Possession/modification by simple juxtaposition

[Besten, J.B. den]

I am having a little trouble with  Andy Spewncer's definition of what he is looking for:

"by maens of pure juxtaposition, without any other morphosyntactic marking (agreement, adpositions, case marking etc.)"  

Does "pure juxtaposition" mean: 'solely by means of juxtaposition'? -- If so Khoekhoegowab (Nama), which is AN without any special marking on either N or A) does not count as the type of language Andy is looking for since it is

DP - (DI 'possessive') -  N - pgn-marker - (A: 'dependent case')

However, if "pure" in "pure juxtaposition" only foreshadows "without any other morphosyntactic marking, etc." then - provided the optional presence of posessive DI does not hurt -- it is still unclear to me whether Khoekhoegowab may count as  the type of language required. On the one hand, Khg does not mark the possessor as GEN or DAT (in fact there are no such cases in Khg); on the other hand Khg  marks the possessor as NOM, which boils down to the absence of the audible case marker A. NOM is asssigned to non-inverted subjects, to possessors (except for 1Sg and 2Sg possessives, which are possessive pronouns) and to the objects of de-nominal postpositions. Therefore, possessive constriuctions in Khg look as follows:

DP-(*A) - (DI) - N-pgn-marker-(A)

Maybe the definition could be refined -- which may have consequences for the languages suggested up till now2.

Hans den Besten





-----Original Message-----
From: Discussion List for ALT [mailto:LINGTYP at LISTSERV.LINGUISTLIST.ORG] On Behalf Of Spencer, Andrew J
Sent: vrijdag 21 november 2008 12:02
To: LINGTYP at LISTSERV.LINGUISTLIST.ORG
Subject: Possession/modification by simple juxtaposition

We are interested in finding languages that express possession and adjectival modification in the same way, namely, by means of pure juxtaposition, without any other morphosyntactic marking (agreement, adpositions, case marking etc.). In other words, we're looking for languages with the following construction types (head-initial/head-final; the linear order doesn't matter to us and the language doesn't have to have a consistent head position):

good book:             book good // good book
the name of (the) boy: name [(the) boy] // (the) boy name
Gwen's book/mother:    book/mother Gwen// Gwen book/mother

A language which is close to what we're looking for is Spoken Welsh (head-initial):

good book:
llyfr   da     
book    good   
                       
the name of the boy:
enw   y    bachgen
name  the  boy

the name of a boy:
enw    bachgen
name   boy
                       
Gwen's book:   llyfr Gwen   
Gwen's mother: mam Gwen   

However, in Welsh, adjectives take the soft mutation when they modify FEM.SG nouns, so this isn't a 'pure' example of the language type we're looking for.

If you know of a language with these properties please contact Andrew Spencer: spena at essex.ac.uk.
We'll put together a synopsis of the replies. (You don't need to reply to the whole list.)

Andrew Spencer, University of Essex
Irina Nikolaeva, School of Oriental and African Studies