Contact-induced language change and remnants of the former system – the verbal agreement of Koĩc (English: Sunuwar; Tibeto-Burman) Dörte Borchers, Karl-Franzens-University, Graz, Austria

Koĩc ([kɔĩts]; English: Sunwar; Nepali: सनु वु ार Sunuvār) is a Kiranti language spoken in eastern Nepal. With about 38 000 speakers, Koĩc is one the bigger languages of the 123 languages of Nepal.1 Koĩc once had a biactantial agreement system with transitive verbs agreeing with agents and patients (Carol Genetti 1988), a typical feature of Kiranti languages. In modern Koĩc, transitive verbs agree with the agent but not with the patient. As most speakers of Koĩc are bilingual with Nepal's official language Nepali (Indo-Iranian), changes in the grammatical system of Koĩc that result in structures similar to those of Nepali are likely to have been caused by language contact. Besides the mentioned loss of verbal object agreement, the loss of marking dual number in nouns is another example of such a contact induced change. The latter change occurs increasingly in places where frequent language switching between Koĩc and Nepali is the rule. A comparison of modern Koĩc verbal paradigms and the obsolete verbal paradigms with biactantial agreement collected by Genetti (1988) shows rather regular similarities in form and function between old and new person and number markers. Such a comparison shows in addition the retention of a structural difference in reference to a first person singular agent as opposed to reference to non-first person singular speech act participants. In the equivalent Nepali paradigms, there is no such difference between indicating first person agents or other agents. A brief overview of changes in the Koĩc language from a biactantial to a monoactantial agreement marking system will be followed by a presentation of the morphological and semantic continuities between the older and newer person marking system of Koĩc. Finally, possible reasons for the special status of the marking of a first person singular agent will be discussed.

References Borchers, Dörte 2008. A Grammar of Sunwar. Descriptive grammar, paradigms, texts and glossary (Languages of the Greater Himalayan Region, Volume 7). Leiden: Brill Central Bureau of Statistics 2012. National Population and Housing Census 2011 (National Report). Volume 01, NPHC 2011. Kathmandu: Government of Nepal, National Planning Commission Secretariat (https://unstats.un.org/unsd/demographic/sources/census/wphc/Nepal/Nepal- Census-2011-Vol1.pdf; accessed 21 March 2017) Commissioner for Linguistic Minorities 2011. 48th Report of the Commissioner for Linguistic Minorities in India (July 2010 to June 2011). Allahabad: Ministry of Minority Affairs, Government of India (http://nclm.nic.in/shared/linkimages/nclm48threport.pdf; accessed 14. April 2017) Genetti, Carol 1988. Notes on the structure of the Sunwari transitive verb, in: Linguistics of the Tibeto-Burman Area11.2; 62-92

1 The figures for languages of Nepal and speakers of Koĩc come from the Census of 2011 (Central Bureau of Statistics 2012). Gebrauch und Dynamik der Regionalsprache in Ost-Lothringen

Mit dem Ende des Mittelalters waren die Sprachgrenzen in Europa konsolidiert und haben sich seitdem nur wenig verändert. Indessen hielten Annexionsbewegungen, Staatenbildung und Grenzverschiebungen von Staaten bis ins 20. Jahrhundert an. Dabei orientierten sich die neuen Grenzziehungen nun nicht an sprachlichen Räumen, sondern durchquerten und zerteilten sie vielmehr. Dies betrifft das Aufeinandertreffen von Germania und Romania in besonderem Maße. So verläuft die Staatsgrenze zwischen Deutschland und dem nordöstlichen Frankreich seit 1957 (wieder) südlich von Merzig – Saabrücken – Zweibrücken – Landau (Pfalz). Bezogen auf die sprachlich- dialektale Ebene liegen somit große Teile des Mosel- und Reinfränkischen im deutschen Staatsgebiet, während sie in Ost-Lothringen Teil des französischen Diasystems sind. In den letzten 150 Jahren hat dieses Gebiet vier Herrschaftswechsel erlebt. Frühere Maßnahmen zur Französisierung Ost- Lothringens zeigten nur geringe Effekte. Jedoch haben die Ereignisse v.a. des 2. Weltkriegs ein kollektives Trauma bei der Bevölkerung ausgelöst, das eine eigenständige Identität auf der Basis ihres germanophonen Dialekts massiv erschwerte. In der Folge wurde die Förderung des Französischen mit wenig Widerstand hingenommen und in vielen Familien die Weitergabe des Lothringer Platt aufgegeben (Dorner 2012). Im Rahmen eines Projekts am Institut für Deutsche Sprache werden zur Zeit umfassende Datenerhebungen durchgeführt, die den Gebrauch der Regionalsprache und seine soziosituativen Bedingungen in einem spezifischen mehrsprachigen Kontext dokumentieren. Eine solche Dokumentation erscheint angesichts der bedrohten Lage umso dringlicher. Die Daten umfassen neben dialektalen Übersetzungstests (Wenkersätze) und stdd. Vorlesetests („Nordwind und Sonne“) auch freie Rede in sprachbiographischen Interviews und in Freundesgesprächen. Vor allem letzteres Setting erlaubt Einblicke auf das Nebeneinander von Lothringer Platt und Französisch. Die Erhebungen im gesamten germanophonen Gebiet Ost-Lothringen sowie von mehreren Generationen ermöglichen zudem Analysen der diatopischen und intergenerationellen Variation. Im Beitrag sollen erste Ergebnisse der Analysen vorgestellt werden. Dabei zeigt sich ein heterogenes Bild bezüglich der intergenerationellen Weitergabe und dem Erreichen von Sprachlagen im standardnahen Bereich, das bisweilen mit der Wahrnehmung der Staatsgrenze als Handlungsgrenze in Zusammenhang stehen.

Dorner, Daniela (2012) „Social actors and the language policy and planning process: A case study from German-speaking Lorraine (France).“ In: Studer, Patrick/Werlen, Iwar (Hrsg.) Linguistic Diversity in Europe. Current trends and Discourses. Berlin/Boston: de Gruyter: 157-175.

Intergenerationelle Weitergabe der Regionalsprache in Ost‐Lothringen

Mit dem Ende des Mittelalters waren die Sprachgrenzen in Europa konsolidiert und haben sich seitdem nur wenig verändert. Indessen hielten Annexionsbewegungen, Staatenbildung und Grenzverschiebungen von Staaten bis ins 20. Jahrhundert an. Dabei orientierten sich die neuen Grenzziehungen nun nicht an sprachlichen Räumen, sondern durchquerten und zerteilten sie vielmehr. Dies betrifft das Aufeinandertreffen von Germania und Romania in besonderem Maße. So verläuft die Staatsgrenze zwischen Deutschland und dem nordöstlichen Frankreich seit 1957 (wieder) südlich von Merzig – Saabrücken – Zweibrücken – Landau (Pfalz). Bezogen auf die sprachlich‐dialektale Ebene liegen somit große Teile des Mosel‐ und Reinfränkischen im deutschen Staatsgebiet, während sie in Ost‐Lothringen Teil des französischen Diasystems sind. In den letzten 150 Jahren hat dieses Gebiet vier Herrschaftswechsel erlebt. Frühere Maßnahmen zur Französisierung Ost‐Lothringens zeigten nur geringe Effekte. Jedoch haben die Ereignisse v.a. des 2. Weltkriegs ein kollektives Trauma bei der Bevölkerung ausgelöst, das eine eigenständige Identität auf der Basis ihres germanophonen Dialekts massiv erschwerte. In der Folge wurde die Förderung des Französischen mit wenig Widerstand hingenommen und in vielen Familien die Weitergabe des Lothringer Platt aufgegeben (Dorner 2012). Im Rahmen eines Projekts am Leibniz‐Institut für Deutsche Sprache werden zur Zeit umfassende Datenerhebungen durchgeführt, die den Gebrauch der Regionalsprache und seine soziosituativen Bedingungen in ihrem spezifischen mehrsprachigen Kontext dokumentieren. Neben dialektalen Übersetzungstests und standarddeutschen Vorlesetests umfassen die Daten auch freie Rede in Freundesgesprächen und in sprachbiographischen Interviews. Die Erhebungen im gesamten germanophonen Gebiet Ost‐Lothringen sowie von mehreren Generationen ermöglichen zudem Analysen der diatopischen und intergenerationellen Variation. Bisher liegen über 90 Stunden Datenmaterial von 53 Informantinnen und Informanten vor. Im Beitrag sollen erste Ergebnisse der Inhaltsanalyse der Interviews vorgestellt werden. Dabei ergibt sich folgendes Bild: Die Sprecher der Geburtenjahrgänge 1928‐1960 hatten zwar größtenteils Französisch als Unterrichtssprache in der Schule, wuchsen aber in einem dialektsprachigen Elternhaus auf. In den folgenden Generationen (*1961‐1996) wird der Faktor "Region" relevant: Während im Osten der autochthone Dialekt noch recht verbreitet war, wuchs im Westen kaum jemand mit ihm auf.

Dorner, Daniela (2012) „Social actors and the language policy and planning process: A case study from German‐speaking Lorraine (France).“ In: Studer, Patrick/Werlen, Iwar (Hrsg.) Linguistic Diversity in Europe. Current trends and Discourses. Berlin/Boston: de Gruyter: 157‐175. III. Vielfaltslinguistik (Köln) – Abstract Paul Compensis, Universität zu Köln

Clitic doubling in Molise Slavic. A corpus-based classification.

Molise Slavic is a Slavic micro-language in Italy, closely related to istrian-ikavian dialects. Although belong to the Western South Slavic branch, Molise Slavic exhibits the syntactic construction clitic doubling (CD). CD (sometimes called object doubling) is a construction in which a direct or indirect object is doubled with a clitic form of a pronoun. This construction type is well-known in languages of the Balkan sprachbund (Albanian, Bulgarian, Greek, Macedonian and Romanian) and in some Romance varieties (Spanish, Southern Italian etc.). In some languages, CD might be an obligatory process, dependent on definiteness (Macedonian), topicality (Bulgarian) or else, while in other languages it can be a facultative construction for emphasis (cf. Tomić 2008). Since a common genetic source can be excluded and it cannot be explained by Balkan linguistic convergence either, it is more likely a contact phenomenon. The first systematic account of Molise Slavic already mentions frequent (but optional) instances of clitic doubling (cf. Rešetar 1911: 233). In his extensive work on Molise Slavic, Breu highlights that the construction perfectly replicates the Italian or Southern Italian dialectal model, especially in the case of fronted objects/clitic topicalization (1). However, instances can be found in which the object does not occur fronted in the sentence (2). Also, it seems that some examples traditionally interpreted as clitic doubling are indeed instances of clitic left dislocation (3):

(1) […] ka tvoju nevistu ju ne moram či nišča […]. [Clitic topicalization] […] REL your-ACC.SG.F daughter-in-law-ACC.SG.F she.ACC NEG can-PRS.1SG make-INF nothing ‘[…] that I cannot do anything to your daughter-in-law […]’. (Breu, Piccoli 2011: [2.5.1]-17)

(2) Bihu bolina […] oni ti ga ne artiraša bolin […]. [Clitic doubling proper] be-IPRF.1SG ticket-NOM.PL DEM.DIST-NOM.SG.M you-DAT.SG he.ACC.SG NEG take-away- IPRF.3SG ticket-ACC.SG.M ‘There were these (meal) tickets […] he did not take the ticket away from you.’ (Breu 2011: [1.2.1]-301)

(3) Mèdz kvindal, ti ga mečahu one zgora pleči. [Clitic left dislocation] Half quintal, you-DAT.SG it.ACC put-IPRF-1SG they-NOM.PL on shoulder-GEN.SG.M ‘Half a quintal, they put it on your shoulder.’ (Breu, Piccoli 2011: [1.2.1]-275)

This raises the question which underlying mechanism enforces the doubling of an object. So far, no systematic account of CD in Molise Slavic exists. Based on a sub-corpus extracted from Breu, Piccoli 2011 (recordings of spoken language; n(tokens)=60.219; n(types) = 7.023), I investige instances of the phenomenon in order to identify its function and role in the language system. In this study, I present a first classification of the phenomenon in Molise Slavic along three dimensions: I investigate if there are functional differences (a) between clitic topicalization (1), clitic doubling proper (2) and clitic left dislocation (3), (b) between doubling of direct objects and indirect objects and (c) between the doubling of a noun or a full-form pronoun. Based on the material, I examine if every sub-type of clitic doubling is motivated by the same mechanism (topicalization, prominence features or core syntax). The classification is especially relevant in regard of the question of optionality vs. obligatoriness and if constraints on the use of CD exist. This study draws on typological insights from clitic doubling research in other languages. The discussion of emergence of the phenomenon (internal vs. external) is not addressed in the talk.

Breu, Walter, and Giovanni Piccoli. 2011. Südslavisch unter romanischem Dach: Die Moliseslaven in Geschichte und Gegenwart im Spiegel ihrer Sprache. Teil I. Texte gesprochener Sprache aus Acquaviva Collecroce. München. Rešetar, Milan. 1911. Die serbokroatischen Kolonien Süditaliens: Mit 22 Abbildungen im Text. Wien. Tomić, Olga. 2008. Variation in clitic-doubling in South Slavic. In: Microvariation in syntactic doubling. Ed. Sjef Barbiers, Olaf Koeneman, Marika Lekakou and Margreet Ham. Leiden. 443-468.

