The ancient Indo-European languages, such as early Vedic or (Homeric) Greek, are usually considered to be characterized by a high degree of lability. According to the communis opinio , they had a considerable number of labile verbs or verbal forms that could be labile, cf. rudrā॔ r̥tásya sádaneṣu vāvr̥dhuḥ ‘Rudras have grown [intransitive] in the residences of the truth’ ~ índram ukthā॔ni vāvr̥dhuḥ ‘The hymns have increased [transitive] Indra’. This paper offers a general overview of the Vedic verbal forms for which labile patterning is attested. I will argue that, for most of these forms, the secondary character of lability can be demonstrated. Thus, for many labile forms with middle inflection (in particular, forms belonging to the present system), labile patterning results from the polyfunctionality of the middle diathesis (self-beneficent / anticausative). The secondary transitive usages of some fundamentally intransitive verbs such as puṣyati ‘prospers; makes prosper’ originates from the syntactic re-analysis of content accusative constructions of the type ‘X prospers (in) Y’→‘X makes Y prosper’. I will further demonstrate that, within the Old Indo-Aryan period, we observe the decline of the labile type. Already in the second most ancient Vedic text, the Atharvaveda, we find very few labile forms. Thus, most of the active perfects which show labile syntax in the Rgveda are either attested in intransitive usages only, or in transitive usages only, or do not occur at all. I will also discuss the main mechanisms of the loss of labile pattern in Old Indo-Aryan.