I found another passage that may shed light on the pāsāda. In MN 12 we have:
Seyyathāpi, sāriputta, pāsādo, tatrāssa kūṭāgāraṃ ullittāvalittaṃ nivātaṃ phusitaggaḷaṃ pihitavātapānaṃ.
Suppose there was a stilt longhouse with an upper storey plastered inside and out, draft-free, with bolts fastened and windows shuttered.
If we take this at face value, then the actual enclosed upper storey of a pāsāda was referred to as kūṭāgāra. Of course, this makes perfect sense, but it does raise some questions. Are all kūṭāgāras in fact the upper storeys of pāsādas? It seems unlikely, but we should look out for it. And how best are we to translate it in such cases? Ven Bodhi has “upper chamber” which seems inadequate.
But it’s not entirely clear to me this reading should be accepted. The description of the luxurious kūṭāgāra is stock (eg. AN 8.30). Has it simply been dragged along to fill out the description of a luxurious dwelling, which here is applied as a metaphor for heavenly pleasure?
Yes, it seems that that is what it means. My doubts are because I’m not aware of anywhere else that a kutagara is said to be the upper storey of a pasada; they are usually considered to be separate kinds of buildings. But the sanskrit link you showed is interesting, so it seems it does have this meaning.
Yes, I get the impression that pāsādas were very diverse. In the Vinaya there area number of buildings with different names such as hammiya and aḍḍhayoga, which according to the commentary are all types of pāsādas. I suppose there is no reason to think the kūṭāgāras would be any different. When you read the description of a kūṭāgāras as “plastered inside and out, draft-free, with bolts fastened and windows shuttered,” it is easy to think it refers to some sort of masonry structure. But I think most, perhaps all, plastered buildings in those days were of the wattle and daub variety. Having this sort of structure on piles should be no problem. It is only when we move into the age of masonry buildings - bricks, stones, and the like - that still-houses become impracticable. As far as I know, there is only one brick building mentioned in the suttas, the giñjakāvasatha at Nādika, e.g. at DN16 and MN31, as well as a few other places. The way it is describes, “the brick house at Nādika,” makes it seem as if it was something of a curiosity and rarity in its day.
I think the renderings “upper chamber” or even “upper storey” are both a bit problematic. They seem to suggest there is a lower storey, when in fact there is not. Why not just “a stilt house with a peaked roof, plastered inside and out …”?
thank you @anon29387788 for the correction, i actually was uneasy about the spelling, but i was too lazy to do a dictionary lookup and having looked at Ven @Sujato’s i decided it was correct (sorry, bhante)
And further in random observations, in MN 37 we have the mention of Sakka’s Palace of Victory, which has a hundred towers, each one of which has seven hundred kutagaras, and in each kutagara there are seven nymphs, each with seven attendants. Here it must mean something like “chambers”, although Chalmers has “stories”.
This, I think, conclusively shows that the antepura, unless otherwise specified, is the rājantepura. It also shows that the gate is a place where the hoi polloi would gather, i.e. it was accessible to everyone. Entrance within was, presumably, guarded by the king’s guards.