An empirical study on universals of nominal classification systems

Nominal classification systems are traditionally divided into gender systems and several classifier systems. Based on this tradition, linguists provide descriptions (Allan 1977; Corbett 1991; Aikhenvald 2000; Senft 2001; Grinevald 2002; Kilarski 2012; Di Garbo 2014; Kramer 2015) and suggest universals (Greenberg 1966) about the nominal classification systems. However, more and more linguists agree nowadays on the view that there are no crucial differences between gender and classifier systems (Contini & Kilarski 2013; Corbett & Fedden 2015; Singer 2016; Passer 2016; Fedden & Corbett 2017; Seifart 2018). I report here an examination of universals about nominal classification systems with a database of 200 languages and explore the possibility of proposing new universals and generalizations. This 200 language sample is a worldwide core sample using the Genus-Macroarea- Method according to Miestamo et al. (2016). It covers most of the world at standard densities. The languages are chosen regardless of whether they have nominal classification systems and which kind of system they have. The database contains general information about the language and a range of properties of the nominal classification system in the language. If the language has more than one nominal classification system, each system has one record in the database. If the language has no nominal classification system, only general information about the language is noted. I test preliminarily eight Greenbergian Universals (Greenberg 1966) about gender systems. Some universals are confirmed. Meanwhile for some other universals serious counterevidence has turned up. Universal 31 says that if either the subject or object noun agrees with the verb in gender, then the adjective always agrees with the noun in gender. Languages like Alamblak, Ama (Papua New Guinea), Barasano, Berik, Burushaski, Kiowa, and Yuchi do not have gender agreement in the adjective though they have subject or object indexes. Thus Universal 31 is highly questionable. Universal 36 states that if a language has the category of gender, it always has the category of number. This universal is confirmed insofar as every gender language in the database also has the number category. However, not only the gender languages but also the classifier languages and languages which do not have any nominal classification systems tend to have number category. Haspelmath (2005) underlines the importance of the number category that only 28 out of 133 languages (21.05%) do not have any nominal plural. This study provides not only a clear distribution of nominal classification systems in the world's languages but also investigates possible principles of human languages on the basis of real universals about nominal classification systems. Reference Aikhenvald, Alexandra. 2000. Classifiers: A typology of noun categorization devices. !Oxford: Oxford University Press. Allan, Keith. 1977. Classifiers. In: Language. Vol. 53, 284–310. Contini-Morava, Ellen & Marcin Kilarski. 2013. Functions of nominal classification. In: ! Language Science. Vol. 40, 263–299. Corbett, Greville G. 1991. Gender. Cambridge: Cambridge University Press. Corbett, Greville G, Sebastian Fedden. 2016. Canonical Gender. In: Journal of !Linguistics, 52 (3), 95–531. Di Garbo, Francesca. 2014. Gender and its interaction with number and evaluative !morphology: An intra- and intergenealogical typological survey of Africa. !Stockholm University Dissertation. Fedden Sebastian, Greville G. Corbett. 2017. Gender and classifiers in concurrent !systems: refining the typology of nominal classification. In: Glossa: a jounral of !general linguistics. 2 (1), 1–47. Greenberg, Joseph H. 1966. Some universals of grammar with particular reference to the !order of meaningful elements. In: Joseph H. Greenberg (eds.): Universals of !language. Cambridge: The M.I.T. Press. Grinevald, Colette. 2002. Marking sense of nominal classification systems. In: Wischer, !Ilse & Gabriele Diewald (ed.): New Reflections on Grammaticalization. !Amsterdam / Philadelphia: John Benjamins. Haspelmath, Martin. 2005. Occurrence of nominal plurality. In: Haspelmath, Martin & !Matthew S. Dryer (eds.): The World Atlas of Language Structures. Oxford: Oxford !University Press. (Also availabe online at http://wals.info/chapter/34). Kilarski, Marcin. 2012. Nominal classification: A history of its study from the classical !period to the present. Amsterdam / Philadelphia: John Benjamin. Kramer, Ruth. 2015. The morphosyntax of gender. Oxford: Oxford University Press. Miestamo, Matti, Dik Bakker & Antti Arppe. 2016. Sampling for variety. In: Linguistic !Typology. Vol. 20(2), 233–296. Passer, Matthias B. 2016. The Typology and Diachrony of Nominal Classification. !University of Amsterdam Dissertation. Seifart, Frank. 2018. The semantic reduction of noun universe and the diachrony !of nominal classification. In: McGregor, William B. and Søren Wichmann (ed.): The !diachrony of classification systems. Amsterdam & Philadelphia: John Benjamins. Senft, Gunter (ed.). 2001. Systems of nominal classification. Cambridge: Cambridge !University Press. Singer, Ruth. 2016. The Dynamics of Nominal Classification: Productive and Lexicalised !Uses of Gender Agreement in Mawng. Boston / Berlin: Walter de Gruyter. Se zokwasi - a register of public speech in Komnzo

Christian Döhler, Universität zu Köln

This paper describes a speech register called se zokwasi in Komnzo. Komnzo is spoken by around 250 speakers in the village of Rouku in the south of Papua New Guinea (Döh- ler, 2018). Se zokwasi can be translated as ‘bark language’ and it is performed during all night dances when a person picks up a bark torch and circles the dancing ground while the dancers are resting. The descriptive part of the paper is based on two recordings of se zokwasi from 2012 and 2013 in Rouku village. I will use subtitled video and audio to illustrate the features of se zokwasi.

In the literature, registers have been approached from two very different angles. The first concentrates on corpus analyses and characterizes a register in terms of linguistic structure (Biber, 1995). The second has been developed by Halliday and it highlights the importance of the context of a situation (Halliday and Hasan, 1976). In the paper, I will follow Hoenigman (2015) and show that a combination of both approaches is not only possible, but results in the richest characterization of a register. As a working definition, I adopt Agha’s definition of registers as “a linguistic repertoire [emphasis in the original] that is associated, culture-internally, with particular social practices and with persons who engage in such practices” (2004: 24).

References

Agha, A. (2004). Registers of language. In: A. Duranti (Ed.), A companion to linguistic anthropology, pp. 23–45. Malden, Mass.: Blackwell. Biber, D. (1995). Dimensions of register variation: A cross-linguistic comparison. Cam- bridge: Cambridge University Press. Döhler, C. (2018). A grammar of Komnzo. Berlin: Language Science Press.

Halliday, M. and R. Hasan (1976). Cohesion in English. London: Longman. Hoenigman, D. (2015). The talk goes many ways - Registers of language and modes of performance in Kanjimei, East Sepik Province, Papua New Guinea. PhD thesis, Australian National University. Nominale Klassifikation: eine kanonisch-typologische Sicht Die Entdeckung neuer bemerkenswerter nominaler Klassifikationssysteme verlangt nach einer Typologie, die Genus und verschiedene Arten von Klassifikatoren einschließt. Die Idee des Gegensatzes zwischen Genus und Klassifikatoren wurde von Dixon (1982) klar zum Ausdruck gebracht, aber Seifarts (2005) Darstellung des Miraña mit klaren Merkmalen von Genus und Klassifikatoren erschwert die Aufrechterhaltung dieses Gegensatzes. Das Ngan’gityemerri (Reid 1997), welches zeigt, dass Klassifikatoren sich durch Grammatikalisierung zu Genera entwickeln können, liefert weitere Beweise für die Schwierigkeit einer kategorischen Unterscheidung. Wenn wir die Trennung zwischen Genus und Klassifikatoren aufgeben, erhalten wir ein klareres Bild der Variationbreite möglicher Systeme. Durch die Zerlegung der traditionell jeweils mit Genus und Klassifikatoren in Zusammenhang gebrachten Eigenschaften sehen wir, dass diese in vielfältigen Kombinationen vorkommen. Deshalb nehmen wir eine kanonische Perspektive ein, wobei wir ‚kanonisches Genus‘ als Messbasis für die Kalibrierung des theoretischen Raums der nominalen Klassifikation verwenden. Gemäß dem Kanonischen Genus-Prinzip „each noun has a single gender value“ (Corbett & Fedden 2016: 503; siehe Dixon 1982: 166). Zwei Kriterien sind diesem Prinzip untergeordnet. Kriterium „konstant“: kanonisch verlangt ein Nomen dieselbe Genuskongruenz in allen Domänen. Nomen mit unterschiedlicher Kongruenz (lexikalische Hybride) sind nicht kanonisch. Deutsch ist hier nah am kanonischen Ideal, während Russisch viele Hybride besitzt wie vrač ‘Arzt/Ärtzin’, das in unterschiedlichen Domänen maskuline oder feminine Kongruenz erlaubt. Kriterium „lexikalisch“: das Genus eines Nomens geht eindeutig aus dem Lexikoneintrag hervor. Dem kanonischen Ideal am nächsten kommen strikt semantische Systeme wie Bagvalal (männlich, weiblich, sonstig; Kibrik 2001: 64–66); desweiteren finden wir Systeme wie Mawng (Singer 2016), wo Rekategorisierung ohne weiteres möglich ist. Am weitesten entfernt vom kanonischen Ideal sind Systeme wie Burmesisch, wo unterschiedliche Numeralklassifikatoren benutzt werden können, um verschiedene semantische Aspekte des Referenten hervorzuheben. Kanonische Kongruenz, auf der kanonisches Genus basiert, steuert folgende Kriterien bei. Kriterium „obligatorisch“: da kanonische Kongruenz obligatorisch ist, ist dies auch eine Komponente von kanonischen Genussystemen wie Russisch oder Chichewa. Weniger kanonisch sind Systeme mit optionaler Kongruenz wie Ngan’gityemerri (Reid 1997); und Klassifikatoren unterschiedlicher Art sind oft optional (ihr Gebrauch hat meist besondere Effekte). Kriterium „spezifisch“: kanonische Systeme benutzen den spezifischsten Merkmalswert, der mit dem Kontroller der Kongruenz assoziiert ist, z.B. Mian. Weniger kanonische Systeme zeigen „superclassing“, z.B. Bininj Gun-wok (Evans 1997) und Jingulu (Meakins & Pensalfini 2016). Ungewöhnlicherweise ist hier der kanonische Fall auch häufig. Kriterium „orthogonal“: wie alle anderen morphosyntaktischen Merkmale ist Genus kanonisch, insofern als es sich orthogonal zu den Wortarten verhält. Man findet in der Tat Systeme, in denen fast alle Wortarten kongruieren (Archi ist ein extremes Beispiel), und am anderen Ende der Skala Systeme, wo Kongruenz eingeschränkt ist, wie z.B. im North Ambrym, wo nur relationale Klassifikatoren entsprechende Flexion zeigen (Franjieh 2016). Bei der sorgfältigen Kalibrierung mit Hilfe dieser Kriterien stellen sich traditionelle Genussprachen und traditionelle Klassifikatorensprachen in mancher Hinsicht als sehr ähnlich heraus, in anderer Hinsicht jedoch als sehr unterschiedlich; korrekterweise: Russisch, Archi, Mawng und Ngan’gityemerri sind äußerst ähnlich in Teilen ihres nominalen Klassifikationssystems und äußerst unterschiedlich in anderen. Unsere Typologie erfasst sowohl die große Diversität von Systemen (größer als noch vor einigen Jahren vermutet) als auch die Ähnlichkeit der Phänomene, die sich als Variationen ähnlicher Themen zeigen.

◆ Corbett, Greville G. & Sebastian Fedden. 2016. Canonical gender. Journal of Linguistics. 52.495-531. ◆ Dixon, R. M. W. 1982. Nominal classification. In: R. M. W. Dixon Where have all the adjectives gone? and other essays in semantics and syntax, 157–233. Berlin: De Gruyter. ◆ Evans, Nicholas. 1997. Head classes and agreement classes in the Mayali dialect chain. In: Mark Harvey & Nicholas Reid (eds) Nominal Classification in Aboriginal Australia, 105-146. Amsterdam: Benjamins. ◆ Franjieh, Michael. 2016. Indirect possessive hosts in North Ambrym: Evidence for gender. Oceanic Linguistics 55.87-115. ◆ Kibrik, Aleksandr E. (ed.) 2001. Bagvalinskij jazyk: Grammatika: Teksty: Slovari. Moscow: Nasledie. ◆ Meakins, Felicity & Rob Pensalfini. 2016. Gender bender: Disagreement in Jingulu Noun Class Marking. In: Felicity Meakins & Carmel O’Shannessy (eds) Loss and renewal: Australian languages since colonisation, 425-450. Berlin: Mouton de Gruyter. ◆Reid, Nicholas. 1997. Class and classifier in Ngan’gityemerri. In: Mark Harvey & Nicholas Reid (eds) Nominal classification in Aboriginal Australia, 165–228. Amsterdam: Benjamins. ◆ Seifart, Frank. 2005. The Structure and Use of Shape-based Noun Classes in Miraña (North West Amazon). Nijmegen: Radboud University Ph.D. ◆ Singer, Ruth. 2016. The Dynamics of Nominal Classification: Productive and Lexicalised Uses of Gender Agreement in Mawng. Berlin: De Gruyter Mouton.

1

The reconstruction of noun classification in Ghana-Togo-Mountain languages Ines Fiedler (Humboldt-Universität zu Berlin)

The presentation compares the synchronic gender systems of most of the Ghana-Togo- Mountain languages, revealing huge differences with respect to their design and complexity, in order to establish an improved reconstruction of the proto-system. This reconstruction aims at improving the existing one (Heine 1968) by including new data and by applying a systematic cross-linguistically oriented approach to the analysis of gender systems as proposed by Güldemann (2000) and, with a particular focus on Niger-Congo, Güldemann and Fiedler (2018). This approach consistently distinguishes the four following gender- related concepts: (a) agreement classes as the ultimate basis of the system of genders, (b) gender (classes) as reflecting the classification of nouns in the lexicon/reference domain, (c) noun (form) classes as the partial reflex of genders hosted by the nouns themselves and (d) declension classes as reflecting the morphological variation of nouns triggered by number, gender etc.. Even though Heine (1968) included agreement forms in the description of the synchronic systems, his reconstruction seems to rely mainly on noun form classes, so that it does not reflect the agreement-based proto-system of genders but rather that for declension classes. The new theoretical approach also contributes to the reconstruction of gender systems in Niger-Congo in general.

References: Güldemann, Tom and Ines Fiedler. 2018. Niger-Congo “noun classes” conflate gender with declension. In Francesca Di Garbo & Bernhard Wälchli (eds.), Grammatical gender and linguistic complexity, 85-135. Berlin: Language Science Press. Güldemann, Tom. 2000. Noun categorization systems in Non-Khoe lineages of Khoisan. Afrikanistische Arbeitspapiere 63:5-33. Heine, Bernd. 1968. Die Verbreitung und Gliederung der Togorestsprachen. (Kölner Beiträge zur Afrikanistik, vol. 1). Berlin: Dietrich Reimer.

The syntactical behaviour of personal names

Corinna Handschuh (Christian-Albrechts-Universitat¨ zu Kiel)

Proper names and common nouns show a number of differences in their grammatical behav- ior. Though individual languages vary greatly with respect to how visible these differences are ranging from distinct forms of the article in a number of Austronesian languages (Lynch et al. 2011:37) to a preference for the pre-nominal position as possessors in German the evidence for a special grammatical status of names has become too numerous for linguistis to further treat them as just “some type of noun”. Personal names, as the prototypical name type, are the class of names that deviates the most from common nouns (Anderson 2003:365, 373; Van Langen- donck & Van de Velde 2016:33). Thus, this paper concentrates on personal names, yet pointing out interesting contrast with other types. Presumably, every area of grammar exhibits some distinct behavior between personal names and common nouns in some language. This paper will illustrate this for the domain of noun phrase syntax. Names show a tendency to exhibit restrictions on their ability to co-occur with modifiers in a number of languages. Another pattern, very prominent at least in the Austronesian languages, is that names seldom occur as NP-heads but are rather used in apposition to common nouns serving as the head of the phrase instead. I will present data from a crosslinguistic survey of the grammatical properties of personal names and illustrate the different syntactic patterns found in the languages of the world. More specifically, the types of nominal modifiers that can or cannot occur with personal names in individual languages as well as the distribution of names as heads versus non-heads (e.g. in apposition to the head) of a noun phrase will be illustrated.

References

Anderson, John. 2003. On the structure of names. Folia Linguistica 37(3–4). 347–398. Lynch, John, Malcom Ross & Terry Crowley (eds.). 2011. The Oceanic languages. London: Routledge. Van Langendonck, Willy & Mark Van de Velde. 2016. Names and grammar. In Carole Hough & Daria Izdebska (eds.), The Oxford Handbook of names and naming, 17–38. Oxford: Oxford University Press.

