But this is incredibly obscure. I mean, no-one has ever heard of a phrase like ekaį¹ samayaį¹ bhagavÄ sÄvatthiyaį¹ viharati. O, wait ā¦ </sarcasm>
Thanks for pointing out what I should have recognized immediately!
Perhaps the nuance is to be found in a Sanskrit usage, where it can mean the āobscuredā moon. In its current, and perhaps original, context, the verse is set on an uposatha, and earlier in the text we have several references to the sky:
abhikkantÄ, bhante, ratti; nikkhanto pacchimo yÄmo; uddhasto aruį¹o; nandimukhÄ« ratti
So itās not us covered by the clouds, itās the sky. And we could translate:
It rains when the sky is hidden,
not when itās revealed.
So reveal whatās hidden,
and it will rain no more.
I doubt if itās germane to the translation, but it is interesting to note that, if we are correct, vivarati here has exactly the opposite of its ancestral Vedic meaning. One of the primal Vedic myths is of Vritra, the ādragonā (= ahi) who imprisons the waters, bringing drought to the land. Indra slays the dragon (vivarati) releasing the waters. The meaning of vritra as āconstrictorā is still felt in the Pali nÄ«varaį¹a.
While this is no more than a side note, it does point to another curious feature of the verse which is usually overlooked. Normally, especially in agrarian cultures, rain is a good thing. But here it seems to be considered a bad thing. However, it is not so straightforward. When Vritra holds back the waters you have a drought. But if he releases them all at once you get floods.
And in one place where ativassati is used with a clear contextual meaning, thatās exactly what it means. In Mil 6.3.1 we have ativassena dhaƱƱaį¹ vinassati.
Obviously ātoo muchā is the most common meaning of ati. Equally, it might mean āon or overā as is usually interpreted here.
CPD hedges its bets, giving āexcessive rainā for ativassa, but āto rain violently, to rain into or throughā for ativassati, attested only in this verse (Cone has more references). It does seem a little unnecessary to posit two senses of a word for one context, when ārain too muchā works perfectly well.
Comm. is ambiguous. It says that the positive use of ativassati does indeed mean too much (ativiya vassati), but in the negative form nÄtivassati the ati is a mere prefix.
The sense of ārain intoā seems to be derived from the Ghatikara Sutta:
Atha kho, mahÄrÄja, Ävesanaį¹ sabbantaį¹ temÄsaį¹ ÄkÄsacchadanaį¹ aį¹į¹hÄsi, na devotivassi
But this has variant readings. Perhaps we should read abhivassati with Mil here. (See Cone under ativassati and abhivassati).
In any case, the meaning of ārain on or intoā is relevant only when the channa is the covered thing that the rains falls on. If it is, instead, the covered sky, then ārain too muchā would seem to be better:
It rains too much when the sky is hidden,
not when itās revealed.
So reveal whatās hidden,
and it wonāt rain too much.
Oh, and I just noticed, that passage from the Ghatikara Sutta is relevant here, too. It says ÄkÄsacchadana, which BB renders āopen to the skyā, but literally it is āsky-coveredā, i.e āthe sky was its only roofā. (cp. English āsky-cladā=naked). This would seem to agree with our interpretation of channa.