Freitag, 12.00 Uhr
Recent discussion of syntactic change treating the development from Old High Germn (OHG) to Modern High German (MHG) was somehow modified by Tomaselli (1995) who focussed on the V3 phenomenon. V3 was lost in MHG, and even in OHG it was not as common as the well known pattern V1,V2 or V-last. However, there are well attested cases of V3 in OHG which might be interesting for the understanding of language change.
Tomaselli notes, following Lippert (1974:15) that "the finite verb may shift to third position given the presence of a pronominal element in 2nd position." (Tomaselli 1995:346) This is demonstrated in (1):
Dhes martyrunga endi dodh uuir findemes mit urchundin dhes heilegin
chiscribes.
His martyrdom and death we demonstrate with evidence of the Holy Writing
(Lippert 1974:52)
Contrary to traditonal analyses which assume a minimal sentence structure of only CP and VP, Tomaselli proposes a "head-medial IP" between CP and VP whose head I° hosts the finite verb with the adjoint subject clitic.
I don’t want to argue against the analysis of medial IP. My point is, that there are cases of V3 which can’t be explained by only the assumption of an IP, as there are V3 clauses initiated by a full subject - DP (2) or by adverbials (3).
[Christus auur [sus quham fona fater ziuuaare so selp so...
Christus but so comes from father for sure so self so
(Isidor 126f)
[Bidhiu huuanda [dhaz ziuuaare ist ubarhepfendi angilofirstandan ioh
iro chiuuizs,...
because of this that really is jumping over the angels brain and their knowledge
(Isidor 117f)
At least in (2), the inversion [sus quham] gives us evidence for a CP-structure. Thus, it can be concluded, that there must be at least one more position in front of CP.
In order to find out about the characteristics of this position, one can turn to recent findings in synchronic theory about The Fine structure of the Left Periphery , to use Rizzi’s terminology (1995). There is for example the work of Mueller & Sternefeld (1993) and Haftka (1994) who both propose the sequence CP - TopP for Modern German, while E. Hoekstra (1993) assumes a tri-partition of CP into WhP - TopP - AgrP for Modern Dutch. All these proposals argue by using facts from embedded clauses, but the resulting sequence CP -TopP doesn’t combine with the findings in the main clauses, e.g. it’s not convincing to analyse in (2) the subject-DP Christus as a constituent in [Spec, CP] followed by a topic sus (so).
In order to explain the syntactic complexity in OHG, I will assume a sentence initial subspecified Topic-Phrase. The minimal clausal structure then would be TopP - IP - VP, where IP is the place for the subject clitic and the finite verb as in Tomaselli’s analysis and where TP hosts the finite verb in topicalization structures which are well attested in OHG. I claim that TopP is a recursive projection (serving as starting point of the grammaticalization process), so that we yield the sequence of TopP - TopP - IP - VP which I take to be the representation of (2).
TopP is considered to be the entrance of new relevant discourse material, and thus a typical locus of reanalysis and grammaticalization. At the workshop, I will try to show in more detail how it interacts with the already existing grammatical CP (or WhP) and how new structures, e.g. some adverbial subordinate clauses, might be established within the system of OHG.
zum Programm der AG 6
zur alphabetischen Übersicht der Abstracts
zur zeitlichen Übersicht