1 Anti-Markierungen in den Sprachen der Welt und ihre funktional-adaptive Erklärung

Es gibt seit den 1970er Jahren eine Reihe von Termini in der typologischen Linguistik, die das Präfix “anti-“ verwenden, von denen insbesondere “Antikausativ” und “Antipassiv” weithin bekannt sind. In diesem Vortrag möchte ich eine Reihe von “Anti”-Konstruktionen vergleichend besprechen, die aus meiner Perspektive parallel zum Antikausativ sind, und am Ende noch kurz besprechen, warum es sich bei der Relation zwischen Antipassiv und Passiv anders verhält. Im folgenden sehen wir fünf Fälle von parallelen “Anti”- Konstruktionen:

Kausativ Türkisch koş- ‘rennen’ koş-tur- ‘rennen lassen’ Antikausativ Arabisch ʕallam- ‘lehren’ ta-ʕallam- ‘lernen’

(vgl. Haspelmath 1987)

Possessiv O’odham mi:stol ‘Katze’ ñ-mi:stol-ga ‘meine Katze’ Antipossessiv Huastec ʔu-čukul ‘mein Bauch’ čuk-lek ‘Bauch’

(vgl. Stiebels 2006)

Plural Deutsch Kind Kind-er Antiplural Walisisch plant ‘Kinder’ plent-yn ‘Kind’

(vgl. Haspelmath 2017)

Lokativ Russisch Moskva ‘Moskau’ v Moskve ‘in Moskau’ Antilokativ Bagvalal hũs:ač’ ‘in Khushtada’ hũs:aɬ han ‘Khushtada’

(vgl. Daniel 2019)

Reflexiv Russisch ee knigu ‘ihr Buch’ svoju knigu ‘ihrREFL Buch’

Antireflexiv Finnisch ruoka-nsa ‘ihrREFL Essen’ hänen ruoka-nsa ‘ihr Essen’

Nicht all diese Phänomene kommen gleich oft vor, aber sie scheinen alle auf den ersten Blick paradox: Ein Wort trägt eine grammatische Markierung, die bezeichnet, das eine (erwartete) Bedeutung NICHT vorliegt. Aber bei genauerem Hinsehen fallen sie alle unter dieselbe Verallgemeinerung:

(1) subklassenbedingte Kodierung Wenn verschiedene Subklassen von Ausdrücken verschieden lange Kodierungen einer grammatischen Bedeutung zeigen, dann finden sich die längeren Kodierungen in den häufigeren und erwarteteren Fällen, und die kürzen (bzw. Null-) Kodierungen in den selteneren und unerwarteteren Fällen.

Das sieht man etwa daran, dass nicht alle Verben Antikausativ-Entsprechungen haben, sondern vor allem diejenigen, die eher transitiv verwendet werden. Und nicht alle Substantive haben Antipossessiv-Markierungen, sondern vor allem diejenigen, die meistens possediert sind. Antilokative kommen vor allem bei Ortsnamen vor, die meistens lokativisch verwendet werden. In dem Vortrag werde ich zeigen, dass diese Kodierungen funktional-adaptiven Wert haben und deshlab gut erklärt werden können, also die scheinbare Paradoxie aufgelöst werden kann.

Literatur

Daniel, Michael. 2019. Bagvalal place names: A case study in part-of-speech typology. LTU... Haspelmath, Martin. 1987. Transitivity alternations of the anticausative type (Arbeitspapiere Des Instituts Für Sprachwissenschaft N.F. Nr. 4). Cologne: Universität zu Köln. Haspelmath, Martin. 2017. Explaining alienability contrasts in adpossessive constructions: Predictability vs. iconicity. Zeitschrift für Sprachwissenschaft 36(2). 193–231. doi:10.1515/zfs-2017-0009. Stiebels, Barbara. 2006. From rags to riches: Nominal linking in contrast to verbal linking. In Dieter Wunderlich (ed.), Advances in the theory of the lexicon, 167–234. Berlin: Mouton de Gruyter.

Birgit Hellwig (Universität zu Köln)

Child language documentation and the adult language

This talk explores what language documentation and description can gain by taking into account child language and child‐directed speech. While our documentary and descriptive efforts remain focused on the adult language, there are nevertheless good reasons to extend these efforts to include the language used by and with children: speech events involving children constitute essential parts of the natural linguistic behavior of speech communities (and thus warrant documentation in their own right); and investigations into language transmission will contribute to our understanding of processes of language endangerment and shift (and thus contribute to language strengthening and revitalization efforts) (e.g., Child Language Research and Revitalization Working Group 2017; Eisenbeiß 2006; Kelly et al. 2015; Kelly & Nordlinger 2014). And we can add a further argument to this list: the insights that child language and child‐directed speech offer into the metalinguistic knowledge of adult speakers.

The talk focuses on child‐directed speech and discusses its potential for providing insights into the adult language. Drawing on data from an on‐going documentation project on Qaqet (a non‐Austronesian language from Papua New Guinea; see http://qaqet.phil‐fak.uni‐koeln.de/), it is argued that this register constitutes a promising new data type which is set to inform language documentation and description.

The Grammaticalisation of motion verbs in Mayan languages: A continuum Nicole Hober (University of Bremen, [email protected]) The envisioned talk focusses on the grammaticalisation of motion verbs in Mayan languages and thus compares the verb-framed spatial deictic systems of languages within the same language family. The research forms part of a larger project dedicated to the description of spatial interrogatives and their functionally corresponding declarative counterparts from a cross-linguistic perspective (cf. Stolz et al. 2017; Robbers & Hober 2018). This talk shall take a closer look at the grammaticalisation continuum of spatial markers where “languages make use of zero spatial morphology or of full repertoires of spatial markers with a lot of options in between” (Robbers & Hober 2018: 400). Taking into consideration theories of and research on grammaticalisation informing this research (cf. Heine, Claudi, & Hünnemeyer 1991; Croft 2003; Hancil & König 2014), the spatial deictic systems of nine Mayan languages, namely Yucatec Maya, Mopán Maya, Tabasco Chontal, Chol de Tila, Tzotzil Zinacantán, Chuj, Kaqchikel, Akateko de San Miguel Acatán, and Mam de Huehuetenango, are examined and compared. To this end, the present contribution addresses the following research questions:

i. How is motion, especially in deictic expression, encoded in the respective Mayan languages? Are there differences between the interrogative and declarative side of the system? ii. Is there evidence of grammaticalised markers in the respective Mayan languages? How are the markers derived and what syntactic and semantic function do they fulfil? Are they obligatorily or optionally employed in the encodings of motion, considering both deictic and non-deictic expression?

Accordingly, it will be shown that one may identify three major categories of spatial deictic systems among the Mayan languages: (1) those using (multi-)verb constructions with verbal members of equal status, (2) those with (weakly) grammaticalised secondary motion verbs, and (3) systems with fully grammaticalised markers, which function as genuine directionals and are not restricted to physical movement. Examples of each type of the same Bible passage are given below.

[Luke 13:31] ‘Leave (and go away) from here.’ (1) Mopán Maya Joc'u('ol)-en wa'ye'

leave-INTR.IMP here (2) Tzotzil de Zinacantán loc‘-an ech‘-el li’-to-e

leave-IMP.2SG DIR:away-NF here-LOC-ENC (3) Mam de Huehuetenango t-ex-a tzalu ex txi'-ya

A2-DIR:out away-A2.ENC here DIR:out away go-A2.ENC The project as a whole and the investigation of the grammaticalisation continuum of Mayan motion verbs in particular offers interesting insights to typologists, areal linguists, Mayanists, and researchers of grammaticalisation. The overall project builds on existing work and aims towards a comprehensive understanding of spatial deixis and grammaticalization paths of motion verbs in the world languages.

Abbreviations

A = set A pronouns; DIR = directional; ENC = enclitic; IMP = imperative; INTR =intransitive; LOC = locative; NF = non-finite; SG = singular

Sources

[MOPNT] Le'ec Ada' U T'an A Dios A Tumulbene (Mopán Maya) © 2012, Wycliffe Bible Translators, Inc.

[TZOZNT] Sc'op riox: ja' li' yaloj ti chac' jcuxlejaltic ta sbatel osli ti cajvaltic Jesucristoe (Tzotzil de Zinacantán) © 2009, Wycliffe Bible Translators, Inc.

[MAMH] Mam de Huehuetenango Bible. © 1993. Sociedades Bíblicas Unidas.

References

Croft, William. 2003. Typology and Universals, 2nd edn. Cambridge: Cambridge University Press.

Hancil, Sylvie & König, Ekkehard. 2014. Grammaticalization – Theory and Data. Amsterdam: John Benjamins.

Heine, Bernd, Claudi, Ulrike, & Hünnemeyer, Friedericke. 1991. Grammaticalization: A Conceptual Framework. Chicago: University of Chicago Press.

Robbers, Maja and Nicole Hober. 2018. Verb-framed spatial deixis in Mesoamerican languages and the increasing complexity of SOURCE via Spanish de, Language Typology and Universals 71(3), 397-423.

Stolz, Thomas, Nataliya Levkovych, Aina Urdze, Julia Nintemann, and Maja Robbers. 2017. Spatial Interrogatives in Europe and Beyond: Where, Whither, Whence. Berlin: Walter de Gruyter.

Zur Nominalflexion im Nuristani Denise Hübner, Universität zu Köln ([email protected])

Die Nuristani-Sprachen1, von circa 100.000 Sprechern in der Hindukuschregion gesprochen, sind eine separate Gruppe des Indoiranischen, einem Hauptzweig der indogermanischen Grundsprache. Darunterfallen: Kati, Prasun (Wasī werī), Kalaṣa-alā, Tregami und Aṣkun (nach

STRAND 1973: 302). Diese sind keine Dialekte (JETTMAR 1975: 30; COSERIU 1980: 115), sondern eigenständige Sprachen mit Dialekten bzw. Dialektgruppen. Ihre interne Klassifizierung ist mindestens ebenso problematisch, wie ihre externen genetischen Beziehungen.

Obwohl die Nuristani-Sprachen viele linguistische und auch kulturell-religiöse Merkmale des Indoiranischen bewahren und damit eine unersetzbare Quelle für vergleichende Studien darstellen, sind sie in ihrer Gesamtheit vielfach von der (Sprach-)Wissenschaft vernachlässigt worden. Dabei erlaubt das gegebene Textkorpus eine synchrone Analyse, die der diachronen vorangeht und den Weg zu einem rekonstruierbaren Ur-Nuristani ebnet. Teil der Analyse müssten neben dem Lexikonbestand vor allem auch die Laut- und Formenlehre sein. Flexion findet sich innerhalb der Nuristani-Sprachen sowohl beim Verbum, als auch beim Nomen.

Der zu präsentierende Untersuchungsgegenstand ist die Flexionsmorphologie des Substantives in vorerst zwei Nuristani-Sprachen, das sind Prasun und Kalaṣa-alā. Die morphologischen Dimensionen des Nomens umschließen in beiden Sprachen Kasus, Numerus und Genus, aber ihre Flexionskategorien unterscheiden sich. Während beide Sprachen beim Numerus zwischen Singular und Plural unterscheiden, unterscheidet das Kalaṣa-alā zwei Genera (maskulin und feminin). Im Prasun gibt es die Genus-Differenzierung beim Substantiv nur in Bezeichnungen von Menschen und Haus- oder Nutztieren (DEGENER 2015: 14) und die Feminin-Motion bleibt auch noch zu klären. Zudem variieren die Anzahl und funktionale Beschaffenheit der Kasus und ihre vermeintlich synonyme Austauschbarkeit mit anderen Kodierungsstrategien morphologischer Flexionsunterschiede, wie Prä- und Postpositionen.

Einhergehend mit den nominalen Flexionsparadigmata sind die zu bestimmenden Nominalklassen beider Einzelsprachen und ggf. ihrer allomorphischen Varianten.

1 Ehemals Kafiri-Sprachen, d.h. die „Sprachen der Ungläubigen“. Diese Bezeichnung zu gebrauchen, gehört es heutzutage zu vermeiden (STRAND 1973: 297).

Literaturverzeichnis

COȘERIU, Eugenio. 1980. „‚Historische Sprache‘ und ‚Dialekt‘“. In: Göschel, Joachim (Hrsg.) Dialekt und Dialektologie: Ergebnisse des internationalen Symposions „Zur Theorie des Dialekts“. Wiesbaden: Steiner.

DEGENER, Almuth. 2015. Materialien zur Prasun-Sprache des afghanischen Hindukusch. 2 Bd. Cambridge, Massachusetts: Harvard University Press.

JETTMAR, Karl. 1975. „Die Region des Hindukusch.“ (Die Regionen der Menschheit, 4.1.) Stuttgart.

STRAND, Richard. 1973. „Notes on the Nūristānī and Dardic Languages.“ In: Journal of the American Oriental Society 93, 297-305. - 1997-2001. Nuristān. Hidden Land of the Hindu-Kush. (Website http://nuristan.info/ zuletzt aufgerufen am 14.10.2018)

Cognate objects in Arabic A structural and functional analysis of data from the Qur’an and from Modern Written Arabic

Empirical studies on cognate objects (further CO/COs) in different languages show that CO-constructions form a heterogeneous group both in terms of their syntax and their semantics (cf. Pereltsvaig (2002) on Hebrew, Nakajima (2006) and Kim & Lim (2012) on English, Glaude & Zribi-Hertz (2014) on Haitian Creole, and Akkuş & Öztürk (2017) on Sason Arabic). It has been pointed out that in contrast to earlier approaches, COs cannot be considered either only as arguments or exclusively as adjuncts (for COs as arguments in English and French cf. Macfarland (1995), in English and Japanese cf. Matsumoto (1996) and for COs as adjuncts in English cf. Jones (1988), in English and German cf. Moltmann (1989)). The question of the status of COs as arguments or adjuncts is accompanied by the discussion whether only intransitive verbs enter CO-constructions and if among the intransitive verbs only unergative verbs or also unaccusative verbs license COs (cf. Levin & Hovav (1995), Nakajima (2006)). Pereltsvaig (2002) and Kim & Lim (2012) (i.a.) argue that both ap- proaches (COs as arguments vs. adjuncts) are compatible by adopting two types of COs ((a) ‘arguments’ or ‘referential-COs’ vs. (b) ‘adjuncts’ or ‘eventive-COs’) and claim that languages differ in whether they allow only CO-type (a) or both. In my talk I will discuss the syntactic and semantic properties of COs based on data from Qur’an (Classical Arabic) and from Modern Written Arabic (cf. Badawi et al. (2004), Ryding (2005) (i.a.)) and argue that (i) in (Classical and Modern Written) Arabic both types of CO-constructions ((a) and (b)) occur with unergative, unaccusative and transitive verbs, (ii) in a CO-construction the nominal head of the CO needs not be etymolog- ically related to the lexical predicate of the construction; rather nominal derivations of other semantically similar verbs can also occur as the head of the CO, and (iii) diachronically in Arabic ‘eventive-COs’ (Kim & Lim 2012: 33) represent a productive path for the lexicalization of adverbs (from an accusative-marked (verbal)noun as head of the CO to an adverb with a fossilized accusative-marking).

