Participles and the General Valency Hypotheses

[Andreas Nolda]

Dear colleagues,

unfortunately, I overlooked Oliver Teuber's (2005: 126-129; passim) 
analysis of adnominally used participles in German, which is also 
couched in a variant of the IL framework. His analysis presupposes a 
verb semantics that is non-standard in IL: the meaning of a verb such 
as "lesen" ('to read') is a relation between a situation s, an
'agent' x1, a 'patient' x2, and the situation's time t. (If assumed at 
all, the General Valency Hypothesis would have to be adapted 
accordingly.) The meaning of the corresponding participle "gelesen" 
('read') is then a relation between the 'patient' x2 and a time t2 
that is the time of the state resulting from s (the 'Nachzustand');
"s" itself is bound by an existential quantifier in the scope of the 
lambda operator binding "x2" and "t2".

As far as I can see, this analysis runs into similar problems as 
Marie-Hélène Viguier's analysis, carried over to German. In both 
cases, the situation or event variable is existentially bound, such 
that the interpretation of temporal, locative, or manner modifiers 
cannot be applied to it in a compositional way. (Note that the time 
expressed by temporal modifiers of participles is the time of the 
situation or event expressed by the corresponding verb: _der gestern 
gelesene Brief_ means 'the letter that was read yesterday' and 
not 'the letter that was yesterday in the state resulting from being 
read').

Best,

Andreas Nolda

References
Teuber, Oliver (2005). "Analytische Verbformen im Deutschen: Syntax
   -- Semantik -- Grammatikalisierung". Germanistische Linguistik:
   Monographien 18. Hildesheim: Olms.
-- 
Dr. Andreas Nolda           http://www.linguistik.hu-berlin.de/~nolda/

Humboldt-Universität zu Berlin
Philosophische Fakultät II
Institut für deutsche Sprache und Linguistik