Abstract
The distinction of passive and middle voice in Tocharian shows clear correspondences with the Proto-Indo-European state of the art. In the functional perspective, the usage of the middle voice ranks Tocharian together with Hittite, Old Indo-Aryan or Ancient Greek.
TA verbs trıku, wıyu, and spärksa exhibit a combination of a passive stem and active ending. I suggest that these forms represent an innovation, possibly restricted to these and semantically close change-of-state verbs.