1 References

Akkuş, Faruk & Balkız Öztürk. 2017. On Cognate Objects in Sason Arabic. University of Pennsylvania Working Papers in Linguistics 23(1). 31–55. Badawi, El Said, Michael Carter & Adrian Gully. 2004. Modern Written Arabic – A Comprehensive Grammar. New York: Routledge. Glaude, Herby & Anne Zribi-Hertz. 2014. Verb cognates in Haitian Creole. The Linguistic Review 31(2). 241–294. Jones, Michael Allan. 1988. Cognate objects and the Case Filter. Journal of Linguistics 24. 89–111. Kim, Jong-Bok & Joo-Young Lim. 2012. English Cognate Object Construc- tion: A Usage-based, Construction Grammar Approach. English Language and Linguistics 18(3). 31–55. Levin, Beth & Malka Rappaport Hovav. 1995. Unaccusativity: At the syn- taxlexical semantics interface. Cambridge: MIT Press. Macfarland, Talke. 1995. Cognate Objects and the Argument/Adjunct Dis- tinction in English. Illionis: Ph.D. dissertation, Northwestern University, Evanston. Matsumoto, Masumi. 1996. The syntax and semantics of the cognate object construction. English Linguistics 13(1). 199–220. Moltmann, Fredericke. 1989. Nominal and clausal event predicates. Papers from the 25th Annual Regional Meeting of the Chicago Linguistic Society 300–314. Nakajima, Heizo. 2006. Adverbial cognate objects. Linguistic Inquiry. Lin- guistic Inquiry 37(4). 674–684. Pereltsvaig, Asya. 2002. Cognate Objects in Modern and Biblical Hebrew. In Jamal Ouhalla & Ur Shlonsky (eds.), Themes in Arabic and Hebrew Syntax, vol. 53 Studies in Natural Language and Linguistic Theory, 107– 136. Dordrecht: Kluwer Academic Publishers. Ryding, Karin C. 2005. A Reference Grammar of Modern Standard Arabic. Cambridge: Cambridge University Press.

2 Reported intentionality in Trans-Himalayan: A case study of Monsang (South-Central)

In a number of Trans-Himalayan languages, the verb ‘say’ has grammaticalized into a quotative marker with a range of extended functions. According to Saxena (1988; 1995), these extended functions include causal, purpose, and conditional subordination, evidentiality, marking questions, indicating the deliberateness of an action, or marking onomatopoetic expressions. Saxena argues that this “quotative complex” is only found in languages that have been in long- standing contact with Indic languages. This idea builds on previous literature that considers versatile quotative constructions a characteristic of the South Asian linguistic area (Kuiper 1974; Klaiman 1977). The present paper examines within Trans-Himalayan the status of a particular type of (semi-)grammaticalized quotative-based construction: ‘reported intentionality’ (cf. McGregor 2007; Spronck 2015). Although a systematic comparative study has not been carried out yet, a preliminary survey has identified the construction in languages from different areas, different branches, and different typological profiles within Trans-Himalayan. This includes languages that do not appear to have been in any but the most recent contact with Indic languages. This study focuses on Monsang, where both synchronic and diachronic reported intentionality constructions are examined, based on a 10,000+ word corpus of various genres. Monsang does not have a history of immediate contact with Indic languages and yet makes extensive use of quotative constructions. In Monsang, we commonly find direct speech employed to express the causes or reasons as well as the purposes underlying the actions of third person participants. In addition to the synchronic functions of quotative constructions, Monsang also has several diachronic quotative constructions that reconstruct back to reported intentionality. Example (1) shows the desiderative construction of the language, which involves a desiderative suffix -níŋ as well as an auxiliary té, which derives from té ‘say’. Examples (2) and (3) illustrate two future constructions: in (2) the general, unspecified future with -váŋ, and in (3) the immediate future with -róŋ. Reported intentionality(-derived) constructions as in (1) appear to be more widespread in Trans-Himalayan as examples exist from languages as diverse as Burmese, Karbi, Galo, or Wadu Pumi. The development from quotative to future constructions as in (2) and (3) appears to be more unusual in this family. Parallels outside Trans-Himalayan exist, however, for example in Papuan languages (Reesink 1993), African languages (Aaron 1996; Botne 1998; Güldemann 2008), or the Amazonian isolate Aikanã (Van der Voort 2009). This paper argues that reported intentionality-type constructions are genuinely Trans-Himalayan, and suggests that this provides us a new perspective on the versatility of synchronic and diachronic quotative constructions in the family.

Examples (1) sá-níŋ kí-té-náʔ eatːI-DESID 1SG-AUX-IPFVːTR ‘I want to eat’

(2) sá-váŋ kí-té (3) sá-róŋ kí-té-náʔ eatːI-FUT 1SG-AUX eatːI-IMMED.FUT 1SG-AUX-IPFV:TR ‘I will eat’ ‘I’m about to eat’

References Aaron, Uche E. 1996. “Grammaticization of the Verb ‘say’ to Future Tense in Obolo.” Journal of West African Languages 26 (2): 87–93. Botne, Robert. 1998. “The Evolution of Future Tenses from Serial ‘Say’ Constructions in Central Eastern Bantu.” Diachronica 15 (2): 207–30. Güldemann, Tom. 2008. Quotative Indexes in African Languages: A Synchronic and Diachronic Survey. Berlin: de Gruyter Mouton. Klaiman, M. H. 1977. “Bengali Syntax: Possible Dravidian Influence.” International Journal of Dravidian Linguistics 4 (2). Kuiper, F. B. J. 1974. “The Genesis of a Linguistic Area.” International Journal of Dravidian Linguistics 3 (4): 135–53. McGregor, William B. 2007. “A Desiderative Complement Construction in Warrwa.” In Language Description, History and Development: Linguistic Indulgence in Memory of Terry Crowley, edited by Jeff Siegel, John Lynch, and Diana Eades, 27–40. Amsterdam: John Benjamins Publishing. Reesink, Ger P. 1993. “‘Inner Speech’ in Papuan Languages.” Language and Linguistics in Melanesia 24 (2): 217–25. Saxena, Anju. 1988. “On Syntactic Convergence: The Case of the Verb ‘Say’in Tibeto-Burman.” In Annual Meeting of the Berkeley Linguistics Society, 14:375–88. Berkeley, CA. ———. 1995. “Unidirectional Grammaticalization: Diachronic and Cross-Linguistic Evidence.” STUF-Language Typology and Universals 48 (4): 350–72. Spronck, Stef. 2015. “Reported Speech in Ungarinyin: Grammar and Social Cognition in a Language of the Kimberley Region, Western Australia.” Ph.D. dissertation, Canberra: Australian National University. Van der Voort, Hein. 2009. “Reduplication and Repetition of Person Markers in Guaporé Isolates.” Morphology 19: 263–86.

Vielfaltslinguistik, Köln, Institut für Linguistik, März 2019

Sibylle Kriegel Aix Marseille Univ, CNRS, LPL, Aix-en-Provence, France

The grammaticalization of reciprocity in Seychelles Creole

Seychelles Creole is one of the few creole languages with a grammaticalized reciprocity marker that is completely innovative compared with the French lexifier (for example: Plant ek plant osi i ed kanmarad, enn a lot – ‘The plants also help each other’). The marker kanmarad (< Fr. ‘camarade’) can be considered to be “freshly grammaticalized” material. This specificity is mentioned in the grammars of Seychelles Creole (Bollée 1977, Choppy 2009, Corne 1977) and in several articles (eg. Michaelis & Haspelmath, to appear) but its evolution, synchronic distribution, and semantics have never been analyzed in any detail. In the first part of this contribution, I will compare data from spoken and written corpora of Seychelles Creole with the data published in the Atlas of Pidgin and Creole Language Structures (Haspelmath et al. 2013). Only seven of the 69 represented languages have a special reciprocal marker based on ‘companion’ like Seychelles Creole. I then will analyze in depth some of the Seychelles Creole data with the use of kanmarad in Modern Mauritian Creole and in the diachrony of Mauritian and Seychelles Creole. This comparison will allow me to trace back the grammaticalization process of the marker, a process that is not very advanced in Mauritian Creole. In the last part, after having excluded the influence of the substrate languages, I will show that the evolution of kanmarad can be considered to be an instance of accelerated functionalisation in accordance with Haspelmath & Michaelis (2017) who claim that creoles show more accelerated functionalisation than their lexifiers. With respect to the possible reasons for the acceleration of functionalization of kanmarad, I will discuss the two hypothesis discussed by Haspelmath & Michaelis (2017), the “loss and replacement hypothesis” and the “extra transparency hypothesis” favored by the authors. I will suggest a third possibility that takes into account the very specific sociolinguistic situation of Seychelles where Creole has been the first national language since 1981.

Bollée, Annegret (1977): Le créole français des Seychelles. Esquisse d’une grammaire - textes - vocabulaire, Tübingen: Niemeyer.

Choppy, Marie-Thérèse (2009): Gramer Kreol Seselwa, Mahé: Lenstiti Kreol.

Corne, Chris (1977): Seychelles Creole Grammar, Tübingen: Narr.

Haspelmath, Martin & the APiCs Consortium (2013): “Reciprocal constructions”, in: Michaelis, Susanne Maria & Maurer, Philippe & Haspelmath, Martin & Huber, Magnus (edd.), The Atlas of Pidgin and Creole Language Structures, Oxford, Oxford University Press, 2013. Haspelmath, Martin & Michaelis, Susanne Maria (2017): “Analytic and synthetic. Typological change in varieties of European languages”, in: Buchstaller, Isabelle & Siebenhaar, Beat (2017) (edd.), Language Variation – European Perspectives VI. Selected papers from the Eighth International Conference on Language Variation in Europe (ICLaVE 8), Leipzig, May 2015, John Benjamins, Amsterdam etc.

Michaelis, Susanne & Haspelmath, Martin (to appear in 2019): “Grammaticalization in creole languages: Accelerated functionalization and semantic imitation”, in: Malchukov, Andrej & Walter Bisang, Handbook of areal patterns of grammaticalization and cross-linguistic variation in grammaticalization scenarios, Berlin: De Gruyter Mouton.

Michaelis, Susanne Maria & Maurer, Philippe & Haspelmath, Martin & Huber, Magnus (éds.) (2013): Atlas of Pidgin and Creole Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.

Abstract zum Vortrag am 28./29. März 2019 3. Konferenz „Vielfaltslinguistik“ in Köln

Enkhmaa Narmandakh Ludwig-Maximilians-Universität München

Referenten und referentielle Ereignisvorstellungen im Khalkha-Mongolischen

Die mongolische Sprache (in Mongolischer/Uigurjin-Schrift: , Mongɣol kele; in kyrillischer Schrift: Монгол хэл, Mongol xel) ist die Amtssprache der (Äußeren) Mongolei und basiert auf den Khalkha- Dialekten. Die mongolischen Sprachen gliedern sich in den Zentral-, Ost- und West-Zweig sowie eine Gruppe der mong. Randsprachen. Außerhalb der Mongolei wird sie in ihren Varietäten in Teilen Russlands und Chinas sowie in Zentralasien gesprochen.

Strukturell gesehen gehört sie dem Typus suffix-agglutinierender und dependent-marking Sprachen an und zeichnet sich durch ihre links-verzweigende Syntax aus. Folgt man der These von Ramstedt (vgl. Ramstedt 1952: 85) weist das Mongolische eine stark nominalisierte Struktur auf. So zeigen nicht nur die sogenannten Teilsätze, welche konverbial und partizipial gebildet werden, sondern auch finite Sätze (außer Imperative und Optative) nominale Eigenschaften (vgl. Verbalnomina).

In meinem Vortrag möchte ich anhand von Daten aus dem Khalkha-Mongolischen auf die Funktionen der verschiedenen Typen von Nominalphrasen eingehen, welche sich syntaktisch in drei Haupttypen gliedern lassen: Zeit-/Raumangaben (urida ‚front-DAT.LOC‘, xoino ‚behind-DAT.LOC‘), Referenten (Grammatische Relationen) in einfachen Sätzen (Qasar exedee zolgov ‚Qasar mother-DAT-POSS meet-PST‘) sowie Referenten in komplexen Sätzen (Bold morinoo unaǰ odloo ‚Bold horse-POSS ride-CVB.IPFV go.away-PST). Hierbei möchte ich nicht nur die morphosyntaktischen Besonderheiten aufzeigen, sondern auch die funktionalen Dimensionen eines Nomens bzw. einer Nominalphrase aus einer kognitive-semantischen Perspektive diskutieren.

Literatur:

- Janhunen, Juha (ed.) 2011. The Mongolic Languages. London: Routledge (Routledge language family series, 5). - Narmandakh, Enkhmaa 2018. A Cognitive Approach to Event Structures in Middle Mongolian based on the Corpus ‘The Secret History of the Mongols’. Dissertation LMU, München (in Druck). - Narmandakh, Enkhmaa 2018. Vergleichende Untersuchung des Deutschen und Mongolischen im Bereich der Syntax: Die Nominalphrase. LMU, Magisterarbeit (unveröffentlicht). - Poppe, Nicholas 1970. Mongolian. Language Handbook. Washington, D.C.: Library of Congress (Center for Applied Linguistics). - Ramstedt, Gustaf John 1952. Einführung in die Altaische Sprachwissenschaft. II. Formenlehre. bearbeitet und herausgegeben von Pentti Aalto. Helsinki: Suomalais-Ugrilainen Seura (Suomalais- Ugrilaisen Seuran toimituksia). - Schulze, Wolfgang 2012. Radical Experientalism/Cognitive Typology, 1–6. (http://schulzewolfgang.de/material/radical_experientialism.pdf, access date: 12/01/2018)

Abstract

Vielfaltslinguistik – 3. Konferenz zu Typologie und Feldforschung 28.-29. März, Universität zu Köln

Die vielen Gesichter der deiktischen Derivation im Julut (Sudan)

Ulrike Nüsslein, Universität Erfurt

Das Julut ist eine kordofanische Sprache (Niger-Kongo Sprachfamilie), die zusammen mit Tima und Katla die Katla-Sprachgruppe bildet. Alle drei Sprachen weisen etwa 10 derivationelle Extensionen auf, die zum Teil sehr verschieden sind. Unter den Gleichartigen befindet sich ein Kausativ, Kompletiv und Deixis/Ventiv. Die deiktische Derivation im Julut erfüllt jedenfalls über den Ventiv hinaus noch diverse andere Funktionen.

Nach einer Vorstellung und Einordnung des Julut sollen die funktionalen Bereiche bestimmt werden, auf die die deiktische Derivation Einfluss hat. Diese sind Phora, Valenz, Transitivität und Aspekt.

