Boundaries and Bridges: Language Contact in Multilingual Ecologies 9781614514886, 9781614516842

Citation preview

Kofi Yakpo and Pieter C. Muysken (Eds.) Boundaries and Bridges

Language Contact and Bilingualism

Editor Yaron Matras

Volume 14

Boundaries and Bridges Language Contact in Multilingual Ecologies Edited by Kofi Yakpo and Pieter C. Muysken

ISBN 978-1-61451-684-2 e-ISBN (PDF) 978-1-61451-488-6 e-ISBN (EPUB) 978-1-5015-0114-2 ISSN 2190-698X Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. © 2017 Walter de Gruyter Inc., Boston/Berlin Cover image: Anette Linnea Rasmus/Fotolia Typesetting: Compuscript Ltd., Shannon, Ireland Printing and binding: CPI books GmbH, Leck ♾ Printed on acid-free paper Printed in Germany www.degruyter.com

Preface The present volume grew out of a project at Radboud University, Nijmegen, Netherlands, on the dynamics of language contact in Suriname (2009–2013), funded by Advanced Grant 230310 of the European Research Council (ERC) entitled “Traces of Contact” to Pieter Muysken. At the time, Kofi Yakpo was a postdoctoral researcher at Radboud University coordinating the Surinamese part of the project. Some of the material in this book was presented at a seminar in Amsterdam Zuidoost in October 2012. We are grateful to participants of this seminar, including Enoch Aboh, Margot van den Berg, Stanley Hanenberg, Maarten Mous, Paul Tjon Sie Fat, Donald Winford, and Solace Yankson for their comments. We are also grateful to the members of the Nijmegen Languages in Contact group, including Suzanne Aalberse, Joshua Birchall, Loretta O’Connor, Pablo Irizarri van Suchtelen, Harald Hammarström, Gerrit Jan Kootstra, Olga Krasnoukhova, Francesca Moro, and Neele Müller for earlier comments. We want to acknowledge the support of the Centre for Language Studies (CLS) and the Department of Linguistics of Radboud University, Nijmegen, Netherlands, the Freiburg Institute for Advanced Studies (FRIAS) of the AlbertLudwigs-Universität in Freiburg, Germany, and the Linguistics Programme of the University of Hong Kong for our work along the way. Some of the editing for this volume took place at the Stellenbosch Institute for Advanced Studies (STIAS) at the Wallenberg Research Centre at Stellenbosch University. We want to thank the reviewers for this volume including Peter Bakker, Thomas Conners, Leonie Cornips, Swintha Danielsen, Spike Gildea, Stephen Matthews, Sandro Sessarego, Jeff Siegel, Nicolien van der Sijs, and the authors, some of whom read each other’s work. Robbert van Sluijs deserves a special word of thanks for helping with the final checking and the formatting of the book to make it ready for publication. Sophie Villerius commented on several versions of the conclusions. We have tried to uniformize spellings as much as possible in this book, following practices in Suriname. Hence Suriname (not Surinam), Trio (not Tiriyo), Jodensavanne (not Jewish Savannah), and alternating Sranantongo and Sranan (but not Sranan Tongo). However, we have used Saramaccan (not Saamaka), because this is how many linguists know the language.

Table of contents Preface v List of contributors

ix

Overview and background Kofi Yakpo and Pieter Muysken Language contact and change in the multilingual ecologies of the Guianas 3 Robert Borges The people and languages of Suriname

21

Case studies Kofi Yakpo Creole in transition: Contact with Dutch and typological change in Sranan 57 Robert Borges The Maroon creoles of the Guianas: Expansion, contact, and hybridization 87 Kofi Yakpo Out of India: Language contact and change in Sarnami (Caribbean Hindustani) 129 Sophie Villerius Developments in Surinamese Javanese

151

Luis Miguel Rojas-Berscia with Jia Ann Shi 179 Hakka as spoken in Suriname Sergio Meira and Pieter Muysken Cariban in contact: New perspectives on Trio-Ndyuka pidgin Eithne B. Carlin Language contact in Southern Suriname: The case of Trio and Wayana 229

197

viii 

 Table of contents

Konrad Rybka Contact-induced phenomena in Lokono (Arawakan)

257

Pieter Muysken The transformation of a colonial language: Surinamese Dutch

283

Comparative perspectives Robert Borges, Pieter Muysken, Sophie Villerius and Kofi Yakpo The tense-mood-aspect systems of the languages of Suriname

311

Bettina Migge From grammar to meaning: Towards a framework for studying synchronic language contact 363 Kofi Yakpo and Pieter Muysken Multilingual ecologies in the Guianas: Overview, typology, prospects 405 References Author index 433 Subject and language index

439

383

List of contributors Robert Borges did his undergraduate work at Rhode Island College in Providence, RI, ­obtained a Master’s in African Linguistics from Leiden University, and a PhD from Radboud ­University, with a thesis titled The Life of Language: dynamics of language contact in Suriname. ­Subsequently he co-edited Surviving the Middle Passage: the West Africa-Surinam Sprachbund (De Gruyter 2014) with Pieter Muysken and Norval Smith. Currently Robert is postdoctoral ­researcher at University of Warsaw. Eithne Carlin is lecturer in Amerindian linguistics at the Leiden University Centre for Linguistics (LUCL). She obtained her Ph.D. in Linguistics in 1993 from the University of Cologne, Germany, and a B.A. (Hons) and M.A. from Trinity College Dublin. She is author of A Grammar of Trio, a Cariban Language of Suriname (Peter Lang, 2004) and co-edited (with the late Jacques Arends) the Atlas of the Languages of Suriname (2002, Leiden and Kingston: KITLV Press and Ian Randle Publishers). Sergio Meira is a Brazilian linguist, with a PhD from Rice University. He has worked at Leiden University, the Max Planck Institute for Psycholinguistics and Radboud University. Currently he is a researcher at the Muséu Goeldi in Belém, Brazil. He carries out descriptive and comparative research on the Tupian and Cariban language families, and on the Trio (or Tiriyo) language in particular. Bettina Migge is Associate Professor in the University College Dublin School of Irish, Celtic Studies, Irish Folklore and Linguistics. She studied at the University of Hamburg, the University of Yaounde, the Free University of Berlin and The Ohio State University (Columbus) where she obtained her Master’s and PhD. She is the author of Creole formation as language contact: The case of the Suriname Creoles (John Benjamins) and Exploring Language in a Multilingual Context: Variation, Interaction and Ideology in Language Documentation (with Isabelle Léglise, Cambridge University Press). Pieter Muysken has been Professor of Linguistics at Radboud University since 2001, after teaching in Amsterdam and Leiden. He has published widely in the area of creole studies, code-switching and language contact, and Andean linguistics. Luis Miguel Rojas-Berscia taught Mandarin and obtained an undergraduate degree in linguistics from the Pontificia Universidad Católica del Perú in Lima and a Master’s degree at Radboud ­University. He is currently writing a doctoral dissertation at Radboud University on variation in the Peruvian Amazonian Cahuapanan language Shawi. Konrad Rybka graduated from the University of Warsaw, Poland. He subsequently completed a Research Master and a Ph.D. in linguistics on the Arawakan language Lokono (Suriname) at the University of Amsterdam. He is currently a postdoctoral researcher at the University of California at Berkeley working on language contact between Lokono, Kari’na and Warao in the Guianas.

x 

 List of contributors

Jia (Ann) Shi 石佳 obtained her master’s degree in linguistics from Radboud University. Sophie Villerius obtained a master’s degree in linguistics from the University of Amsterdam and is currently writing a doctoral dissertation on Surinamese Javanese at Radboud University. Kofi Yakpo is Assistant Professor in linguistics at The University of Hong Kong. He obtained a PhD in linguistics from Radboud University Nijmegen, a Magister Artium in linguistics, social anthropology, and political science from the University of Cologne, and an MBA from the University of Geneva. He has published on creole linguistics, language contact, descriptive and documentary linguistics, and socio-political aspects of multilingualism.

Overview and background

Kofi Yakpo and Pieter Muysken

Language contact and change in the multilingual ecologies of the Guianas 1 Introduction This book deals with multilingualism, language contact, language change and convergence in the Guianas of South America, with a focus on Suriname. The Guianas are a very complex region. The national identity of the countries in the Guianas involves both a sense of common destiny and of multiple ethnic affiliations. In this sense it presents a condensed microcosm of the Latin American and Caribbean quest for identity that we find in works as apart geographically, if not intellectually, as José Vasconcelos’ La Raza Cósmica in Mexico (1925), José Carlos Mariátegui’s 7 Ensayos de Interpretación de la Realidad Peruana in Peru (1928), or Jean Price-Mars’ Ainsi parla l’oncle in Haiti (1928) (see also the overview of essays on ethnicity and boundaries gathered in Oostindie 1996). We have named our volume Boundaries and bridges because it reflects at the same time the maintenance of ethnic and linguistic boundaries, through the languages involved, but also the numerous instances of cross-linguistic influence across these boundaries. It illustrates the point that in the complex multilingual and multi-ethnic area of the Guianas, the languages spoken have been part of an effort of groups to keep themselves apart, as boundaries, but have also undergone numerous changes in the presence of other languages, and thus form bridges. The Guianas, or any part of them, do not form a single language community, but rather a chain of interacting and intersecting communities, which have very diverse and complex relations among themselves. Hence the term multilingual ecologies in our subtitle. However, these cases of cross-linguistic influence are very diverse in nature, and involve many parts of language. They result from different contact scenarios and include maintenance, shift, and creation. Our main focus in this book will be the Republic of Suriname, but it is useful to consider Suriname for a moment in the more global context of the Guianas. The term Guianas refers to a region bordering on the Amazon basin but straddled along the Caribbean coast of northern South America. The tropical coastal areas were discovered to be suitable for sugar plantations, as we will discuss below, and the inlands are mostly tropical rainforest, which are currently being exploited. The southern part of the Guianas is a plateau, separating the coastal plains from the Amazon basin proper. DOI 10.1515/9781614514886-001

4 

 Kofi Yakpo and Pieter Muysken

Different parts of the Guianas were occupied by five European colonial powers: from west to east these were Spain, England, the Netherlands, France, and Portugal. This has resulted in the fact that nowadays, they correspond to five different political entities: the provinces Amazonas, Bolivar, and Delta Amacuro of Venezuela, the Republic of Guyana, the Republic of Suriname, the French colony Guyane, and the state of Amapá in Brazil. The histories of these five entities are necessarily quite different, but there are a number of common threads as well: the continued presence of different Amerindian peoples particularly belonging to the Cariban and Arawakan language families, the deportation from Africa and exploitation of enslaved Africans in the seventeenth through the nineteenth century, struggles for independence as well as internal strife and some border disputes. In addition to the Amerindians, people of European, and African origin, there is also the presence of Asian-descended populations. Their immigration started around the middle of the nineteenth century. Thus, the region is thoroughly Tab. 1: Overview of the five Guianas in terms of their history and ethnic composition Venezuelan Guyana Guayana

Suriname

Guyane

Amapá

Status

Three states in Venezuela

Spanish

Republic French colony (independence in 1975) Former Dutch colony Dutch French

State of Brazilian Federal Republic

Official language Creole languages

Republic (since 1970, independence 1966) Former British colony English Guyanese English Creole, Berbice Dutch, Skepi Dutch

Guyanese French Creole, Haitian, Maroon Creole languages

Guyanese French Creole (Lanc-Patuá), Karipuna French Creole

Amerindian languages

Arawakan, Warao

Arawakan, Warao

Sranantongo, Maroon Creole languages, Haitian, Guyanese English Creole Arawakan, Cariban, (Warao)

Arawakan, Cariban, Tupí-Guaraní

Asian languages

Chinese

Guyanese Bhojpuri

Other languages

English

Sarnami, Javanese, Kejia (Hakka) Portuguese, English

Kejia (Hakka), Hmong, Vietnamese Portuguese, Spanish, Dutch

Arawakan (Palikúr), Tupí-Guaraní (Karipuna, Wayampí), Galibi (Cariban), Iapamá (isolate)

Portuguese



Language contact and change in the multilingual ecologies of the Guianas 

 5

multi-ethnic and multi-lingual. In Tab. 1 above, we give a broad overview of the five Guianas in terms of their history and ethnic composition. Suriname, the focus of the present book, is linguistically and ethnically the most diverse of the five Guianas, as we will illustrate in the following sections. Through processes of convergence, it has also become a linguistic area in its own right.

2 Linguistic areas and contact linguistics in the Guianas This collective volume looks at the dynamics of a specific type of linguistic area from a fresh perspective. Based on case studies, it strives to integrate common concepts in contact linguistics such as borrowing, contact-induced change, language maintenance and shift, creolization, koineization, genealogical differentiation, typological change, and areal convergence. The most common contact situations described in the linguistic literature involve (1) synchronic and diachronic studies of language contact with one language uni-directionally exerting influence on another (e.g. the studies in Thomason and Kaufmann 1988, Thomason 2001, Matras and Sakel 2007a, 2007b); (2) studies in areal linguistics addressing the outcomes of contact between more than two languages. Here the time-depth and lack of documentation often make it difficult to describe with more accuracy the processes leading to contact-induced changes and their directionality (e.g. Aikhenvald and Dixon 2007; Nicolai and Zima 2002). This study adds to the types of studies listed above in looking at the situation of multiple language contact, characterized by multilevel interactions between more than two languages, involving Sranantongo, Dutch, and multiple other languages, and simultaneous multidirectional change. Furthermore, it studies contactinduced change in a selection of languages that are typologically and genetically highly diverse, and describes the emergence of complex linguistic areas over a five hundred year period, taking into account both synchronic variation, and changes at various time depths, involving shifting sociolinguistic configurations. The volume offers detailed information on different grammatical domains (among which tense, mood, aspect and argument realization) across various case studies, and in doing so, allows a thorough reassessment of the two important notions of borrowability and stability of linguistic elements and structures in language contact. The studies and findings of this volume have a high potential for generalization. Multiple language contact and change involving more than two languages is little described. However, it probably represents the most common type of contact ecology in most parts of the (non-Western) world, where fast demographic and socio-economic change, and multilingualism have led to equally rapid linguistic change.

6 

 Kofi Yakpo and Pieter Muysken

This book attempts to draw the outlines of language contact and change in multilingual ecologies more clearly, and it proposes a framework for comparable studies. The case studies are centred on the Guianas in South America, and Suriname in particular. In this region, we can see the formation and transformation of a linguistic area unfolding before our eyes. The area has been recognized by anthropologists, historians, and linguists alike as an object of intense study for several reasons: –– The socio-history, culture, and demographic development of this region, within a diachronic depth of about five centuries, are exceptionally well documented; –– The Guianas, and Suriname in particular, boast a higher number of “new” languages and cultures that have arisen through language contact, than many other regions of the world; these have been the subject of very central theoretical debate. But many aspects of these languages still remain to be studied; –– Currently more than twenty genetically diverse languages are spoken in the region; –– There is an astounding typological spread with representatives of the following major linguistic groupings: Afro-Caribbean English-lexifier Creoles, Indo-Aryan, Indo-Germanic, Austronesian, Sinitic, Arawakan, Cariban; –– Language contact is multidirectional and minimally involves two main donor or target languages, namely a dominant Creole language and European colonial languages; –– The impact of the main target languages is temporally layered, as well as sociolinguistically diverse with correspondingly differentiated impacts on the domains of phonology, lexicon and morphosyntax, as well as style and register; –– We have identified various hotspots of contact and structural convergence between the languages studied and we draw numerous comparisons with other contact scenarios in the wider region and other parts of the world.

3 Multilingualism and language contact in Suriname Suriname has been the scene of complex and overlapping population movements throughout its history. In this section, we give a brief overview of how these movements have driven the development of multilingualism in the recent history of Suriname and in present times. Patterns of community-wide multilingualism have probably characterized the societies of Suriname from well before colonial conquest. Linguistic diversity in



Language contact and change in the multilingual ecologies of the Guianas 

 7

Suriname has increased significantly since the beginning of the colonial period, reaching a peak in contemporary Suriname and ushering in the type of extensive language contact that characterizes the country today. For detailed overviews of multilingual Suriname see Charry, Koefoed and Muysken 1983; Carlin and Arends 2002 and the chapter by Borges in this volume, which also includes a historical overview. Some key events in the history of Suriname with sociolinguistic significance are presented in Tab. 2. Tab. 2: Some key events in the history of Suriname and their sociolinguistic significance (taken from Yakpo and Muysken 2014) Date

Event and its demographic significance

1200–1500s Migratory movements in the Guianas 1650 Establishment of an English colony in Suriname 1652– Beginning of the deportation and enslavement of West Africans in Suriname 1665

1667

1685

1730 1804–1816 1844–1854 1863

1853

Contact-related aspects Extensive contact between Warao, Cariban and Arawak languages Varieties of English brought to Suriname

Gradual creation of an English lexicon coastal Creole language which would develop into Sranan in the latter part of the 17th century Arrival of the Portuguese Jewish Varieties of Portuguese and quite possibly planters from Brazil, possibly with Portuguese-based Creole brought to some enslaved Africans Suriname Suriname becomes a Dutch Varieties of Dutch brought to Suriname colonial possession as an elite language; speedy end to the presence of English Emergence of the Saramaccans as Creation of the Saramaccan (or Saamaka) a separate ethnic group language out of West-African languages, a Portuguese and the English lexicon pidgin/Creole Emergence of the Ndyuka as a Creation of Ndyuka out of West-African separate ethnic group languages and the English lexicon Creole English occupation English superstrate influence on Sranan lexicon limited to few words Enslaved Africans were allowed to Sranan texts created; consolidation of a learn how to read and write written register of the language Emancipation of the enslaved Increased presence of Sranan speakers in rural population and the urban centre Paramaribo dismantlement of the traditional plantation system First arrival of Chinese indentured Hakka and other Chinese languages laborers and traders brought to Suriname

8 

 Kofi Yakpo and Pieter Muysken

Tab. 2 (continued) Date

Event and its demographic significance

Contact-related aspects

1876

Dutch introduced into schools as only medium of instruction and part of universal education

1873–1916

Arrival of Indian indentured laborers

1890–1939

Arrival of Javanese indentured laborers Independence of the Republic of Suriname

The beginnings of urban Dutch-Sranan multilingualism in a slightly larger population; begin of prestige loss of Sranan. The adoption of Dutch as an L2 by increasing numbers of Surinamese causes the creation of a profoundly Sranan-influenced Surinamese Dutch. Varieties of north Indian languages brought to Suriname, such as Bhojpuri, Magahi and Maithili Varieties of Javanese brought to Suriname

1975

1990s

Economic development

Symbolic break with the former colonial power, possibility for autonomous developments in Surinamese Dutch, stronger Dutch influence on Sranan due to circular migration Netherlands-Suriname Gradual influx of Haitians, Brazilians, Chinese

3.1 Pre-colonial contact and creolization in the colonial period Taking pre- and early colonial times as a starting point, there were originally three indigenous language families represented on the territory of present-day Suriname, namely Warao, Carib, and Arawak (cf. Hoff 1995). Particularly striking is the partly convergent development within the Arawak and Cariban languages, including the creation of a sixteenth century Carib Coastal Pidgin (Taylor and Hoff 1980). Convergence must have been the consequence of multilingualism in a situation of language maintenance, probably with both Carib and Arawak enjoying similar degrees of prestige. With the beginning of the European colonization of Suriname in the seventeenth century, the linguistic situation becomes more complex. The Netherlands ends up being the sole colonial power in 1667. The establishment of a plantation economy leads to the deportation from the western seaboard of Africa and enslavement of an estimated total of approximately 350,000 Africans by the Dutch between 1675 and 1803 (Postma 1990). Various interlocking linguistic processes played a role in the emergence of the Creole languages of Suriname, among them the present-day lingua franca Sranantongo. Language creation led to the rise of early Creole varieties largely drawing



Language contact and change in the multilingual ecologies of the Guianas 

 9

on first Portuguese, then English superstrate lexicon, and as well as grammatical features from African substrate languages (cf. e.g. Huttar 1983; Huttar, Essegbey and Ameka 2007; Winford and Migge 2007). High mortality rates under the brutal laboring conditions on Dutch-owned plantations made it impossible for the enslaved African population to replenish itself through natural growth (Arends 1995). Therefore most sources agree that creolization in Suriname must have been gradual, involving a long period of multilingualism in the emerging Creole, and African and European languages (Selbach, Cardoso and Van den Berg 2009). Language creation must therefore have been accompanied both by gradual language shift (to the Creole and for some Dutch) by Suriname-born Africans as well as maintenance of African languages among African-born Africans and Surinameborn children. African languages have only survived into the present in a fossilized form in the ritual languages Kumanti, Ampuku and Papa (Thoden van Velzen and van Wetering 1988; Thoden van Velzen and van Wetering 2004). The Creole languages of Suriname, however, thrived and have differentiated into the three distinct clusters of Sranantongo, Western and Eastern Maroon Creole (Smith 1987; Smith 2002). Amongst these, Sranantongo has spread beyond the coastal belt into the interior to become the most-widely spoken Creole of the country. The indigenous languages of Suriname have undergone quite fundamental contact-induced changes since colonization as well, both through contact with each other (Carlin 2006, this volume) as well as with Sranantongo and Dutch (Rybka, this volume).

