Dear Rene,

I will respond below.

rsalm wrote:

>1. avijjaa and upanisaa are nouns, both feminine. The ending of the first word has been elided, as is normal in compounds, and it must be determined from the context if 'upanisaa' is either (a) nom. sg. fem; (b) nom. pl. fem.; or (c) nom. pl. masc. agreeing with sa'nkhaaraa (in a bahubbiihi).
>
>

Exactly. Context is key here. It makes it easier to see this point by
looking to the rest of this passage in which my previous example of
"sa.laayatanuupaniso phasso" occurred, because it gives an immediate
clue that something is clearly different than the usual declension of
"upanisaa." Another clue comes with "upaadaanuupaniso bhavo," etc.

avijjuupanisaa sa’nkhaaraa, sa’nkhaaruupanisa.m vi~n~naa.na.m,
vi~n~naa.nuupanisa.m naamaruupa.m, naamaruupuupanisa.m sa.laayatana.m,
sa.laayatanuupaniso phasso, phassuupanisaa vedanaa, vedanuupanisaa
ta.nhaa, ta.nhuupanisa.m upaadaana.m, upaadaanuupaniso bhavo,
bhavuupanisaa jaati, jaatuupanisa.m dukkha.m, dukkhuupanisaa saddhaa,
saddhuupanisa.m paamojja.m, paamojjuupanisaa piiti, piituupanisaa
passaddhi, passaddhuupanisa.m sukha.m, sukhuupaniso samaadhi,
samaadhuupanisa.m yathaabhuuta~naa.nadassana.m,
yathaabhuuta~naa.nadassanuupanisaa nibbidaa, nibbiduupaniso viraago,
viraaguupanisaa vimutti, vimuttuupanisa.m khaye~naa.na.m

>2. The compound is a kammadhaaraya (Kd). A Kd requires that both components be in the same case and that one qualify the other. Both conditions are met here. The cpd. is nominative, and 'avijjaa' modifies 'upanisaa': to the question "What kind of cause?" the answer is, "The cause of ignorance." Though the Kd is sometimes referred to as an "adjectival compound" (Budh/2.40), it can be made up of two nouns, as we have here. Perniola (164) gives four kinds of such Kd. This kind would be a nominal Kd of "apposition."
>
>

Right again. Though instead of "the cause of ignorance," a better
translation would be either "cause which is ignorance" or "ignorance as
cause" because "the cause of ignorance" sounds too much like a tappurisa
with a genitive case relation between the nouns. In relation to types of
Kammadhaaraya compounds, you might appreciate this "Compound reference
sheet" that I have compiled from conversations here in this group.
http://www.lioncity.net/alanmcclure/resources/Pali_compounds.htm
In this sheet you will see the four kinds of Kd. of which you speak
above, namely:

*1) adj+substantive(noun)*
*2)* *adjective (or adverb)+adjective(or participle)*
*3)* *substantive+adj *
*4) substantive+substantive*

*English:*
1) Black-bird: a bird that is black
2) Soft-spoken
3) Ice-cold: cold like ice
4) Girl-friend: friend who is a girl
*
Paa.li:*
1) Akaalamegho: untimely cloud
2) Sammaapaṭipanna: rightly disposed, rightly seen (as in right view)
3) Paa.nasama: lit: ‘The same as life', hence: 'dear as life'.
4) Raajisi: “king-sage” (there is elistion of the –an)

>3. In this case, the compound has been turned into an adjective and modifies 'sa'nkhaaraa.' This is what makes it a bahubbiihi (Bh). In a Bh, something essential for the meaning is found outside the compound itself. As Warder puts it (137), "The Bh class of compounds consists of those whose meanings are subordinate to the meanings of words other than the members of the compounds themselves." In this case, the relation between ignorance and cause is united only in the term 'sa'nkhaaraa.' It is sa'nkhaaraa which have ignorance as cause.
>
>

Exactly. We are not talking primarily about "ignorance as cause," the
main focus of the phrase is the subject 'sa"nkhaaraa' which is then
qualified with the bahubbiihi of 'avijjuupanisaa.'