Im Hauptteil der Präsentation werden die einzelnen Funktionen konkretisiert und anhand von Beispielen illustriert. Der erste Teil bespricht die ventive Funktion, die üblicherweise auf Bewegungsverben angewendet wird. Tritt sie zu anderen Verbklassen hinzu, entsteht eine Mehrfachsituation („alloying“ Alamin, Dimmendaal, Schneider-Blum, 2012). Zweitens hat die deiktische Derivation auch anaphorische Funktion. Drittens wird die valenzerhöhende Funktion des sogenannten „lokalen Applikativ“ erörtert. Der vierte Abschnitt behandelt die aspektuelle Funktion atelische Verben in telische zu überführen.

Den Abschluss bilden Überlegungen, wie solch zum Teil verschiedene Funktionen von einem einzigen derivationellen Mittel erfüllt werden können und wie diese sich zu einem Gesamtbild zusammenfügen.

Alamin, Suzan, Dimmendaal, Gerrit J., Gertrud Schneider-Blum. 2012. Finding your way in Tima. In Angelika Mietzner and Ulrike Claudi (eds.), Directionality in Grammar and Discourse: Case Studies from Africa, pp. 9- 33. Cologne: Rüdiger Köppe. Indexical and iconic prosody in Anāl (Trans-Himalayan (TH); Manipur, India) A widespread phenomenon in many Tibeto-Burman languages of South-East Asia, North-East India and China is a pronunciation of a syllable with an extra-high pitch and/or extreme length, used to impart it a distinct “emphatic” interpretation. It has been briefly described as “paralinguistic emphasis” for a number of languages (e.g. Matisoff 1994:117 for Lahu). This phenomenon has evident cross-linguistic parallels (Ogden 2012; Niebuhr 2010). However, no study addressed it in detail for any TH language and detailed research on parallel phenomena in other languages is also scarce (e.g. Simard 2013). This study examines the phenomenon in Anal, a South-Central (Kuki-Chin) TH language. It is based on 260 examples collected from 1.5 hours of narratives/interviews and 25 minutes of spontaneous speech. The deviant prosody can be divided into two groups. The first group is characterised by an extreme deviation of the pitch (falsetto voice) and/or considerable lengthening of the syllable. This is found almost exclusively on grammatical morphemes (suffixes) denoting: (i) an edge degree or precision on the scale of quantity or time (e.g. distributive, augmentative and durative verbal suffixes). Consider Figure 1, where the affected syllable is the augmentative suffix déː. While the average pitch of the rest of the intonation unit is 220Hz, the affected suffix is over 500Hz high and the extreme length is evident. (ii) depictive/mimetic communication: ideophones and verbal ideophonic suffixes, as in Figure 2. (iii) distal demonstrative lé, accompanied by a pointing gesture (iv) emotive suffixes (e.g. a pity-expressing suffix) and interjections The second group of examples is characterised by a deviation and often lengthening of the last syllable of the intonation unit. This results in a syllable much higher and longer than the rest of the unit, yet much lower than the falsetto-voice examples above. This usage is a contact- establishing, addressee-engaging prosody found in appeal particles, agreement- or back- channelling (Figure 3) seeking particles, vocatives etc. Remarkably, it also grammaticalized as the question-marking prosody, replacing the question particle =mo found throughout the group. While the first group is closely related to the concept of “expressive communication”, discussed primarily for ideophones (see e.g. Sicoli 2014 and the references therein), the second group appears to represent a case of “indexical performance structures” (Noel Aziz Hanna and Sonnenhauser 2013). I propose to unify the view of the two for Anal: pitch deviation indexes the speakers’ engagement-triggering appeal to the interlocutor. This suffices for the indexing appealing-vocative particles of the second group. In the case of the first group the speaker may additionally express in an iconic way their affective attitude with respect to the content, requesting an interpretive work on the side of the listener as for the reasons for this involvement (cf. Nuckolls 1996:96; Sicoli 2014:451). Hence, these phenomena demonstrate the conventionalized role of iconic and indexical communication in language. Remarkably, this prosody in Anal occurs on a restricted set of suffixes and even appears to have grammaticalized as question-marking. The study also sheds light on the notion of “emphasis” suggesting that in the current case the actual phenomenon is addressee-engagement by dedicated non-arbitrary means.

Figure 1

Figure 2

Figure 3 References: Matisoff, James A. 1994. “Tone, Intonation, and Sound Symbolism in Lahu: Loading the Syllable Canon.” In Sound Symbolism, edited by John Ohala, Leanne Hinton, and Johanna Nichols, 115–29. Cambridge: Cambridge University Press. Niebuhr, Oliver. 2010. “On the Phonetics of Intensifying Emphasis in German.” Phonetica 67 (3): 170–98. https://doi.org/10.1159/000321054. Noel Aziz Hanna, Patrizia, and Barbara Sonnenhauser. 2013. “Vocatives as Functional Performance Structures.” In Vocative!, Addressing between System and Performance, edited by Barbara Sonnenhauser and Patrizia Noel Aziz Hanna, 283–303. Berlin, Boston: De Gruyter Mouton. https://doi.org/10.1515/9783110304176. Nuckolls, Janis B. 1996. Sounds like Life Sound-Symbolic Grammar, Performance, and Cognition in Pastaza Quechua. Oxford Studies in Anthropological Linguistics 2. New York: Oxford University Press. Ogden, Richard. 2012. “Making Sense of Outliers.” Phonetica 69 (1–2): 48–67. https://doi.org/10.1159/000343197. Sicoli, Mark A. 2014. “Ideophones, Rhemes, Interpretants.” In Ideophones: Between Grammar and Poetry, edited by Katherine Lahti, Rusty Barrett, and Anthony K. Webster, 3:445–454. Pragmatics and Society 5. Amsterdam/Philadelphia: John Benjamins. doi.org/10.1075/ps.5.3.08sic. Simard, Candide. 2013. “Prosody and Function of ‘Iconic Lengthening’ in Jaminjung.” SOAS Working Papers in Linguistics 16. http://www.soas.ac.uk/linguistics/research/workingpapers/volume-16/.

Sound Comparisons: Eine Onlinedatenbank zur Erforschung phonetischer Vielfalt

Ludger Paschen1, Paul Heggarty1

1MPI für Menschheitsgeschichte, Jena

Mit über 90.000 Aufnahmen einzelner Wörter, von denen für aktuell (Stand Dezember 2018) etwa die Hälfte enge phonetische Transkriptionen vorliegen, bietet Sound Comparisons einen einzigartigen Einblick in die sprachliche Vielfalt von mehr als 600 größtenteils Nicht-Standard-Varietäten aus 11 Sprachfamilien. Neben dem europäischen Sprachraum mit den “üblichen Verdächtigen” (Germanisch, Romanisch, Slawisch, …) sind auch Südamerika mit Quechua, Mapundungun und anderen Familien sowie die Insel Malakula (Vanuatu) mit verschiedenen ozeanischen Sprachen vertreten. Auf der Website soundcomparisons.com haben Nutzer die Möglichkeit, die Aufnahmen anzuhören, nach verschiedenen Kriterien zu filtern und ganze Datensätze inklusive der Transkriptionen herunterzuladen; eine vorherige Anmeldung oder Registrierung ist nicht erforderlich. Die Transkriptionen folgen neueren Initiativen zur Standadisierung phonetischer Umschrift in Unicode (Anderson et al. 2019). Jede Aufnahme kann zudem durch eine eindeutige URL zitiert werden. Durch die intuitiv zugängliche mehrsprachige Oberfläche spricht die Website nicht nur das Fachpublikum an, sondern vermittelt auch interessierten Laien ein Gespür für inter- und intradialektale lautliche Diversität und die Beziehung zwischen sprachlichen Varietäten. In dem Vortrag sollen die Grundfunktionen der Datenbank vorgestellt und seine Entstehungsgeschichte näher beleuchtet werden. Anhand ausgewählter Beispiele werden zudem einige relevante Anwendungsgebiete diskutiert. Diese umfassen neben phonetischen Distanzmessungen (McMahon 2007) und dem Überprüfen von Hypothesen zur Sprachgeschichte (Heggarty et al. 2010) auch die Darstellung von Dialektkontinua sowie von Divergenz- und Konvergenzprozessen. Nicht zuletzt bietet Sound Comparisons auch eine erste Archivierungsmöglichkeit für lautsprachliche Felddaten.

Anderson, C., Tresoldi, T., Chacon, T., Fehn, A.-M., Walworth, M., Forkel, R., List, J.-M. 2019. A Cross-Linguistic Database of Phonetic Transcription Systems. Erscheint in: Yearbook of the Poznań Linguistic Meeting.

Heggarty, P., Maguire, W., McMahon, A. 2010. Splits or waves? Tree or webs? How divergence measures and network analysis can unravel language histories. In: Philosophical Transactions of the Royal Society B 365, 3829–3843.

McMahon, A., Heggarty, P., McMahon, R., Maguire, W. 2007. The sound patterns of Englishes: representing phonetic similarity. In: English Language & Linguistics 11(1), 113–142. 3. Vielfaltslinguistik 2019 in Köln

Konferenz zu Typologie und Feldforschung

Vorschlag zu einem möglichen Vortrag:

Zu funktional „recycelten“ Suffixen / Die vielen Leben eines Suffixes (in Deutsch)

In meinem Vortrag auf der 3. Vielfaltslinguistik-Tagung möchte ich von den vielen diachronen „Leben“ und Funktionsänderungen eines Suffixes (-s-) berichten. Nach mutmaßlicher Grammatikalisierung (?) in geraumer Urzeit, tritt dieses Suffix (*-s-) als solches recht plausibel rekonstruierbar zunächst als INVERSE Markierung innerhalb eines hierarchischen Personenindexierungssystems auf, um wohl erst nach dem Urindogermanischen zu den indogermanischen Tochtersprachen hin eine Personalendung sowohl für die 2. Person (vedisch -s-) als auch für die 3. Person Singular zu werden (Pooth 2018). Ich werde die mögliche Vorgeschichte aber nur anreißen. Im vedischen Sanskrit finden wir dasselbe Suffix als Personalendung der 2. Person Singular, während es dort nur noch im Aorist regelmäßig die 3. Person Singular Optativ markiert (Kümmel 2018). Ich werde auch kurz zeigen, dass es unmöglich ist, diese zu trennen. In meinem Vortrag werde ich dann hauptsächlich nachweisen, dass es innerhalb des Vedischen zu einem Teil des Stammsuffixes des sog. „Prekativs“ (= Aorist-Optativstamms) -yās- fusioniert. Dabei werde ich eine Parallele anbringen, die das verwandte hethitische Suffix -š- betrifft, das innerhalb des Hethitischen von einem Personenindex der 2. Sg. und 3. Sg. ausgehend zu einem Teil einer längeren Personalendung wird. Das Ganze hat wichtige Implikationen für die Hypothesen zur Vorgeschichte des Stammsuffixes -s- des vedischen sigmatischen Aoriststamms (Watkins 1962, 1969). Wir können am Ende genau sehen und sicher sein, dass die bislang angenommenen Veränderungen und Formerweiterungen innerhalb der indogermanischen Sprachen existieren. Ich werde damit dann im Anschluss kurz einiges allgemeiner zu den Funktionsänderungen von Markierungen sagen, die nichts mit Grammatikalisierung und funktionalen Grammatikalisierungpfaden zu tun haben können, und wo sich die Funktionsveränderung sozusagen „aus dem funktionalen Nichts“ via „Recycling von unproduktiver Restmorphologie“ innerhalb des Paradigmas ergeben hat.

Ich würde den Vortrag in Köln gerne auf Deutsch halten.

Referenzen Kümmel, Martin Joachim (2018): Anatolisches und indoiranisches Verbum: Erbe und Neuerung, 100 Jahre Entzifferung des Hethitischen, in: Morphosyntaktische Kategorien in Sprachgeschichte und Forschung Akten der Arbeitstagung der Indogermanischen Gesellschaft vom 21. bis 23. September 2015 in Marburg. Herausgegeben von Elisabeth Rieken unter Mitwirkung von Ulrich Geupel und Theresa Maria Roth. Wiesbaden: Reichert, 239ff. Pooth, Roland A. (2018): Transitivity Direction in Proto-Indo-European: Direct, Inverse, and Intransitive Verb Forms, Manuskript, eingereicht bei Historische Sprachfoschung, vorveröffentlicht auf https://leidenuniv.academia.edu/RolandPooth Watkins, Calvert (1962): Indo-European Origins of the Celtic Verb. The Sigmatic Aorist. Dublin: Dublin Institute for Advanced Studies. Watkins, Calvert (1969): Indogermanische Grammatik. Bd. 3: Formenlehre. 1. Teil: Geschichte der indogermanischen Verbalflexion. Heidelberg: Winter.

Towards a typology of coverb constructions in Papuan languages Complex predicates consisting of an inflected verb in combination with an unmarked non-verbal element have been repeatedly pointed out to be a characteristic feature of Papuan languages, found in numerous Papuan languages across different families. Consider the following three examples from Duna (Isolate), Imonda (Border), and Yali (Dani). (1) a. hata no anda re su yesterday 1SG house eye hold:PFV ‘yesterday I watched the house’ (DUNA; San Roque 2008: 246) b. ehe-m hël-fe-ual-fan 3-GL kill-do-DL-PFV ‘he has killed the two of them’ (IMONDA; Seiler 1985: 114) c. kinang kulupnu=roho laha earth shake=ADV.PART go:3SG.IM.PST ‘the earth shook’ (YALI; field notes) These kinds of constructions have been called ‘verb compounds’ (Haiman 1980), ‘secondary verbs’ (Bromley 1981), ‘coverb’ constructions (Fedden 2011), and, probably most wide spread in the Papuan literature, ‘verb adjunct phrases’ (Pawley 2000, among others). Following Schultze- Berndt (2000), I will call them COVERB CONSTRUCTIONS (CVCs). Papuan languages make use of coverb constructions (CVCs) in many different ways, and great cross-linguistic variation can be observed with respect to both the inflected verbs, as well as the non-verbal elements that participate in the construction. Thus, some languages only use one or two very generic verbs in the inflected verb slot (e.g. ‘do’ and ‘say’ in Mian (Fedden 2011)), while others draw from a much larger set of verbs (e.g. 12 major verbs in Enga (Lang 1975)). Likewise, languages differ with respect to what elements can occur in the non-inflected slots. In some languages, one will primarily find nouns, adjectives, and/or adverbs (cf. (1a)), in other languages the coverb can never occur in any other position or function and thus constitutes a word class on its own (cf. (1b)). In yet other cases, coverbs constitute a distinct word class, but allow for non-finite marking, such as, e.g., participle marking, as in (1c). Many languages exhibit all three options. We also find variation – cross-linguistically as well as language internally – as to how ‘tight’ CVCs are, i.e. whether (and if, what kind of) material is allowed to intervene between the coverb and the inflected verb. Though CVCs are often mentioned in grammatical descriptions of single languages, unlike for serial verb constructions, only little comparative work is available. In this paper will give a first overview of CVCs in Papuan languages, showing more and less prototypical instances of CVCs and setting them off from other kinds of multi verb constructions and complex predicates.