3.2 The abolition of slavery and the Asian languages of Suriname The full abolition of slavery in 1873 after a transitional period of ten years of forced labor prompted the Dutch colonial regime to “import” indentured laborers from Asia, as in other plantation economies throughout the Caribbean and elsewhere in the colonial world in order to substitute for slave labor (Saunders 1984; Kale 1998). Through these arrangements, a total of about 30,000 (male and female) laborers were transshipped to Suriname from northern India between 1873 and 1916 (Damsteegt 1988: 95). A total of about 30,000 laborers arrived from Java (Indonesia) between 1890 and 1939 (Bersselaar, Ketelaars and Dalhuisen 1991). A third, much smaller wave of migrants arrived from Guangdong province of southern China from the 1850s onwards as laborers and traders, numbering only about two thousand but constituting an important community in economic terms (Fat 2009: 52). These migratory movements brought about a fundamental transformation of the previously established demographic constellation in Suriname. A country

10 

 Kofi Yakpo and Pieter Muysken

with a largely African-descended population with relatively small Indigenous American and European components in the mid-nineteenth century had acquired an Asian-descended population numbering nearly half the size of the population by the turn of the twenty-first century. Hence in the 2004 national census about 27% of the total Surinamese population of half a million self-identifies as “Hindoestaans” (Indian-descended) and 15% as “Javaans” (Javanese-descended) while the category “others” of 6% subsumes amongst others the Chinese-descended population and the Indigenous peoples of Suriname. Self-identified “Kreolen” and “Marrons” (both African-descended) Surinamese make up 18% and 15% respectively of the population. The substantial number of Surinamese who self-classify themselves as “mixed” (12%) or leave their ethnicity unreported (6%) is indicative of a growing proportion of Surinamese either claiming a mixed heritage of various constellations or rejecting ethnic labelling altogether. The migratory mass movements of the indenture period have been equally transformative for the linguistic situation in Suriname as they have been for the demography of the country. Various northern Indian language varieties merged to form the koiné Sarnami, the community language of the Indian-descended population of Suriname (Damsteegt 1988, Yakpo, this volume). Besides change due to contact with Sranantongo and Dutch, some degree of koineization also affected the Javanese language since it arrived in Suriname (cf. Vruggink 1987). This is probably also due to the fact that a small but not insignificant part of the “Javanese” population of Suriname had its origins elsewhere in the Indonesian Archipelago than Java (Gobardhan-Rambocus and Sarmo 1993). Contrary to Sarnami, there are indications that Javanese is not as vital anymore as it still was in the second half of the twentieth century and that there is an ongoing language shift, particularly by speakers below twenty to Sranantongo and Dutch. The language of the Chinese community was, for a long time, chiefly Hakka (also called “Kejia”). But Cantonese and more recently Mandarin have played important roles as prestige languages within the community and there is an ongoing language shift to Sranantongo and Dutch (Tjon Sie Fat 2002).

3.3 Sranantongo and Dutch as lingua francas Sranantongo and Dutch play a special role in Suriname: they are the only languages extensively used outside of their traditional speaker communities (principally the Afro-Surinamese population of the coastal belt). Within the four hundred years or so since its creation by enslaved Africans on the European plantations of Suriname, Sranantongo has evolved into a multi-ethnic dia-system used as a



Language contact and change in the multilingual ecologies of the Guianas 

 11

lingua franca by the ethnically diverse population of the coast. The language has also made inroads into the interior where it shares a common space with various Maroon Creoles (Migge 2007; Migge and Léglise 2011, 2013) and Indigenous languages. Sranantongo served as the primary donor of lexical material to the Asian languages of Suriname during the indenture period, when knowledge of Dutch was not yet as widespread within these communities as it now is. Nowadays Sranantongo plays the role of a donor language together with Dutch. Sranantongo is the only language of Suriname that virtually every Surinamese has at least some knowledge of, however in growing competition with Dutch. It should be pointed out that the expansion of Sranantongo is solely a consequence of an incremental growth because the language has not benefited from state support of any kind whatsoever since it was abolished as a language of instruction in 1876 (cf. Tab. 2). This stands in stark contrast to the development of Dutch, which has also witnessed a considerable growth in speaker numbers throughout the 20th century due to sustained institutional and elite support. Since colonial times, Dutch has been the sole language of government business and parliamentary affairs, and the de facto language of education at the primary, secondary and tertiary levels. It has remained the language of upward social mobility and high prestige and is extensively used by officialdom and by coastal Surinamese in a variety of registers. One of the consequences of this disposition is that Dutch has witnessed a fundamental transformation within the last hundred years or so. From being a language of the colonial administration and a relatively small Dutch-educated elite, it has been appropriated by larger sections of Surinamese society. In the process, Dutch has engaged on a trajectory of its own and today plays an important role as a donor language to Sranantongo and other languages of Suriname. At the same time, Dutch has itself become a recipient language for lexical (cf. De Bies, Martin and Smedts 2009) and structural borrowing from Sranantongo (De Kleine 1999). Our sociolinguistic interviews show widespread competence in (varieties of) spoken Dutch with Surinamese of diverse class backgrounds hence beyond the traditional patterns of upper and middle class use of Dutch inherited from the colonial period. Together with Sranantongo, Dutch is also a target for language shift from traditional community languages such as Javanese, Sarnami and Hakka. The hierarchical superposition of Dutch to Sranantongo and the other languages of Suriname is being driven by a similar set of ideological, political and economic factors as in other postcolonial societies (cf. Omondi and Sure 1997; Heine 1990; Veiga 1999 for the status quo of colonial and African languages in African nations). The widespread assumption and acceptance of the “superior” status of Dutch in Suriname is reflected in often negative and self-denigrating

12 

 Kofi Yakpo and Pieter Muysken

attitudes of speakers towards the non-European languages they speak and in the corresponding language practices. However, the social and functional division of labor between Dutch and Sranantongo outlined above has also led to Sranantongo enjoying a large amount of covert prestige. In many contexts, using Sranantongo is an act of identity assertion, defiance and resistance against norms transmitted through Dutch, with all its problematic associations with elitism, the colonial past and a post-colonial present.

3.4 Contemporary data on multilingualism in Suriname Determining the size of speaker communities in present-day Suriname is not easy in the absence of a comprehensive linguistic survey. Chapter 2 by Borges provides a detailed overview of the highly complex and still rapidly changing current situation. Sranantongo and Dutch constitute the two main axes of multilingualism. These two languages show the highest total percentages of self-reported “most often” and “second language” uses. At the same time they manifest the largest differences between “most often” and “second language” uses. The differences in social function between these two most widely spoken languages of Suriname transpire in the significant differences in percentage of “most spoken”. The percentage of 9% for Sranantongo for “language spoken most often” is surprisingly low, particularly in comparison to an equally surprisingly high score of 46.6% for Dutch. We attribute these percentages to prevailing language attitudes in Suriname that result from the functional and prestige differences between these languages referred to in the preceding section. Hence the high prestige of Dutch leads to over-reporting of use as “language spoken most often”, while the, the low prestige of Sranantongo leads to underreporting of use as a primary language. As for the other languages listed in Tab. 2, the lower percentages in the “second language” column seem to point to these languages largely functioning as in-group “ethnic” languages. For Sarnami for example, the relation of “most spoken” (about 75% of the total) and “second language” (about 25% of the total) may well be indicative of a partial language shift to Dutch and Sranantongo, or at least a certain decline in use. The same holds for Javanese. We have seen that language creation has been of primordial historical importance for the rise of linguistic diversity in the country. In the present context, we find the maintenance of community languages alongside language shift to the two dominant languages, Sranantongo and Dutch.



Language contact and change in the multilingual ecologies of the Guianas 

 13

4 Data collection methods in the present volume The studies reported on this book rely for the largest part on field data collected in Suriname in 2011–12 as part of the ERC project “Traces of Contact” at the Centre for Language Studies at Radboud University Nijmegen. The corpus contains recordings in eight Surinamese languages: The Creole languages Sranantongo, Ndyuka, Kwinti and Saramaccan, as well as Sarnami, Surinamese Javanese, Surinamese Hakka and Surinamese Dutch. Comparative data has been collected in India, the Netherlands, West Africa and Mauritius. The corpus consists of a total of about hundred and fifty hours of data, of which the recordings of Sarnami and its control groups in India (Awadhi, Bhojpuri, Maithili, Magahi) and Mauritius (Mauritian Bhojpuri) make up about thirty hours. All unreferenced examples in this paper stem from our own field data. The chapters of Carlin, Rybka, and Migge rely on data collected in the course of many years of field research in the Guianas. The data was collected according to a unified methodology in order to allow comparison across varieties and languages. Data collection methods involved the use of broad (story-based) and narrow (video clip-based) visual stimuli on the one hand and (semi-)structured interviews on specific topics on the other. Elicitation was complemented by recordings of natural discourse. In Suriname, we also conducted about fifty sociolinguistic interviews in Sranantongo on the backgrounds of speakers and their attitudes vis-à-vis the languages they speak. We are much in favor of approaches employing quantitative analyses based on large diachronic and synchronic corpora in order to differentiate between codeswitching and borrowing, as well as between “normal” variability and contact-induced change (e.g., Van Hout and Muysken 1994; Poplack, Zentz and Dion 2012). However, when working with less documented languages, as in the case of Suriname, one is in a less fortunate position. There is a lack of sizeable corpora of diachronic data for all languages but Sranantongo (cf. http: //suca.ruhosting.nl), and the collection and handling of even modest corpora of synchronic data involves considerable efforts. It seems then that only a mixed strategy is feasible. This involves quantitative investigations based on smaller corpora and extrapolation based on in-depth morpho-syntactic investigations of particular structural areas.

5 Theoretical background Language contact and mutual borrowing of lexical items and structures in the languages of Suriname is a consequence of widespread multilingualism. We will therefore first review some of the concepts related to language contact and multilingualism

14 

 Kofi Yakpo and Pieter Muysken

that we will be referring to. The typology of language contact that Thomason and Kaufman (1988) propose provides for three principal contact scenarios. We define scenario as the organized fashion in which multilingual speakers, in certain social settings, deal with the various languages in their repertoire. In the maintenance scenario the language that borrows (henceforth the recipient language), from another language (henceforth the donor language) continues to be spoken by its speaker community, i.e. it is maintained. The literature shows that there is a large range of variation in maintenance scenarios. In some cases of maintenance, the recipient language may undergo more moderate lexical and structural transfer from a donor language. Other cases of maintenance show extensive transfers of phonological features, lexical material and structural patterns (e.g. Hainan Cham, whose Austronesian typological profile has been significantly altered due to contact with Sinitic, cf. Thurgood and Li 2003). The classification of a scenario as involving maintenance may also be theory-dependent. For example, a strong position on relexification – i.e. the mapping of one language’s semantic and morphosyntactic properties onto another’s phonological shapes – may in fact be seen as an extreme case of maintenance of the language providing the semantic and morphosyntactic content. Such a position is implicit in Lefebvre’s (1993, 1998) interpretation of the rise of Haitian Creole. In the second scenario suggested by Thomason and Kaufman (1988), shift, a community leaves behind its traditional language and shifts to another language, usually due to the socio-economic and/or political dominance of the community speaking the language shifted to. Contact effects in shift scenarios may be very similar to those encountered in maintenance scenarios. Studies have shown that intermediary stages of language shift and obsolescence (cf. e.g. the case study in Aikhenvald 2012) show the same kind of heavy structural and lexical borrowing that may characterize maintenance scenarios in which a recipient language is not threatened by language loss (for an illustrative example, cf. Gómez-Rendón 2007). A principal difference between shift and maintenance is pointed out by Van Coetsem (2000): Language shift involves a change in directionality of borrowing (termed “agentivity” by Van Coetsem) between a recipient language and a source language. Hence during a shift, contact effects chiefly manifest themselves through structural rather than lexical influence from a shifting community’s traditional language which is usually still spoken alongside the dominant language by some proportion of the shifting community. In a maintenance scenario, however, the traditional language of the community remains the dominant language and lexical borrowing is usually far more common, than or at least as common as structural borrowing from the donor language. The distinction is



Language contact and change in the multilingual ecologies of the Guianas 

 15

also relevant for Suriname, which features a range of maintenance scenarios of varying depth or extensiveness of contact. The third major scenario proposed by Thomason and Kaufman (1988) involves the creation of new linguistic systems composed of elements of contributing languages. Creolization as one type of language creation is particularly important in the linguistic trajectory of Suriname. In the Surinamese creolization scenarios, European superstrate languages (English and Portuguese) provided most of the lexicon while several African substrate languages provided some lexicon and substantial parts of the grammatical and phonological systems. Next to genetic inheritance from contributing languages, creolization in Suriname also seems to have involved various degrees of restructuring of the input languages driven by linguistic-cognitive factors – the respective contribution ascribed to either of the two factors being subject to theoretical preferences (Alleyne 1980; Lefebvre 1998; McWhorter 2005; Bickerton 2009). However, in the context of Suriname, other scenarios play a role as well. Language creation in Suriname concerns not only creolization but also koineization as diachronic and synchronic processes. We understand koineization as a less pervasive type of language creation in that there is less restructuring of the input languages involved in the creation of the koiné, as has been amply observed in cases of dialect contact (cf. e.g. Auer 1998b and the classic study of the rise of the Indic koiné of Fiji by Siegel 1985, 1987). The literature suggests that typological proximity and mutual intelligibility are the chief linguistic reasons responsible for the more modest restructuring of an interlanguage or koiné with respect to its input languages (cf. e.g. the studies in Braunmüller 2009; Kühl and Petersen 2009). Stable multilingualism over some generations, as in Suriname, can lead to structural convergence between the various languages spoken in the same geographical space (Winford 2003). In the process, the languages in contact may become more similar by mutual accommodation, i.e. bidirectional change, for example by adopting a compromise on the basis of already existing common structures. In this paper, we employ “convergence” in a broader sense, however, as a cover term for the multiple contact scenario characteristic for Suriname. Here, borrowed structures may stem from the two dominant donor languages Dutch and Sranantongo simultaneously, and these two languages may interact in their influence on a recipient language. Due to this circumstance, it is often difficult to attribute instances of contact-induced change in a language like Sarnami to a single source. In our classification of contact phenomena in the Surinamese languages we rely on models that differentiate between the borrowing of a specific form or matter (morphemes and their phonological shapes) and structure or pattern (morphosyntactic and semantic structures without the corresponding forms).

16 

 Kofi Yakpo and Pieter Muysken

The latter phenomenon has been also been referred to in the literature (with varying degrees of overlap in meaning) by terms like “calquing” (Haugen 1950), “metatypy” (Ross 1996), “grammatical replication” (Heine and Kuteva 2003; Heine and Kuteva 2010), “pattern replication” (Matras and Sakel 2007; Sakel 2007), “rule borrowing” (Boretzky 1985) and last but not least “relexification” (e.g. Muysken 1981; Lefebvre 1993). We also refer to the code-copying model proposed by Johanson (1992). Such approach not only allow operationalizing these two fundamental types of borrowing, it also allows yet finer distinctions of structural borrowing. Borrowing of patterns allows us to differentiate for example, between the replication of lexical versus grammatical structures. It may also encompass cases of partial replication in which a donor language pattern undergoes adaptation, i.e. is grammaticalized to fulfill functions in the recipient language that differ to some degree from those attested in the donor language (Heine and Kuteva 2003; Meyerhoff 2009). The differentiation between the borrowings of forms (matter) and structure (pattern) also leaves room for identifying combinations of matter and pattern borrowing, in which a form and its morphosyntactic and semantic specifications are carried over into another language. As we move on, we will see that both types of borrowing and combinations between them can be found in our Surinamese corpus. Before moving on to the next section with the chapters in this book, we wish to point out that we share the general understanding that the outcomes of multilingualism and language contact are of course not solely determined by linguistic factors. Socio-economic, political, cultural and demographic factors, the timedepth of cultural and linguistic contact between communities and so forth, are at least as important in fashioning the processes and outcomes of contact between languages (Myers-Scotton 1993; Roberts 2005; Gómez-Rendón 2008). Our focus in this paper is on the linguistic as much as the extra-linguistic factors of contactinduced language change in Suriname.

6 Chapters in the present volume In The people and languages of Suriname, Robert Borges reviews the various contact processes in the multilingual Guianas. Five processes are considered: (1) the partly convergent development within the Arawakan and Cariban languages (2) the introduction of the languages of three competing colonial powers: English, Dutch, French, and Portuguese; (3) the introduction of West African Gbe languages and languages of the Central African Kikongo cluster with enslaved Africans.



