One Voice Heard Twice

Likeness between two traditions can be evidence of independent convergence only when the likeness could not have been borrowed. The translation-strain method rewards low strain as the strong signal, but low strain is exactly what historical transmission produces: when a word, a lineage, a translator, or a modern habit of universalizing carries language across traditions, the borrowed claim shows little strain because the vocabulary itself traveled. So low strain is ambiguous. It can mean a stable shared pattern, or it can mean one voice heard twice. Worse, the direction often inverts: genuinely independent recurrence tends to come with more strain, because each tradition reached a similar place by its own road and its own words, while borrowed agreement comes with less. The decisive close-reading case is Gauḍapāda's non-origination beside Madhyamaka non-origination: it is among the lowest-strain Hindu-Buddhist pairs, and it is among the most historically connected. Therefore strain cannot carry evidential weight until provenance is checked first; without a contact-and-lineage gate, the strain score reports its own opposite.

Existing translation-strain work treats strain as a load test where surviving overlap suggests a stable cross-tradition pattern. This narrows that claim by showing the load test is undefined without a provenance gate, because the very signal it rewards, low strain and surviving role-similarity, is also the signature of direct transmission. The sharper move is the inversion: independent convergence is more likely to be rough than smooth, so smoothness should lower confidence rather than raise it whenever historical contact is possible. The genetic, areal, and independent trichotomy is old in linguistics, which keeps novelty modest; the distinct contribution is applying it to invert the evidential reading of strain and to require provenance before any convergence score is allowed to strengthen a claim.

The single-axis inversion is too crude. Strain is multi-level, and borrowing can lower one level while raising another: Gauḍapāda borrows the logic of non-origination at low strain yet bends it hard to seat a permanent Self, producing high strain at the metaphysical level. So provenance interacts with strain level by level rather than uniformly lowering it. Historical contact also does not prove that a specific likeness was borrowed; two traditions can keep diverging or even independently re-converge after contact. And the genetic, areal, and independent distinction is well established in comparative method, so this may be application rather than discovery. The claim should be weakened if coding shows low-strain pairs distributed independently of historical contact, or if documented borrowing more often raises strain than lowers it.