Commentaries can be manifold in ancient India. They have different purposes and form, but they all share some characters:

• they have a given text as their main interlocutor/they are mainly about a given text
• like with Origene’s commentaries, they are a genre in its own right, not a minor specialisation for authors at their beginnings (Sakai 2015, section 4, suggests that authors in fact needed to have already become acknowledged authorities before being entrusted with the honour of composing a commentary on an influential text.)
• they are characterised by a varied but strong degree of textual reuse
• they allow for significant degrees of innovation (This is evident in the case of the Navya Nyāya commentaries on the NS. Outside the precinct of philosophy, juridical commentaries often reflect the recent juridical developments much more than the original text they are commenting upon.)

As for reuse, one might object that reuse is much more present in later commentaries such as the Seśvaramīmāṃsā than in earlier ones, such as the Śābarabhāṣya. However, the ŚBh does indeed quote extensively from a previous commentator and the fact that contemporary readers do not recognise many other reuses does not rule out the possibility that Śabara did in fact extensively reuse but, as it is often the case throughout Indian philosophy, without marking his reuses, as he could assume that his audience would have recognised what was happening. In other words: the seeming increase in the amount of textual reuse from, e.g., the 2nd c. to the 14th. could be due more to the increase of our awareness of reuse. In this connection, it is worth remembering that:

• commentaries are also an important source for the retrieval of (written or oral) texts which would otherwise be lost

Commentaries bring us back within a close analysis of a text, often even in an advanced classroom milieu. Thus, they need to evoke important textual authorities, including the ones which happen to be fashionable at their time and might have been lost or never recorded in script.

By contrast, commentaries diverge sharply as for other characters. So much, that even the first item listed above might need to be re-conceived in a plural form, with texts entangling at the same time various others, as it can comment on various different texts and discuss with various others, named and unnamed ones. The landscape of Indian philosophy (perhaps of all philosophical traditions?) is complex and invariably entangled.

What is your experience with the genre of commentaries? Do you have counter-examples or would you rather agree with my preliminary assessment?
For a first attempt towards the definition of the genre of commentaries, see here. I am grateful to Ramakrishna Bhattacharya for his feedback on this post (see below).