The case of combination variants like dadhi and dadhy is used by Nyāya authors as an evidence of the fact that words are produced and modified. Mīmāṃsā authors, who think that language is without beginning, need to respond to that and explain therefore that dadhy is not a modification of dadhi, but an alternative word, and both are used in specific phonetic contexts.
Veṅkaṭanātha in his commentary on PMS 1.1.16 elaborates thereon and explains that they are described as archetype and ectype of each other for pedagogical reasons only (in order not to further multiply the number of words to be learnt). At this point, he faces two very different objections.
The first opponent says that the archetype-ectype relation could be reverted according to a different grammatical analysis. This probably means that dadhy could be considered as the archetype and dadhi as the ectype. Veṅkaṭanātha answers that one should choose the grammatical analysis based on its pedagogical merits, and the one suggested by the opponent is not pedagogically easier.
The other opponent says that the ectype-archetype relation is real and based on the similarity between the two. The “similarity” is not further elaborated upon, but we can guess something more about it through Veṅkaṭanātha’s reply. Veṅkaṭanātha answers that if similarity were the ground for real archetype-ectype connections, then there would be no way not to avoid over- and under-extensions. On the one hand, one could over-extend it to other cases of similarities, like yogurt (dadhi) and jasmine flowers, that are similar insofar as they are both white, although they are not considered to be archetype and ectype of each other. On the other hand, cow-dung and beetles are dissimilar, but are considered one the ectype of the other (beetles are believed to be a transformation of cow-dung).
Now, my problem regards a terminological choice. The first opponent says: vyākaraṇāntareṇa prakṛtivikṛtivaiparītyam. In his answer to the second opponent, Veṅkaṭanātha says na ca sādṛśyāt prakṛtivikṛtibhāvaḥ śaṅkhyaḥ, vaiparītyasyāpi prasaṅgāt. However, vaiparītya in the first case seems to be the opposite of what should be the case (the inversion of dadhi and dadhy as archetype and ectype). By contrast, in the second case vaiparītya seems to indicate just a different set of consequences. Comments welcome!
(Cross-posted on my personal blog, elisafreschi.com)
elisa, you have set a hard problem here, a puzzle for the specialist: the terms and the context are not familiar to me. Firstly, archetype/ ectype sounds neoplatonist to me: is it really appropriate here? or perhaps framing the problem in an unhelpful way?
Secondly, the term vaiparītya is generated from viparita, which featured in an earlier debate involving Bartrhari and Mandana Miśra, concerning error as a possible source of knowledge. Allen Wright Thrasher gave a close commentary in The Advaita Vedānta of Brahma-Siddhi, which I cited in one of my first posts here; now @ p. 17.
In this perspective, is the second term an ‘ectype’ of the first, or rather an exception to the rule of the first, thus an ‘error’ which also marks the limits of the rule? That would give Veṅkaṭanātha moving on into close linguistic analysis, in keeping with his focus on texts.
Thirdly, with regard to the two forms of the key term, I find the ending -asya to be ambiguous between a genetive and a contraction of as-ya, with the sense of ‘that which’ : and here that could mark Veṅkaṭanātha making a pragmatic turn, from signifier to signified.
I find Patañjali making frequent use of that contraction, so establishing a distinctively pragmatic realism. In the time of Veṅkaṭanātha, Hatha yoga was coming under commentary, and I think Veṅkaṭanātha would have understood the Yoga sūtra better than most.
Looking back now to my original query about compounds, I am surprised to find that the Sanskrit prefixes seem to derive from Munda, and not any Indo-European ‘archetype’; Panini’s rules assume generation by suffixing; and the classic sa- and nir- then enter from Buddhist discourse! Patañjali does not use them.
. . .with klapa, but rather bija, where the context turns to herbalism and Southern, not Vedic plants; and the syntax runs curiously into computation and astronomy, through the bija-corrections. All also not Indo-European in any ‘archetypal’ sense.