Choeroboscus' text is corrupt and has κινῶ: διὰ τοῦ ι. παρὰ γὰρ τὸ κιινεῖν, which does not make any sense. Lentz justifies the correction and assumes that, although this etymology is not explicitly given as Herodian's in our sources, it goes back to him. It is a derivational etymology explaining a motion verb from another one. It implies one formal manipulation, the insertion of [n]. As often, the shorter form is assumed to be the etymon and the longer one the derived form. From the semantic point of view, Greek etymologists did not take into account the fact that κινέω is transitive and means "to set in motion", to move something", whereas κίω is intransitive "to go", so that the etymon matches semantically the middle κινέομαι: this is a consequence of their lack of interest for the problem of diathesis. The lemma is given in the active form, which is the rule, even if the etymology refers to a meaning which is not that of the active but of the middle or passive. The etymology is basically correct from a modern point of view: κίω and κινέω belong to the same root, even though the one is not derived directly from the other