Dear Jan,

the domain in which you are specializing is fascinating and some rudimentary knowledge of it is useful, even indispensable, for art historians and for other Sanskritic śāstras including kāvya -- should hence evoke a more than lukewarm interest of Indologists and readers of this list...

Encouraged by this positive evaluation, I am sharing my reply with the list.

It is indeed "the nub" of my argument to take "sahita asya to represent sahitaḥ asya rather than sahite asya" because the latter option, as we have seen, leads nowhere except for turning an author elsewhere eager to communicate meaning accessible to 99 percent of his contemporaneous public, into an obscurantist who can only be fathomed by 1 percent

With respect, though, we haven't actually seen that. You claimed it, and I said that even if it were so, etc. In fact, I would still like to know on what you base that claim, as I cannot see it at all.

-- plus a whole tribe of philologists centuries later who are ready to accept suddenly, ad hoc, a Vedic sandhi, etc., in what otherwise seems to be a śāstric text in impeccable classical Sanskrit.

The sandhi problem is the same in either case, as far as I can see. Sahitaḥ + asya ought by standard practice to have resulted in sahito 'sya.

Next, trying to think in line with your argument and in the wider context of your interpretation, what could be the syntax of the verse? 

Apparently there are two options separated by vā. The expression yadgṛhe asks for a corresponding term, in the first option tatra, in the second option, I would suggest, tena (gṛhena, rather than, in your interpretation, janmalagnapatinā).

But what would that mean? In that house which is aspected by that [house] does not make any sense. And conventionally it is the planets that are invested with 'sight' or aspect.

Both options lead to the same result: asya ... labdhis. If tatra and tena go with yadgṛhe, asya probably does not refer to yadgṛhe -- here I would modify my previous suggestion: could it go with janmalagnapatir? Could the janmalagnapatir, the lord of the first house, have something to do with the own body and hence with aṅgasukham?

The body is considered the portfolio of the first house, and, by extension, of the ruler of the house. So the latter is a special case of the former: the ruler of the first house would affect the body, but so would planets occupying or aspecting the first house itself.

Then, if both options lead to the same result, what could be the precise difference between them? In the first, the janmalagnapati is said to be uttamavīrya, and the house where he resides is (positively) aspected

But the 'positively' is not in the text. The fact of the house being aspected by the strong ruler of the ascendant is itself considered positive, regardless of the type of aspect.

-- whether tatra ... dṛṣṭe is a locative absolute or whether it is in direct concord to yadgṛhe. In the second, in my suggested reading, he is not necessarily uttamavīrya, but at least sahita with "that", i.e. with that house. MW 1095 col. 1 gives "(in astron.) in conjunction with (instr. or comp.)" for sahita. In western astrology, "in conjunction with" is always a matter of plus or minus 2-3 degrees, so that a planet can be in conjunction with another planet or even a house even if it is not squarely coinciding or residing in it, even if it is just outside that house: possibly Indian astrology is here more black and white,

Both Indian astrology and ancient and classical astrology generally are indeed clearer in their distinctions than modern western astrology, but that is slightly beside the point: the present question is of a more technical and, to some extent, terminological nature. If a planet were within a few degrees of the first house but not actually in it, it would necessarily be in some other house: either the second (which relates primarily to wealth and assets: dhanabhāva) or the twelfth (which concerns losses and the like: vyayabhāva). This is true irrespective of which method is used for dividing the houses: there are no empty spaces between them. It would be highly counter-intuitive for Yādavasūri to state pleasures of the body as the only result to be expected from a planet actually placed in either the second or the twelfth house, even if it were close to the juncture with the first.

The way I understand the syntax of the verse under consideration, the correlates of yadgṛhe are, first, tatra (with tatra ... dṛṣṭe ... vā sahit[e] indeed being a locative absolute, tatra standing in for tasmin for metrical reasons -- slightly inelegant, but not unusual), and second, asya. Tena refers to the agent of the two passive participles, which has to be something other than the house itself and therefore can only be the ruler of the ascendant. The conclusion of the verse bears this out: if [the planet] is in the first house (tanau), there is pleasure of the body, which is the domain of the first house. The ruler being sahita 'joined' thus clearly means being placed in a house (locative).

I take your point about not neglecting particles and so on, but in a metrical text, I think a madhyamā pratipad is the wisest option. Sometimes these little ca, tu, hi, etc., really are just there to fill out a line, especially if the writer is an indifferent verse-maker, or in a hurry, or temporarily uninspired. I speak from experience. :-)

Best wishes,
Martin