Language contact and change in the multilingual ecologies of the Guianas 

 17

This led to the emergence of the various Creoles; (4) after the abolition of slavery indentured laborers were brought in, speaking northern Indian languages, Kejia, Cantonese, and Javanese; these languages underwent leveling; (5) with increasing regional migration in recent times speakers of Guyanese Creole, Haitian Creole, Brazilian Portuguese, and Mandarin have come to Suriname. Currently Surinamese Dutch is the dominant language, transformed from a metropolitan standard language to a local interethnic urban variety. Kofi Yakpo presents a systematic overview, in Creole in transition: Contact with Dutch and typological change in Sranantongo, of how the expression of spatial relations and grammatical relations in Sranantongo has undergone typological change from a more West African typology inherited from the substrates to a more Dutch-like system. Sranantongo is the only language of Suriname spoken to some degree by the vast majority of the Surinamese population, irrespective of their class, ethno-linguistic and regional background. Robert Borges, in The Maroon Creoles of the Guianas: Expansion, contact, and hybridization provides an account of the historical developments, up to the present, of the Maroons and Maroon languages in Suriname and French Guiana. Following the arrival of enslaved Africans, groups of individuals fled their captivity, taking creole varieties with them as they established independent communities outside the plantation area. The relative isolation of the Maroon groups provided a setting in which the Maroon languages diverged substantially from the Creole varieties associated with the plantation area. This isolation came to a gradual end, leading to increased contact between Maroon varieties on the one hand, and Sranan and Dutch, on the other, and to the leveling of Maroon ­varieties. In Out of India: Language contact and change in Sarnami (Caribbean Hindustani), Kofi Yakpo describes how Sarnami of Suriname, the only IndoAryan koiné of the Caribbean that still enjoys a stable speaker community, has diverged from its north Indian contributing languages due to contact with Sranan and Dutch. The study shows that head-initial order has increased or supplanted head-final order inherited from Indo-Aryan in some domains (i.e. SVO, AuxV, NRel), while other domains have remained head-final (e.g. NAdp, AdjN). Sophie Villerius, in Developments in Surinamese Javanese, shows how Javanese was brought to Suriname during the colonial labor trade within roughly the same period as Sarnami and still serves as a community language to a sizeable portion of the Javanese-descended population of Suriname. This chapter addresses effects of contact with Dutch and Sranan that have so far remained unstudied. In an exploratory paper Luis Miguel Rojas-Berscia with Jia Shi in an exploratory paper entitled Hakka as spoken in Suriname, analyze recordings of Hakka or Kejia Chinese. An ethnic Chinese population began constituting itself in

18 

 Kofi Yakpo and Pieter Muysken

­ uriname in the mid nineteenth century, when the Dutch colonial regime started S importing Asian indentured labor as a substitute for African slave labor. This first cohort of Chinese migrants largely stemmed from Hakka/Kejia-speaking villages in the Fuidung’on Region. In the 1960s, a second wave consisting of acculturated Fuidung’on Hakka chain migrants joined the Chinese population already present in Suriname via Hong Kong. The third and latest migratory wave began in the 1990s after the People’s Republic of China (PRC) eased restrictions on emigration. The paper Cariban in contact: New perspectives on Trio-Ndyuka pidgin by Sergio Meira and Pieter Muysken, analyse the Trio-Ndyuka pidgin of the interior of Suriname, and situate the pidgin in the context of Carib/non-Carib ­contacts during the last five or six centuries. In earlier important work by Huttar and Venantie (1997) the pidgin was looked at from the Ndyuka perspective. Here the Carib perspective is focused upon. In Language contact in Southern Suriname: The case of Trio and Wayana, Eithne Carlin presents an in-depth overview of different types of language contact phenomena involving the Amerindians of southern Suriname and nonAmerindian groups, over a period of approximately 400 years. Overall, most of the language contact has resulted in lexical borrowing, but this process has been quite complicated. Konrad Rybka, in Contact-induced phenomena in Lokono (Arawakan) shows that a number of contact-induced phenomena have affected a variety of Lokono spoken in Suriname. Setting out from an account of the different layers of old nominal borrowings, it moves on to focus on synchronic examples involving hybrid verbal paradigms. This category shift is a sign of a new contact situation in the history of Lokono – the widespread multilingualism in Lokono, Sranantongo and Dutch makes it possible for the languages to exert influence on one another far beyond simple lexical borrowing. Pieter Muysken, in The transformation of a colonial language: Surinamese Dutch analyzes the structural evolution of Dutch in Suriname since the mid seventeenth century. Historical records show that the language was used by Dutch colonists, and albeit fewer, free and enslaved Africans alike from the very beginning of its implantation in Suriname and the Dutch Antilles. In the course of this development, Surinamese Dutch has also undergone dramatic changes due to substratal influence, mainly from Sranantongo. Robert Borges, Pieter Muysken, Sophie Villerius and Kofi Yakpo describe in The tense-mood-aspect systems of the languages of Suriname how the expression of Tense, Mood, and Aspect (TMA) has received a great deal of scholarly attention by contact linguists, in particularly by those interested in pidgins and creoles. Thus there is a wealth of data on the expression of TMA in monolingual varieties of these languages, and on how these expressions emerged, but we know



Language contact and change in the multilingual ecologies of the Guianas 

 19

very little about their behavior in multilingual practices. This chapter compares the use of TMA markers in several types of contact languages (creoles, koinés, Surinamese Dutch) in their respective monolingual and multilingual settings. Bettina Migge in From grammar to meaning: Towards a framework for studying synchronic language contact, explores the dynamic nature of language contact. Taking a language or system-based perspective, research on language contact tends to focus on describing the structural effects of language contact on particular structural sub-domains. Contact mechanisms and processes are inferred from macro-social data and structural linguistic comparative data alone. Such an approach creates homogenizing and one-dimensional contact scenarios. This chapter critically examines the viability of this approach for understanding synchronic contact settings, and drawing on contact between Maroon Creoles, Srananatongo, Dutch and other languages, shows how people make use of language contact to negotiate different social and interactional meanings. In the final chapter Multilingual ecologies in the Guianas: Overview, typology, prospects, Kofi Yakpo and Pieter Muysken discuss to what extent and in which way the grammatical systems of the various languages covered in this volume have converged. Are the changes these languages have undergone the result of contact or are they motivated by other factors? How stable are specific structural features and areas in this particular contact setting? The crucial aspects of borrowability and stability of linguistic features during contact are also addressed here.

Robert Borges

The people and languages of Suriname1 1 Introduction Like most places in the western hemisphere, the past five hundred or so years has brought the territory now known as Suriname drastic demographic changes under the administration of multiple colonial powers. With a number of Amerindian peoples occupying the territory, Suriname was already alive with multiple civilizations and complex intergroup relations when the Spanish claimed the territory in 1499. The place remained a sort of no-man’s land, as far as Europeans were concerned, leaving the territory open for claims by other powers. Various unsuccessful attempts at settlement were made by different groups of Europeans, but the first successful settlement is attributed to the English Lord Willoughby, who equipped a group to relocate from Barbados to Suriname in 1651. Willoughby’s settlement was conquered by a fleet of ships from Zeeland led by Abraham Crijnssen in 1667 and was kept following a tentative end of hostilities between the English and the United Dutch Provinces with the treaty of Breda later that year. Except for another brief period of British occupation at the beginning of the nineteenth century, Suriname remained a Dutch possession until its independence in 1975. Settlement in Suriname was heterogeneous; it consisted of not only English and Dutch, but also other Europeans, such as French, Portuguese, and Germans, a strong contingent of Sephardic and Ashkenazi Jews, and of course, African slaves. These groups’ interactions laid the foundations of diversity in Suriname, which was subsequently augmented by Indian, Indonesian and Chinese groups, leading to varying cultural and linguistic practices in modern Suriname. This chapter provides an overview of the historical events and migratory processes that have culminated in the demographic makeup, language practices, and language attitudes in Suriname as we know it.

1.1 Country profile Suriname is the smallest sovereign territory in South America, with 163,820 km2 (63,251 mi2) (ABS 2009). To the north, Suriname’s coastline is adjacent to

1 A previous version of this text has appeared in Borges 2014. DOI 10.1515/9781614514886-002

22 

 Robert Borges

the Atlantic Ocean. The country shares borders with Guyana to the west, French Guiana to the east, and Brazil to the south. There are two disputed land areas, the New River Triangle between the New and Kutari Rivers, where both Suriname and Guyana stake claim, and the area between the Litani and Marowini Rivers, claimed by both Suriname and French Guiana. Suriname and Guyana also dispute the placement of their border in relation to the Corentijn River (currently on the western bank of the river, rather than the thalweg, or deepest central point), which also delimits the boundary between the two countries’ maritime territory. Although international courts have ruled in favor of Guyana’s maritime claims, all these disputes persist (Menon 1978; Donovan 2003; Aizenstatd 2011). At just a few degrees from the equator, Suriname is host to a warm climate (temp. range 2008, 23.1C (73.5F) – 30.7C (87.3F) (ABS 2009)), though land formation features, rather than climate, are responsible for ecological and forest diversity in Suriname. Although the majority of the country’s interior is covered by tropical forests, there is also a savannah belt with open grasslands and dry/ deciduous forest. The old coastal plain consists of sandy ridges, wetlands, and swamps. Mangrove forests, mudflats, estuaries, and coastal beaches are found on the young coastal plain (see also Biodiversity Steering Committee 2006: 5; Namdar 2007: 2 for more further details). A number of river systems cross Suriname. Flowing from south to north, and to the west in the north of the country, river systems tend to act as means of transportation and communication by indigenous peoples, rather than as borders as by colonial authorities. The population of Suriname is just over half a million people (517,052 as of mid 2008 (ABS 2009), and 539,910 in 2011 (ABS website)) and the overwhelming majority of the population is settled on the coastal plain. Despite its relatively small population, Suriname is characterized by a high cultural and linguistic diversity. Apart from the native Amerindian population, Surinamese people trace their roots to places in Europe, Africa, India, Indonesia, and China. The movements and circumstances surrounding the populating of the Surinamese territory and the various groups’ linguistic contributions are summarized in the following section. Section 3 provides a discussion of language attitudes and langauge practices among Surinamese people.

2 History and the influx of the country’s people and languages Much work has been done on the history of Suriname, both generally and of several individual peoples in particular. The following overview is meant



The people and languages of Suriname 

 23

Map 1: Suriname and disputed territories, retrieved from Wikipedia Commons http://upload.wikimedia.org/wikipedia/commons/e/eb/Suriname1991_Karte_umstrittene_ Gebiete.jpg July 19, 2013

to acquaint the unfamiliar reader with the origins of the ethnolinguistic diversity present in the country. Much of what is presented here is covered in greater detail elsewhere, and the curious reader is urged to consult the sources referred to for additional information. The following subsections will

24 

 Robert Borges

focus on the particulars of individual groups of people and their contribution to Suriname’s linguistic landscape insomuch as possible. These will be presented in a rough chronological order. As the reader will notice however, so much of each group’s developments are dependent on the others, such that discussing them in complete isolation would not portray a realistic picture of historical developments in the territory.

2.1 Amerindians Details of pre-Columbian Amerindian civilizations in the Guianas are not well understood. Much of what is known about the early Amerindians comes from accounts of explorers and early settlers in the area. Archaeological evidence tells us that Arawakan groups were engaging in some subsistence agriculture in the area at least a millennium before the Spanish began exploring the area (Janssen 1980). At the time of the Spanish arrival to the Guianas, another group, the Caribs, had apparently recently arrived in the area, hostilely taking over other groups’ territory and resources (Boven and Morroy 2000: 377). In the seventeenth and eighteenth centuries, several coastal groups were found in the Guianas, some of which also occupied neighboring areas in South America and the Caribbean: Arawakan groups – Lokono, Sepayo; Cariban groups – Kari’na, Parakoto, Yaos, Nepuyo; isolate – Warao (Carlin and Boven 2002: 18). Early attempts by the Spanish to settle the Guianas, e.g. Cayenne in 1568, were met with hostilities and stopped by the Kari’na the Parakoto. Subsequent attempts made by the Dutch, English, and French were also thwarted by Amerindians, e.g. English and French settlement destroyed in 1645 by Kari’na (Buddingh 1995: 10; Carlin and Boven 2002: 16–19). Soon after, however, it became clear to some Amerindians that, unlike the Spanish, non-Spaniards were primarily interested in trade rather than the spread of religion and subjugation of natives. A commercial relationship ensued, whereby all coastal Amerindian groups engaged in trade with the English, and later, the Dutch. Amerindians provided goods such as wood, hammocks, wax, balsam, spices, and slaves in exchange for firearms, cloth, machetes, knives, fishhooks, combs, and mirrors (Nelemans 1980a: 21; Carlin and Boven 2002: 17). In 1651, two Kari’na chiefs travelled to Barbados to negotiate with the English about their settlement plans in the Amerindians’ territory (Carlin and Boven 2002: 18). The English would then be able to establish themselves in what would be the first permanent settlement in what would become Suriname (see section 2.2). As the English laid the foundations for what would eventually become large-scale sugar cultivation, the Amerindians also used Europeans as pawns



The people and languages of Suriname 

 25

in their own conflicts amongst each other, and especially as advantageous allies in their own strife with the Spanish. The English were aware of their precarious position in the new settlement and did all they could to appease all groups of Amerindians in order to secure their economic interests. However, their efforts were hardly sufficient. The Amerindians became so dissatisfied with the colonists that just as the Dutch were taking over the colony, they began a series of ‘fierce’ attacks against the colonists in an effort to drive them out completely. These Indian Wars would have likely ruined the colony if not for van Aerssen van Sommelsdyk’s peace efforts which culminated in a 1684 treaty declaring all Kari’na, Arawak, and Warao to be free and unenslaveable (Arends 2002: 121; Carlin and Boven 2002: 19). Contact with Europeans and African slaves was not without consequence for the Amerindians. Amerindian settlement was much more extensive in the seventeenth century than it is today. Disease was a major factor in depopulation of Amerindian groups and their movement away from coastal areas (Carlin and Boven 2002: 15, 19, 20). Although the Dutch colonial officials often called upon Amerindians to hunt runaway slaves, those who escaped to the forests en masse in the early eighteenth century drove Amerindian groups further south as they established their own free societies (Carlin and Boven 2002: 16, 29). By the Mid eighteenth century, Maroons had reached the Tapanahony River and invited the Wayana to relocate further downriver in order to facilitate trade (Carlin and Boven 2002: 23; Thoden van Velzen and Hoogbergen 2011). By the end of the eighteenth century, Maroons had completely monopolized commerce with Amerindians, blocking Europeans’ access to them (Carlin and Boven 2002: 21, 26). A recurring characteristic of Amerindian societies, at least among those in the Guianas, is seemingly constant migration, regroupment, and amalgamation of smaller groups (Carlin and Boven 2002: 16). Naturally, this makes tracing the history, migratory paths, and settlement patterns of individual non-literate peoples notoriously difficult. Several groups mentioned above, such as the Warao, died out in the twentieth century and are no longer found in Suriname. Most recently, beginning in the 1960s, several of the Cariban groups of the interior have conglomerated on the Sipaliwini River. Though they continue to make ethnic distinctions among themselves, Akuriyo, Sikïiyana, Tunayana, and Mawayana all speak the Trio language and identify as Trio to non-Amerindians, leaving those languages moribund (Carlin and Boven 2002: 37). Surinamese Amerindian peoples and their approximate location as of the year 2000 are presented in Tab. 1 (Carlin and Boven 2002: 37). In fact, none of the Amerindian languages are faring well in the face of pressure to use other languages like Dutch and Sranan. As far as we can tell,

26 

 Robert Borges

Tab. 1: Surinamese Amerindian peoples and their approximate location. Arawakan Cariban

Lokono Mawayana Kari’ña Trio Wayana Akuriyo Sikïiyana Tunayana / Katuena

Coastal area Sipaliwini River Coastal area Palumeu, Tapanahony, and Sipaliwini Rivers Palumeu, Tapanahony, Marowijne, and Lawa Rivers Tapanahony and Sipaliwini Rivers Sipaliwini River Sipaliwini River

Amerindian languages have been and continue to be in-group languages in Suriname. Except for a trade pidgin that developed in the eighteenth century between the Trio and Ndyuka Maroons (Huttar and Velantie 1997; Borges 2015; Meira and Muysken, this volume), Amerindian languages tend to be recipients, rather than sources, of linguistic material in their contacts with Maroon languages, Sranan, and Dutch (see Carlin, this volume, for a discussion of borrowing patterns between Amerindian languages).

2.2 The English The English are credited with establishing the first permanent settlement in the territory that would become Suriname, though their administrative authority in the area did not last even two decades. They had already begun exploring the Guianas in the sixteenth century and throughout the first half of the seventeenth century, the English competed with the Dutch and the French to secure an economic foothold in the Guianas (Sijpesteijn 1854; Goslinga 1971; Arends 2002: 115; Carlin and Boven 2002: 16; van der Meiden 2008). In 1630, for example, sixty Englishmen under Captain Marshall settled along the western Bank of the Suriname River, just upstream from modern Domburg, to grow tobacco. They reported that there were already some Portuguese or French in the area (Sijpesteijn 1854: 3). The settlement did not last long, and Marshall tried again in 1643 without success (Arends 2002: 115). It was not until 1651 that a group of settlers arrived in the area of Paramaribo. Lord Willoughby, then governor of Barbados, equipped two ships with his own funds and settlers from Barbados, where they established themselves close to remnants of previous European settlement attempts and with the permission of the local Amerindians (Sijpesteijn 1854: 7; Carlin and Boven 2002: 18; van der



The people and languages of Suriname 

 27

Meiden 2008: 19). Though the English settlers came from Barbados, it is likely that they originated in, or at least came via other English colonies in the Caribbean, such as St. Kitts, Nevus, and Montserrat (Arends 2002: 117). The English brought an established method of sugar cultivation and experienced planters and slaves from their other colonies in the Caribbean. They began to set up sugar growing operations right away without a typical phase of ‘settlement society’ (Arends 1995: 237). Both indentured workers and African slaves provided labor in the English period, though there was still a relatively high ratio of Europeans to Africans between 1651 and 1680, from 1:1–1:3 (Arends 1995: 259; Migge 2003: 28). The English were augmented by a number of settlers of other European nationalities. In 1654, for example, a group of French was displaced from Cayenne by Amerindians; they were well received by Europeans in Suriname, which consisted of approximately 350 English at the time (Sijpesteijn 1854: 8; Arends 1995: 259). The settlement consisted of a large proportion of people with different nationalities, such as Dutch, German, and French (van Lier 1971: 38; van der Meiden 2008: 13), which resulted in a range of non-standard, L2 varieties and pidgin varieties of English among planters and laborers (Migge 2003: 29–30; van den Berg 2007: 6). By 1663, the population had grown to approximately 4,000, slaves included (Goslinga 1971: 425; van den Berg 2007: 5; van der Meiden 2008: 22). Also in that year, the settlement was declared a new colony “Willoughbyland”, whereby planters would be required to pay a certain percentage of goods to England (Meiden 2008: 19–20). As a result, several hundred people left the colony by 1665, which by now had 40–50 productive sugar plantations (Sijpesteijn 1854: 8; van der Meiden 2008: 21–22). The new policies combined with infighting among the planters, slave uprisings, marronage and Amerindian revolts brought the new colony to a state of chaos, leaving it vulnerable to aggression by other European powers (Carlin and Boven 2002: 19). In February of 1667, Paramaribo was besieged by several ships and nearly 400 soldiers from the Dutch province of Zeeland, commanded by Abraham Crijnssen. When the English refused orders to surrender, Crijnssen and his soldiers attacked abruptly. They allowed the remaining English soldiers to leave and settlers to either remain neutral in any further conflict with the English or to sell their property and leave (Sijpesteijn 1854: 8–10; van Lier 1971: 19). Willoughbyland was officially traded for Nieuwe Nederland (New York) and renamed Suriname with the signing of the Treaty of Breda in July 1667. This arrangement was later ratified with the Treaty of Westminster in 1674 (Goslinga 1971: 426). English settlers who chose to stay were guaranteed equal rights in the colony under Zeelandic administration (Sijpesteijn 1854: 8–10).

28 

 Robert Borges

Despite the obstructive policies of the new Dutch administration, many English did choose to leave, taking some of their slaves purchased before 1667 with them (Smith 2010). Between 1668 and 1680, approximately 600 English and 1,700 slaves left Suriname for Barbados and Jamaica. Another 39 English planters stayed behind after the treaty of Breda and some of those who left later returned (Arends 1995: 236). It is also a possibility that ‘new’ English arrived in Suriname after 1680 (Arends 2002: 119–121). The English administrative influence did not end with the Dutch takeover of Suriname. They reoccupied Suriname from 1799 to 1816, during which a new group of planters established themselves in the Nickerie area and England abolished the slave trade in all its colonies (Arends 2002: 126).2 The linguistic legacy of the English is most notable in the etymological origins of the majority of the Surinamese creoles’ lexical inventory. While English has not been maintained as a community language by any significant group of Surinamese, its influence on the creoles’ vocabulary is paramount. Smith (1987: 119) reports that basic vocabulary of Sranan and Ndyuka is approximately 75% derived from English, and approximately 50% in Saramaccan.

