EALing Paris fall School 2006 course description


Peter Svenonius

CASTL, University of Tromsø

Anatomy of the Category P
(Preposition, Postposition, Particle)

E-mail: Peter.Svenonius@hum.uit.no
Project web site: www.hum.uit.no/MRA/
Personal web page: www.hum.uit.no/a/svenonius/

Postal Address:
CASTL/Department of Languages and Linguistics
Faculty of Humanities
University of Tromsø
N-9037 Tromsø
Norway

Course Description

This is a cross-linguistic investigation into locative and directional expressions, with a focus on their relationship to adpositions (prepositions and postpositions, the category P). "Local" case systems like those of Finnish and Hungarian will be examined, as will languages which make extensive use of relational nouns to express locative concepts. The aim of the course will be to develop some sense of the range and limits of cross-linguistic variation in this domain, with an eye toward characterizing the nature of universals, whether they are syntactically autonomous or cognitively grounded.

Day-to-day Program
(click on lecture title for handout)

Monday: Introduction

Cross-linguistic expressions of location; the universality of the category P In approaching an unfamiliar language, it is difficult to ask the question whether it has lexical items of the category P before we have a cross-linguistically valid definition for P. At the same time, it is difficult to develop a cross-linguistically familiar definition for P if we have not examined a lot of unfamiliar languages. Therefore, a methodologically useful initial approach is to look at how languages express the notions expressed using Ps in languages for which the category has been clearly identified, such as English. When we do this, we find first of all that adpositions are common cross-linguistically; that in some languages, many adpositions are nounlike, derived from nouns, or difficult to distinguish from nouns; and that in other languages, many adpositions are verblike. Furthermore, we find that many languages have complex local case systems which function like adpositions, and that other languages have applicative morphemes which can be attached to the verb to introduce arguments which satisfy the kinds of functions that are satisfied by PPs in other languages. Thirdly, we find that some languages (in particular those with local case systems or applicatives) have been described as having very few or even no adpositions.

Tuesday: The Extended Projection of P

Cross-linguistically, adpositional expressions can be divided into those expressing Place, or static location, and those expressing Path, a change in location over time (occasionally, languages distinguish location of states from location of activities, but in this coarser description these would be subtypes of Place). It can be shown that Path dominates Place in the syntax, that Path is further from the Ground DP (“Landmark,” the complement of P) in the morphology, and that Path is compositionally composed from a Place expression rather than the other way around. These three facts reduce to a single one if a unified morphology-syntax-semantics representation is adopted.
 

Wednesday: The Emergence of AxPart

Many languages have a set of about a half-dozen to a dozen and a half locative expressions, often called relational or locative nouns, which typically translate into front, back, top, bottom, side, and so on. These I call Axial Parts, following Jackendoff (who in turn builds on Marr). I argue that they represent a distinct functional category below Place, and show how they contribute to the interpretation of locative expressions.

Thursday: Licensing Directional Interpretations

As noted by Talmy and Carter, not all languages are like English in freely allowing Path expressions to be combined with manner of motion verbs to express directed motion. Languages like French, Spanish, Japanese, and Korean are much more restrictive in licensing directional interpretations. I analyze these restrictions using the extended projection of P motivated in the previous lectures.