory companion. Gould & Lewontin's counterweight travels with this doctrine EVERYWHERE: the vertebrate retina is wired backwards, the recurrent laryngeal nerve takes its detour. Nature is full of frozen accidents. 'Nature does it this way' is a hypothesis generator and never a proof.
		MW records that in mythology prakrti is 'a goddess, the personified will of the Supreme in the creation'. The doctrine concerns an evidential relation to the observed biosphere. Personification imports a deity, and the corpus's GAIA entry records that the literal claim that the Earth is an organism is INADMISSIBLE -- with the same heavy doctrinal freight noted there: 'the name is a deity'.
		A motto asserts the doctrine instead of naming it, and asserts it in the undefended form. The concept is 'defensible in exactly one form' and carries a mandatory counterweight; a motto carries neither the form nor the counterweight.

Back-translation test — status: NOT-RUN

ADAPTATIONIST_COUNTERWEIGHT —

Store: META · Domain: natura · Confidence: No rendering is proposed in any of the four non-English registers, and the primary source is NOT-CONFIRMED (not read in this pass). The entry's usable content is its ambiguity_notes and forbidden_mistranslations. Recording the unwalked provenance chain is part of the entry, not a gap in it: NOT-CONFIRMED is a fence on this corpus's own homework and is stated as such. · Reviewer: NONE - PENDING EXPERT REVIEW

Gould & Lewontin's mandatory companion to nature-as-authority: not every trait is an adaptation, because phylogenetic inertia, drift, developmental constraint, pleiotropy and historical contingency all produce features that are not optimal solutions to anything.

register	form	literal gloss	status
sanskrit	—	—	NOT-ATTEMPTED
latin	—	—	NOT-ATTEMPTED
english	the adaptationist counterweight	Gould & Lewontin's mandatory companion to nature-as-authority: not every trait is an adaptation, because phylogenetic inertia, drift, developmental constraint, pleiotropy and historical contingency all produce features that are not optimal solutions to anything.	—
hindi	—	—	NOT-ATTEMPTED
spanish	—	—	NOT-ATTEMPTED

Ambiguity notes

• THE PROVENANCE IS AT ONE REMOVE AND IS FENCED. This entry's content is CONCEPTS.json's characterisation of Gould & Lewontin's position. The primary -- standardly cited as Gould & Lewontin, 'The Spandrels of San Marco and the Panglossian Paradigm', Proc. R. Soc. Lond. B 205 (1979) -- WAS NOT READ IN THIS PASS. That attribution is therefore NOT-CONFIRMED: the chain is known and unwalked. Falsifier: read the primary and confirm or strike. This fence is on the CORPUS'S OWN HOMEWORK, not on the state of the literature, and the two must never be merged.
• 'Counterweight' is this corpus's word, not Gould & Lewontin's. It is an engineering metaphor: a mass on the other arm of a lever. The doctrinal content is that the counterweight is MANDATORY and travels with nature-as-authority EVERYWHERE. A register that renders it as 'objection' or 'criticism' makes it optional and separable -- something one might raise, or might not. The concept is that it is not detachable.
• 'Spandrel', the term the primary is famous for, is itself an architectural loan (the space between an arch and its frame) used to name a byproduct of a design constraint rather than an adaptation. Any register attempting it inherits an architectural metaphor from a second language on top of an evolutionary argument in a first. Do not attempt it lightly, and do not attempt it at all without reading the primary.
• The five mechanisms are NOT a rhetorical list and must not be compressed. Phylogenetic inertia, drift, developmental constraint, pleiotropy and historical contingency are five DIFFERENT causes, each producing non-adaptive features by a different route. A register that renders them as 'chance' or 'accident' collapses five mechanisms into one and destroys the argument's force -- drift is stochastic, but developmental constraint and pleiotropy are not chance at all.
• 'Not optimal solutions to anything' is stronger than 'suboptimal' and the difference is the point. A suboptimal solution is still a solution to some problem. The claim is that some features solve NOTHING -- they are byproducts, inheritances, or side-effects. A register that renders it as 'imperfect adaptation' has conceded the adaptationist premise while appearing to state the objection to it.
• The concept is META store: it disciplines commentary, and is never itself a TRUE signal. It cannot be falsified by a measurement of an organism, because it is not a claim about any organism -- it is a claim about what may be inferred from organisms.

Forbidden mistranslations

register	form	why it is wrong
		Imports a standard nature is failing to meet, and hence a designer with intentions who erred. The claim is that some features are not solutions to anything AT ALL -- not that they are failed solutions. 'Mistake' presupposes an attempt.
		Makes it detachable. CONCEPTS.json says it is the MANDATORY companion and that it travels with nature-as-authority EVERYWHERE. An objection is something one may answer and move past; a counterweight is structural and always present. The wrong word here would let the whole discipline be dropped politely.
		Only drift is stochastic. Phylogenetic inertia is history, developmental constraint is mechanism, pleiotropy is genetic architecture, contingency is path-dependence. Collapsing them concedes that non-adaptive features are mere noise -- which is exactly the adaptationist reply the argument is built to block.
		Overshoots in the other direction. CONCEPTS.json's gate is precise: a nature-derived design must beat a tuned baseline on a pre-registered metric or be recorded NEGATIVE. That gate is 'the entire difference between this wing and cargo-cult biomimicry' -- it disciplines the method, it does not abolish it.

Back-translation test — status: NOT-RUN (no candidate)