2.3 Jewish settlement The presence of Jews in Suriname possibly dates from as early as 1639, when there were apparently some settlers who arrived from Holland and Italy (Fontaine 1980: 33). Van der Meiden (2008: 17) puts Jewish settlement beginning at 1643. However, it is unlikely that there was any significant Jewish population in the area before the 1660s. The first Jews are often said to have arrived in the area from Barbados in 1652 with the English. These quickly settled near Cassipora Creek on the upper Suriname River and began to establish what would become the Jodensavanne (Fontaine 1980: 33; van Lier 1982: 19). According to Rens (1982: 30), however, it is rather unlikely that any significant number of Jews came from Barbados. He argues Jews were not allowed to practice their religion in English territories and there are no records of Jews either living in Barbados or migrating from there to Suriname. He estimates that the population of Jews in Suriname was not more than 30 persons at any given time before 1665 (Rens 1982: 30). The next group came from Cayenne in 1665–66 and was later augmented by other Sephardic Jews from Amsterdam, Livorno, and possibly Essequibo. The English colonial government granted Jews a number of special privileges,

2 The slave trade continued illegally until the 1830s in Suriname.



The people and languages of Suriname 

 29

including the right to worship, private civic guard, civil court (which could make binding decisions in cases up to the value of 10,000 pounds of sugar), schools, and ‘churches’ (Fontaine 1980: 34; van Lier 1982: 20). If the previous group had not established the Jodensavanne, these certainly did (Sijpesteijn 1854: 4; Fontaine 1980: 33; van Lier 1982: 19; Arends 2002: 118). It is controversial whether this group had previously been resident in North East Brazil (Pernambuco) and expelled from there by the Portuguese in 1654 (Fontaine 1980: 33; Arends 1999; Ladhams 1999; Smith 1999). There is no evidence that Jews migrated directly from Brazil to Suriname (Rens 1982: 36) and very little evidence that a significant number of those from Cayenne had previously been in Pernambuco (Arends 1999; Ladhams 1999), though Smith (1999, 2015c: 34) maintains that there was in fact a connection between North East Brazil and Suriname, which best explains the Portuguese element in the Surinamese creoles, despite the scant documentary evidence. Putting their origins aside, and whether they brought slaves with them, the Jews, who established the Jodensavanne, brought wealth and an advanced knowledge of tropical agriculture (sugar). Although their numbers remained small for the duration of the period of English rule (approx. 200), they exercised considerable influence on the development of agricultural techniques and the emerging creole languages in Suriname (Goslinga 1971: 425; Rens 1982: 30; Fontaine 1980: 34; van Lier 1982: 19). Van Lier (1982: 19) reports that many Jews left following the colony’s takeover by the Dutch province of Zeeland, although this possibility seems rather unlikely since they were recent arrivals and the English would not have been eager to assist them in relocating over ‘their own’. The Dutch were well known as tolerant of Jewish people and customs, and they actively tried to prevent any but natural-born Englishmen from leaving the colony (Rens 1982: 36–37). The new administration by the province of Zeeland left those privileges afforded to the Jews by the English intact (Fontaine 1980: 33). The Jodensavanne flourished in the late seventeenth and early eighteenth centuries. It was the most prosperous township in the whole colony. By 1694, there were 40 sugar estates and approximately 9,000 slaves owned by Jewish planters. There were 92 Portuguese Jewish families (about 570 individuals) and 10 German Jewish families in Suriname at the time (Fontaine 1980: 34; van Lier 1982: 19). The Portuguese Jews maintained their own separate community within Suriname, though they mixed socially with other planters and had a favorable reputation before the mid eighteenth century (van Lier 1982: 21). During the French invasion of 1712, Admiral Cassard demanded an enormous fee to keep his men from plundering the plantations. Jews paid the bulk of this demand and never recovered economically. Further, the introduction of beet sugar, economic hardships in Europe, and the rise of anti-Semitism in Suriname meant that loan capital was no longer available to Jews (Fontaine 1980: 36). In

30 

 Robert Borges

the colony as a whole, those estates that did not keep up with their interest payments were sold off by bankers. The Jews were particularly affected by this and lost approximately half of their collective property in the second half of the eighteenth century (van Lier 1982: 22–23). Many left the Jodensavanne for the economic bustle of Paramaribo (Fontaine 1980: 36). The Jodensavanne was almost completely abandoned by the 1830s (van Lier 1982: 19). The Sephardic Jews continued to use Portuguese (and to some extent Spanish) until ‘far into’ the eighteenth century as an in-group language as well as with their slaves (Arends 2002: 119). The non-standard, L2 varieties, and pidgin varieties of English spoken throughout the colony underwent influence by Portuguese (and Spanish), which was more intense in the Sephardic areas (Migge 2003: 29–30). The origin of this Portuguese element, its nature, and the extent of its use and influence in Suriname’s formative years remain the subject of much debate. Arends (1999), Ladhams (1999), and Smith (1999) all propose a number of hypotheses that differ in details (e.g. if any significant proportion of the Cayenne Jews were first in Pernambuco, if they were able to bring slaves along with them, or if a Luso-pidgin or -creole was imported to Suriname), but all maintain that the Jewish presence in Suriname was requisite for the Portuguese elements found throughout the Surinamese creoles. Saramaccan is known to be the “most Portuguese” of the Surinamese creoles, with 35.88% of basic vocabulary of Portuguese origin (c.f. Sranan 3.7%, Ndyuka 5.04% (Smith 1987: 119–120)).

2.4 The Dutch The Dutch were also involved in exploring the Guianas from 1581, when they made their first voyage there (van der Meiden 2008: 17). They made several settlement attempts 1610s and 1630s (Arends 2002: 115). The first attempt was made in 1613 by 50 families of Zeelanders who settled along the Corantijn River to grow tobacco. Their settlement was destroyed and all its residents were killed by the Spanish (Goslinga 1971: 79; de Bies 1994: 7). Some Dutchmen had also attempted to settle Cayenne in 1615, though they abandoned it quickly for the area that would become Suriname (Goslinga 1971: 79). In the area where Paramaribo is now, a trading post had been established by Dirck Claasz and Nicolaas Balistel in 1613 (Buddingh 1995: 10). Apparently none of these settlements lasted for more than a couple of years. After the English had established its colony and Willoughbyland had been conquered by Crijnssen and his fleet in 1667, the freshly renamed colony of Suriname remained under the jurisdiction of the Dutch province of



The people and languages of Suriname 

 31

Zeeland (Arends 2002: 119; van der Meiden 2008: 13). Although some English planters left following seizure of their colony, the English continued to play an important role in the colony until at least the end of the seventeenth century (van der Meiden 2008: 13). In this period, the European population was mainly augmented with Dutch convicts, vagabonds, and orphans (Arends 1995: 239). The Dutch also stepped up slave importation (from European to African ratio of 1:3 under the English to 1:12) and large-scale sugar cultivation activities before the end of the seventeenth century. This demographic and economic shift intensified social stratification where out-group interactions were increasingly limited to work related dealings (Arends 1995: 260; Migge 2003: 30). In 1683, the jurisdiction of Suriname was transferred from the province of Zeeland to the Sociëteit van Suriname ‘Society of Suriname’ whose share holders were the city of Amsterdam, the Dutch West India Company, and the van Aerssen van Sommelsdyk family (van Lier 1971: 19; Arends 2002: 119). The Society survived over a century of assorted conflicts, economic decline, and ethnolinguistic stabilization. By the mid eighteenth century, the population of Suriname consisted of approximately 1,500 Europeans and 30,000 slaves, who were largely situated in the countryside; there were still around 700 to 800 people in Paramaribo (Arends 1995: 259; van der Meiden 2008: 14). An economic downturn led to an increase in urbanization of Suriname before the beginning of the nineteenth century. After the liquidation of the Dutch West India Company in 1791, the Society of Suriname lasted just a few years, dissolving in 1795 and transferring jurisdiction of Suriname to the newly established Batavian Republic (van Lier 1971: 20; Arends 2002: 126). Suriname was again lost to the English in 1799, though it was returned to the Dutch with the institution of the Royal House of Orange in 1816 and the colony remained in Dutch hands until its independence in 1975. Despite their role in administering the colony, Dutchmen did not form the majority among Europeans in Suriname until ‘well into’ the nineteenth century (Arends 2002: 119) and their language had served only a marginal role outside official domains of Surinamese society as a whole until the second half of the nineteenth century (Arends 2002: 124). In 1876, the Dutch colonial government implemented compulsory education policy, called leerplicht, whereby most children in the colony were required to attend school with Dutch as the language of instruction (Buddingh 1995: 228). Following the leerplicht, use of languages other than Dutch in schools was forbidden (de Bies 1994: 7). A more general Dutch only policy was enforced on a large scale, particularly in Paramaribo from 1878 (de Kleine 2002: 213). For many, Dutch became associated with upward social mobility and has been adopted by a portion of the population as a main language.

32 

 Robert Borges

2.5 African slaves Slavery has played an integral role in the Guianas since the first Europeans began their explorations and settlement of the area. Before the arrival of the English in 1651, slaves were primarily Amerindian (Arawak), though the existence of African slaves cannot be ruled out during this period. The oldest Saramaccan clan, for example, traces its earliest ancestors to Marshall’s 1630 settlement attempt (Arends 2002: 116–118). The English did import African slaves into Suriname, though the numbers remained relatively low. They relied on a system of indenture at the time in their colonies in the Caribbean. In fact, overpopulation of ex indentured whites in other colonies was among the motives to colonize Suriname. During the English period, the slave population grew to approximately 3,000 individuals, double that of the European population and attrition rates of slaves were very low (Arends 1995: 259; van den Berg 2007: 5; Migge 2003: 28–29). Following the Dutch takeover, the importation of Africans would increase exponentially in order to meet labor demands in the growing plantation colony. Arends (1995) was the first to systematically survey slave imports to Suriname and attempt to link them to African slaves’ ethnolinguistic affiliations. His work, based on extrapolations of data from Postma’s (1990) survey of the Dutch slave trade, suggests a number of trends in the relative numbers of slaves’ origins over time that have later been correlated with linguistic developments in the creoles of Suriname (e.g. Migge 2003; Bruyn 2009; Essegbey et al. 2013). Since Arends’ (1995) article, a new resource has become available for the study of the Transatlantic slave trade, namely the Voyages Database (2008) which is a large scale, searchable, online database that archives all available information on any transatlantic slaving voyages. In the following paragraphs, I compare data on the origin of slaves imported to Suriname presented in Arends 1995 with data extracted from the Voyages database cf. Fig. 1. A few preliminary remarks are necessary. Firstly, there are some terminological differences between Arends (1995, henceforth A) and the Voyages database (2008, henceforth The Vd) regarding African regions. Some of these overlap with only minor differences, allowing for a relatively simple comparison of data. 1. Sierra Leone – defined in The Vd as “Rio Nunes to just short of Cape Mount”. In modern national terms, roughly the southern part of Guinea and modern Sierra Leone. A does not address this region. 2. Windward Coast – defined in The Vd as “Cape Mount south-east to and including the Assini River”. This includes roughly modern Liberia and Ivory Coast and corresponds with A’s use of Windward coast. 3. Gold Coast – “the area east of [the Assini River] up to and including the Volta River” in The Vd. A’s use of Gold Coast only extends to Accra.

The people and languages of Suriname 



 33

1650–1659 1660–1669 Vd

0 100 2000 3000 4000 500 0 6000 700 0 8000 9000 100 0 00

1670–1679 Vd

1650–1679 Vd A 1680–1689 Vd A 1690–1699 Vd A 1700–1709 Vd A 1710–1719 Vd A 1720–1729 Vd A 1730–1739 Vd A 1740–1749 Vd A 1750–1759 Vd A 1760–1769 Vd A 1770–1779 Vd A 1780–1789 Vd A 1790–1803 Vd A

0 Bight of Benin Sierra Leone unspecified

10000

20000

Bight of Biafra West-Central Africa

30000

40000

Gold Coast Windward Coast

Fig. 1: Comparison of slave imports in Arends and the Voyages database.

34 

 Robert Borges

4. Bight of Benin – from the Volta River to the Nun River according to The Vd. In modern terms, this refers to the eastern part of Ghana, Togo, Benin, and western Nigeria. A calls Togo, Benin, and the western part of Nigeria the Slave Coast. 5. Bight of Biafra and Gulf of Guinea islands – in The Vd, “east of the Nun to Cape Lopez”, roughly Eastern Nigeria to the northern part of Gabon. A does not address this area, though his area defined as Loango includes Equatorial Guinea and the northern part of Gabon. 6. Western Central Africa – according to The Vd is the area from Cape Lopez to the Cape of Good Hope, roughly ‘the rest’ of Western Africa. This corresponds, in part, to what A calls Loango, the area from Southern Camaroon to Cabinda, roughly Equatorial Guinea to the Democratic Republic of Congo. The search of The Vd for shipments to Suriname also returned two unspecified categories with respect to regional origin of slaves, “other Africa” which provides no information about embarkation point or regional origin of slaves, and what I call “unspecified” where these fields were simply returned empty. According to the database compilers (Eltis 2013, p.c.; Vos 2013, p.c.), “other Africa” always refers to voyages that acquired slaves along the Windward coast, which unlike other major embarkation points required stops at a number of ports of call where only small numbers of slaves could be negotiated for purchase (Vos 2010). I have conflated the Vd’s “other Africa” and Windward coast figures for comparison with A’s Windward coast figures. The database was queried for shipments with a principle disembarkation region of slaves in the “Dutch Guianas” (this also returned shipments to pre 1667 Suriname) and shipments that did not disembark slaves in Suriname were eliminated. In some senses the data is a bit messy, e.g. a shipment might list multiple points of slave disembarkation, but provides no breakdown of the number of slaves disembarked at each port. In this case, the total number of slaves disembarked was included in the calculations. In some cases of multiple disembarkation points, a numerical breakdown was provided, in which case only the figure for Suriname was added to the calculation. The regional origin of slaves disembarked in Suriname was compiled per decade ranging from 1650 to 1839. Raw results tailored for comparison with A are provided in the appendix of this chapter. The expectation would be that the overall numbers of slave imports to Suriname would increase with the additional sources utilized in the Vd. This is indeed the result for all decades except the 1690s and the 1730s. The overall decrease in slaves disembarked in Suriname in these decades is 344 and 227, respectively, which could easily amount just two shipments missing from returned results of the query. This could be caused e.g. by shipments with a principle disembarkation area other than Dutch Guianas which did actually disembark slaves in Suriname



The people and languages of Suriname 

 35

or shipments from other places in the Americas or Caribbean. These anomalies are negligible among the overall slave imports. Major differences between A and The Vd in the overall imports occur in the periods from 1650 and 1679, and between 1790 and 1803. These increased figures in The Vd, 8,401 (174% of A’s figure) and 20,508 (418%) are the result of the inclusion of data other than that included in Postma (1990), which was Arends’ source. Arends noted in his later article (2002: 121) that the origins of slaves before 1675 are not known in detail, adding that there is some indication that Calibar (Nigeria) was an important source of slaves shipped to Suriname in the early 1670s. Indeed, data presented in A indicates that the origin of 94.6% of slave imports pre 1680 is unknown. The remaining imports are attributed to the Loango area. The Vd provides additional information with regards to slaves’ origins, such that just 13% of imports are listed with an unknown origin. The remaining imports from the Bight of Benin (9.7%), Bight of Biafra (25.7%), Gold Coast (9%), and West Central Africa (42.5%). Slave imports in this period from the Bight of Benin and the Gold coast only began in the 1670s. Arends’ mention of the Calibar’s role in slave imports to Suriname is confirmed by the data extracted from the Vd. Additionally, the new information on slave origins pre 1680 will be of particular interest to those who invoke Mufwene’s (1996) founder principle in their accounts of creolization. In the latter half of the 1660s, both the European and the African population had decreased by two-thirds due to emigration prompted by the Zeelanders’ occupation, and increased death rates from an epidemic. The population began to rise again in the 1670s and from the 1680s, there was a steady increase in both the number of slaves, slave imports and ratio of African slaves to Europeans for the next century (Arends 1995: 259–260; van den Berg 2007: 6–17; Migge 2003: 30–35). Mortality rates of slaves also increased significantly from the 1680s due to the brutal nature of slavery in Suriname. This meant that newly arrived Africans drastically outnumbered locally born slaves; nativization of the African population in Suriname was extremely slow (Arends 1995: 235). Thanks to A’s work, the general consensus is that from 1650 to 1720, the majority of known slave imports to Suriname came from the Slave coast (Gbe languages) and the Loango (Bantu languages) area and just a very small percentage (less than 5%) originated from the Gold Coast (Kwa languages, esp. Akan varieties). The Vd shows us that, in addition, there was a formidable ratio of slaves imported from the Bight of Biafra and Gulf Islands between 1660 and 1690, which means that additional African languages and possibly early Gulf of Guinea Creole varieties may have played a more prominent role in the Early Surinamese contact setting. The Gold Coast would become the major source for Surinamese slaves from 1720 to 1740. In this time Slave Coast slaves were still well represented and Loango imports dwindled.

36 

 Robert Borges

After 1740, the Windward Coast (Kru, Souther Mande, and Southern Atlantic languages) became a major source area for slaves, though Gold Coast slaves and Loango slaves continued to be imported in large numbers (Arends 1995: 243, 278–281). The last major difference between The Vd and A is in the final 13 years of the comparison period (1790–1803), when the Vd indexes 15,601 more slaves disembarked in Suriname than Arends. In the entire comparison period, total slaves disembarked in Suriname rise 27,695 individuals from A’s number of 181,591 in the Vd, and for all years, up to 1830, the Vd provides a total of 221,776 slaves disembarked in Suriname. Despite this, at its height in 1775, the slave population was just shy of 60,000, and just before abolition in the 1860s African/Afro-Surinamese population did not exceed 40,000 (Arends 1995: 259; 2002: 122–23). The slave trade was officially abolished in 1808 in all English colonies, which also applied to Suriname, as it had recently come under the English (Buddingh 1995: 179), though it continued illegally into the 1830s (Arends 2002: 126). Labor to man the plantations remained a concern for the planters. With the end of their labor supply, planters and the government took measures to keep better track and improve the well being of existing slaves. In the 1820s, for example, efforts were made to register all the colony’s slaves (Budding 1995: 182), From 1839, slaves could no longer be sold as individuals; families were kept together (Lamur 1987: 39). In 1856, a bill passed in the Dutch house by which slavery was to be abolished in all its colonies (van Lier 1971: 179). For Suriname, all slaves were officially declared free persons as of July 1, 1863, though they were required to continue working on a plantation of their choice for the next ten years (van Lier 1971: 180–181). Following this period of government supervision, many of these former slaves opted to leave the plantations for wage labor in Paramaribo or engage in small scale agriculture, leaving the planters to search for a new source of labor to man their plantations (Buddingh 1995: 220–222; Arends 2002: 127). With the increased importation of African slaves from 1680, there was decrease in both access to English varieties spoken in the colony as well as in motivation for Africans to acquire them (Migge 2003: 30). As a result of this constant influx of African slaves continued to speak African languages in-group contexts until the late eighteenth century (van den Berg 2013). These Africans and their languages are also partly responsible for the formation of the creoles spoken in Suriname, though the early groups of Gbe and Kikongo speakers appear to have exerted the most influence on the emerging creole due to their founding role (Arends 1995: 250; Arends 2002; Migge 2003: 35; Mufwene 1996). Examination of data provided in the Voyages database (2008) tells us that slaves originating in the Bight of Biafra and Gulf of Guinea Islands were more prominently represented in Suriname’s early years.



