Compound Case Markers in Australian Languages

Abstract

In several Australian languages, it is possible for nominals to carry more than one inflectional case marker. This can be due to adnominal multiple case where two or more cases are assigned to a nominal. This type has been known as Suffixaufnahme. The recent book Double Case (Plank 1995) gives a good survey of this topic (including data from Australian languages). A further possibility is derivational multiple case (compound cases). Here a case marker forms an oblique stem (founding form) that may attract further case markers. The use of a ligative (ease spacing) can be seen as an interesting mixture of (adnominal) double case and compound case. This paper presents the results of a pilot study that includes several languages from the Pama-Nyungan and the Tangkic family. 1. INTRODUCTION. A good survey of multiple case in Australian languages is given in Dench and Evans 1988. Following a classification proposed in Austin 1995, three situations can be distinguished where multiple case may occur. The present paper is a pilot study of a special type of multiple case marking, namely, compound cases. In many Australian languages, a certain [CASE.sub.1] that the nominal X already be marked for [CASE.sub.2], so that the pattern X-[CASE.sub.2]-[CASE.sub.1] is found. We give an example from Kalkatungu. Here we find matyumpa 'kangaroo', matyumpathu 'kangaroo + ERG', and matyumpathu[LANGUAGE CHARACTER NOT REPRODUCIBLE IN ASCII]u 'kangaroo + CAUS'. The causal suffix -[LANGUAGE CHARACTER NOT REPRODUCIBLE IN ASCII]u requires that the noun be marked by the ergative suffix -thu. The most common situations seem to be the following (more details are given in section 4): (1) The outer [CASE.sub.1] is a local case. (2) The inner [CASE.sub.2] is a local case, or a case genitive/dative, or an oblique case. (3) In some languages th ere is a split for only some nominals does [CASE.sub.1] require [CASE.sub.2], while for others it does not. (4) For pronouns it seems to be the rule that several cases require an oblique stem that is related to the genitive or dative form. In section 2, a brief account is given on the three major subtypes of multiple case marking: derivational, adnominal, and referential. Some first examples of derivational case (= compound cases) are included. The following case-marking patterns within the noun phrase may be distinguished. A language is called phrase-marking if the case marker is attached to some word of the noun phrase. There are three subtypes: final-marking (the last word of the NP is marked for case), head-marking (the head of the NP is marked for case), and free marking (any word of the NP may be marked for case). A language is called word-marking if all words in the NP are marked for case. The use of a ligative (case spacing in Dench and Evans 1988) is mentioned as occurring in each language in the sample. It is noted that, in the languages in the sample, pronouns distinguish two or three case forms to express the core relations S, A, and O, and typically use an oblique stem as a founding form for other cases. To mak e the appearance of compound cases less exotic, examples from Estonian and Slovak are discussed briefly. No attempt is made to extend this study further beyond Australian languages, although compound cases are to be found in Uralic, Caucasian, Dravidian, and other languages. Section 3 is the main part of the present study. The result of a search based on a sample of 20 Australian languages is presented. These languages are Anguthimri, Diyari, Djabugay, Djapu, Gumbaynggir, Guugu Yimidhirr, Kalkatungu, Madhimadhi, Margany/Gunya, Mbabaram, Ngiyambaa, Nyawaygi, Panyjima, Pitta-Pitta, Uradhi, Wargamay, Watjarri, Yankunytjatjara, Yaygir, and Yukulta The location of the languages reveals some regional features, but further investigation is needed to reach conclusions on areal diffusion or genetic relationship. …