Basque's Surdéclinaison

[Valdeut]

Ah, I hadn't noticed that it was the same author. His use of quotations makes me a little wary, but otherwise okay: you've convinced me that "surdeclinaison" is a real linguistics term (especially since that book is more recent than Double Case).

It appears to be an established term at least in vasconic linguistics, especially in French but also in English to a lesser extend. And at least in French, it seems to be used occasionally beyond Vasconic. See this example about Burushaski:
https://books.google.se/books?id=fjU7M2mdQ2oC&pg=PA22

I'm also not entirely sure what the term means, though. If I'm not mistaken, the Burushaski process mentioned is basically a form of compound case, very much similar to the multi-morphemic cases of some Daghestanian languages like Tsez. This seems to me to be very different from Basque.

Grandsire-Koevoets gives an incredibly broad definition of "surdéclinaison" in his presentation:
"The ability some languages have to take an already inflected word, and inflect it again, giving it a different function in the phrase and/or sentence."

It seems like this definition would fit a wide variety of processes. I'm not sure that surdéclinaison is a coherent linguistic phenomenon, but the term is certainly somewhat established.

What I'm not yet convinced of, though, is whether hypostasis and surdeclinaison are actually different phenomena from suffixaufnahme or suffixhaufung. Perhaps I am missing something, but they both seem to just be headless constructions of suffixaufnahme (or suffixhaufung-- the distinction being erased in headless phrases anyway), which would just be a language-by-language thing regarding which ones allow headless constructions and which don't... there's no need to posit an entirely new species of suffix-stacking, right? Many languages allow headless relative clauses, but we still just call them "relative clauses".

I think there may be a slight difference between headless suffixaufnahme and hypostasis. Suffixaufnahme is basically agreement with the head and this agreement can remain when the head is elided. Apparently "In all languages, if Suffixaufnahme occurs with head present, it also occurs with head missing." (p. 469) In languages with headed suffixaufnahme, it also makes sense to talk about headless suffixaufnahme. I'm not sure how headless constructions work in languages with suffixhäufung.

On the other hand, there are languages like Basque, Archi and Huallaga Quechua that only have this kind of double case marking in headless constructions. I think this is true of Moden Georgian as well, unlike Old Georgian. Although this looks the same as headless suffixaufnahme, I think there may be a slightly different mechanism behind it. Unlike suffixaufnahme, it's not really agreement, it's more of a strategy to derive a free-standing NP from an adnominal construction.

But some authors do treat hypostasis as a subtype of suffixaufnahme. Others maintain that "[h]ypostasis is a completely different phenomenon from agreeing case".
https://books.google.se/books?id=k_GtOx ... &q&f=false

[zompist]

Do you happen to have an example handy from Huallaga Quechua? In Ayacuchano and Cuzequeño, the only double case marking I know of is after the genitive, e.g.

tayta-pa-pi
father-gen + loc
at my father's

[Valdeut]

Do you happen to have an example handy from Huallaga Quechua? In Ayacuchano and Cuzequeño, the only double case marking I know of is after the genitive, e.g.

tayta-pa-pi
father-gen + loc
at my father's

That example looks very much like the type of hypostasis formation discussed above, which involve case marking after an adnominal case (so usually the genitive) when there is no explicit head.

I based the statement about Huallaga Quechua on Plank (1995) "Double Case" cited above (p. 469):
“In particular, while in the majority of Suffixaufnahme languages the construction occurs both with present and with elided heads, and while there are also languages—such as Archi (Kibrik) or perhaps Huallaga Quechua (McCreight 1988: 164)—where Suffixaufnahme occurs only as hypostasis, i.e. in elliptical constructions, there are no languages where Suffixaufnahme is restricted to headed constructions.”

McCreight 1988 is:
McCreight, Katherine L. 1988. Multiple Case Assignments. Ph.D. dissertation, Massachusetts Institute of Technology.

There are some examples from Huallaga Quechua throughout the book but they are not examples of hypostasis. Apparantly, Huallaga Quechua has more traditional suffixaufnahme when the possessor is separated from the possessed.

(p. 47)
“
Huallaga Quechua (Andean, Andean-Equatorial)
hipash-nin-ta kuya-: Hwan-pa-ta
daughter-3Poss-Acc love-1 Juan-Gen-Acc
'I love Juan's daughter' (only if possessor is separated from possessed;
otherwise No. 8:
Hwan-pa hipash-nin-ta kuya-:
Juan-Gen daughter-3Poss-Acc love-1)

While the primary relates to its secondary in terms of person, the secondary, identified as such by its genitive, indicates its primary in terms of the case encoding the primary's external relation (accusative in this example).”

(p. 86)
“Another moot question about Evenki and Even is whether secondaries in the comitative are adnominal, as the hosts of prototypical Suffixaufnahme are supposed to be. Comitatives in general are prone to vacillating between adnominal and clause-level status, not unlike secondaries in various local cases hosting Suffixaufnahme in Australia. They may also shade into coordinate conjunctions, on occasion producing what is the illusion of Suffixaufnahme rather than the real thing, as in Huallaga Quechua (Weber 1989: 218, 227, 348-350):

Hwan-ta-wan Pablu-ta rika-shka-:
Juan-Acc-Com Pablo-Acc see-Perf-1
'I saw Pablo and/with Juan'
”

(p. 89 – same example as p. 47)
“Prototypical in these respects, though in addition marking the primary for relatedness-indication, was Huallaga Quechua (74), where secondaries practice Aufnahme only when separated from primaries that they normally precede, thus conforming to pattern No. 9, X-y Y-sec-x, rather than No. 8, X-y Y-sec (Weber 1989).

a. Hwan-pa hipash-nin-ta kuya-:
Juan-Gen daughter-3Poss-Acc love-1
b. hipash-nin-ta kuya-: Hwan-pa-ta
daughter-3Poss-Acc love-1 Juan-Gen-Ace
'I love Juan'sdaughter'
“

[zompist]

Thanks for typing all that in. The genitive examples match the dialects I know.

The -wan example is new to me, but it may just not come up in the grammars I have— their -wan examples are in subject position.