The people and languages of Suriname 

 37

2.6 Maroons As long as enslaved Africans were brought to the New World, some attempted to flee. Success rates for escape and avoiding recapture varied per colony. Suriname was among the most likely places where a slave could escape and maintain his freedom in the Caribbean. Slaves had been rebelling against their bondage, rising up, and presumably forming Maroon communities since the English period (Hoogbergen 1983: 76; Carlin and Boven 2002: 19). The earliest group that we know of, the Karboegers, was of mixed Amerindian and African origins and spoke a Cariban language, Kari’na. This group was already well established by the end of the seventeenth century. The Karboegers were protected under the 1684 treaty with the colonial government which acknowledged the rights of certain Amerindian groups, and were subsequently employed in the bush patrols intended to catch runaway slaves (Carlin and Boven 2002: 19; Price 2012: 12). The Karboegers have long since disappeared, however Suriname is currently home to one of the largest Maroon communities in the whole of the Caribbean and South America. Generally speaking, massive slave uprisings were unusual in Suriname, though occasionally slaves did leave a plantation en masse, often killing the owner or overseer (Lamur 1987: 17–18; Migge and Léglise 2013: 76). Slaves often left plantations from time to time as a means of protest, staying for a time in the nearby forest, or to visit family on other plantations. This phenomenon is known as petit marronage, and planters generally accepted this type of behavior as part of the system; usually these folks would return after a time due to hunger or some other factor (Hoogbergen 1983: 77). Those slaves who wished to remain away from plantation life permanently did so in several ways: to escape and join an already established group, to be recruited / kidnapped by an existing group, or to become established gradually in stages (Hoogbergen 1983: 78). Marronage in stages, first involved the establishment of a kapuwari ‘garden plot’ where – mostly single men – would spend the days working, fishing and trapping small game. These Kapuwarimen were still heavily reliant on contacts with the plantations and sometimes returned to the slave quarters for the night (Hoogbergen 1983: 78). The next step was the formation of small groups, not usually more than 8–10 individuals, who survived on the produce of their kapuwari and by hunting and trapping fish and small game. These groups would also establish provision grounds and shelters further from the plantations in the rainforest and coastal wetlands. Once the new provision grounds produced sufficient amounts, the settlements near the plantations would be abandoned – the final stage. These Maroons would then supplement their numbers by relocating family still on plantations, or by abducting women from plantations (Hoogbergen 1983: 79–80).

38 

 Robert Borges

From the rainforest and coastal marshes they would continue to take measures against plantations, raiding for supplies and new recruits. Marronage became a large problem for the colony. Planters relied on patrols of the colonial government to deal with runaways, and also took matters into their own hands, attempting to recapture or kill runaways with the help of their own armed slaves (Hoogbergen 1983: 75). Due to the financial burden of fighting the Maroons, the colonial government concluded a peace treaty with three groups of Maroons, the Ndyuka, Saramaccan and Matawai, in the 1760s. According to the agreement, the newly ‘pacified’ Maroons were to put an end to plantation raids and return new runaways to the colonial government. In return, they were to receive a number of essential goods on a yearly basis. These treaties effectively ended the possibility for slaves to escape to established groups, since ‘pacified’ groups were meant to return runaways, and the second cycle of Maroon groups that formed quickly thereafter depended on a high degree of secrecy for their survival. Many runaways were executed in the years following the peace treaties in an unsuccessful attempt to deter runaways by setting examples of those who tried (Hoogbergen 1983: 76–78). Tab. 2: Known Maroon groups, based on Smith (2002). Group

Time of flight

Current location

Karboegers Saramaccan Matawai Ndyuka Aluku Kwinti Pamaka Brosu

English period ca 1690 ca 1700 1712 Before 1760 ca 1760 1800 1820

Extinct or assimilated Upper Suriname River Saramacca River Tapanahony, Marowijne, and Cottica Rivers Lawa River Coppename River, Saramacca River Marowijne River Assimilated to Ndyuka

A number of groups emerged in the early history of Suriname. Known groups are presented, along with their approximate time of marronage, and their location in Tab. 2. The Saramaccans and the Matawai, commonly referred to collectively as Central Maroons, fled plantations of Sephardic Jews on the upper reaches of the Suriname river, also known as the Jodensavanne, between 16903 and 1710 and initially formed a single group (Green 1974: 31; Migge and Léglise 2013: 77). The

3 1690 is the year that these groups were first mentioned in archival material; therefore the nascent group must have been in existence for some time before warranting attention from the colonial government.



The people and languages of Suriname 

 39

other groups, Ndyuka, Aluku, Kwinti, and Pamaka, fled at different time periods in the eighteenth century from the north-eastern part of the plantation area, the area around e.g. the Cottica, Commewijne, and Tempati Rivers (Migge and Léglise 2013: 78–83). The flight and isolation of the Maroons was decisive along cultural and linguistic lines. As each group left the plantations, they took with them a variety of plantation creole, which could then develop relatively freely from Sranan and influence from other languages like Dutch. Intelligibility is low between Central and Eastern Maroon languages, largely as a result of a higher degree of Portuguese derived elements in the former as a result of their origins in the Jodensavanne. Until relatively recently, Maroon languages were largely used for in-group communication within the communities. Construction of the Afobaka Dam and the civil war, the Binnenlandse Oorlog, drove a large number of Maroons to Paramaribo and other coastal centers, causing wider recognition and the use of, particularly Eastern Maroon varieties, by non-Maroons (cf. Léglise and Migge 2006; Migge and Léglise 2013). For additional information about the histories of individual groups, see e.g. Price (1983) for Saramaccan, Green (1974) or de Beet and Sterman (1981) for Matawai, Thoden van Velzen and Hoogbergen (2011) for Ndyuka, Hoogbergen (1989) or Bilby (1990) for Aluku, van der Elst (1975) or Hoogbergen (1992) for Kwinti, Lenoir (1973) for Pamaka.

2.7 The Hakka In anticipation of the end of slave labor, the colonial government began seeking alternative options to fill the impending labor deficit on the plantations. In 1853, the first Chinese contract laborers (18 individuals) were imported to Suriname from Java. In subsequent years, up to 1870, several thousand workers were imported directly from China’s Guangdong Province (Buddingh 1995: 213; Tjon Sie Fat 2002: 233–234). Poor social and economic conditions in Southern China provided the impetus for contract labor (Tjon Sie Fat 2009: 67). The vast majority of these men4 were Hakka, a not entirely straight forwardly definable ethnic group, who were speakers of some variety of the Kejia language which is a dialect continuum related to the Min and Yue languages (Tjon Sie Fat 2002: 239; 2015).

4 Just 3% Of all migrants from China in this period were women (Tjon Sie Fat 2009: 68).

40 

 Robert Borges

Most Chinese workers neither renewed their contracts nor stayed in the agricultural sector after the initial period of indenture. As a result, the colonial government ended indenture contracts for the Chinese (van Lier 1971: 183; Tjon Sie Fat 2002: 234). Surinamese indenture contracts did not include an option for paid passage back to China, and the vast majority stayed. It is estimated that across the Caribbean, only 0.1% of indentured workers paid their return to China (Tjon Sie Fat 2009: 68). This initial group quickly established social and economic networks, engaging largely in retail trade and other mercantile activities (Tjon Sie Fat 2009: 68). Due to the lack of Chinese women in Suriname, many of these men took Creole wives; it is unlikely that Kejia played any significant role in these mixed households (Tjon Sie Fat 2002: 234). Although the Hakka imported to Suriname were rather insignificant numerically and perhaps ethnolinguistically, they formed the kernel of what would become Chinese identity and supported chain migrations from other Hakka in the Guangdong lasting until the 1930’s. Preceding WWII, migration from China stopped and Suriname became isolated from China. In the 1960s and 70s immigration from Southern China and Hong Kong resumed to some extent, but as a result of the Cultural Revolution in China (to which the Surinamese were not privy) there was a certain amount of cultural and linguistic difference between these and the old Chinese in Suriname (Tjon Sie Fat 2002: 234–236; 2009: 66). From the time of the earliest Chinese immigrants in Suriname, Kejia was the prominent Chinese language in Suriname until the 1960s (Tjon Sie Fat 2002: 236). In Suriname, a local variety of Kejia developed, which, compared to the baseline variety has a reduced tonal system, archaic vocabulary, and a high number of Sranan loanwords (Tjon Sie Fat 2002, 2015; Rojas-Berscia and Shi, this volume). Many ethnic Chinese have assimilated to Surinamese society and no longer speak Kejia. Resumed migration from the 1960s, and a third wave of immigrants from China, beginning in the 1990s, has brought additional Chinese languages to Suriname, and Kejia has since been replaced by varieties of Mandarin and Cantonese as the main Chinese language spoken in Suriname.

2.8 Indentured East Indians After the inadequacy of the indentured workers from China, the colonial government continued to ensure a cheap labor supply to man the plantations following the abolition of slavery. A deal was negotiated with the British to ensure indentured labor from India in 1870. The first ship with indentured workers from India arrived in 1873, and through 1916, approximately 34,000 (male and female) laborers were shipped to Suriname from India (van Lier 1971: 217; Marhé 1985: 7). Laborers were recruited from a wide area in central and eastern north India, Uttar Pradesh, Bihar



The people and languages of Suriname 

 41

and Jharkhand (Damsteegt 1988; Damsteegt 2002: 255). Approximately 17% of the migrants were Muslim (Buddingh 1995: 215). Following their indenture, the majority of Indians (approx. 66%) remained in Suriname (van Lier 1971: 8–9, 217). Contract laborers were required to work on a full time basis, earned wages regulated by the colonial government, and were entitled to free accommodation and a plot of land for their own use (van Lier 1971: 219). Children from ages ten to fifteen years old were required to work half time, which caused conflict with Suriname’s educational requirement. An additional problem was that from 1876 the government required education to be given in Dutch. Special schools for Hindustani children were opened in 1890 to overcome the language barrier and scheduling conflicts (Buddingh 1995: 229). The schools provided lessons in the children’s own languages until 1906, after which Hindustani children were integrated into the regular school system (van Lier 1971: 193). The initial agreement included a paid return to India following the period of indenture however workers forfeited their right to a free passage home by settling as farmers. The government became increasingly interested in expanding its population, and in 1895, the colonial government revised their agreement to allow the retention of right of return despite a period of independent farming following indenture, but offered a premium of 100 florins to renounce right and take up residence and work on a government faming settlement, rent free for 6 years (van Lier 1971: 234). Hindustanis (and to a lesser extent Javanese) were subsequently responsible for the expansion and economic viability of independent small-scale agriculture, and thus began the transition of Suriname from a large scale plantation economy in the early twentieth century (van Lier 1971: 221). The Hindustani community displayed a high degree of entrepreneurship and the desire to get off government plots and become independent landowners. They produced high quality crops for markets and rapidly expanded their economic prospects (van Lier 1971: 235), though they remained relatively inactive politically until the 1940s (van Lier 1971: 339). In the years leading up to the onset of the Second World War, the build-up of foreign troops and a defense industry to protect against possible German hostilities led to rapid urbanization by Hindustanis, leaving the agricultural sector for defense work (Buddingh 1995: 274). In the area where the majority of Indian laborers came from, several languages, including Awadhi, Bhojpuri, and Magahi, are spoken and comprise a linguistic continuum. These languages, mixed and leveled, forming a koiné before extraction from India, resulting in a language that is not like any present day Indic language found in India (Kishna 1983; Damsteegt 2002: 249, 254). Since there was no majority language among the indentured Indians, structures and forms common to their languages prevailed in their language, Sarnami (van Lier 1971: 217; Damsteegt 2002: 249–255). “Simplification” is not a tenable hypothesis in explaining the formation of Sarnami. While some forms can be considered simple compared

42 

 Robert Borges

to Bhojpuri, compared to Awadhi on the other hand, Sarnami likewise is more complex. Non-Indic languages appear to have played no role in the formation of Sarnami (Damsteegt 2002: 261). Sarnami is an in-group language par excellence which displays a relatively low level of prestige versus other Indic languages such as Hindi and Urdu, as well as Dutch (Marhé 1983; Damsteegt 2002: 251).

2.9 Javanese labor While recruitment of contract laborers from India was successful, there was a growing opposition to Indian contract labor because the colonial government found it undesirable that such a large percentage of its labor force remained British subjects (de Waal Malefijt 1960: 24). Being British subjects, the Hindustanis could theoretically undermine the Dutch colonial government’s and planters’ authority with complaints of mistreatment to the British council (Hoefte 1998: 19). As a result, the government continued to look for additional sources of labor. The decision was made to tap the labor pool of the island of Java, which was part of the Dutch colony of Oost-Indië and at the time had an extremely high population density (Hoefte 1998: 26). Between 1891 and 1839, over 32,000 Javanese were contracted to work in Suriname, of which approximately 7,500 returned to their home country before the Second World War and another 1,000 in 1954 (de Waal Malefijt 1960: 24; van Lier 1971: 218; Hoefte 1987: 3). Conditions for Javanese contract laborers were much the same as those of the Hindustani immigrants (van Lier 1971: 220). Laborers were contracted for a term of five years, and were to work six days per week, either seven hours per day in an agricultural field or ten hours per day in a factory. Employers were to provide free medical care, shelter, provisions for the first three months on loan, and paid passage to the home country following indenture. Contracts could be renewed for an additional year at a time for a bonus of 20 Guilders each year. The government also encouraged Javanese workers to stay in Suriname with the same 100 florins and a plot of land to renounce their passage home as offered to the Hindustanis (Hoefte 1998: 29). Contrary to the Hindustani immigrants, Javanese did not flourish economically in their new environment. They had a tendency to be satisfied remaining on government farm settlements and producing only what they needed to get by (van Lier 1971: 235). For some time, they remained the most ‘backward’ [achtergebleven] group in Suriname with extremely poor living conditions (Buddingh 1995: 240–241). The Javanese themselves did not have positive attitudes about Suriname; many Javanese men reported being tricked into their indenture and women reported being sold into indenture (Hoefte 1998: 31–35). Like the other groups brought to Suriname, the Javanese also brought their language along to their new country. Due to the isolation of Surinamese Javanese



The people and languages of Suriname 

 43

from Java, a local variety of the Javanese language developed. Surinamese Javanese is an in-group language and its use is severely declining in recent years. Many Sranan and Dutch loanwords can be found in Surinamese Javanese, and perhaps the most salient feature of the language is the disuse of formal registers employed systematically in Java (de Waal Malefijt 1975; Wolfowitz 2002). There is also some evidence of morphosyntactic influence from Sranan on Surinamese Javanese. In particular, the Surinamese variety makes use of serial verb constructions for describing motion events (Lestiono 2012).

2.10 New immigrants and post colonial developments In the years leading up to and following Suriname’s independence from the Netherlands in 1975, fears that the country would quickly become a failed state gave way to mass emigration of Surinamese. In that year, the peak year for emigration, around forty thousand Surinamese left for the Netherlands, the majority of which were Creoles and Hindustani from the Paramaribo district (Choenni and Harmsen 2007). In 2010, some 345.000 individuals in the Netherlands were classified as Surinamese by the Central Bureau of Statistics in the Netherlands (2011). A second major migratory movement within and out of Suriname was triggered by the civil war which ravaged large parts of the interior between 1986 and 1992 (Hoogbergen and Kruijt 2005; de Vries 2005). Amerindians and Maroons alike shared a similar fate: depopulation of traditional territories, war trauma, undermining of traditional political structures, and interrupted education. Violence and economic deprivation led to refugee movements to places such as Paramaribo, French Guiana and the United States, along with a renewed trend for emigration to the Netherlands. A large number of Surinamese Maroons left the country for its eastern neighbor. This led to the firm establishment of the Saramaccans, Ndyuka, and Pamaka in French Guiana leading to heightened crossborder mobility (Migge and Léglise 2013). The connections established by Maroons in Paramaribo during the war and the subsequent economic take-off in the interior led to an increase in ease and frequency of circular migrations between the interior and the coast (Léglise and Migge 2006; van Stipriaan 2011, 2015). One consequence of these demographic developments has been the establishment of patterns of circular migration between the Netherlands and Suriname for diverse motives such as business and work, family and leisure, and arts and culture. This has led to a transatlantic Surinamese economic, cultural, and linguistic area, in which goods, people, ideas, and, in particular, language practices are continuously exchanged across the Atlantic (cf. Oostindie and Schoorl 2011). A second consequence has been the establishment of Sranan as an important

44 

 Robert Borges

and highly visible heritage language in the Netherlands and its acquisition of local characteristics and influence on other language varieties in Suriname (e.g. straattaal ‘street language’, see Cornips 2004). Concentration of Maroons in centers in Paramaribo and western French Guiana has also been leading to increased contact and convergence among the various Maroon creoles and Sranan (Léglise and Migge 2006; Migge and Léglise 2013: chapters 7–9). A number of internal and external factors have led to new migratory currents into Suriname since the 1990s, most prominently including, Brazilians and a number of different ‘Chinese’ peoples. Chinese immigration in Suriname began to rise again in the 1990s, following the People’s Republic of China easing restrictions on emigration. These new Chinese are non-Hakka and refuse to learn Kejia, while the Surinamese Hakka often do not speak any other Sinitic language (Tjon Sie Fat 2015). Brazilians are famous in Suriname for their involvement in artisanal gold mining. A relatively large number of Brazilians, estimated between 20,000 and 40,000 individuals, have entered Suriname since 1995 (de Theije 2008: 152). Most migrants come from the northern districts of Pará and Maranhão (de Theije 2008: 153) in what began as semi-legal, or illegal, small time mining operations in the interior of Suriname. Brazilian migration is a ‘come-and-go’ type affair involving largely undocumented peasants and laborers (de Theije 2008: 152). Due to increasing settlement in Paramaribo, there has been a sharp increase in the economic visibility of other Brazilians who cater to the miners – sale of mining equipment, gold smelters, bars and restaurants, and brothels (de Theije and Heemskerk 2009: 13). These miners, or garimpeiros, who began to trickle into Suriname from the 1980s, are responsible for the introduction of modern mining equipment, such as hydraulics bulldozers, excavators, etc, to the interior (de Theije and Heemskerk 2009: 14). Recent waves of immigration into Suriname have further diversified the linguistic landscape. In some areas of Paramaribo, Brazilians tend to congregate in large neighborhoods where Portuguese is highly present in both spoken and written form, but the Surinamese government has done little to introduce Portuguese as a foreign language in the educational system. The impact of Portuguese on the Surinamese linguistic landscape is understudied. Some non-Brazilian merchants in areas heavily populated by Brazilians have taken it upon themselves to learn enough Portuguese for commercial interaction (Borges 2011, field notes) and some Maroons have also begun to use Portuguese at different levels in their interactions with miners in the interior (de Theije p.c.). New immigrants from China come from a number of ethno-linguistic backgrounds, and along with their own languages, they often use Mandarin or Putonghua as a lingua franca among them. There is a sharp ethnolinguistic rift between these new immigrants and the old Hakka in Suriname and the new Chinese have spurred a severe restructuring of “Chinese-ness” (if such a thing exists) in Suriname (see Tjon Sie Fat 2009).

The people and languages of Suriname 



 45

2.11 Language development in Suriname: summary and conclusions The various historical events which spawned movements of people into, within, and out of Suriname that have had an impact on the contemporary linguistic landscape of Suriname have been described in the preceding sections and are summarized in Tab. 3. The intention of this section has been to orient the reader with the complex origins of Suriname’s ethno-linguistic composition. In the last five hundred years or so, Suriname has been home to a number of ethnic groups whose languages Tab. 3: Development of Suriname’s linguistic landscape. Event

Languages involved

Linguistic process / outcome

Amalgamation and decline of indigenous populations Arrival of European planters and African slaves Mass escapes of slaves

All indigenous languages

Shift and death

English, Portuguese, and Dutch varieties. Akan, Gbe and Kikongo varieties. Plantation creole

Language creation, creolization

Institution of Dutch only policy in Education

Dutch

Arrival of Asian contract laborers WWII economic turn

Sarnami, Javanese, Kejia

Independence from the Netherlands and introduction of national politics The Binnenlandse Oorlog and subsequent urbanization patterns Arrival of new immigrant groups

All Surinamese languages

All Surinamese languages

Maroon languages

Language divergence; differentiation of Maroon creoles. Saramaccan, Matawai, Ndyuka, Aluku, Kwinti, Pamaka Increased access and prestige of Dutch among the Afro-Surinamese (non-Maroons); divergence of Surinamese variety of Dutch Diversification of Suriname’s linguistic landscape Increased access and prestige of Dutch among non-Afro-­ Surinamese; divergence of Dutch variety of Sranan Expansion of Sranan and Dutch; convergence and language shift; increased access to European Dutch Levelling of Maroon creoles in/ around Paramaribo

Brazilian Portuguese, Haitian Diversification of Suriname’s Creole, Mandarin, Cantonese, linguistic landscape and other ‘Chinese’ languages

46 

 Robert Borges

are involved in a range of contact related linguistic processes, from simple lexical borrowing to the complexities of structural interference and creolization, and from language creation to language endangerment and death.

