The Syntax and Acquisition of German PATH
(2015) 23rd Conference of the Student Organization of Linguistics in Europe (ConSOLE XXIII) — Location: Paris Diderot- Paris 7 (7.January.2015)
Files
GermanPath-Console.pdf
Restricted Access - Adobe PDF
- 1.35 MB
Details
- Authors
- Abstract
- Romance and Germanic languages are widely considered to typologically differ in the expression of motion (Talmy, 1975, 1985, 2000), as Romance languages express the Path of motion predominantly in the verb root, whereas Germanic languages do so by the means of ‘satellites’ (e.g. prepositions, particles, adverbs), while the verb root conflates manner of motion. Hence, some German prepositions are able to express directionality, whereas French prepositions do not, as path is expressed by the verb root. This is illustrated in (1) –(2) for French and German, the languages under scrutiny in this study. (1) Antoine entre dans la maison (en courant). (2) Anton läuft in den Wald. A. enters in the house (running) A. runs into the-ACC wood The linguistic description of motion events has been the topic of much debate in recent years, leading to various accounts in different frameworks (s. Asbury et al. 2008 for an overview), with some aiming to provide an explanation for the above-mentioned differences between Romance and Germanic languages (a.o. Puigdollers 2010; Zubizarreta &Oh 2007), however to this day none has been able to provide a unified comparative analysis that also delivers a satisfactory account for German PATH expressions which can take a number of different shapes, making use of a wide array of structures such as prepositions, postpositional and circumpositional elements as represented in (3)-(5), and the phenomenon of German ‘case-changing prepositions as contrasted in (2) vs. (6). (3) Anton läuft in den Wald hinein. > circumposition: postposition = copy of base preposition A. runs in the wood-ACC DEIX-into (4) Tim stieg auf das Dach hinunter. > circumposition: postposition ≠base preposition Tim climbed onto the roof-ACC DEIX-down (5) Hans fällt die Treppe hinunter. > postposition Hans falls the stairs-ACC DEIX-down (6) Anton läuft in dem Wald. Anton runs in the wood-DAT The aim of this paper is threefold; first the differences between French and German in the realisation of PATH will be linked to a difference in the syntactic structure of the PP. The second claim is that there is no such thing as polysemy in spatial prepositions. In German, where a class of prepositions seems to be able to assign either dative, as in (6), or accusative, as in (2), case in correlation with a locative (6) or directional (2) meaning, we argue that the dative case assigned by the preposition is the default (and the only assigned prepositional) case, while the accusative case is the by-product of syntactic P-to-V incorporation, an operation that is not possible in French as path is constrained to the verbal root and, as such, is part of a morphological unit within the verb and not the consequence of a productive morpho-syntactic incorporation process. Finally, we will show that our theoretical claims are supported by empirical evidence from second language acquisition of German path by native speakers of French. We adopt Real Puigdoller’s (2010) Extended Structural Ambiguity Hypothesis (ESAH) (based on Gehrke 2008), which states that ambiguity between a directional and a locative meaning is due to structural configurations and not lexical representations of the preposition, the latter being thus only locative. This applies to French prepositions and explains the difference between the sentences (7) and (8), in which only (7) can be interpreted as directional due to the verbal semantics which conflates manner-of-motion but also an inherent path, (8) is only locative: (7) La petite fille court dans la maison. (8) Jean tombe dans la piscine. The little girl runs in the house J. falls in/into the pool As mentioned above, polysemy does not exist either in the case of directional/locative prepositions in German. Those prepositions that are traditionally considered as ‘case changing prepositions’ or ‘Wechselpräpositionen’, do not actually change case assignment (e.g. (2)vs.(6)) and with it their meaning (see discussion below), but are inherently dative assigning, locative prepositions (see Abraham, 2001 and van Riemsdijk, 2007, 2012 for similar conclusions). In light of these observations, we claim that the differences in the realisation of path are due to different syntactic structures in French and in German. Assuming, in accordance with standard reasoning, that parametric variation follows from different properties of functional categories, we postulate the following complex syntactic structure for German PPs (9), which allows the accommodation of the variety of prepositional and postpositional German P-elements (for a similar analysis see Gehrke (2008) and Jackendoff (1983; 1990), who argue for an internal hierarchy within spatial PPs in which Path dominates Place and van Riemsdijk (1990), who stipulates a head-final pP in German for postpositional elements). The structure for French on the other hand, is a reduced one, giving rise to the locative character of its prepositions; the verb