[Lingtyp] Lingtyp Digest, Vol. 89, issue 2: Phonological differences of alienable vs. inalienabl, possession

[Sebastian Löbner]

Hi Martin,

thank for your replay. I feel that my contruibution you relate to is in 
need of a bit more of explanation. Thank you for pointing out sources of 
possible misunderstanding.

In the abstract of your 2017 article on explaining alienability 
contrasts, you say that “alienability contrasts in adnominal possessive 
constructions should not be explained by iconicity of distance, but by 
predictability due to the higher relative frequency of possessed 
occurrences of inalienable nouns.“ I agree, and I think your general 
standpoint is compatible with what I had in mind when I said there is a 
semantic explanation for the observed crosslinguistic tendency that 
alienable possessive constructions exhibit more complex or more salient 
expression. What I am arguing for in my theory of concept types and 
determination is this:
(1) Nouns (in fact CNPs in general) come with a lexically and/or 
compositionlly determined concept type either as sortal, individual, 
relational, or functional (a subclass of relational).
(2) Types of determination such as definite, demonstrative, indefinite 
etc. select for N/CNPs of certain concept types as input and render NPs 
of certain types as output; for example, simple indefinite 
determination, demonstratives, numerals, plural select for sortal nouns.
(3) The determination in an NP/DP is “congruent” if it is applied to a 
N/CNP of a concept type it selects: for example a relational noun with 
possessive determination presents a case of congruent determination;
(4) In principle, N/CPS of any conceptual type of N/CNP can be used with 
any type of determination; however, in the “incongruent” case of 
mismatch of N/CNP type and the input requirement of the determination 
applied to it (for example, a sortal noun with definite or possessive 
determination), the N/CNP is in need of some conceptual /shift/ or other 
in order to meet the input requirement of the respective type of 
determination. Therefore, incongruent determinations in general involve 
increased processing compared to congruent determination.


This predicts that
(a) There is an expectable crosslinguistic tendency of marking the 
incongruent cases more saliently in order to express the extra construal 
load; this need not mean a difference in terms of the numbers of 
morphemes, but may include phonetic differences such as vowel length or 
quality, or other means of contrasting; this is not the same as 
Ortmann’s iconicity claim you argue is questionable.
(b) The processing time for NPs with incongruent determination is 
longer, due to the execution of the additional shift.
(c) N/CNPs will occur more frequently with congruent than with 
incongruent determination because the lexical entries will be adjusted 
as to minimize semantic processing effort for the determination.

All these predictions turned out to be correct (see references in my 
2011 paper). CTD explains these tendencies merely as such; they derive 
from the functional principle of maximizing expressivity. The 
antagonistic principle of economy, along with probably quite a number of 
other factors, such as syntactic uniformity of NPs/DPs, explains why 
these are /only/ tendencies. Languages without possessive splits fit 
perfectly in this picture: there are so many conceptual shifts involved 
in semantic composition, for example metaphorical or metonymical shifts, 
which will not surface morphologically.
I do think that this theory provides a semantic explanation for the 
existence and form of determination splits such as alienability or 
definiteness splits. It only claims tendencies.

Sebastian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20220207/1416bcc7/attachment.htm>