3 Language attitudes and practices There has been no systematic study of language practices in Suriname. Several studies, such the census from Suriname’s General Statistics Bureau, and Kroon and Yagmur (2010) have attempted to quantify language use, though they have largely noted subjective speaker assessments, i.e. what speakers think and say they do, rather than what they actually do. These studies therefore, contain valuable information on language attitudes among Surinamese people. When Surinamese are asked about the languages they use, they name the languages they prefer. The 2004 census of the General Statistics Bureau is the only census so far to list Surinamese households of all districts by language use. Households were asked to name the “language spoken most often” and the “second language spoken”. The resulting figures are reproduced in Tab. 4. Sranantongo and Dutch are the only languages that display large differences between the languages used “most often” and the “second language”. These languages also appear to display Tab. 4: Languages spoken in households. Language spoken most often Language Dutch Sranan Sarnami Javanese Maroon languages* Others No 2nd language** Unknown Total

Second language spoken

Number 57.577 11.105 19.513 6.895 18.797

In % 46,6 9,0 15,8 5,6 15,2

Number 29.163 45.634 8.121 6.846 2.493

In % 23,6 37,0 6,6 5,5 2,0

Total % 70,2 46,0 22,4 11,1 17,2

6.501 NA

5,3 -

4.030 23.754

3,3 19,2

8,6 19,2

3.075 123.463

2,5 100

3.422 123.463

2,8 100

5,3 200

*Named: Saramaccan, Aucan, Paramaccan in the census; **NA= Not applicable (Source: SIC 213–2005/02. Zevende algemene volks- en woningtelling in Suriname, landelijke resultaten, volume I, demografische en sociale karakteristieken [Seventh general population and household census in Suriname, national results, volume 1, demographic and social characteristics]. Paramaribo: Algemeen Bureau voor de Statistiek)



The people and languages of Suriname 

 47

the largest discrepancy between actual language behavior and reported behavior when compared to ethnic composition in Suriname. Note that the census does not provide a complete listing of languages spoken in Suriname as it lumps together the Eastern and Central Maroon languages under one heading and does not even list the Indigenous languages of the country (cf. Carlin and Arends 2002 for a complete overview of languages). Such high scores for Dutch as a home language, and low scores for Sranan, especially as a primary language, do not tally with the admittedly impressionistic observations I have made and those of other linguists who have worked on the languages of Suriname (e.g. de Bies, p.c.; Eersel, p.c., Migge, p.c.; Yakpo, p.c.). They therefore seem unlikely indicators of actual language use, particularly in the area outside of Paramaribo. Another recent survey (Kroon and Yagmur 2010) provides the top fourteen home languages reported in a large sample of primary and secondary school students across Suriname in a study supported by the Nederlandse Taalunie (the Dutch language treaty organization of which Suriname, the Netherlands, and Belgium are members). They show that on average, children report the use of more than two languages in their household. Their high figures for Dutch and low figures for Sranan also appear to be inconsistent with impressionistic expectations. Léglise and Migge (2015) explicitly point to the attitudinal nature of the figures rendered by their recent survey conducted amongst school children. They also report the usual combination of unexpectedly high figures of reported Dutch usage and surprisingly low reports of Sranan usage, even after inquiring specifically about unreported use of Sranan. The figures in all three surveys appear to report somewhat more balanced scores for other languages, suggesting that people have strong attitudes about Dutch and Sranan. Dutch, as a highly prestigious language in Suriname is overrepresented among all three surveys discussed here, while Sranan appears to be known and used by the vast majority of Surinamese is highly underreported (see also Essed-Fruin 1983; Charry 1983; Eersel 1983). High scores for Dutch may also tell us something about the perceived presence of Dutch in multilingual interactions in households, which may range from the sparse use of single Dutch lexical items, light to heavy code-mixing with Dutch, or monolingual use of Dutch. Similarly, covert prestige of Sranan means that its perceived presence within households is lower. Another way in which Sranan and Dutch usage differ from the other languages in Suriname is that they are the most frequently used languages for interethnic communication among Surinamese, and index social distance, depending on how they are used and by whom (Westmaas 1983). The other languages provided in Tab. 3 are similar in that they are primarily used for in-group communication. In some cases, like Sarnami, the 2008 census data nicely shows

48 

 Robert Borges

a correspondence between those claiming “Hindoestaan” ethnicity and Sarnami as a mother tongue in some districts. However, this type of correspondence does not work consistently for all in-group languages or in all districts of Suriname. The data on “mother-tongue” background alone could allow the conclusion that Sranan is also chiefly an in-group language. It also gets exceedingly low scores as a mother tongue, namely 1% and 2% respectively for those who self-identify as “Hindoestaan” (Indo-Surinamese) and Surinamese Javanese respectively. Even more surprisingly, Sranan gets a low score of 13% as a mother tongue within the Afro-Surinamese group - with more than 80% claiming Dutch as a mother tongue. Although a strong allegiance with Dutch and a symbolic rejection of Sranantongo as a primary language (cf. Essed-Fruin 1983) may be typical of these two (peri-)urban districts, research in the district of Saramacca for example, revealed older patterns of multilingualism beyond the Sranan–Dutch axis. Quite a few people with whom Yakpo and Hanenberg conducted extended sociolinguistic interviews in 2011 claimed that their parents or grandparents above fifty years of age had a good command of Sarnami and Javanese although they did not self-identity as the corresponding ethnicity. Even though qualitative and quantitative data are not easily correlated, these observations demonstrate the complex relationship between language attitudes and actual practices and the difficulty researching them. As a consequence of the respective overt and covert prestige of Dutch and Sranan, these languages function as “attractors”, exerting influence on the other languages in a non-reciprocal or unbalanced way. Sranan and Dutch act as agents of change and targets of convergence and language shift, providing lexical items and structural patterns to both each other, as well as other languages of Suriname. The presence of Afro-Surinamese language practices and features is highly visible in the Sranan and Dutch spoken by all ethno-linguistic groups of Suriname. For example, de Bies et al.’s Dictionary of Surinamese Dutch (2009) exhibits a large number of Sranan loans and calques. Since for a large proportion of Surinamese history Afro-Surinamese language practices have dominated the linguistic space of Suriname thereby establishing a “standard-setting” role (cf. Mufwene 1996 on the “founder effect”). Most people in Paramaribo have a good command of Dutch and the percentage of native Dutch speakers has increased substantially since the second half of the nineteenth century due to its increasing position as a prestige language and its institutionalization as sole language of education. As a direct result of widespread multilingualism in Suriname, however, other Surinamese languages have had an effect on the variety of Dutch spoken there (de Kleine 2002: 209). Surinamese Dutch is considered by de Bies to be the result of language acculturation within the construct of Surinamese society, and



The people and languages of Suriname 

 49

is particularly reflected in the lexicon of the language (1994: 9). De Bies states: “(...) Dutch in Suriname is still Dutch, although vocabulary is largely supplemented with words that are specific to the Surinamese context and culture. Overall, the Surinamese have enriched Dutch with their contribution” (1994: 7–8).5 Many Surinamese speak Sranan as a second or third language on a daily basis, but will not admit to this when asked. Sranan usage is also expanding, but this fact is only reflected in the correspondingly high second language allegiance. Structural patterns from Sranan have influenced a number of other Surinamese languages. Furthermore, the increase in access to Dutch and its prestigious position in Surinamese society means that Dutch also exerts a demonstrable influence on most other languages in Suriname, including Sranan.

4 Conclusion In its strategic and prosperous position as a plantation colony, first of the English and later the Dutch, Suriname has attracted many who either went voluntarily to make their fortune, or were taken to the colony to provide cheap labor. In this chapter, I have outlined the complex processes that have led to Suriname’s current diverse linguistic landscape, which as Eersel (1983: 169) notes, is still developing. Available information suggests that language attitudes and practices are at least partly responsible for patterns of linguistic variation and language change over time in Suriname.

5 The original passage reads “(...) want het Nederlands in Suriname is nog steeds Nederlands, zij het dan dat de woordenschat van het Nederlands op grote schaal werd aangevuld met “woorden” die de specifiek Surinaamse context en culturen benoemen. Al bij al heeft de Surinamer het Nederlands verrijkt met zijn Surinaamse bijdrage.”

1760–1769

1750–1759

1740–1749

1730–1739

1720–1729

1710–1719

1700–1709

1690–1699

1680–1689

1670–1679

1660–1669

816 14.60% 3460 33.68% 3736 53.33% 5201 60.71% 5035 64.34% 4183 40.85% 7009 40.28% 478 2.01% 752 2.87% 690 1.88%

Bight of Benin

1126 40.06% 1034 18.50% 1125 10.95%

Bight of Biafra

421 4.91% 174 2.22% 5008 48.91% 9573 55.01% 1839 7.75% 7359 28.04% 5570 15.18%

758 13.56% 175 1.70%

Gold Coast

204 0.78%

Sierra Leone 980 34.86% 2591 46.35% 5095 49.60% 3049 43.53% 2111 24.64% 1780 22.74% 632 6.17% 544 3.13% 7193 30.31% 7344 27.98% 15410 41.98%

West–Central Africa

Tab. A1: Slave imports to Suriname by principle region of embarkation from the Voyages database.

Appendix: Slave imports to Suriname

276 1.59% 5941 25.03% 9178 34.97% 13434 36.60%

Windward Coast

8281 34.89% 1408 5.36% 1601 4.36%

705 25.08% 391 6.99% 417 4.06% 220 3.14% 834 9.74% 837 10.70% 417 4.07%

unspec.

2811 100.00% 5590 100.00% 10272 100.00% 7005 100.00% 8567 100.00% 7826 100.00% 10240 100.00% 17402 100.00% 23732 100.00% 26245 100.00% 36705 100.00%

total imports

50   Robert Borges

Total

1820–1829

1810–1819

1800–1809

1790–1799

1780–1789

1770–1779

32116 14.48%

470 2.07%

286 1.07%

Bight of Benin

Tab. A1 (continued)

11222 5.06%

212 2.61% 7056 31.11% 624 61.60%

45 0.17%

Bight of Biafra

46376 20.91%

4587 17.10% 3803 68.37% 4509 55.59% 2600 11.46%

Gold Coast

3587 1.62%

296 5.32% 982 12.11% 1716 7.57% 389 38.40%

Sierra Leone

61473 27.72%

8400 31.32% 462 8.31% 1069 13.18% 4813 21.22%

West–Central Africa

45014777 20.30%

12849 47.91% 904 16.25% 597 7.36% 1834 8.09%

Windward Coast

21991402 9.92%

1192 100.00%

654 2.44% 97 1.74% 742 9.15% 4193 18.49%

unspec.

221776 100.00%

26821 100.00% 5562 100.00% 8111 100.00% 22682 100.00% 1013 100.00% 1,192 100.00%

total imports

 The people and languages of Suriname 

 51

1720–1729

1710–1719

1700–1709

1690–1699

1680–1689

1650–1679

A

Vd

A

Vd

A

Vd

A

Vd

A

Vd

A

Vd

3460 33.68% 3854 39.44% 3736 53.33% 3147 42.82% 5201 60.71% 5587 70.55% 5035 64.34% 5020 68.98% 4183 40.85% 2695 28.11%

816 9.71%

Bight of Benin

1125 10.95%

2160 25.71%

Bight of Biafra

5008 48.91% 6261 65.31%

421 4.91% 657 8.30% 174 2.22%

325 3.33%

1.70%

175

9.02%

758

Gold Coast

Sierra Leone 3571 42.51% 260 5.38% 5095 49.60% 4561 46.67% 3049 43.53% 2999 40.81% 2111 24.64% 1147 14.48% 1780 22.74% 1589 21.84% 632 6.17% 251 2.62%

West–Central Africa

Tab. A2: Slave imports to Suriname. Arends (1995, A) and the Voyages database (2008, Vd) compared. Windward Coast 1096 13.05% 4574 94.62% 417 4.06% 1032 10.56% 220 3.14% 1203 16.37% 834 9.74% 528 6.67% 837 10.70% 668 9.18% 417 4.07% 380 3.96%

unspec.

9587

10240

7277

7826

7919

8567

7349

7005

9772

10272

4834

8401

total

52   Robert Borges

1780–1789

1770–1779

1760–1769

1750–1759

1740–1749

1730–1739

A

Vd

A

Vd

A

Vd

A

Vd

A

Vd

A

Vd

Tab. A2 (continued)

690 1.88% 320 0.90% 286 1.07%

7009 40.28% 5602 31.78% 478 2.01% 478 2.18% 752 2.87%

Bight of Benin

45 0.17%

Bight of Biafra

4840 19.38% 3803 68.37% 1165 26.57%

17.10%

4587

9573 55.01% 9462 53.67% 1839 7.75% 1626 7.41% 7359 28.04% 5125 20.08% 5570 15.18% 6001 16.89%

Gold Coast

296 5.32%

204 0.78%

Sierra Leone 544 3.13% 1097 6.22% 7193 30.31% 4941 22.53% 7344 27.98% 5927 23.22% 15410 41.98% 12686 35.71% 8400 31.32% 8193 32.80% 462 8.31% 500 11.41%

West–Central Africa

62.02%

2719

276 1.57% 5941 25.03% 3796 17.31% 9178 34.97% 11959 46.85% 13434 36.60% 14003 39.42% 12849 47.91% 11293 45.21% 904 16.25%

1.59%

276

Windward Coast

654

1.74%

97

2.62%

2.44%

654

7.08%

2517

4.36%

1601

9.86%

2517

5.36%

1408

50.57%

11093

34.89%

8281

6.76%

1192

unspec.

4384

5562

24980

26821

35527

36705

25528

26245

21934

23732

17629

17402

total

 The people and languages of Suriname 

 53

Total

1790–1803

A

Vd

A

Vd

Tab. A2 (continued)

32116 15.35% 26703 14.71%

470 2.29%

Bight of Benin

7418 3.54%

4088 19.93%

Bight of Biafra

44472 21.25% 36599 20.15%

5205 25.38% 1137 23.34%

Gold Coast

3279 1.57%

2779 13.55%

Sierra Leone

58101 27.76% 45053 24.81%

2510 12.24% 902 18.52%

West–Central Africa

44320 21.18% 46698 25.72%

1738 8.47% 2652 54.44%

Windward Coast

19580 9.36% 26538 14.61%

180 3.70%

18.13%

3718

unspec.

100.00%

181591

100.00%

209286

100.00%

4871

20508

total

54   Robert Borges

Case studies

Kofi Yakpo

Creole in transition: Contact with Dutch and typological change in Sranan 1 Introduction Most Surinamese today acquire a heterogeneous variety of Sranan characterized by extensive admixture with Dutch. The analysis of a corpus of contemporary Sranan reveals variation in the expression of spatial relations and the realization of arguments in ditransitive constructions. Both domains feature syntactic rearrangements and semantic changes that replicate Dutch structures. Pattern replication has led to alterations in the frequency and distribution of Sranan elements and structures, as well as innovations with Sranan and Dutch borrowed elements fulfilling new, previously unattested functions. Sranan is undergoing a substantial typological shift from more substrate-oriented Kwa-like structures to ones similar to those found in the West Germanic superstrate Dutch. Societywide multilingualism involving Dutch, Sranan and often additional languages provides the socio-linguistic backdrop to contact-induced variation and change in Sranan.1 Voices in Suriname say that Sranan is being fundamentally transformed by contact with Dutch. There is a feeling among Surinamese linguists and language observers that the more monolingual variety of dipi Sranan, “elaborate Sranan”, of their parent and grandparent generation is no longer spoken by younger speakers of the urban and peri-urban zone of Paramaribo (e.g. Eersel, p.c.; D. France Oliviera, p.c.; Tjon Sie Fat, p.c.). This variety, they say, is being pushed aside by a heavily mixed variety of Sranan featuring a growing Dutch-derived and a reduced English- and African-derived vocabulary. The general sentiment is that competence in dipi Sranan is on the wane, that the “gaps” are being filled with Dutch elements and structures, and that an informal and colloquial register has become generalized as the only one available to many speakers. Sranan seems

1 I am indebted to Gracia Blanker, Henna Blanker, Renata De Bies, Hein Eersel, Stanley Hanenberg, Jit Narain and D. France Olivieira, without whose support and advice the research on which this study relies could not have been carried out. I owe particular gratitude to our Surinamese collaborating researcher D. France Olivieira, who transcribed the Sranan data. I am also grateful to Soraya Renjaan, who transcribed the Surinamese Dutch corpus. The linguistic annotation of the data was done by me with the SIL software FieldWorks Language Explorer (FLEx). I also wish to thank Bettina Migge for her valuable comments on a first draft of this chapter. DOI 10.1515/9781614514886-003

58 

 Kofi Yakpo

to be undergoing a more far-reaching contact-induced transformation than was the case in preceding centuries. Part of this change coincides with the expansion of plurilingual practices involving Sranan, Dutch, and often, a third Surinamese language like Sarnami, Javanese or Ndyuka (e.g. Migge 2015; Léglise and Migge 2015), and the consolidation of a mixed Sranan-Dutch code as an unmarked norm (Yakpo 2015). Sranan-Dutch mixing patterns that have already been conventionalized in Sranan are carried over, often through holophrastic borrowing and calquing into languages other than Sranan, thus representing manifestations of “second order code-switching” together with these third languages (Meeuwis and Blommaert 1998; Auer 1998a: 16). The use of plurilingual modes of speaking Sranan however varies between speakers and situations, depending on factors like domain, discourse participants, and so forth. We must also assume that some Sranan speakers acquire and use different styles of Sranan next to each other, some of which are more monolectal and homogenous, while others are more heterogeneous and mixed. Opting for one or the other variety in accordance with pragmatic conventions will therefore constitute an important part of an individual’s plurilingual competence in Sranan. A possible difference between present and past plurilingual practices is that the mixed heterogeneous style is becoming the dominant form in the panoply of available styles while the more homogenous variety is becoming a minority practice with fewer people today learning and making use of it. There is almost certainly also a geographic dimension to the prevalence of these plurilingual practices, which still awaits investigation. Our data seems to indicate, for example, that Sranan spoken in the more rural district of Coronie is on average less interspersed with Dutch than that of urban Paramaribo, and may indeed also be more conservative with respect to the changes described in this chapter. Various works have shown how features carried over from specific substrates, from the lexifier English, from the colonial superstrate Dutch and internal grammaticalization processes have shaped specific sub-systems of Sranan and the other creoles of Suriname, e.g. the verbal system (Migge 2003; Winford and Migge 2007; Borges et al. 2014, this volume; Essegbey, van den Berg and van de Vate 2013), the copula system (e.g. Arends 1986) and the nominal system (e.g. Bruyn 1995; van den Berg 2014). The analysis of contact-induced change in Sranan in this chapter covers aspects that have not been (fully) explored by previous work (e.g. Essegbey and Bruyn 2002; Yakpo and Bruyn 2015). In this chapter, I provide evidence for typological change in Sranan through contact with Dutch. I focus on two domains, namely (1) the grammar of space (locative constructions), and (2) argument structure (ditransitive constructions). I conclude that change involves the processes of “pattern borrowing” and some “matter borrowing” (Sakel 2007)



Creole in transition: Contact with Dutch and typological change in Sranan 

 59

from Dutch. The outcomes of these contact processes are alterations of the frequency and the distribution of Sranan elements and combinations of elements. On a whole, the changes confirm the impressions in Suriname about Sranan becoming more Dutch-like. At this point, it is however unclear whether this convergence is reflective of some degree of attrition and shift to Dutch by sectors of the urban population of Paramaribo, or whether the changes are taking place within the context of stable multilingualism. The data on which this chapter is based was gathered as part of the ERCfunded “Traces of Contact” Project (principal investigator Pieter Muysken) by me, Stanley Hanenberg, and Robert Borges in Suriname between 2010 and 2012 for seven Surinamese languages (Sranan, Ndyuka, Saramaccan, Kwinti, Sarnami and Surinamese Javanese, as well as Surinamese Dutch). The Sranan corpus on its own consists of about 60,000 words of elicited and naturalistic data. Examples from our Surinamese Dutch corpus (about 15,000 words) are also used in this chapter. About 60% of the corpus consists of data obtained via the use of focused elicitations, parallel texts (e.g. Mayer 1969) and director-matcher tasks covering the domains of spatial relations, tense-aspect-mood, grammatical relations and event integration. The remaining 40% consist of more naturalistic data including sociolinguistic interviews and informal conversations. In section 2, I discuss the plurilingual practices that characterize the linguistic scenario in which the changes described for Sranan have taken place. Section 3 provides a detailed overview of contact-induced developments in the expression of spatial relations, focusing on locative constructions. Section 4 looks at change in ditransitive construction and “transfer events” in particular. Section 5 addresses the outcomes and processes of contact-induced change observed in this study and section 6 concludes this chapter.

