[Lingtyp] query: Metaphoricity and Agreement in Genitive Constructions

[David Gil]

Dear all,

I am interested in testing a hypothesis regarding a correlation between 
the respective directionalities of metaphoricity and agreement.

Let X Y be a construction in which Y is the source of a metaphor (the 
metaphorical description) and X its metaphorical target (the thing being 
described by the source).

Hypothesis:

IF a relationship of morphological agreement obtains between X and Y, 
THEN X is the controller of agreement and Y is its target (but not vice 
versa).

(Terminological note: it is inconvenient that both theories of metaphor 
and theories of agreement use the same term "target".It should be kept 
in mind that there is no connection between the two usages of the term - 
in fact, the hypothesis suggests that the two usages fall on opposite 
sides of the correlation.)

Commonplace examples upholding the correlation are cases of an NP in 
construction with an adjectival or verbal predicate, as in (1), and a N 
in construction with an adjectival or verbal attribute, as in (2).(The 
examples are in French so as to illustrate the agreement.)

(1) Ton idée est verte

(2) Une idée verte

However, attributive genitive metaphors, as in (3), pose a potential 
challenge to the hypothesis.

(3) Heart of stone

Cross-linguistically, in languages where there is agreement in genitive 
constructions, it is the possessor (or G) that controls the agreement 
and the possessum (or N) that is its target (the so-called 
"head-marking" pattern).Accordingly, in such languages, in the 
equivalent of, say, (3), the metaphorical target 'heart' would also be 
the agreement target', in violation of the proposed hypothesis.

My query therefore is:are there languages with agreement in genitive 
constructions in which metaphorical interpretations are available in 
such constructions (in violation of the hypothesis)?

Hebrew provides prima facie reason to suspect that there may not be any 
such counterexamples.Hebrew has two genitive constructions, the first, 
as in (4/5a), without agreement, the second, as in (4/5b), with agreement:

(4)(a)Halev šel moše

DEF:heart of Moses

(b)Libo šel moše

heart.CONSTR:3SGM of Moses

(5)(a)Lev šel even

heart of stone

(b)Libo šel even

heart.CONSTR:3SGM of stone

While (4a/b) are interpreted literally, (5a) has a metaphorical 
interpretation.Crucially, though, in (5b), the metaphorical 
interpretation is no longer available — the construction doesn't make 
sense.What seems to be happening here is that the agreement in (5b) is 
preventing the metaphorical interpretation from occurring, and thereby 
providing seemingly strong support for the hypothesis.

But Hebrew is just one language.So I'd be interested in knowing whether 
similar facts hold cross-linguistically.A counterexample to my 
hypothesis would be a language that allows an agreeing genitive 
construction such as in (5b) to bear a metaphorical interpretation:Is 
anybody familiar with such cases?Of course, I would also greatly 
appreciate examples of languages with agreeing genitives that do not 
allow them to bear metaphorical interpretations, as these would provide 
additional support for the hypothesis.

Thanks,

David

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20190916/51974e1d/attachment.htm>