The Navya-Nyāya logicians of medieval Bengal developed a technology that most mental-model builders still lack: a formal grammar for the difference between a concept and an instance of that concept. They called the distinction vyāpti — the invariant pervasion between a property and its locus — and they were obsessive about it because they knew that most bad reasoning isn't logical error. It's category drift: you form a model at one level of abstraction and then silently apply it at another, never noticing the slide. George Pólya, the Hungarian mathematician who spent his career studying how problem-solvers actually fail, found the same culprit in his 1945 'How to Solve It': the moment of false analogy, where a solver maps a new problem onto a familiar structure not because the structures match, but because the vocabulary sounds similar. The Navya-Nyāya logicians would have recognized Pólya's failure mode immediately. Their prescription was to force every inference to make its scope explicit — to name exactly which class of things your rule applies to before you apply it. For writers and thinkers who use frameworks to generate arguments: the discipline isn't choosing better models. It's annotating the ones you already use with their precise domain of application, so the model stops traveling silently into territory where it was never meant to go.

Pick one framework you've used more than three times this week. What is the smallest class of problems it was originally designed to handle — and how far outside that class have you already applied it?

Drawing from Navya-Nyāya (New Logic school of Indian philosophy) — Gaṅgeśa's successors in the Navya-Nyāya tradition, cross-referenced with George Pólya (Stanford, 'How to Solve It', 1945)