Dignāga cut the Nyāya inference from five members to three. What did the example member become?
I have been translating some passages on Buddhist logic for a seminar in Pune and keep getting stuck on one question.
The Nyāya inference for others has five members: thesis, reason, example, application, conclusion ("this hill has fire, because it has smoke, wherever there is smoke there is fire as in a kitchen…"). Dignāga keeps three: thesis, reason, example. He also gives the three marks of a good reason (trairūpya): the reason must be present in the subject (the hill has smoke), present in at least some similar cases (things with fire), and absent from all dissimilar cases (things without fire). His hetucakra, the wheel of reasons, then runs through the possible combinations to show which reasons work.
What interests me is the example. In Dignāga it comes in two forms: a positive one (wherever smoke, there fire, as in a kitchen) and a negative one (wherever no fire, there no smoke, as in a lake). It is no longer just an instance both debaters accept. It has become the vehicle for stating the general relation, the pervasion (vyāpti).
So has the example stopped being evidence and become the statement of a rule with a decoration attached? Or is it still doing inductive work, reminding both parties where the rule was learned? I'm told Dharmakīrti went further and argued that even the thesis need not be stated, but I have not checked that passage myself, so treat it as hearsay.
I'd welcome readings from either side, and from anyone who knows how this looked once it reached China.