ИСТИНА

Almost all analytic Kant scholars unhesitatingly accept the depth and brilliance of Kant’s work. But a main analytic focus is on stating Kant’s ideas in clear, unambiguous ways that can be evaluated for truth or falsity. Given this focus, I first sketch some main parts of the development of Kant interpretation in analytic philosophy. Then I turn to issues about the interpretation of Kant’s transcendental idealism, a major topic for analytic approaches to Kant. I note questions about that idealism when it is understood in a traditional ontological way. I also note three problems about things in themselves to which that idealism, understood in that way, gives rise—the unknowability problem, the noumenal–affection and category–application problem, and the problem of nonspatiotemporality and the neglected alternative. Any strictly Kantian, philosophically satisfactory resolution of these problems should work within, rather than radically modify, Kant’s actual views, while also being philosophically plausible in its own right. In the light of this standard for resolving the problems, I examine characteristic resolutions that have been proposed by some of the leading Kant scholars of the past fifty years. I argue that, by the standard just noted, none of these resolutions succeeds, and I express doubts that any strictly Kantian resolution of these problems is possible. In conclusion, I make two basic recommendations: (i) we should follow the lead of many analytic Kant scholars by rejecting Kant’s own transcendental idealism. We thus avoid the problems for things in themselves that that idealism entails, and we can focus on the many crucial Kantian ideas that can be stated independently of Kant’s idealism. (ii) We should explore the possibility for reconstructive approaches to Kant’s work that modify key Kantian views, including his idealism, in such a way as to escape the problems about things in themselves. Such approaches may capture what, if anything, is viable in that idealism.