Proof — Verse 1.147

शब्दः संस्कारहीनो यो गौरिति प्रयुयुक्षिते ।
तमपभ्रंशमिच्छन्ति विशिष्टार्थनिवेशिनम् ॥ १४७ ॥

The word devoid of the refinement — when 'gauh' is what one intends to employ — that they call apabhramsa, though it settles on the distinguished meaning.

Endorsed axioms (each cite is verbatim commentary)

• {"kind": "predication", "name": "samskara-hina", "of": "apabhramsa"}“samskara-hina, lacking the refinement grammar names (1.131's sabda-samskara, here in its technical sense)”
• {"kind": "relation", "name": "nivesinam", "from": "apabhramsa", "to": "visistartha"}“visistartha-nivesinam — the fallen form *settles on the distinguished meaning perfectly well*; the hearer gets the cow”
• {"kind": "predication", "name": "intention-relative", "of": "apabhramsa"}“prayuyuksite ('when one intends to employ') marks the definition as intention-relative: the same phonetic form is apabhramsa only against a meant sadhu target”
• {"kind": "relation", "name": "asrita", "from": "apabhramsa", "to": "sadhu"}“the apabhramsa conveys, but conveys *through* the sadhu form it deviates from (as 1.152 will say), like a mispronounced name that reaches its bearer only because the right name is recoverable through it”

Denials

• {"kind": "predication", "name": "determined-by-success", "of": "apabhramsa"}“Corruption is thus defined against intention and norm, not against communicative success”

Rejected readings (refuted by compiled theorems)

• “Since corrupt forms successfully communicate meaning, there is no difference between them and correct forms.” — Ignores the explicit doctrine: corruption is defined by norm and intention, not by communicative success. The axioms establish that apabhramsa both lacks samskara AND settles on meaning as independent features. Success does not erase the norm-violation.
• “The form 'gavi' is simply a variant pronunciation of 'gauh', with neither more correct than the other.” — Erases the intention-relative definition captured in the axioms. The contested point makes explicit: the same phonetic form is apabhramsa only against a meant sadhu target. This reading treats gavi and gauh as symmetric variants, missing that apabhramsa-status depends on what the speaker intends to say.
• “Corrupted utterances like 'gavi' fail to reach their hearers or leave them confused about the meaning.” — Contradicts the nivesinam axiom: the fallen form settles on the distinguished meaning perfectly well; the hearer gets the meaning. Apabhramsa is communicatively successful by definition, even while being corrupt (norm-violating). These are independent properties in the Bhartṛhari framework.

Lean module

/- Verse 1.147 — generated from data/contracts/1.147.json. DO NOT EDIT.
   Axiom cites are verbatim commentary quotations, validated by
   tests/test_contracts.py. -/
import VakyaVallari.Adequacy

namespace VakyaVallari.Verses.V1_147

open VakyaVallari

def apabhramsa : Entity := ⟨"apabhramśa", Sorta.linguisticItem⟩
def samskara : Entity := ⟨"saṃskāra", Sorta.power⟩
def visistartha : Entity := ⟨"viśiṣṭārtha", Sorta.property⟩
def intention : Entity := ⟨"prayukti", Sorta.cognition⟩
def sadhu : Entity := ⟨"sadhu", Sorta.linguisticItem⟩

def contract : Contract :=
  { axioms := [ Claim.predication "samskara-hina" apabhramsa
    , Claim.relation "nivesinam" (Node.ent apabhramsa) (Node.ent visistartha)
    , Claim.predication "intention-relative" apabhramsa
    , Claim.relation "asrita" (Node.ent apabhramsa) (Node.ent sadhu) ]
  , denials := [ Claim.predication "determined-by-success" apabhramsa ]
  , reported := [] }

def accepted : Reading :=
  { claims := [ Claim.predication "samskara-hina" apabhramsa
    , Claim.relation "nivesinam" (Node.ent apabhramsa) (Node.ent visistartha)
    , Claim.predication "intention-relative" apabhramsa ] }

theorem accepted_adequate : contract.Adequate accepted := by decide

namespace Counterexamples

/- 'Since corrupt forms successfully communicate meaning, there is no difference between them and correct forms.'
   Why rejected: Ignores the explicit doctrine: corruption is defined by norm and intention, not by communicative success. The axioms establish that apabhramsa both lacks samskara AND settles on meaning as independent features. Success does not erase the norm-violation. -/
def ignore_norm_criterion : Reading :=
  { claims := [ Claim.identity apabhramsa sadhu ] }
theorem ignore_norm_criterion_inadequate : ¬ contract.Adequate ignore_norm_criterion := by decide
#guard contract.licenses ignore_norm_criterion = false

/- "The form 'gavi' is simply a variant pronunciation of 'gauh', with neither more correct than the other."
   Why rejected: Erases the intention-relative definition captured in the axioms. The contested point makes explicit: the same phonetic form is apabhramsa only against a meant sadhu target. This reading treats gavi and gauh as symmetric variants, missing that apabhramsa-status depends on what the speaker intends to say. -/
def gavi : Entity := ⟨"gavi", Sorta.linguisticItem⟩
def elide_intention_dependence : Reading :=
  { claims := [ Claim.identity gavi sadhu ] }
theorem elide_intention_dependence_inadequate : ¬ contract.Adequate elide_intention_dependence := by decide
#guard contract.licenses elide_intention_dependence = false

/- "Corrupted utterances like 'gavi' fail to reach their hearers or leave them confused about the meaning."
   Why rejected: Contradicts the nivesinam axiom: the fallen form settles on the distinguished meaning perfectly well; the hearer gets the meaning. Apabhramsa is communicatively successful by definition, even while being corrupt (norm-violating). These are independent properties in the Bhartṛhari framework. -/
def deny_communicative_power : Reading :=
  { claims := [ Claim.predication "asamveditartha" apabhramsa ] }
theorem deny_communicative_power_inadequate : ¬ contract.Adequate deny_communicative_power := by decide
#guard contract.licenses deny_communicative_power = false

end Counterexamples

end VakyaVallari.Verses.V1_147