root being the only possibility of expressing path, no decomposition is needed. Thus the French PP contains just the PLACEP (10). (9) German PP: [pP [PathP P [PlaceP in [DP den Wald ]]]] (10) French PP: [PlaceP dans [DP la maison ]] Concerning German and its seemingly ‘case-changing prepositions’, as mentioned above, the dative is the default-case, we claim that it is the result of P-to-V incorporation in the sense of Baker (1988). The so-called ‘case-changing’ preposition is base generated in PLACEP and incorporates into PATHP. The operation is triggered by an affix feature. From there, according to whether the construction contains one prepositional root or two different ones, there are two ways for the base preposition to incorporate into V: (i) the base preposition incorporates overtly- as in (2), (3) and (5), through the postpositional p-head (followed by excorporation of V), while in the case of (3), leaving a phonetically realised reduplicated twin behind in its base position, or (ii) the base preposition itself incorporates covertly through Spec pP and into Spec VP as in (4). In this case, the base preposition is not phonetically realised in the incorporated position but in its initial position, like auf in (4). In both cases the preposition in pP combines with the deictic adverb hin-/her- in an intermediate step. As a side effect of the incorporation operation, the preposition loses its ability to assign oblique case (dative), and consequently accusative is assigned by v in (2)-(5). Finally, we will provide empirical evidence for this analysis from adult L2/L3 acquisition data illustrating the difference between French and German PATH expressions concerning the (complex) structure of the PP, dative as default case and incorporation. Indeed, a narrative elicitation task carried out with 104 learners of German of various levels (A1 to C in the CEFR) proved that a PLACE-semantics is privileged by learners in the early stages of acquisition, A1 thus significantly differing from the other learner groups (p< 0.01) and the German native speakers (p< 0.01). With advancing proficiency the use of PLACE diminishes and reaches target-language-like rates at level C. By contrast, the exact opposite phenomenon can be observed in the use of PATH structures, which are shunned in the early stages of acquisition in favour of simple bare motion or PLACE structures. However their use increases and eventually the learners come close to but do not entirely reach native-like levels. Concerning the choice of case when using a traditionally so-called ‘case-changing preposition’, learners also behaved differently compared to natives and used significantly (p<0.01) less accusative, especially in the lower proficiency levels. Instead, learners with a background in Romance languages at level A1 and A2 tended to favour the dative case. Finally, learners also showed little to no use of the complex PATH-structures illustrated in (3)-(5), thus differing once again from the natives (p<0.01). The number of occurrences once again goes up with the level of proficiency, however the learners never reach the quantitative or qualitative variety observed in the native data. These observations from acquisitional data provide further evidence for the claims made above, e.g. a difference between French and German in the structure of PPs (simple in French vs. complex for German) on the one hand; the dative case as default- or “only”-case for prepositions associated with spatial meanings on the other. The learners transfer their reduced French PP structure into their learner interlanguage and at the initial stages of acquisition, struggle with the expression of PATH, thus compensating with an overuse of PLACE. The dative case follows from the reduced structure in the learner interlanguage, as PLACE cannot yet incorporate into PATH, leaving the preposition stranded in the PLACEP and unable to incorporate into V to ensure the assignation of accusative. Bibliography: Abraham, W. (2001). Gibt es im Deutschen eine Klasse von Präpositionen mit Doppelrektion? Deutsche Sprache 29 - Asbury, A. et al. (eds.) (2008). Syntax and Semantics of Spatial P. Amsterdam, (a.o.): John Benjamins - Baker, M. C. (1988). Incorporation: A Theory of Grammatical Function Changing. Chicago: U. of Chicago Press - Gehrke, B. (2008). Ps in Motion: On the Semantics and Syntax of P elements and motion events. LOT dissertation series 184 - Jackendoff, R. (1990): Semantic Structures. Cambridge MA.: The MIT Press - Real Puigdollers, C. (2010). A microparametric approach on goal of motion constructions: properties of adpositional systems in Romance and Germanic. Catalan Journal of Linguistics 9 - Talmy, L. (2000). Towards a Cognitive Semantics, Vol 2. Cambridge, MA.: MIT Press - van Riemsdijk, H. (1990). Functional Prepositions. In: Pinkster, H.& I. Greene (eds.). - van Riemsdijk, H. (2012). Discerning Default Datives: Some Properties of the Dative Case in German. In: Grewendorf et al. (eds.).
- Affiliations
Citations
Scheirs, V. (2015). The Syntax and Acquisition of German PATH. 23rd Conference of the Student Organization of Linguistics in Europe (ConSOLE XXIII), Paris Diderot- Paris 7. https://hdl.handle.net/2078.5/40498