Certeza lógica: certeza além da práxis

Abstract

In an impermanent world, is anything fixed? Could we point at the landscape and proudly declare, like Russell, "The grass is green"? In Über Gewißheit (On Certainty), Wittgenstein develops G. E. Moore's disconcertingly direct objection to the immaterialist skeptical doubt, concluding that the grass is indeed green, though not in the expected sense. Something is fixed, but not the reality pinpointed by Russell and Moore; rather, it is the conditions of our language games, and thus of doubt itself. Our thesis is a logical reconstruction of the conception of certainty investigated by Wittgenstein and inspired by Moore. Our central negative dispute is the rejection of the normative conception of certainty, proposed in Über Gewißheit and widely adopted in the exegetical literature; the positive one is a reconstruction of certainty as a new form of analyticity, dual to Moore's paradox or, broadly speaking, as the logical limit of the conceivable. In particular, we will argue that the grammatical conception of certainty is circular because it bases language games on a language game; and inconsistent, for taking certainties as unequivocal rules, a contradiction in terms. We also contest the very concept of unequivocity, which we have shown to be logical necessity “in disguise”. This suggests that a logicist resumption is a rather consistent development of the language model in Philosophische Untersuchungen (Philosophical Investigations), if we recognize that the possibility of a language game is conditioned by certain facts. Thus, we outline a logical view of certainty, inspired by Tractatus Logico-Philosophicus, in which certainties are analytical presuppositions, and the domain of bipolarity includes propositions and rules, given the mutual dependence between doubt and equivocity. Finally, we apply these results to some established topics of the epistemological debate. In particular: we show that the normative vision of certainty conflicts with its externality, which we outline in a reliabilist key; whereas the grammatical or normative approach fails as an antiskeptical argument, but a more satisfying objection is possible, taking certainty under a logical key; finally, we argue that such a logical reading is also a more appropriate objection to Methodist challenges to the foundational status of certainty, and to those that aim at its coherentist aspect, such as the input problem.