ADV = adverbial; DL = dual; GL = goal; IM = immediate; PART = participle; PFV = perfective; PST = past; SG = singular

Bromley, M. 1981. A grammar of Lower Grand Valley Dani. Canberra: Pacific Linguistics. Fedden, S. 2011. A grammar of Mian. Berlin: Mouton de Gruyter. Haiman, J. 1980. Hua: A Papuan language of the Eastern Highlands of New Guinea. Amsterdam: John Benjamins. Lang, A. 1971. Nouns and classificatory verbs in Enga (New Guinea): A semantic study. PhD thesis, ANU. Pawley, A., S. P. Gi, J. S. Manjep & J. Kias. 2000. Hunger acts on me: The grammar and semantics of bodily and mental process expressions in Kalam. In V. P. de Guzman & B. W. Bender (eds.), Grammatical analysis. Morphology, syntax and semantics. Studies in the honor of Stanley Starosa. Honolulu: University of Hawai’i Press. San Roque, L. 2008. An introduction to Duna grammar. PhD thesis, ANU. Schultze-Berndt; E. 2000. Simple and complex verbs in Jaminjung. A study of event categorisations in an Australian language. Nijmegen: MPI Series Seiler, W. 1985. Imonda, a Papuan language. Canberra: Pacific Linguistics. Towards a better understanding of the relations of spoken and written Romani Melanie Schippling, Universität zu Köln

Spoken and written forms of a language are never identical, and a substantial body of research investigates the relation of spoken and written language. The PhD project presented in this talk contributes to this research by focusing on a language without a long-known tradition of writing, aiming to gain a better understanding of the relations of spoken and written language in Romani. Romani is a traditionally oral language (cf. e.g. Matras & Elšík 2006: 53) of Indo-Aryan origin spoken in areas which are characterised by a literacy-based culture. Bakker and Matras (1997: vii) state that “Romani is a minority language in all societies where it is spoken, but in terms of numbers of speakers Romani actually ranks among the more widely spoken languages of the world.” Historical written material in Romani is rather rare compared to its long-attested history and size of the speaker group. Matras and Elšík (2006: 53) point out that also nowadays

there is no form of standard Romani, even though written publications and internet correspondence in individual dialects of the language have been expanding rapidly since the early 1990s. There is not even a globally accepted prestige dialect. Every form of Romani is therefore a ‘dialect’.

In fact, speakers often employ an orthography resembling the notation conventions of their L2 or contact language(s). Identity-establishing functions and the question of representation of the language itself intertwine in the complex relations of spoken and written usage of Romani. To account for this plurality, the PhD project will examine multiple aspects of these relations, including:

- Investigations of attitudes of speakers towards oral and written language and their practice of using these modes with suitable methods such as sociolinguistic interviews and questionnaires. - Based on the results of these investigations, an informed analysis of similarities and differences between the two modes drawing on a method used in previous research (e.g. Maas 2010): recording oral language use, afterwards asking speakers to write down what they said.

While presenting initial findings, this talk also discusses challenges attributed to the project laying special emphasis on a collaborative approach.

References

BAKKER, PETER & YARON MATRAS. 1997. “Introduction.” In: Yaron Matras & Peter Bakker & Hristo Kyuchukov (eds.), The Typology and Dialectology of Romani. (Amsterdam Studies in the Theory and History of Linguistic Science. Series 4 – Current Issues in Linguistic Theory 156.) Amsterdam et al.: John Benjamins. vii-xxx.

MAAS, UTZ. 2010. “Orat und Literat. Grundbegriffe der Analyse geschriebener und gesprochener Sprache.” Grazer Linguistische Studien 73. 21-150.

MATRAS, YARON & VIKTOR ELŠÍK. 2006. Markedness and Language Change. The Romani Sample. (Empirical Approaches to Language Typology 32.) Berlin et al.: de Gruyter.

Towards subject-predicate agreement in Vera’a (Oceanic)

We investigate the choice between pronoun and zero anaphor for the expression of subjects in narrative texts from Vera’a (Oceanic; Schnell 2011, 2016). We relate this choice to a putative process of (re-)grammaticalisation of subject-predicate agreement typical of Oceanic languages (Ross 2004), whereby subject pronouns would eventually become the default referential choice, turning into agreement markers (Corbett 2003). This process interacts with older agreement morphology that has been reanalysed as exponents of tense, aspect, mood, polarity (TAMP) categories that have since deteriorated in morphological substance (François 2009 ). While 1 st and 2 nd person subjects are categorically expressed by a pronoun (Schnell 2018), we focus here on the alternation in 3 rd person subjects. We exclude from this investigation full NPs, assuming that their use is relatively well explained in terms of accessibility theory (Ariel 1990) and related approaches to discourse structure.

Non-lexical 3 rd person subjects are predominantly expressed by a pronoun too, but zero form is still attested in 26% of cases. A mixed-effect generalized linear regression model (Table 1) identifies six significant factors for the use of a pronoun, namely anaphoric distance and antecedent function, animacy and number, and form and agreement properties of co-present tense, aspect, mood, polarity (TAMP) morphology within the verbal predicate. Antecedent distance and function are relevant in restricting zero subjects largely to same-subject clause chains. Within this context, we find zero subjects to be most likely with singular subjects and prospective aspect, where subject agreement is co-expressed (unlike in any other TAMP category) by a formally unreduced marker, ex (1), in contrast to the respective plural form, ex (2), and many other high-frequency TAMP markers, ex (3).

Although our synchronic corpus investigation cannot provide any conclusive evidence for or against specific diachronic developments, our findings do square with the hypothesis of re- grammaticalizing subject-predicate agreement, so that pronouns are practically the default non-lexical form of expression, leaving zero anaphor only for those context where older agreement is co-present and formally fully transparent. Our findings suggest that a putative process of re-grammaticalizing subject agreement is spurred primarily by purely morphological changes (Barth & Kapatsinski 2017; Bybee & Thompson 2007) rather than functionally motivated (Givón 1976; Ariel 2000).

1

EXAMPLES (1) gōsuwō ne kalraka ne rōw lē =n nanara lumasag […] rat PROSP :3 SG get.up PROSP :3 SG jump LOC =ART tree.sp ontop

ne rōw kal lumasag kēkē ne sag 'i PROSP :3 SG jump up ontop high PROSP :3 SG sit DEL

‘Rat got up, jumped ontop of the nanara (trunk), jumped up ontop and sat down there.’ GAQG.028

(2) duru =k kalraka duru =k tēk mē di so 3DU =PROSP :NSG get.up 3DU =PROSP :NSG speak DAT 3SG QUOT

‘Then they (two) got up and said to him:’ HHAK.117

(3) di =m sag sur suwō di =m kur sa =n gako wova'al ē 3SG =PRF sit down downwards 3SG =PRF gnaw EMPH =ART stalk pawpaw DEM 3

‘He sat down and gnawed (through) that very pawpaw stalk.’ GAQG.078

TABLE 1. Results of mixed-effects generalized linear regression Log likelihood: −575.5 AIC: 1173.1 BIC: 1230.8 Speaker variance: 0.28 ± 0.53 Estimate SE Z p Intercept -0.17 0.41 -0.42 0.67 Antecedent distance = 2+ clauses 1.74 0.31 5.63 0.00 Function of antecedent = object 0.10 0.37 0.27 0.79 Function of antecedent = other 1.51 0.33 4.55 0.00 Number = non-singular 1.25 0.29 4.35 0.00 Animacy = human -0.83 0.34 -2.45 0.01 Animacy = inanimate 0.69 0.30 2.28 0.02 TAMP form = particle 0.54 0.22 2.44 0.01 TAMP person = no inflection -1.24 0.37 -3.31 0.00 TAMP person marking by TAMP form = no inflection by particles 1.42 0.46 3.07 0.00 Note : Positive coefficients are associated with higher pronoun expression

REFERENCES

Ariel, Mira. 1990. Accessing noun phrase antecedents. London: Croom Helm. Ariel, Mira. 2000. The development of person agreement markers: from pronouns to higher accessibility markers. In Michael Barlow & Suzanne Kemmer (eds.), Usage-based models of language , 197–260. Stanford, CA: CSLI. Barth, Danielle & Vsevolod Kapatsinski. (2017). A multimodal inference approach to categorical variant choice: construction, priming and frequency effects on the choice between full and contracted forms of am, are and is . Corpus Linguistics and Linguistic Theory 13(2), 1-58. Bybee, Joan & Sandra Thompson. 2007[1997]. Three frequency effects in syntax. In Joan Bybee (ed.), Frequency of use and the organization of language , 269–278. Oxford: Oxford University Press.

2

Corbett, Greville G. 2003b. Agreement: The range of the phenomenon and the principles of the Surrey Database of Agreement. Transactions of the Philological Society 1001(2). 155–202. François, Alexandre. 2009. Verbal aspect and personal pronouns: The history of aorist markers in north Vanuatu. In Alexander Adelaar & Andrew Pawley (eds.), Austronesian historical linguistics and culture history: A festschrift for Robert Blust , 179–196. Canberra: Pacific Linguistics. Givón, Talmy. 1976. Topic, pronoun, and grammatical agreement. In Charles N. Li (ed.), Subject and topic , 149–188. New York: Academic Press. Ross, Malcolm D. 2004. The morphosyntactic typology of Oceanic languages. Language and Linguistics 5(2). 491–541. Schnell, Stefan. 2011. A grammar of Vera’a, an Oceanic language of North Vanuatu. PhD dissertation (druckreife Fassung). Kiel: University of Kiel. Schnell, Stefan. 2016. Vera’a. In Geoffrey Haig & Stefan Schnell,.MultiCAST: Multilanguage corpora of annotated spoken texts . Language Archive Cologne. https://lac.uni- koeln.de/de/multicast/ (accessed 23 September 2017). Schnell, Stefan. 2018. Whence subject-verb agreement? Investigating the role of topicality, accessibility, and frequency in Vera’a texts. Linguistics 56.4, 735-780. DOI: https://doi.org/10.1515/ling-2018-0010

3

Michael Schulze, PhD student, Humboldt-Universität zu Berlin Vielfaltslinguistik III, 28.-29. März 2019, Universität zu Köln

Abstract: “Prefix-dropping” in the Mel languages (Niger-Congo)

The Mel languages of Sierra Leone, Guinea and Liberia – traditionally being a part of the Atlantic subfamily of Niger-Congo (see Greenberg 1963) – are nowadays considered by many scholars to be an independent branch of Niger-Congo (cf. e.g. Dalby 1965; Hammarström et al. 2018). The Northern Mel languages (Temne and Baga languages) show a prefixing gender and noun form morphology with a tendency towards alliterative agreement (see Wilson 1961; for the separation of gender and noun form morphology see Güldemann & Fiedler forthcoming). This morphosyntactic type is well-known from Bantu languages and was reconstructed for Proto-Bantu (Meeussen 1967). Given the overwhelming amount of research done on Bantu languages, it has been almost taken for granted that the Proto-Niger-Congo system must have been of this type, too. Since more and more data from Niger-Congo groups outside Bantu are coming in, this view must be seriously challenged. One phenomenon related to this debate is the absence of the noun form prefix of head nouns in noun phrases extended by modifiers. This feature has already been described for Aghem (Hyman 1979) and Dakarkari (Hoffmann 1967), like the Bantu languages members of the Benue-Congo branch of Niger-Congo. A thorough analysis of the data on the Southern Mel languages Mani (Childs 2011), Krim (Pichl 1972), Sherbro (Rogers 1967) and Kisi (Childs 1995) shows that “prefix- dropping” is prevalent in these languages, too. This talk will give an overview and offer possible explanations how the feature could have evolved diachronically. Although it seems somewhat self-evident that the prefix has been simply omitted in extended noun phrases, it cannot be excluded that the obligatory prefix-marking of the head noun in all contexts is a secondary development.

Michael Schulze, PhD student, Humboldt-Universität zu Berlin Vielfaltslinguistik III, 28.-29. März 2019, Universität zu Köln

References Childs, George Tucker. 1995. A grammar of Kisi, a southern Atlantic language (Mouton Grammar Library 16). Berlin & New York: Mouton de Gruyter. Childs, George Tucker. 2011. A grammar of Mani (Mouton Grammar Library 54). Berlin & Boston: De Gruyter Mouton. Dalby, David. 1965. The Mel languages: A reclassification of southern ‘West Atlantic’. African Language Studies 6. 1-17. Greenberg, Joseph H. 1963. The languages of Africa. Bloomington: Bloomington University Center in Anthropology, Folklore, Linguistics und Den Haag: Mouton. Güldemann, Tom & Ines Fiedler. Forthcoming. Niger-Congo “noun classes” conflate gender with declension. In Francesca Di Garbo & Bernhard Wälchli (eds.), Grammatical gender and linguistic complexity, 85-135. Berlin: Language Science Press. Hammarström, Harald, Robert Forkel und Martin Haspelmath. 2018. Glottolog 3.3. Jena: Max Planck Institute for the Science of Human History. Available online at http://glottolog.org, Accessed on 2018-12-05. Hoffmann, Carl. 1967. An outline of the Dakarkari noun class system and the relation between prefix and suffix noun class systems. In Manessy, Gabriel (ed.), La classification nominale dans les langues négro-africaines, 237-259. Paris: CNRS. Hyman, Larry M. 1979. Phonology and Noun Structure. In Larry M. Hyman (ed.), Aghem Grammatical Structure (Southern California Occasional Papers in Linguistics 7), 1-72. Los Angeles: Department of Linguistics/University of Southern California. Meeussen, Achille E. 1967. Bantu grammatical reconstructions. Africana Linguistica 3. 79-121. Pichl, Walter J. 1972. The Krim language in Sierra Leone. Pittsburgh: Duquesne University. Rogers, Henry E. 1967. The phonology and morphology of Sherbro. New Haven: Yale University, PhD thesis. Wilson, William André Auquier. 1961. An outline of the Temne language. London: School of Oriental and African Studies (SOAS), University of London. Titel: Sprechpausen und die Entstehung von Affixen: Evidenzen aus Sprachdokumentationskorpora zu zehn Sprachen und aus typologischen Datenbanken

Abstract: Affixe entstehen historisch durch Fusion von Funktionswörtern mit Inhaltswörtern. So können z.B. Kasussuffixe aus Postpositionen entstehen. Allerdings kann dies nur geschehen, wenn zwischen den entsprechenden Funktionswörtern und Inhaltswörtern nicht regelmäßig Sprechpausen auftreten. Die Wahrscheinlichkeit des Auftretens von Pausen in bestimmten Positionen kann also "Affixasymmetrien" erklären, z.B. die bekannte „Suffixpräferenz“, d.h. das es - typologisch gesehen - weniger häufig Präfixen als Suffixen gibt (Himmelmann 2014). In diesem Vortrag werde ich eine Studie vorstellen, in der einerseits Pausenwahrscheinlichkeiten vor Verben vs. Nomen typologisch untersucht werden. Dies geschieht anhand von Sprachdokumentationskorpora von zehn Sprachen: Baure, Bora, Chintang, Englisch, Even, Hoocąk, Niederländisch, Nǁng, Sakha und Texistepec. Diese Korpora dokumentieren spontanen Sprachgebrauch und sind von Experten annotiert worden z.B. für Wortarten und Wortgrenzen. Zur Identifikation von Pausen sind Transkription und Audiosignal automatisch zeitaligniert worden (Strunk, Schiel & Seifart 2014). Ich zeige, dass direkt vor Nomen häufiger Pausen auftreten als direkt vor Verben (was frühere Beobachtung bestätigt bzgl. Pausenwahrscheinlichkeiten in 0,5-Sekunden Zeitfenstern vor Nomen und Verben, s. Seifart et al. 2018). Der Grund dafür scheint zu sein, dass Nomen im Diskurs häufiger neue Informationen liefern als Verben. Nach Nomen und Verben gibt es kein entsprechendes Muster. Andererseits zeige ich eine typologische Tendenz, dass typologisch gesehen an Nomen weniger häufig Präfixe auftreten als an Verben. Dies geschieht anhand des Vergleichs von gebundenen vs. freien Markern vor Verben vs. Nomen in der Autotyp-Datenbank (Bickel et al. 2017). Die Verbindung dieser beiden Beobachtungen legt nahe, dass Pausenwahrscheinlichkeiten auch die Asymmetrie bezüglich verbaler vs. nominaler Präfixe erklären können.

