Why the Commentators ended up with the theory of khaṇa
(moments). I suspect this stems from how the Commentaries interpret 2 key messages in the Abhidhamma.
Firstly, in the Vibhaṅga's treatment of DA, there is an alternate to the 3 lives presentation, where all the 11 links are compacted into a per-citta
basis (see s.248 to 255 of the Paṭiccasamuppādavibhaṅgo). Then, there is the presentation used in the Dhammasaṅgaṇī, where you have relationships laid out in a very standard pattern -
Yasmiṃ samaye (insert type of citta under discussion), tasmiṃ samaye A hoti, B hoti, C hoti etc etc
What's striking is that the Vibhaṅga and the Dhammasaṅgaṇī both employ the relative clause structure of "Yasmiṃ samaye ....tasmiṃ samaye ..
..". The Comy would treat both ya
locatives as being temporal locatives, meaning that whatever is in the principle clause (the ta
clause), they must all be concurrent and simultaneous with the subordinate ya
clause. Everything has to happen all at once in each citta
. This includes the unfortunate side effect from this -
Yasmiṃ samaye akusalaṃ cittaṃ uppannaṃ hoti somanassasahagataṃ diṭṭhigatasampayuttaṃ rūpārammaṇaṃ vā saddārammaṇaṃ vā gandhārammaṇaṃ vā rasārammaṇaṃ vā phoṭṭhabbārammaṇaṃ vā dhammārammaṇaṃ vā yaṃ yaṃ vā panārabbha, tasmiṃ samaye avijjāpaccayā saṅkhāro, saṅkhārapaccayā viññāṇaṃ, viññāṇapaccayā nāmaṃ, nāmapaccayā chaṭṭhāyatanaṃ, chaṭṭhāyatanapaccayā phasso, phassapaccayā vedanā, vedanāpaccayā taṇhā, taṇhāpaccayā upādānaṃ, upādānapaccayā bhavo, bhavapaccayā jāti, jātipaccayā jarāmaraṇaṃ.
has to arise concurrently with its decay. The occassion (samaya
) in this case has to be compressed by the Comy to the smallest possible hypothetical to accomodate this reading of DA on a per citta
basis. As a logical inference, this seems reasonable if one decides to read both the locatives as being temporal locatives. But are they necessarily so?
While "Yasmiṃ samaye ....tasmiṃ samaye.
..." will be correctly interpreted as importing simultaneity most of the time, there are examples from the suttas where the tasmiṃ
is not functioning as a temporal locative, but as a referential locative (see AN 9.39 where the same construction is used to describe the jhanas and thinking). Warder gives copious example of the relative pronoun ya
and the demonstrative pronoun ta
carrying different inflectional senses, and there is no reason to suppose that the Comy was correct in reading both locatives as being temporal ones.
So, if the Comy had not followed the Sarvastivadins in reading the samaya
as time, but had instead read these passages as -
On whatever occassion A arises, then with reference to that occassion X, Y, Z
I suspect the Comy would have totally dispensed with the theory of moments. X, Y and Z would then not be need to be simultaneous with A, nor with one another.