Footsteps, Nice, 2010
The first post about my infatuation with Vasubandhu, the founder of the Yogācāra school of Buddhism, is here. It gives a brief overview of who he was and what he was about. Please remember that I'm totally new to this stuff and probably misunderstanding much of it, so don't take any of it too seriously; my primary intent in writing this is to clarify my own thoughts about this stuff. Since I don't read Sanskrit, I'm reading this translation and commentary by Stefan Anacker.
Vasubandhu's Vāda-vidhi, translated as "A Method for Argumentation" is the first attempt in Indian philosophy to describe a formal system of logic. Since no complete manuscript has survived, the text I'm reading is based on reconstructions from quotations by Vasubandhu's students, in particular the great logician Dignāga, as well as translations to Chinese and Tibetan.
I was surprised and thrilled to discover that Vāda-vidhi isn't really about logic. It does describe a simple formal structure for argumentation, but that structure is grounded in the author's ideas about cognition. In particular, Vasubandhu's methods for distinguishing fallacious arguments from valid ones rely heavily on his theory of cognition. These methods are simpler and arguably less rigorous than the ones described in the formal logic derived from Aristotle, but they're also highly practical and, I think, are pretty good at avoiding discussions about the number of angels that can dance on the head of a pin. Where Aristotle takes the P and Q in his "If P, then Q" type syllogisms as given, Vasubandhu devotes a considerable amount of this short text to exploring the relationship of the objects in an argument to our ways of cognizing them.
Vasubandu's argument structure is very simple:
Thesis: "This mountain is fire-possessing."He then goes on to describe a number of logical fallacies, which he classifies into three types: reversed, incorrect or unreal, and contradictory. He then moves on from the trivial example above to thornier ones involving properties like arising-instantaneously-upon-an-effort and non-eternality, and shows the applicability of his system for these more complex cases.
Justification: "Because this mountain is smoke-possessing."
Invariable concomitance: "If a thing is smoke-possessing, it is always fire-possessing."
Example: "Like a kitchen."
Counter-example: "And not like a lake."
Overall, his formal system of argumentation is simple and practical, and especially well-suited for the quick back-and-forth of the verbal debates that were very much in vogue in Vasubandhu's day. He had a reputation for being a ferocious debater, and I have no doubt that it was well-deserved—nor that Vāda-vidhi is the fruit of a lifetime's experience of matching wits with the sharpest minds in a subcontinent full of them. However, I found his ideas on cognition a good deal more interesting.
Underlying Vasubandhu's treatise on logic is an unstated premise that the objects in the argument structure—the mountain, the kitchen, the lake, their smoke-possessingness, fire-possessingness, or lack thereof, have no independent existence. Instead, they only come into existence provisionally, when cognized. He further breaks down our process of cognition into direct perception, such as perceptions of pleasure, pain, sound, or sight, and inferred perception, such as the perception of a mountain as fire-possessing when it is observed to be smoke-possessing.
Knowledge through inference can be specified as an observation coming when the means-of-evidence is directly observed, and the invariable concomitance between it and what can be inferred is remembered. One does not occur unless something else is directly known. Otherwise there is no inference.
We can never be absolutely certain about anything, because we can only make inferences based upon our perceptions, which can be misleading, and memory, which is unreliable. He goes on to give examples of problems with cognition, such as a false cognition-of-silver arising from looking at mother-of-pearl, and cognition of objects that do not exist, such as a luminous circle that is perceived when a torch is hurled in an arc.
This makes the example and counter-example so vital to the argument. Any thesis can be disproved by showing that the proposed invariable concomitance is not, in fact, invariable. The last part of Vāda-vidhi is devoted to methods that can be used to distinguish logical fallacies from valid arguments. In my opinion, Vasubandhu does not quite manage to resolve these problems; there appears to be a great deal of room left for pontification about the validity of invariable concomitances, examples and counter-examples; our more familiar categorization of fallacies would come in very handy there.
The more I read Vasubandhu, the more I'm impressed by him. His insights into what makes us tick are deep indeed, and he describes and analyzes them with a rigorous and truly brilliant intellect. He's a very refreshing reminder that the "going beyond words" Buddhist teachers often mention is anything but an abdication of intelligence or a descent into a mushy fluff of anything-goes. He's also way more accessible to a 21st century dilettante than I feared. I can't wait to get to the next part, titled Pañcaskhandhaka-prakaraņa, "A Discussion of the Five Aggregates."
Incidentally, the photo at the head of this posting is a nice example of Vasubandhu's false perception. When I saw those footsteps, I assumed that someone who had been swimming or wading in the sea was walking barefoot on the Promenade, since I thought there was an invariable concomitance between the marks in the photo and wet bare feet. When I caught up with the individual in question, it turned out that I was slightly off the mark, and a more careful application of Vasubandhu's method would have led me to the correct conclusion. I leave the correct chain of inference and the error in my initial conclusion as an exercise for the reader.