Referenzen: Bickel, Balthasar, Johanna Nichols, Taras Zakharko, Alena Witzlack-Makarevich, Kristine Hildebrandt, Michael Rießler, Lennart Bierkandt, Fernando Zúñiga & John B. Lowe. 2017. The AUTOTYP typological databases. Version 0.1.0. https://github.com/autotyp/autotyp-data/tree/0.1.0. Himmelmann, Nikolaus P. 2014. Asymmetries in the prosodic phrasing of function words: Another look at the suffixing preference. Language 90(4). 927–960. doi:10.1353/lan.2014.0105. Seifart, Frank, Jan Strunk, Swintha Danielsen, Iren Hartmann, Brigitte Pakendorf, Søren Wichmann, Alena Witzlack-Makarevich, Nivja H. de Jong & Balthasar Bickel. 2018. Nouns slow down speech across structurally and culturally diverse languages. Proceedings of the National Academy of Sciences 115(22). 5720–5725. doi:10.1073/pnas.1800708115. Strunk, Jan, Florian Schiel & Frank Seifart. 2014. Untrained Forced Alignment of Transcriptions and Audio for Language Documentation Corpora using WebMAUS. In Nicoletta Calzolari, Khalid Choukri, Thierry Declerck, Hrafn Loftsson, Bente Maegaard, Joseph Mariani, Asuncion Moreno, Jan Odijk & Stelios Piperidis (eds.), Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC 2014), 3940–3947. Reykjavik: European Language Resources Association (ELRA). http://www.lrec-conf.org/proceedings/lrec2014/pdf/1176_Paper.pdf.

Towards diachronic typology of partitive expressions Ilja A. Seržant

A partitive expression is a construction that encodes part-whole relationship between two referents (a cup of that tea). In this talk, I focus on what kind of meanings and functions these constructions develop in addition to or, sometimes, instead of, the original part-whole meaning. More specifically, I focus on partitive expressions that can be used without the head constituent, henceforth PE, (like ‘of that tea’). In many languages, these expressions undergo very versatile developments. Thus, they enter the domain of indefiniteness and non-specificity which is typically handled by determiners otherwise, cf. Russian ja kupil konfet (1SG buy.PST.M.SG candy.GEN.PL) ‘I bought (some) candies.’ In this example, the candies concerned are indefinite and often non-specific. The indefiniteness of the PE-encoded referents is sometimes related to discourse backgroundedness: the PE-marked NP typically has low cataphoric potential; it is part of a larger focus constituent and never argument focus itself, etc. Another function of PEs is to code weak quantification of the NP’s referent and, in some languages, of the hosting VP which has immediate implications for its aspectual interpretation, cf. Finnish Hän avasi ikkunaa (3SG open.PST.3SG window.PART.SG) ‘(S)he was opening the window’ or ‘(S)he somewhat opened the window.’ The PE-marked NP’s referent ‘window’ is affected holistically while the activity of opening the window is quantified (‘somewhat’). Historically immediately related to (i) is predicate negation (cf. Muller 1997 on French; Kiparsky 2011 on Finnish; Timberlake 1986, Padučeva 2006, Rakhilina, ed. 2000, Harves 2013 or Guiraud-Weber 2014 on Russian; Arkadiev 2017+ on Lithuanian; Miestamo 2014 on cross-linguistic patterns). In some of these languages, it has developed into a syntactic rule requiring the PE marking on direct objects and existential subjects across the board if the negation scopes over the predicate, cf. Lithuanian Jo/*Jis nėra namuose (3SG.GEN/*3SG.NOM NEG.be.PRS.3 home) ‘He is not at home.’ Finally, intensional verbs, i.e. those verbs which allow the opaque (narrowest scope) reading of the object’s referent (cf. Quine 1960: §32; Zimmermann 1993), may favour (cf. Huumo 2013) and, subsequently, generalize the PE marking of their object as did the Lithuanian verb ieškoti ‘to seek, to look for’, which now requires the PE marking on its object. Very often partitives develop into pseudo-partitives. While a true partitive construction (a piece of that cake) is the one that codes a subset (a piece) from a clearly defined superset (that cake) under the exclusion of the complement (all other pieces of that cake), a pseudo- partitive expression is the one in which a construction formally analogical to a true-partitive construction violates the Partitivity Constraint. The latter requires the superset to refer to sets identifiable by both the speaker and the hearer. A pseudo-partitive expression, in turn, has no restricted superset. The latter is replaced by a kind-referring expression (cf. Yesterday I had a cup of tea). With a kind referring expression, neither the complement nor the superset can be defined in terms of sets. While the main line of research on partitivity and pseudo-partitivity has been concerned with the description of how the PEs apply in these different domains (cf. Huumo 2010), less work has been done to uncover the historical and areal development of the polysemy, i.e., how these meanings are historically interrelated. An important study of the diachrony of partitives encoded by the preposition de/dē in Romance is Carlier & Lamiroy (2014) based on evidence from Latin, Old Spanish and Old French, and, finally, the modern successors. These authors distinguish five stages in each of which they describe the semantic properties and syntactic properties. To our knowledge, this is the only study that provides a complete picture while others only treat particular aspects or diachronic stages. The present paper will elaborate on the diachronic aspects of partitives along these lines. References

Arkadiev, Peter 2017+: Objektnyj partitive otricanija: areal’no-tipologičeskaja perspektiva. (The partitive of the negated object: an areal-typological perspective). Carlier A. & B. Lamiroy 2014: The grammaticalization of the prepositional partitive in Romance. In: S. Luraghi & T. Huumo (eds.), Partitive Cases and Related Categories. Berlin, Boston: de Gruyter Mouton, 477–519. Guiraud-Weber, Marguerite 2003: Eščo raz o russkom genitive otricanija: vzgljad so storony, Russian Linguistics 27, 363-384. (Once again about the Russian genitive of negation: looking from outside) Harves S. (2013). The genitive of negation in Russian. Language and Linguistics Compass, 1– 16. Huumo, Tuomas 2010: Nominal aspect, quantity, and time: The case of the Finnish object. Journal of Linguistics 46, 83–125. Huumo, Tuomas 2013: On the many faces of incompleteness: Hide-and-seek with the Finnish partitive object. Folia Linguistica 47(1), 89–111. Kiparsky, Paul 2001: Structural case in Finnish, Lingua 111, 315–376. Miestamo, Matti 2014: Partitives and negation: A cross-linguistic survey. In: S. Luraghi & T. Huumo (eds.), Partitive Cases and Related Categories. Berlin, Boston: de Gruyter Mouton, 63–86. Muller, Claude 1997: De partitif et la négation. In: D. Forget, P. Hirschbühler, Fr. Martineau & M. L. Rivero (eds.), Negation and Polarity. Syntax and Semantics. Amsterdam, Philadelphia: John Benjamins, 251–270. Padučeva, Elena V. 2006: Genitiv dopolnenija v otricatel’nom predloženii, Voprosy jazykoznanija 6, 21-43. (The object genitive in negated sentences) Rakhilina, Ekaterina V. (ed.) 2000: Ob”ektnyj genitiv pri otricanii v russkom jazyke. Moscow: Probel. (The genitive of negation on objects in Russian) Seržant, Ilja A. 2015a: Independent partitive as a Circum-Baltic isogloss, Journal Language Contact 8, 341-418. Timberlake, Alan 1986: Hierarchies in the genitive of negation. In: R. Brecht & J. Levine (eds.),Case in Slavic. Columbus (OH): Slavica, 338–360.

Predicative possession beyond HAVE-constructions: towards taking stock of BELONG and OWN across languages Thomas Stolz & Nataliya Levkovych (University of Bremen)

Ever since the seminal studies by Hansjakob Seiler in the 1970-1980ies possession has developed into one of the favourite topics in linguistic typology. There is as yet a plethora of dedicated publications which focus on different aspects of the grammar of possession. It suffices to mention Heine (1997) as an influential monograph which covers the entire domain of linguistic possession. Koptjevskaja-Tamm (2002) gives an account of adnominal possession – an issue investigated more recently also by Haspelmath (2017). Predicative possession is payed attention in several typologically-minded studies. There is e.g. the book- length study by Stassen (2009). Mazzitelli (2015, 2017) inquires into predicative possession in Belarusian and Lithuanian and provides an areal survey for the languages of the Circum- Baltic region. Stassen (2005) traces the geo-linguistic distribution of predicative possessive constructions in the world’s languages. Michaelis (2013) looks at the situation in Pidgin and Creole languages. What all of these and many more studies have in common is that they limit the scope of the research to proper HAVE-constructions, i.e. they take account of functional equivalents of sentences like English Georgina has a new car. However, we know from Seiler’s work (Seiler 1983) that there are further kinds of predicative possession (other than those of the HAVE-type) which have not been studied in sufficient detail yet. Aikhenvald & Dixon (2013) show that it makes sense to study OWN-constructions more closely without, however, putting forward any typologically valid generalisations as to the relation between HAVE and OWN. A further notion that is repeatedly mentioned in the literature on possession is that of BELONG. Usually BELONG is depicted as the mirror image of HAVE. Some authors like Heine (1997) claim that the opposition HAVE ≠ BELONG is universal and each language has formal means to distinguish the two possessive categories. In contrast there are also voices like Baron & Herslund & Sørensen (2001) who cast doubt upon the tenability of this universal. On the basis of Clark (1978) and Ultan (1978) it seems reasonable to assume that one branch of BELONG-constructions is widely common among the world’s languages, namely so called substantival possesstion corresponding to English The car is hers / Georgina’s. What is still largely unknown is the distribution of BELONG-constructions which involve lexical verbs of possession (like English The car belongs to her / Georgina).

The talk is intended to convince field linguists and typologists to look beyond HAVE- constructions when they carry out their descriptive linguistic or cross-linguistic projects. If we want to understand the grammar of possession in individual languages as well as cross- linguistically, we need to widen the scope of our research so that OWN-constructions and BELONG-constructions can be adequately integrated into the general picture. In support of this hypothesis we will present and discuss data especially from Mesoamerican and Micronesian languages.

References

Aikhenvald, Alexandra Y. & Dixon, R.M.W. (eds.). 2013. Possession and ownership. A cross-linguistic typology. Oxford: Oxford University Press. Baron, Irène & Herslund, Michael & Sørensen, Finn (eds.). 2001. Dimensions of possession. Amsterdam, Philadelphia: Benjamins. Clark, Eve. 1978. Locationals: existential, locative, and possessive constructions. In: Joseph H. Greenberg (ed.) Universals of human language. Volume 4: Syntax. Stanford: Stanford University Press, 85–126. Haspelmath, Martin. 2017. Explaining alienability contrasts in adpossessive constructions: predictability vs. iconicity. Zeitschrift für Sprachwissenschaft 36 (2), 193–231. Heine, Bernd. 1997. Possession. Cognitive sources, forces, and grammaticalization. Cambridge: Cambridge University Press. Koptjevskaja-Tamm, Maria. 2002. Adnominal possession in the European languages: form and function. Sprachtypologie und Universalienforschung 55 (2), 141–172. Mazzitelli, Lidia Federica. 2015. The expression of predicative possession. A comparative study of Belarusian and Lithuanian. Berlin, Munich, Boston: De Gruyter Mouton. Mazzitelli, Lidia Federica. 2017. Predicative possession in the languages of the Circum-Baltic area. Folia Linguistica 51 (1), 1–60. Michaelis, Susanne Maria (and the APiCS Consortium). 2013. Predicative possession. In: Susanne Maria Michaelis & Philippe Maurer & Martin Haspelmath & Magnus Huber (eds.) The atlas of Pidgin & Creole language structures. Oxford: Oxford University Press. 306–309. Seiler, Hansjakob. 1983. Possession as an operational dimension of language. Tübingen: Narr. Stassen, Leon. 2005. Predicative possession. In Martin Haspelmath & Matthew Dryer & David Gil & Bernard Comrie (eds.), The world atlas of language structures. Oxford: Oxford University Press, 474–477.