2 Plurilingual practices involving Sranan The contact-induced structural changes described in this chapter are rooted in plurilingual communicative practices that are part of the “normal way of speaking” in Suriname. The typical characteristics of plurilingual speech involving Sranan are the insertion of Dutch content words and free function words, the frequent alternation between Sranan and Dutch clauses and sentences, as well as lexical and structural calquing of Dutch elements and collocations, and viceversa, of Sranan elements in Dutch stretches of discourse. Much of the admixture of Dutch, even when it occurs on the spur of the moment, follows established routes of morphosyntactic adaptation in Sranan. Likewise, the admixture of

60 

 Kofi Yakpo

certain (classes of) Dutch elements and collocations in Sranan discourse often occurs with a higher likelihood than others, thereby constituting patterns in a heterogeneous mixed variety of Sranan. In this section, I will present some of the characteristics of this mixed Sranan code in the mould of which typological change in Sranan is taking place. Codeswitching is the most conspicuous plurilingual practice. It plays an important role in the emergence of the heterogeneous Sranan variety that is becoming the norm in Suriname, and probably lies at the origin of some of the structural changes that Sranan has undergone. The following excerpt stems from a procedural interaction, in which speaker 1 (SP1, male) asks speaker 2 (SP2, female) to relate to him a recipe for the preparation of afingi, ‘cassava dumplings’ (Sranan elements are in italics, Dutch elements in bold italics, Eastern Maroon Creole elements in bold italics and underlined): Excerpt 1: SP1 1 Yu srefi sabi meki en? 2sg self know make 3sg.indp ‘Do you yourself know how to make it [cassava dumplings]? SP2 2 Ja, heel lekker ook. yes whole tasty also ‘Yes, (it’s) really good too.’ SP1 3 Yu kan taygi mi fa yu e meki en? 2sg can tell 1sg how 2sg ipfv make 3sg.indp ‘Can you tell me how to make it?’ SP2 4 So afingi na geraspte cassave san den sma so dumpling foc grated cassava what def.pl person 5 e poti ini wan matapi. ipfv put in one cassava.tube ‘So afingi is grated cassava that people put into a cassava tube.’ switi kasaba? SP1 6 Gewoon a just def.sg sweet cassava ‘Just the sweet cassava?’ SP2 7 Iya, dan yu o rasp en, te yu rasp yes then 2sg fut grate 3sg.indp temp 2sg grate 8 en, dan yu o poti en ini wan 3sg.indp then 2sg fut put 3sg.indp in one 9 matapi fu puru a vocht uit, cassava.tube prep remove def.sg humidity part



Creole in transition: Contact with Dutch and typological change in Sranan 

 61

10 fu yu kan droog en uit, dan prep 2sg can dry 3sg.indp part then 11 yu o haal en door wan zeef. 2sg fut take 3sg.indp part one sieve ‘Yes, then you’ll grate it, when you’ve grated it, then you’ll put it into a cassava tube to take the humidity out, for you to dry it out, then you run it through a sieve.’ 12 Te yu haal en door a zeef, temp 2sg take 3sg.indp part def.sg sieve 13 dan ga je het een beetje sprenkelen met een then go you it a bit sprinkle with a beetje water. bit water ‘When you’ve run it through the sieve, then you’ll sprinkle it a bit with a bit of water.’ 14 Dan ga je balletje van maken. then go you little.ball from make ‘Then you’ll make little balls out of it.’ 15 En uit den bal di yu o meke de, and out 3pl ball sub 2sg fut make there 16 dan yu o puu pikin-pikin balletjes. 2sg fut remove red-small little.balls ‘And out of those balls that you’ll make there, then you’ll remove lots of little balls (from them).’ A first characteristic of the language used in excerpt 1 is that it is, in fact, a trilingual text. SP2, a 23 year old teacher residing in the Surinamese capital Paramaribo self-identifies as a person of Ndyuka (an Eastern Maroon Creole language) stock. There are at least two words that are Ndyuka in form, namely meke ‘make’ (line 15), and puu ‘remove’ (line 16). The corresponding Sranan forms are meki and puru. Phonological variation and mixing in a single stretch of discourse between cognate forms like meki (Sranan) and meke (Ndyuka), as well as the use of forms intermediate between Sranan and Ndyuka and homophonous dia-morphs like de ‘there’ (line 15) are characteristic not only for varieties of Sranan used by some Ndyuka speakers. It is also one of the hallmarks of an emerging koiné used in the interior of Suriname and in western French Guiana (see Migge and Léglise 2013; Migge 2015; Borges et al. 2014: 123–130, this volume). Besides that, excerpt 1 is replete with Dutch elements. This includes alternations at clause boundaries, in which the principal language switches completely

62 

 Kofi Yakpo

to Dutch (lines 2, and 13–14) and back to a mix of Sranan and Ndyuka (line 15). Beyond that, excerpt 1 features insertions of constituents. We find the inserted clause-peripheral Dutch adverbs gewoon ‘just’ (line 6) and dan ‘then’ (lines 8, 10) and the Dutch nouns vocht ‘humidity’ (line 9), zeef ‘sieve’ (line 11) and bal ‘ball’ (line 15), all of which are preceded by one of the Sranan determiners a ‘def.sg’, den ‘def.pl’ and wan ‘one’. We also find the insertion of a Dutch noun phrase, namely geraspte cassave ‘grated cassava’ as a complement to the Sranan focus marker and identity copula na ‘foc’ (line 4). The insertion of Dutch noun-­modifier constituents is common where such often conventionalized collocations are also frequent in Dutch-based discourse (cf. Khakimov 2015). The insertion of verbs follows a conventionalized pattern whereby an invariant inflected Dutch verb form, based on the most frequent form (1sg, 1sg/2sg in inverted questions, imperative) of the most frequently used Dutch TAM category (present tense), is used throughout the Sranan person paradigm in finite (i.e. rasp ‘grate’ in line 7, haal ‘take’ in lines 11 and 12) and in non-finite contexts alike (i.e. yu kan droog ‘you can dry’ in line 10). Sranan speakers also routinely incorporate the separable verbal particles of Dutch verbs in the postposed position following Dutch grammatical norms (the equivalent of English particles like out in a collocation like find out). These particle verb constructions can involve collocations of two Dutch elements, such as droog en uit ‘dry it out’ (Dutch in bold), in line 10. We also find mixed collocations, featuring a Sranan semantic calque of a Dutch verb, followed by a Dutch particle, as in puru a vocht uit ‘remove/take the humidity out’ (line 9), where the use of puru in this context is calqued on the Dutch verb haal- ‘take, remove’. Excerpt 1 does not contain instances of full calques of Dutch (particle) verbs, but one example found elsewhere in our data is the Sranan collocation go abra literally ‘go over’, which replicates the semantics of a corresponding Dutch idiom with the sense of ‘be about’ as in this story is about X. The same kind of flexibility with respect to the insertion of particle verbs has been observed for other Surinamese languages, e.g. Ndyuka (Borges 2014a) and Sarnami (Borges et al. 2014: 201–204). The continuum of the insertion of (a) fully Dutch items, hence matter transfer, (b) mixed Sranan-Dutch expressions involving partial calquing and (c) full calques of Dutch expressions and hence the transfer of patterns alone is also characteristic for the mixing of other elements in plurilingual discourse involving Sranan. All in all the insertion of adverbs, nouns and verbs is highly conventionalized and adheres to relatively established norms, in accordance with a grammar of code-switching so to say. Code-switching involving Sranan is governed by a multitude of factors, ranging from discourse-pragmatic and information-management ones like turn-taking, commenting and framing (e.g. line 2), to memory effects like retrieval or priming.



Creole in transition: Contact with Dutch and typological change in Sranan 

 63

What is striking regarding excerpt 1 is that such plurilingual practices, whatever their local significance within a given text, are highly conventionalized. This means that Sranan and Dutch, two different ‘languages’, constitute a unitary code to fulfill linguistic functions within a conversation that would be fulfilled by linguistic material, structures, and stylistic devices from a single ‘language’ in a monolectal text. Many of the switching phenomena we can observe are therefore not “interactionally meaningful” (Auer 1998a: 20). They are rather manifestations of the conventionalized mixed code that has come to characterize this unitary use of Sranan and Dutch. I argue elsewhere that this evolving grammar of mixing as well as the discourse-pragmatic motivations for code-mixing are very similar, and in some cases identical across the languages of Suriname (see Yakpo 2015). Thus, a linguistic area has emerged in Suriname, with Sranan as one of its two cornerstones. This area consists of layers of convergence towards Dutch and Sranan as donor and recipient languages to each other, and donor languages to other languages of Suriname. Many conventionalized patterns and elements ultimately of Dutch origin have entered third languages like Sarnami and Surinamese Javanese via Sranan, where they were already established beforehand. Sranan therefore plays a dual role as a conduit for indirect Dutch influence on other languages of Suriname, while simultaneously exerting direct influence on these languages.

3 Contact-induced change in the expression of spatial relations In this section, I cover contact-induced change in the grammar of space in Sranan. Sranan features typological specificities in its grammar of space that set it apart from its lexifier English and its superstrate Dutch. These are the existence of postpositional structures featuring locative or relator nouns, the use of a general locative preposition, and serial verb constructions. I will show that contemporary Sranan has undergone profound restructuring and a typological shift away from these characteristics in its grammar of space due to contact with Dutch. The existing literature, cited where relevant, addresses some of the contact-induced changes described in the following. This study is, however, the first work to provide detailed analyses, explore not yet described aspects of these changes, and corroborate claims to change with statistical evidence based on primary data. Tab. 1 summarizes salient characteristics of locative constructions discussed in the following sections.

64 

 Kofi Yakpo

Tab. 1: Characteristics of locative constructions. Characteristic

N

In %

Examples

(a) Complex locative constructions over total of locative constructions (b) Postpositional structures over total of locative constructions (c) General locative preposition marks Source over total of Source-oriented constructions (d) General locative preposition marks Goal over total of Goal-oriented constructions

91/1335

6.8%

(1), (5)

2/1335

0.1%

(1), (6)

7/73

9.6%

(6)

253/398

63.6%

(12), (13)

Tab. 1 seems to indicate that Sranan is undergoing a transformation of its grammar of space, characterized by the demise of complex locative constructions (a); the loss of postpositional structures (b); a low frequency of the general locative preposition in Source-oriented events (c); and a lower than expected frequency of the general locative preposition in Goal-oriented events (d).

3.1 From postpositional to prepositional structures Earlier varieties of Sranan featured locative constructions involving the simultaneous use of a general locative preposition and a postposed or preposed locative noun. Contemporary Sranan is characterized by the use of locative constructions featuring specific prepositions very similar to corresponding Dutch ones. I argue that contact with Dutch has led to two linked changes in locative constructions. Firstly, the use of postposed locative nouns has been abandoned, and locative nouns are exclusively found in a preposed position before the Ground noun in contemporary Sranan. Secondly, the general locative preposition na ‘loc’ only rarely co-occurs with preposed locative nouns in “complex locative constructions”. The dominant type of locative construction found in the data is, instead, a “simplex locative construction”, which features erstwhile Sranan locative nouns functionally converted to Dutch-style prepositions expressing specific topological relations. Tab. 2 lists the absolute and relative frequencies of simplex locative constructions with each of the five principal locative elements in the corpus.2 The four

2 Tab. 2 lists the occurrences of locative constructions in which the Ground is explicitly mentioned, hence structures like (1) to (5). An example of a structure in which the Ground is not explicitly mentioned is a de na tapu [3sg.sbj cop loc top] ‘It is on top/above.’ Here, the locative element tapu is nominal by default and these structures therefore require the use of a relational element like na.

Creole in transition: Contact with Dutch and typological change in Sranan 



 65

Tab. 2: Frequencies of locative elements in simplex constructions relative to complex constructions. Locative element

ini

tapu

baka

ondro

fesi

Total

Total Total in %

531 94.7%

514 91.6%

158 98.1%

23 88.5%

18 69.2%

1244 93.2%

most frequent locative elements ini ‘in(side)’, tapu ‘on/top’ and baka ‘behind/ back’ occur far less often in complex constructions (5.3%, 8.4%, 1.9%, 11.5% respectively) than in simplex ones (94.7%, 91.6%, 88.5% respectively). The occurrence of the least frequent locative element fesi ‘(in) front (of)’ in complex locative constructions (30.8%) as opposed to simplex ones (69.2%) is significantly higher. However, all but one occurrence of the collocation na fesi stem from a single, elderly speaker (70+ years), an indication that the variation described in this section probably reflects diachronic change. Tab. 2 therefore seems to point to a shrinkage of the functional scope of the general locative preposition na. Example (1) below involves the elements that make locative constructions in earlier varieties of Sranan differ from those of contemporary Sranan: (i) a locative construction introduced by a general locative preposition (na ‘loc’, alternatively realized as a), with vague spatial semantics, and employed in motion and stative events alike; (ii) a postpositional locative noun (tapu) that specifies the Region of a Ground (ede ‘head’); (iii) the Ground noun and the Region noun are linked in a possessive/modifying construction, with the Ground functioning as the possessor/ modifier noun and the locative noun functioning as the possessed/modified noun and syntactic head to the construction (Yakpo and Bruyn 2015: 142, 166): (1)

A dagu kren na a boy ede tapu def.sg dog climb loc def.sg boy head top fu no nati skin. prep neg wet body ‘The dog climbed onto the boy’s head so as not to wet (its) body.’

Very similar constructions in the Niger Congo phylum and the Gbe grouping in particular have been argued to form the template of the Sranan construction (Bruyn 1996; Essegbey 2005; Yakpo and Bruyn 2015). I will henceforth refer to such structures, which involve a general locative preposition and a locative noun as “complex locative constructions”. Complex locative constructions featuring postpositional locative nouns like (1) are found in eighteenth century Sranan sources (see Essegbey and Bruyn 2002; van den Berg 2007) and are attested well into the mid-twentieth century (see e.g. Voorhoeve 1953; Voorhoeve 1962). Such constructions are also well documented and still in use in the Maroon Creole languages (for Ndyuka, see Huttar and Huttar 2003: 531; for Saramaccan, see Muysken 1987;

66 

 Kofi Yakpo

McWhorter and Good 2012: 186). This is additional evidence that the type of locative construction in (1) has undergone change in Sranan and is likely to have been more wide-spread before. In our corpus of contemporary Sranan, such structures have, however, all but disappeared. There are only two instances in the entire corpus. Both instances are produced by the same elderly speaker of 60+ years (cf. example (1) above). In other descriptions of spatial relations, even this speaker instead employs structures like the following one: (2) A kren tapu wan schutting fu kisi 3sg.sbj climb on one fence prep get a froktu. def.sg fruit ‘He climbed on a fence to get (hold of) the fruit.’ Example (2) above differs from (1) in two respects: (i) the general locative preposition na is no longer made use of; (ii) tapu is now a prepositional element, hence found before the Ground (wan schutting) rather than after it and behaves very much like the English preposition on or Dutch op (see Yakpo and Bruyn 2015 for a detailed discussion of the morphosyntactic evidence). Locative elements other than tapu are also used like prepositions. Compare the use of ondro ‘under’ in (3). I refer to structures featuring prepositional uses of erstwhile locative nouns like tapu and ondro in (2) above and (3) below as “simplex locative constructions”: (3) A poti a sturu ondro a bon. 3sg.sbj put def.sg chair under def.sg tree ‘He put the chair under the tree.’ It is very likely that Dutch provides the template for simplex locative constructions in Sranan. Prepositional structures are the default option in Dutch for expressing core spatial relations like ‘under’, ‘in’ or ‘on’. Compare (3) above with example (4) below from our corpus of Surinamese Dutch: (4) Een muis slaapt onder de boom. a mouse sleeps under the tree ‘A mouse is sleeping under the tree.’ (Surinamese Dutch) I now discuss an additional aspect of the contact-induced shift to prepositional structures in contemporary Sranan. There is a third logical possibility, intermediate between (1) and (2) above, which is the use of the general locative preposition na ‘loc’ in tandem with the prepositional use of locative elements like tapu and ondro. Such structures have been recorded as alternatives to postpositional ones like (1) above in Early Sranan since the eighteenth century (Yakpo and Bruyn



Creole in transition: Contact with Dutch and typological change in Sranan 

 67

2015: 140–152) and are attested in our corpus as well. Complex locative constructions featuring preposed locative nouns are also given as the only option together with ones featuring postposed locative nouns (i.e. (1)) by the Surinamese linguist van der Hilst (2014: 165). An example follows: (5)

A e sidon na tapu a fiets èn 3sg.sbj ipfv sit loc top def.sg bike and a e rey en gwe. 3sg.sbj ipfv ride 3sg.indp go.away ‘She’s sitting on the bike and she’s riding it away.’

There is reason to assume that the prepositional locative element (i.e. tapu) in examples like (5) is also a noun-like element, hence a locative noun as in postpositional structures like (1) above. In our corpus of contemporary Sranan, the co-occurrence of the general locative preposition with a prepositional locative element as in (5) is, however, rare. There are however a handful of occurrences of the preposition nin ‘in’ (see also Essegbey and Bruyn 2002: 15), a variant of the Sranan preposition ini ‘in’. Diachronically, nin is a merger of the general locative preposition na ‘loc’ and the locative noun ini ‘in(side)’, hence a complex locative construction. In contemporary Sranan nin is, however, perceived as a mono-morphemic form by speakers and it is therefore counted under simplex constructions in Tab. 2. I now turn to specific developments in the expression of Source, Goal and Place relations providing further evidence for systemic change in the grammar of space of Sranan.

3.2 Changes in the expression of Source, Goal and Place The expression of Goal, Source and Place in Sranan is characterized by a number of features suggestive of contact-induced change. Sranan has moved away from a more isomorphic system in which the functional elements participating in a spatial description denote one particular aspect of it, to a system featuring portmanteau prepositions. The formal correspondences of these changes are a decline in the use of the general locative preposition in motion events in general, and in Source-oriented relations in particular, and the demise of locative nouns (and hence of complex locative constructions) in Goal, Source and Place relations. Sranan has also borrowed the Dutch Sourceoriented preposition uit ‘out of, from’ and there is evidence for lexical calquing of the functions of the Dutch preposition op ‘on’ onto the e­ quivalent Sranan locative element tapu ‘on’.

