In the same seminar on quotations mentioned above (see here and this post), Raffaele Torella proposed to distinguish between several typologies of the usage of quotations:
- quotations as text (texts made only of quotations)
- hidden quotations (large passages which migrate, so to say, from one text to another, without mention of their author. Torella mentioned in this regard the case of the Vaiṣṇava Advaitin Vāmanadatta. Large quotes of him have been reused in Vaiṣṇava Tantras.
- manipulated quotations (e.g., a dualist text may be forced by a non-dualist to justify her school's stances).
- quotations as empty containers (e.g., the no-more-productive Sāṅkhya is reinterpreted by the late Advaitin Vijñānabhikṣu who interprets it as an Advaita school, thus filling quotations of Sāṅkhya texts with Advaita contents.
- made-up quotations (such as, maintains Torella, many of Madhva's ones. I am still perplex about this case since I cannot understand why Madhva should not have, instead, forced a text to support his opinion, manipulating a text (or using it as an empty container) instead of faking one).
- exemplificatory quotations (laukikanyāya, functioning like our proverbs, where the authors is hardly recognisable and has no role at all).
This typological distinction is probably not exhaustive and it conflates various meanings of "quotation", without distinguishing literal quotations from references, but it is a first (and hence useful) attempt to discuss the function of quotations within a text.
In this regard, the Islamist Paola Carusi observed that many occult quotations are such only for us. For their readers/listeners, their origin was obvious. This is obviously the case for quotations out of the Sacred Texts. Even in India, the case I am more acquainted with, one should remember that lay readers did not exist. One would only start listening to a teacher's lesson or read his text if one were his pupil (or a rival's pupil). Hence, texts may presuppose a general acquaintance with the school's basic tenets and with its root texts. In sum, occult quotations should only designate quotations which are intentionally hidden (for instance, because one wants to re-use a passage of an author who has been banned from one's tradition.
Maria de las Nieves Muñiz proposed a very interesting suggestion while discussing Giacomo Leopardi's Zibaldone: there used to be an "interlingua" (inter-language) current in the intellectual milieu of Leopardi and made of texts, ideas, quotes, pictures, myths, etc. We have to reconstruct it in order to make sense of Leopardi's horizon and to understand who has been read and who has been, instead, indirectly absorbed through the interlanguage.
The same scholar also noticed that too often one does not quote exactly the authors one has been more influenced by. One is, so to say, permeated by them and cannot distinguish theirs from one's own thought and words. To explicitly quote, I would summarise, requires a distance.