Stassen, Leon. 2009. Predicative possession. Oxford: Oxford University Press. Ultan, Russell. 1978. Towards a typology of substantival possession. In Joseph H. Greenberg (ed.) Universals of human language. Volume 4: Syntax. Stanford: Stanford University Press, 11–50. ‚Kaffeemäßig‘ – ‚kaffeeähnlich‘ – ‚kaffeehaltig‘: Abgeleitete Adjektive im Kambaata, einer äthiopischen Sprache

Anders als viele verwandte kuschitische Sprachen hat Kambaata eine morphosyntaktisch klar abgrenzbare Wortart Adjektiv. Neben einer großen Anzahl von unabgeleiteten, monomor- phematischen Adjektiven, verfügt die Sprache über mehrere produktive Derivationsprozesse, um Adjektive von Nomen, Verben, Ideophonen und anderen Adjektiven abzuleiten. Zu diesen Prozessen gehört beispielsweise die Attenuativableitung mit dem Morphem -lab (ATT), siehe z.B. mux-á ‚nass‘ > mux-(i)-lab-á ‚ein bisschen nass, feucht‘, die Similativableitung mit -agud (SIM), siehe z.B. hamiil-ú ‚Kohl‘ > hamiil-agud-á ‚kohlmäßig‘, die Proprietivableitung mit -aam (PROP), siehe z.B. az-úta ‚Milch‘ > az-aam-ú ‚milchhaltig‘ und die Agentivableitung mit -aan (AG), siehe z.B. ros- ‚lernen‘ > ros-aan-ch-ú ‚lernend‘. Die aus diesen Derivationen resultierenden Adjektive verhaltenden sich morphologisch und syntaktisch wie Basisadjektive. Im Zentrum meines Vortrags soll die Attenuativableitung stehen. Nach einer kurzen Einführung in die formalen (morphologischen und syntaktischen) Eigenschaften von Attenuativadjektiven, steht die Analyse der Semantik von -lab (ATT) im Vordergrund. In meinen Feldforschungskorpus und in lokalen Kambaata-Publikation kommt die Ableitung mit -lab am häufigsten bei Farbadjektiven vor, daneben außerdem an einem Dutzend anderer Adjektive ohne Farbbezug. Hier drückt -lab einen reduzierten Grad einer Eigenschaft aus (‚ein bisschen [Eigenschaft]‘, ‚fast [Eigenschaft]‘, ‚ziemlich [Eigenschaft]‘), z.B. biishsh-á [ADJ] ‚rot‘ > biishsh-lab-á ‚rötlich‘. Die Attenuativadjektive sind nicht eindeutig negativ oder positiv konnotiert. Außer an adjektivischen Stämmen finden wir -lab an einfachen und erweiterten Verbstämmen – bisher sind 15 verschiedene deverbale Adjektive belegt. Deverbale Attenuativ- adjektive werden normalerweise in adverbialer Funktion gebraucht und drücken eine ungefähre Art und Weise der Ausführung aus, d.h. ‚fast (aber nicht wirklich) [Verb]-end‘, z.B. torr- [V] ‚werfen‘ > torr-lab-á ‚fast werfend/auf werfende Art und Weise ‘, gis- [V] ‚schlafen’ > gis-lab- á ‘fast schlafend, schläfrig, dösig’. Denominale Adjektive mit -lab – von denen bisher 21 verschiedene belegt sind – drücken aus, dass etwas ‚[Nomen]-mäßig‘, ‚fast wie [Nomen]‘ ist, wodurch auf produktive Weise Adhoc-Kategorien gebildet werden können, z.B. weteechch-ú [N] ‚Baumart‘ > weteechch-lab-á haqqá ‚Baum, der ähnlich wie ein weteechchú-Baum ist (hier: einen ähnlichen Duft erzeugt)‘, marf-á ‚Nadel‘ > marf-lab-á ‚nadelmäßiges Ding‘. Die Bedeutung der letzten Untergruppe, d.h. der denominalen Attenuativadjektive, ist auf den ersten Blick kaum von den o.g. Similativadjektiven (abgeleitet durch -agud) zu unterscheiden. In meinem Vortag werde ich jedoch nachweisen, dass Similativadjektive eine äußere Ähnlichkeit in Aussehen (Struktur, Aufbau) ausdrücken, während denominale Attenuativadjektive eine Ähnlichkeit in Charakter (Natur, inneren Eigenschaften) vermitteln, siehe bun-á ‚Kaffee‘ (i) bun-agud-á (SIM) ‘kaffeeähnliches Getränk (was Farbe / Aussehen betrifft)’ vs. > (i) bun-lab- á (ATT) ‚kaffeeähnliches Getränk (z.B. was Geschmack / Zubereitungsart) betrifft‘. Die Analyse von abgeleiteten Adjektiven im Kambaata ist eingebettet in eine Diskussion der Ableitungsmechanismen in verwandten und benachbarten afrikanischen Sprachen.

1 Initial consonant loss and its consequences Evidence from Cape York Peninsula (Australia) J-C Verstraete (University of Leuven) This paper examines the historical loss of root-initial consonants and syllables, and its consequences for stress systems and the structure of roots. Systematic initial loss is relatively rare from a worldwide perspective (Blevins 2007), but it is found in several areas in Australia (Hale 1976a, Alpher 1976, Blevins 2001), including Cape York Peninsula in the northeast. This paper uses data from six Paman (Pama-Nyungan) languages of this region to show that initial loss is not just a typological curiosity, but has wide-ranging consequences beyond the segmental level. The languages studied here belong to the Middle Paman and the Lamalamic subgroups of Paman (Verstraete 2012, 2018). Middle Paman languages show a combination of retention, lenition and loss of initial consonants, as shown in (1) for Umpithamu, while Lamalamic languages show systematic loss of initial consonants, as shown in (2a) for Umbuygamu, and/or initial syllables, as shown in (2b) for Lamalama (Proto-Paman reconstructions: Hale 1976b). (1) a. kuwa ‘west’ ~ *kuwa b. ya’u ‘foot’ ~ *caru c. aangkal ‘shoulder’ ~ *paangkal (2) a. agarr ‘flesh’ ~ *pangkarr b. karr ‘flesh’ ~ *pangkarr The process of initial loss found in these languages is interesting not just as a typological rarity, but also because it has consequences at various levels. First, loss of initial consonants can lead to a reorganization of the system of word stress. I use the case of Umpithamu (Middle Paman) to show that partial loss of initial consonants in this language initiates a cascade of changes resulting in stress shift, from the left-aligned moraic trochees found in many Pama- Nyungan languages, to the right-aligned system found in Umpithamu. This is interesting from a diachronic perspective, because stress shift has often been argued to cause initial loss (e.g. Hale 1964), whereas here it can be shown to be a consequence. Secondly, initial loss can also lead to generalized root expansion, in order to restore a bisyllabic minimum for roots. In Middle Paman and Lamalamic, the net result of systematic initial loss would be a large percentage of monosyllabic roots in the lexicon (or their metrical equivalents in bisyllabic roots with extrametrical initial vowels). I use historical-comparative-evidence to show that in both subgroups, various historical processes of root expansion can be identified that keep the percentage of monosyllables low, ranging from metrically driven leftward and rightward expansion, to large-scale lexicalization of compound structures. In this sense, initial loss can be shown to be the starting point for a range of processes that lead to a fairly radical change in the typological profile of the languages where it occurs. Alpher, B. 1976. Some linguistic innovations in Cape York and their sociocultural correlates. In Sutton, ed. Languages of Cape York. Canberra: AIAS. 84-101. Blevins, J. 2001. Where have all the onsets gone? Initial consonant loss in Australian Aboriginal languages. In Simpson et al, eds. Forty years on: Ken Hale and Australian languages. Canberra: PL. 481-492. Blevins, J. 2007. Endangered sound patterns: Three perspectives on theory and description. Language Documentation and Conservation 1: 1-16. Hale, K. 1964. Classification of Northern Paman languages, Cape York Peninsula. A research report. Oceanic Linguistics 3: 248-265. Hale, K. 1976a. Phonological developments in particular Northern Paman languages. In Sutton, ed. Languages of Cape York. Canberra: AIAS. 7-40. Hale, K. 1976b. Wik reflections of Middle Paman phonology. In Sutton, ed. Languages of Cape York. Canberra: AIAS. 50-60. Verstraete, JC. 2012. Contact-induced restructuring of pronominal morphosyntax in Umpithamu (Cape York Peninsula, Australia). Diachronica 29ː 326-358. Verstraete, JC. 2018. The genetic status of Lamalamic: Phonological and morphological evidence. Oceanic Linguistics 57: 1-30. Poqomchi’ (Mayan) today and 400 years ago: Some remarks on diachronic development

Poqomchi’ is a Mayan language spoken in Guatemala by around 100,000 people. Mayan family is traditionally considered one of the best documented in Latin America. However, Poqomchi’ clearly remains understudied mostly due to its peripheral geographical location. At the same time, this makes Poqomchi’ especially interesting from a perspective of areal contacts. Being genetically a Highland language, Poqomchi’ shares several traits with neighboring Lowland Mayan languages (see Law 2014). Poqomchi’ is also an intriguing object for diachronic research. The first documents written in Poqomchi’ were produced by Dominican missionaries after the Spanish conquest, in the second half of the 16th century. The most important sources on the Poqomchi’ language of that period include the Poqomchi’-Spanish-Latin dictionary written by Dionisio de Zúñiga around 1625, his “Arte” (grammatical description) of Poqomchi’, and three extensive collections of sermons. These manuscripts are stored in libraries and archives in US, but some of them have been digitized. In the talk, I will offer a brief overview of the current state of linguistic research on Poqomchi’, followed by an introduction of the most important colonial manuscripts. After that, I will discuss some individual findings in the diachronic development of different grammatical domains. The phenomena observed in the corpus of colonial texts will be compared to their modern equivalents. Much information about Poqomchi’ and its current dialectal variation was obtained during four fieldwork stays in Guatemala beginning in 2014. For instance, I will argue that the preverbal clitic na= which denotes the future in the modern language, denoted the progressive aspect in colonial Poqomchi’. Compare (1) and (2): (1) xrilov hinah chic angel, narupurinic pam yihal taxah xrilow jenaj chik angel narupuriniik paam yejaal taxaaj x-r-il-ow jenaj chik angel na=ru-pur-in-iik paam PFV-3SG.ERG-see-TR INDEF more angel PROG=3SG.POSS-fly-INTR-NMLZ LOC yejaal taxaaj middle sky ‘He saw an angel flying in the middle of the sky.’ (“Sermones en lengua Pokonchi”, ~1600, folio 173 verso) (2) xa re’ chik i keem na’ab’anam rajlaal i q’iij xa_re’ chik i keem na=’a-b’an-am r-ajlaal i q’iij only more DEF weave FUT=2POSS-do-NMLZ 3POSS-all DEF day ‘Now, you will only weave all the day.’ (Modern Poqomchi’; Vinogradov 2016: 184) In this context, sentence (1) would not make sense if na= has a future reading. Meanwhile, sentence (2) is unequivocally interpreted as a future prediction by the speakers of Poqomchi’ nowadays. Another example concerns negation. Modern Poqomchi’ commonly uses a double negator which consists of the negative preverbal particle ma’ and the enclitic ta or taj analyzed as an irrealis marker. In the corpus of colonial texts, there are no instances of such a double negation. Instead, the simple negative ma’ is used. The enclitic spelled tah is attested, but its meaning seems to be close to the optative modality. This morpheme does not accompany the negator in early colonial texts. I will also discuss the “recent past” marker a- which was probably borrowed from Lowland Cholan Mayan languages into colonial Poqomchi’ and then disappeared; the loss of the ancient definite article e’ which probably stems from a noun meaning ‘thing’; the emergence of a new set of interrogative pronouns; and the conjunction eh ‘and’ borrowed from 16th century Spanish. References Law, Danny. 2014. Language contact, inherited similarity and social difference: the story of linguistic interaction in the Maya lowlands. Amsterdam: Benjamins. Vinogradov, Igor. 2016. El tesoro no recibido por miedo: dos cuentos mayas poqomchi’. Tlalocan 21: 167-215. Highlights of the MelaTAMP projects—methods and results

In the MelaTAMP project, we have been investigating tense, aspect, modality and polarity in seven Oceanic languages of Melanesia since 2016. In this talk, we would like to highlight Language stay Redup. Impf. which methods we have been using in our research and how Daakaka + + − they relate to our results. Nafsan + − − 1) In our work on habitual aspect in four Oceanic languages Mav̋ea + + + (accepted, joined work with two more authors), we look pri- Saliba-Logea − + − marily at corpora from language documentation on Nafsan (South Efate), Saliba-Logea, Mavea and Daakaka to find out how they express habituality. Methods • We exhausted existing descriptions and results of previously conducted questionnaires such as Dahl (1985). • We then identified target contexts in the corpora a) by looking for keywords suchas used to, usually, always in the English translations, and b) based on a set of stories which were present in several languages, which we knew to contain specific habitual contexts. • We imported the corpus data into ANNIS, for optimal search facilities. Results • Three out of four languages had cognate continuous auxiliaries related to full verbs meaning stay, which were frequently used in habitual contexts. • Mavea has an additional imperfective affix, which also featured prominently in expressing ha- bituality. • Three out of four languages used reduplication in order to express habitual aspect. • Those two languages that used both reduplication and imperfective or continuous aspect mark- ing to express habituality sometimes combined both structures. The combination of these two elements was exclusive to habitual contexts. • Both realis and irrealis are compatible with habitual contexts. 2) In our article on mapping irreality in six Vanuatu languages (ac- cepted with two other authors), we investigated the relation between tense and modality by focussing on counterfactual future contexts as in If were to win the lottery tomorrow, I would take a vacation. We investi- gated whether these contexts patterned with counterfactual condition- als of the past (If I had won the lottery, I would have taken a vacation), or with indicative conditionals of the future (If I win the lottery, I will take a vacation). We took the results as an indication whether each lan- guage prioritized the modal domain (counterfactuality) or the temporal domain (future) in these contexts. Figure 1: The counter- Methods: We used storyboards, some of which we designed ourselves factual future (shaded gray) specifically for this study, to elicit counterfactual future conditionals as the intersection between and minimal contrasts to counterfactual past conditionals and indica- the future (solid outline) and tive future conditionals. the counterfactual (dashed Results: Languages show remarkable variation in how they carve outline) out the modal-temporal semantic space, which we modeled with a branching-time framework. Sharing the load ‐ the interplay of verbal and gestural negation in Savosavo Claudia Wegener (Universität zu Köln) & Jana Bressem (Technische Universität Chemnitz)

This talk presents preliminary work on the interplay of the verbal and gestural domains to express negation in Savosavo, a Non‐Austronesian language spoken by about 3.500 people on Savo Island, Solomon Islands (http://dobes.mpi.nl/projects/savosavo/).

Much of the previous research on gestures associated with negation focused on identifying the usage contexts and semantic core as well as the range of possible form variations of relevant gestures and gesture families, usually for better‐studied languages such as English, French, Italian or German (e.g. Kendon 2004, Harrison 2009, Calbris 2011, Bressem and Müller 2014). Some also looked at utterances where both negation gestures and explicit verbal negation co‐occur and studied their temporal alignment (Harrison 2009, 2014). What has not been addressed, to our knowledge, is the question of when and why gestural and verbal negation do indeed co‐occur. In this talk, we will present our preliminary results regarding this question for verbal and gestural negation in Savosavo.

Based on a corpus of about 6h of video data we studied the relation of verbal and gestural negation from two perspectives. First, we analyzed the patterns of use of one particular gesture associated with negation, the “sweeping away” gesture (flat hands moving outwards as if wiping away crumbs from a table), and noticed a strong tendency of this gesture not to occur with explicit verbal negation, i.e. negative morphemes like ghoma ‘no’ or sika ‘don’t’. Instead, it is found mostly with utterances containing implicit (lexical or pragmatic) negation, i.e. for example with lexemes like zui ‘end’ or tabu ‘forbidden’, or utterances that only implicate or presuppose a negation. We then looked at explicit verbal negation, checking every occurrence in our corpus to see if any kind of gesture associated with negation is present, including head movements. Our results show that most instances of explicit verbal negation are indeed not accompanied by any gestural negation, confirming the pattern observed with the “sweeping away” gesture. We then discuss potential factors that could explain those cases where they do co‐occur, as well as general implications of these findings with regard to the question of how the verbal and gestural domain interact and collaborate in communication.

References: Calbris, G. (2011). Elements of Meaning in Gesture. Amsterdam, John Benjamins Publishing Company. Harrison, S. (2009). Grammar, Gesture, and Cognition: The Case of Negation in English, Université Michel de Montaigne, Bourdeaux 3. Ph.D. Harrison, S. (2014). The organisation of kinesic ensembles associated with negation. Gesture 14(2), 117-141. Kendon, A. (2004). Gesture: Visible Action as Utterance. Cambridge, Cambridge University Press.