>This furnishes the translation of the clause: "Thus, indeed, monks, [it is] sa'nkhaaraa [which have] ignorance as cause." Rett and Ole have both given us a shorter and smoother version: "Thus, indeed, monks, sa'nkhaaraa have ignorance as cause." The ending of upanisaa is (c) above, agreeing with sa'nkhaaraa (m. pl.)
>
>
>

Right again.

>This makes sense and I accept it. The rest of this post is a discussion, somewhat academic and a bit involved.
>
>One thing that I find interesting is that (at least on the face of it) Bh's are clauses, not complete statements. To give some examples (Pern/169): paapiccho bhikkhu = 'a monk who has evil desires'; tarunavacchaa gaavi = 'a cow that has a young calf'; antimasariiro puriso = 'a man that carries the last body.' These trnaslations (as given by Perniola) are not complete statements (sentences) but clauses. But this doesn't seem to bother him. He goes one step further and converts a clause into a complete statement (sentence) when the context demands. I notice this in his example that we discussed on an earlier thread (don't worry-I won't get into this one again): Perniola considers 'manopubba'ngamaa dhammaa manosetthaa' also as a Kd à Bd. He translates it literally: "factors that have the mind as the first and as the best." Again, this is not a complete thought, only a clause. But then Perniola immediately interprets it: "i.e., mind is the first and the best of all factors." This is a complete statement.
>
>So, apparently, the translator of Bh's has two choices as determined by context. One is to treat the complex as a clause, e.g.: paapiccho bhikkhu adhaama.m deseti, "The monk who has evil desires preaches untruth." The other alternative is to treat it as a complete statement: bhikkhave, paapiccho bhikkhu vadaami, "monks, the monk has evil desires, I say." I wonder if this is correct.
>
>

I'll have to let others respond to this as I have not had a lot of
experience translating bahubbiihi compounds.

>
>----------
>
>Perniola (169) writes that a tappurisa compound (Tp) can also be turned into a bahubbiihi. (In a Tp, the independent components of the compound are not in the same case.) He gives a couple of examples: (i) Gotamo samaggaaraamo (samagga, 'mutual accord') = 'Gotama whose delight is in mutual accord' (instr. Tp); (ii) a.t.thika-sa'nkhaalika.m sariira.m (a.t.thi, 'bone') = 'the body with its chain of bones' (gen. Tp). We notice here, as before, that these are not literally complete statements. From the discussion above, I presume that these compounds also can also be interpreted as complete statements, e.g.: Bhikkhave, Gotamo samaggaaraamo, "Monks, Gotama delights in mutual accord"; a.t.thika-sa'nkhaalika.m sariira.m, "The body is a chain of bones." I wonder also if this is correct.
>
>-----------
>
>We know that certain compounds can be interepreted in various ways. As I understand 'a.ngulimaala,' for example, this compound can function (a) as a dvanda: ("fingers + garland"); (b) as a Kd ("fingers-garland," answering the question "what kind of garland"); or (c) as a Bh (as the name of a particular person during the time of the Buddha). Context and terminations usually tell us which is meant. I suppose that if both Kd and Tp compounds can be interpreted as Bh and as complete statements, then some interesting grammatical intersections exist. But I haven't studied this enough to know.
>
>

Yes, this is an important point that we have discussed before. If
terminations are unclear, often context is everything, and even then,
sometimes different options seem possible, i.e., tappurisa and/or
kammadhaaraya. As Rett has pointed out before, having some understanding
of the doctrinal peculiarities of Buddhism is helpful for many cases of
this type, but in other cases it is hard to know what was intended
without knowing the way of thinking of the person who created the word.
This is where, as you put it, "interesting grammatical intersections
exist," and probably interesting grammatical debates as well.



All: I'll give my usual disclaimer for appreciating any corrections (by
Rett, Dr. Pind or others) of what I've said above, but I don't think
I've made any comments that fall off the Pali beaten path.

Rene: Please let me know how you find this "Compound Reference Sheet,"
if you have the time to look at it, and I'll do some updating on it
today to incorporate this latest discussion.


Metta,

Alan