Mozi's Practical Approach to Knowledge — Epoche B2
Sources, Not Reasons: What the Mohists Actually Contribute to the Gettier Problem The question Classical Chinese philosophy is often brought into epistemology as a corrective: where Western theory obsesses over justification, the Mohists are said to have measured knowledge by its practical benefit, and so to have sidestepped the puzzles that justification generates. This essay agrees that the Mohist material is relevant to post-Gettier epistemology and disagrees about where the relevance lies [1] . The comparison usually offered rests on Mozi's three standards for testing a doctrine [2] , and those standards do not touch Gettier cases: they are a test for accepting or rejecting a thesis about how to live and govern, not an analysis of what makes a particular person's particular belief knowledge. Worse, a genuinely pragmatic reading of Mohist knowing walks straight into the problem Gettier exposed, since a belief can be useful by accident just as it can be true by accident. The contribution is elsewhere, in the later Mohist Canons, and it is structural rather than pragmatic: the Canons classify knowing by its source , which individuates a case of knowledge by how it was acquired rather than by the reasons its holder can produce. That is the move Western epistemology made after 1963, and the comparison is worth making because it shows the move being made for different reasons. The claim is deniable — a reader who holds that the three standards constitute a pragmatist theory of individual knowledge will reject the first half of it. Justified true belief, and Gettier's cases The analysis under attack in 1963 says: someone knows that p just in case p is true, they believe that p, and their belief is justified. Justification is the requirement that the belief rest on adequate grounds [3] , and its point is to exclude lucky guesses. The analysis is often traced to Plato's Theaetetus [4] , which is misleading and worth correcting. The dialogue examines the proposal that knowledge is true judgement together with an account, raises three readings of "account", finds all three wanting, and ends without a definition. The Meno is closer to the source of the idea: true opinions are valuable but apt to run away, and what ties them down is working out the reason. Edmund Gettier's 'Is Justified True Belief Knowledge?' (1963) shows that the three conditions are not sufficient. Take his second case. Smith has strong evidence that his colleague Jones owns a Ford: he has seen Jones driving one for years, and Jones has just offered him a lift in it. Smith has no idea where his other acquaintance Brown is. He nevertheless notices that a disjunction follows from what he believes, and accepts it: either Jones owns a Ford or Brown is in Barcelona . Now suppose Jones has in fact sold the Ford and is driving a hired car, while Brown, by pure coincidence, is in Barcelona. Smith's disjunction is true. It is justified, since it follows deductively from something for which he has excellent evidence. And Smith plainly does not know it. The structure is general: justification is inherited from a false belief, and truth arrives from an unrelated direction. Mozi and the three standards Mozi (c. 470–391 BCE) founded a school in the Warring States period that was, for two centuries, Confucianism's chief rival. Its ethics is consequentialist: the criterion of what is right, yì , is what produces benefit, lì , for the world at large, and the school's central practical doctrine is jiān ài , concern extended impartially rather than graded by relationship. Its best-known methodological contribution is the three standards, set out in the chapters against fatalism. A doctrine is to be tested by rooting it in the deeds of the ancient sage-kings; by sourcing it in what ordinary people have actually seen and heard; and by putting it to use in government and observing whether it benefits the state and the people. Mozi's target is fatalism, the doctrine that outcomes are fixed regardless of effort, and the three standards are the instrument with which he rejects it. Why the standards do not touch Gettier Notice what the standards are for. Their subject is a yán — a saying, a thesis, a doctrine that can be adopted or rejected by a state. Fatalism is such a thesis; so are the claims about ritual expenditure and offensive warfare that Mozi opposes. A Gettier case is not of this kind. Smith's disjunction about Jones and Brown is nobody's doctrine, has no policy consequences, and no one would think of testing it against the deeds of the ancient kings. The pragmatic reading makes things worse rather than better, and this is the part of the standard comparison that most needs abandoning. Suppose we say that what makes a belief count as knowledge is its tendency to produce benefit. Then a belief that is beneficial by luck is knowledge, which repeats Gettier's problem in a new currency. The Mohists were not naive about this. The reason there are three standards, and not just the third, is that usefulness in application is not by itself a warrant: the first two standards look to what has been recorded and to what people have observed, precisely because a doctrine can be adopted with good effect and still be false. A reading that treats Mohism as replacing truth-conduciveness with utility is not a stronger epistemology than the justification-based one; it is a weaker one. What the later Mohists actually offer The material that repays comparison is in the Canons — the terse, technical chapters preserved in the Mozi and reconstructed by A. C. Graham in Later Mohist Logic [5] , Ethics and Science (1978). These are the work of the school in its later, more analytic phase, and they contain definitions rather than exhortations. Two of them matter here. The Canons distinguish the capacity to know from the achievement of knowing something in particular, which is already a distinction the three standards do not draw. And they classify knowing by where it c