68 

 Kofi Yakpo

3.2.1 The expression of Source The overwhelming majority of speakers in the corpus express Source in a way very similar to Dutch: A single path-incorporating preposition occurs before the Ground noun to mark Source, either fu ‘from’ or the Dutch-derived uit ‘out of, from’. Tab. 3 provides the frequencies of the three most frequent prepositions in Source-oriented constructions (and an additional single occurrence of ini ‘in’) with the two most common Source-trajectory verbs in the corpus: the intransitive motion verb komoto ‘come from’ and the caused-motion verb puru ‘remove’). The table shows that Source is mainly marked by fu and uit with a comparable frequency of 47% and 43% respectively. These two prepositions are therefore each more than four times more common as a means to mark Source than na with its roughly 10%. Taken together the prepositions fu and uit are used in 90% of all instances to mark Source in the corpus. Tab. 3: Marking of Source in locative constructions. Verb/Preposition

na

fu

uit

ini

komoto/komopo ‘come from’ puru ‘remove’ Total

4 3 7

17 17 34

17 14 31

1 0 1

Total in %

9.6%

46.6%

42.5%

1.4%

Total 39 34 73 100%

Sranan speakers in our corpus therefore employ Path-incorporating prepositions to mark Source in the overwhelming number of cases in locative constructions whose structure and semantics are very similar to those of Dutch. Compare the Goal-oriented construction in (1) above with the Source-oriented construction in (6) below. In both examples, we find locative constructions introduced by na, which is lexically unspecified for Path. Typical for complex locative constructions, we also find a locative noun expressing the Region and heading a possessive construction with the Ground as a possessor. (6) A teki a sani na a tafra tapu. 3sg.sbj take def.sg thing loc def.sg table top ‘She took the thing from the table.’ Source-oriented constructions like (6) are, however, rare. It is more common to mark a Source by means of the Sranan multipurpose preposition fu ‘at, from’ (a reflex of the English preposition for), as in (7) or by the preposition uit ‘out of, from’ (8), which has been borrowed from Dutch (Essegbey and Bruyn 2002):



Creole in transition: Contact with Dutch and typological change in Sranan 

 69

(7) A froktu san komoto fu a bon na def.sg fruit what come.out from def.sg tree foc wan maka-maka froktu. one red-thorn fruit ‘The fruit that came off the tree is a thorny fruit.’ (8) Mi si wan man puru swarfu uit a dosu. 1sg see one man remove match out.of def.sg box ‘I saw a man remove a match from the box.’ Constructions involving both fu and uit are highly similar to the corresponding Dutch ones in various ways. For comparison, I provide an equivalent of sentence (8) above from Surinamese Dutch: (9) Een persoon pakt een lucifer uit het doosje. a person takes a match out.of the little.box ‘A person is taking a match from the little box.’(Surinamese Dutch) For one, the locative constructions in both languages share the same linear order, with the preposition uit directly followed by the Ground NP (bon, dosu and doosje in (7), (8) and (9) respectively) – there is no additional locative noun. Secondly, the constructions are semantically similar. Sranan and (Surinamese) Dutch both feature portmanteau prepositions which simultaneously encode Path, Region and the Spatial Relation that holds between a Ground and a Figure. This stands in contrast to the serial verb construction in (10) below, where each constituent denotes one particular aspect of the spatial description (Yakpo and Bruyn 2015: 140–141) and where Path is neither part of the meaning of the general locative preposition na, nor of the locative noun. It is rather the second verb in the series puru ‘remove’ or alternatively, komoto ‘come from’ that provides the sentence with (Source) motion semantics. teki wan swarfu puru/komoto (10) A man def.sg man take one match remove/come.from na ini a dosu. loc inside def.sg box ‘The man took a match from the box.’ Directional serial verb constructions like (10) have been seen as a typological hallmark of Sranan (cf. Bruyn 1995: 241–253; Essegbey and Bruyn 2002; Yakpo and Bruyn 2015). However, the count in Tab. 3 does not contain a single serial verb construction. In addition, our corpus features far more Sources marked by fu ‘at, from’ and uit ‘out of’ than na ‘loc’. I attribute it to contact with Dutch that the vast majority of Source relations is marked by way of these two prepositions rather than na, as in (6), or na in combination with a serial verb construction, as in (10).

70 

 Kofi Yakpo

3.2.2 The expression of Goal The expression of Goal in Sranan also appears to be changing through contact with Dutch. As with Source, the use of complex locative constructions involving locative nouns is very rare. Instead we find locative nouns being used as prepositions indicating specific topological relations (e.g. ini ‘in’ and tapu ‘on’). However, the use of na to mark Goal is still more common than for marking Source. This may point to an emerging Goal-Source asymmetry: Goal and Place are coded by the same markers (i.e. na ‘loc, and specific prepositions like ini ‘in’), while Source is marked distinctly from both Goal and Place (i.e. by the path-incorporating prepositions fu and uit). Tab. 4 shows the frequencies of the four most common Goal marking patterns with the six most common motion verbs. Figures for complex locative constructions (e.g. na ini) are given in brackets, simplex constructions (e.g. ini) are without brackets:3 Tab. 4: Marking of Goal in locative constructions. Verb/Pattern

na

go ‘go to’ kon ‘come to’ fadon ‘fall (in/on) to’ waka ‘walk’ doro ‘arrive at’ lon(we) ‘run (off) to’ Total

203 28 9 6 4 3 253

Total in %

64%

(na +) ini

(na +) tapu

(1) 62 (1) 3 (0) 9 (0) 5 (0) 0 (0) 0 (2) 79

(3) 21 (0) 3 (4) 25 (0) 3 (0) 0 (0) 0 (7) 52

(0.5%) 20%

(1.5%) 13%

ø

Total 0 0 0 0 5 0 5

290 35 47 14 9 3 398

1%

100%

The figures in columns 3 and 4 of Tab. 4 show that complex locative constructions are exceedingly rare in the expression of Goal (0.5% and 1.5%, i.e. 3% of the total number of constructions). Speakers overwhelmingly make use of simplex (prepositional) locative constructions. We therefore see the same clear tendency towards simplex prepositional structures already observed with the expression of Source. The following example shows such a Goal-oriented structure, featuring fadon ‘fall’: (11) Ma di a du dati, a boy fadon ini but sub 3sg.sbj do dist def.sg boy fall in

3 Tab. 4 includes directional serial verb constructions. Hence locative constructions like a waka go na foto [3sg.sbj walk go loc town] ‘he walked to town’ are counted as instances of go ‘go to’ (row 1).



Creole in transition: Contact with Dutch and typological change in Sranan 

 71

a watra. def.sg water ‘But when he did that, the boy fell in(to) the water.’ The use of ini in (11) provides a specific spatial reading. The use of specific prepositions (i.e. ini ‘in’ and tapu ‘on’) however alternates with the use of the general locative preposition na. The latter use is exemplified in sentence (12), uttered by another speaker than (11), yet describing the same scene in the Frog Story (Mayer 1969). Tab. 4 (columns 2 and 3) show that both alternatives are equally common with fadon, with nine occurrences each: (12) Ô, a fadon na watra. intj 3sg.sbj fall loc water ‘Oh, he has fallen in(to) the water.’ The variation between the specific preposition ini and the general locative preposition na is also encountered with the verb go ‘go’, where the marking of Goal by na is more common (60%) than by the two specific prepositions ini and tapu together (83/203). The use of the general locative preposition is particularly common with the generic motion verbs go ‘go’ and kon ‘come’ because they very often feature stereotypical, discursively backgrounded Goals, marked by the general preposition rather than a specific one. Compare (13), where the discovery of the wasp nest is foregrounded, with (14), where there is less of a pragmatic hierarchy between the two clauses: (13) Den go a busi, den si wan bigi waswasi nesi. 3pl go loc forest 3pl see one big wasp nest ‘They went (in)to the forest, (and then) they saw a big waspnest.’ (14) A go ini a busi, a bari. 3sg.sbj go in def.sg forest 3sg.sbj shout ‘todo, todo pe yu de?’ frog frog where 2sg cop ‘He went into the forest, he shouted, ‘frog, frog, where are you?’ The simplex prepositional locative constructions in (11) and (14) mirror corresponding Dutch ones. Compare the following Goal-oriented construction in Surinamese Dutch, disregarding subject-verb inversion: (15) En zo vielen beiden dan in het water. and so fell both then in the water ‘And so both fell in(to) the water.’ (Surinamese Dutch) The parallels between Dutch and Sranan Goal-oriented constructions also extend to semantic aspects. When specific prepositions like ini (ex. (11), (14)) tapu (ex. (2)) and

72 

 Kofi Yakpo

ondro (ex. (3)) mark Goals they remain unspecified for Path (allative/motion-to) in their meaning. In this they differ from the Source prepositions fu and uit (ex. (7) and (8)), which incorporate Path (ablative/motion-from) in their meaning. Dutch also makes a lexical distinction between Path-incorporating (directional) and static, Place-denoting prepositions (Gehrke 2006). The directional preposition naar ‘to’, for example, denotes the end-point of a motion, while the static adpositions in ‘in’, op ‘on’, achter ‘behind’, and voor ‘in front of’ only receive a directional reading when they occur with motion verbs or in particular constructions (e.g. when some of them are used as postpositions). Hence the Dutch example in (15) can potentially be interpreted as either static or directional, just like the Sranan example (12) above (i.e. ‘fell in the water, not on the river bank’). Contemporary Sranan therefore replicates the Dutch Goal-Source asymmetry in which basic Goal and Place relations can be marked by static prepositions and Source is marked differentially, by directional prepositions. The question remains, however, why the general locative preposition na is still more frequent than other prepositions with the locomotion verbs listed in Tab. 4. The table shows that waka, doro and lon/lonwe feature more Goals marked by na on its own (na + ø in Tab. 4) than Goals marked by other locative elements. A possible explanation for the strong presence of na is reinforcement from Dutch. The Dutch preposition naar ‘to(wards)’ and the Sranan preposition na are interlingual (near) homophones. There is also a considerable functional overlap between the two prepositions. Both can be used on their own in Goal-oriented constructions involving motion along a Path and up to a (Region of a) Ground. Both prepositions therefore have a strong tendency to co-occur with high frequency motion verbs like ‘go’, ‘come’, and ‘walk (to)’ whose Goals often do not involve containment (i.e. ‘go in’), or contact (i.e. ‘go on’) or other more specific spatial relations (e.g. ‘go under sth.’). The retention of na in Goal-oriented constructions as opposed to its demise in Source-oriented constructions might therefore also well be contact-related.

3.2.3 The expression of Place I now turn to the expression of the spatial relation of Place, where there is also evidence for contact-induced change. Here too, the frequency of simplex locative constructions involving a single specific preposition (e.g. ini ‘in’, baka ‘behind’) by far outstrips that of complex locative constructions featuring the general locative preposition na and a locative noun. Yet, simplex locative constructions for expressing Place relations are also undergoing change: The use of the general locative preposition alone in order to locate a Figure in a stereotypical location (e.g. na oso ‘at home’) varies with the use of the preposition tapu ‘on’, thus replicating the semantics of the Dutch preposition op ‘on’.



Creole in transition: Contact with Dutch and typological change in Sranan 

 73

Tab. 5: Marking of Place in locative constructions. Verb/locative element(s)

na

(na +) ini

(na +) tapu

de ‘be at’ sidon ‘sit’ tnapu/knapu ‘stand’ didon ‘lie’ Total

31 0 0 1 32

(4) 50 (0) 1 (0) 1 (0) 2 (4) 54

(1) 49 (7) 36 (1) 29 (2) 7 (11) 121

Total in %

13%

(2%) 22% (5%) 50%

(na +) baka

(na +) ondro

(na +) fesi

(0) 2 (0) 0 (0) 5 (0) 0 0 (7)

(0) 2 (0) 0 (0) 3 (0) 0 (0) 5

(1) 0 (1) 0 (0) 5 (0) 0 (2) 5

(0%) 3%

(0%) 2%

(1%) 2%

Total 140 45 44 12 241 100%

Tab. 5 presents a count of Place-oriented constructions involving the four most frequent verbs to participate in such structures in the corpus. Here too, the figures for complex locative constructions are given in brackets, simplex constructions are without brackets. It shows more simplex locative constructions in Place relations than complex ones, continuing the tendency already observed with Source and Goal relations. Simplex prepositional locative constructions are about ten times more frequent than complex ones, both on a whole (79% vs. 8%), and with the most common locative elements ini (22% vs. 2%) and tapu (50% vs. 5%). Example (16) features a Basic Locative Construction (Ameka and Levinson 2007), the most frequent type of simplex Place-oriented construction in Tab. 5 (58% of Place relations (140/241), see the first line under the captions). The most frequent Basic Locative Construction involves the locative-existential copula de followed by a locative complement introduced by tapu ‘on’, compare (16). (16) Kon un taki yu de tapu a wroko, yu come 1/2pl talk 2sg cop on def.sg work 2sg ala-dey wroko, sortu tongo yu e taki moro furu? all-day work type tongue 2sg ipfv talk more much ‘Let’s say if you are at work, at your everyday work, which language do you talk more often? The occurrence of tapu in the Basic Locative Construction in (16) above is noteworthy. In the example, tapu does not literally describe the position of the Figure (yu ‘2sg’) in a superior location in relation to the Ground (wroko ‘work’). In (16), tapu instead marks a general or stereotypical location. The marking of non-specific location by way of tapu is common in the data. Other examples are tapu a uku ‘at/ on the corner’, tapu a skoro ‘at school’, tapu wan dey ‘(on) one day’. The expression of non-specific topological relations is expected to fall within the functional ambit of the general locative preposition na, as in the analogous example (17). Other examples of non-specific uses of na in Basic Locative ­Constructions are de

74 

 Kofi Yakpo

na oso ‘be at home’, de na wroko ‘be at work’, de na skoro ‘be at school’, de na China ‘be in China’). mi de na wroko, nanga (17) Luku doorgaans te look throughout temp 1sg cop loc work with mi chef yere, mi e taki Nederlands, af.en.toe 1sg boss hear 1sg ipfv talk Dutch now.and.then Negerengels. Sranan ‘Look throughout, when I am at work, with my boss, right, I speak Dutch, (and) now and then Sranan.’ The use of tapu to mark a non-specific sense is a carry-over from Dutch, where the equivalent element, the preposition op ‘on’, fulfills a similar range of functions. Examples are op werk ‘at work’, op de hoek ‘on the corner’, op school ‘at school’, op een dag ‘(on) one day’. It is likely that the calquing of such idiomatic uses from Dutch constitutes a port of entry for the use of tapu for other stereotypical or non-specific uses in Sranan in contexts that would otherwise require the use of the general locative preposition na. Other instances in the data in which tapu is used where na would otherwise be expected are tapu wan boto ‘in a boat’ or tapu a dyari ‘in the garden’. Other marginally locative or oblique uses of tapu calqued from Dutch found in the data are stik krosi tapu wan masyin ‘stitch clothing on/with a (sewing) machine’ and mi musu gi yu tu piki tapu a aksi dati ‘I have to give you two answers to that question’. To conclude, Place-oriented constructions in Sranan are characterized by similar tendencies as Source- and Goal-oriented relations. We find a preponderance of simplex locative constructions, which have become far more frequent than corresponding complex ones. We also find a shrinkage of the functional range of the general locative preposition, and an expansion of tapu calqued on similar uses in Dutch. Also noteworthy is the existence of a Goal-Source asymmetry: Place and Goal relations are marked by the same means, namely prepositions with static senses, while Source is marked by motion-incorporating prepositions.

3.3 Conclusion Contact with Dutch has led to the following changes in the expression of spatial relations in Sranan: a) The loss of postpositional structures, the demise of prepositional complex locative constructions and the concurrent consolidation of simplex prepositional structures. b) In Source-oriented constructions: The demise of na, the borrowing of uit from Dutch, and the use of uit and fu as Path-incorporating prepositions.



Creole in transition: Contact with Dutch and typological change in Sranan 

 75

c) In Goal-oriented constructions: The specialization of na to marking the Goals of generic motion verbs like go ‘go’ and kon ‘come’, waka ‘walk’, possibly due to convergence with Dutch naar ‘to(ward). The predominance of non-directional (static) prepositions with more specific meanings with other verbs (e.g. ini ‘in’, tapu ‘on’, ondro ‘under’). d) In Place-oriented relations: The emergence of tapu ‘on’ as a preposition with a general, non-specific sense through calquing of the functions of the Dutch equivalent op ‘on’, and the corresponding shrinkage of the functional range of the Sranan general locative preposition na. The changes manifest in (a) to (d) above and in Tab. 1 show a typological transition in the grammar of space in Sranan. The once dominant pattern in Sranan is one in which (1) prepositions and locative nouns express static relations while directional (serial) verbs alone express motion. This pattern has largely been replaced by one in which (2) portmanteau prepositions can express static and motion relations, thus incorporating Path. Pattern (1) is a Kwa, Niger-Congo and areal West African typological one common to Sranan substrate languages like Fon and Kikongo (Heine, Claudi and Hünnemeyer 1991: 140–143; Creissels 2006; Yakpo and Bruyn 2015: 140–152). Pattern (2) is characteristic of Sranan’s West ­Germanic superstrate Dutch and lexifier English. A second typological change in the grammar of space in Sranan is the development of a Goal-Source asymmetry (Blake 1977; Nikitina 2009; Pantcheva 2010; Zwarts 2010). The earlier Sranan system is symmetrical, with Goal, Source, and Place being obligatorily marked by the general locative preposition na and where necessary, locative nouns. The present system is asymmetrical. Place and Goal are marked by the same elements, namely the general locative preposition na, or erstwhile locative nouns like tapu, now used as prepositions. In contrast, Source is now almost exclusively marked by the Path-incorporating prepositions fu ‘from’ and uit ‘out of, from’, the latter having been borrowed from Dutch. Both typological changes have removed the grammar of space of Sranan away from that of its Niger-Congo substrates and drawn it closer to Dutch. As a consequence, Sranan has become more Germanic in its typological profile.

4 Contact-induced change in ditransitive constructions In this section, I will focus on developments in argument marking. Specifically, I cover the coding of Theme and Recipient, as well as verbal and prepositional semantics in ditransitive constructions. I argue that contact with Dutch is

76 

 Kofi Yakpo

r­ esponsible for the reanalysis of gi ‘give’ as a preposition equivalent to English ‘to’, the virtual absence in the corpus of the take-give serial verb constructions, and the high frequency of ditransitive constructions calqued on Dutch structures and semantics. The statistical tendencies present in the corpus may be indicative of ongoing change and possibly constitute a further area of typological alignment with Dutch.

4.1 Double objects, prepositional objects and serial verbs Contemporary Sranan features ditransitive constructions that are strikingly similar to Dutch ones. Syntactic evidence suggests that the Sranan verb gi ‘give’ today functions as a preposition in ditransitive constructions. An erstwhile SVC of the ‘give’ type has been reanalyzed as a Prepositional Object Construction (henceforth POC). Other developments likely to have been caused or reinforced by contact with Dutch are the emergence of a lexical preference for the verb langa ‘hand over’ in POCs for the description of literal transfer events, the generally high frequency of Double Object Constructions (henceforth DOCs) and the exclusive use of DOCs in non-literal transfer events. There are two ways of realizing the two participants of ditransitive events in Sranan. In the monoverbal DOC the Recipient and the Theme are distinguished by word order alone, and the Recipient invariably precedes the Theme, as in (18). (18) A wan nanga redi trui sori a tra def.sg one with red pullover show def.sg other man wan buku. man one book. ‘The one with the red pullover showed the other man a book.’ The bi-verbal serial verb construction (henceforth SVC) features the inverse order. The Theme immediately follows the verb of transfer and the Recipient is marked by the “verb” gi ‘give’, as in (19). The categorial status of gi ‘give’ in SVCs like (19) below has been the subject of debate in the literature on Sranan. Voorhoeve (1975) and Sebba (1987) assign gi an intermediate status between verb and preposition on the basis of distributional evidence. In the meantime, and for clarity of exposition, I will continue referring to constructions like (19) as “SVCs” (in quotes), before addressing the categorial status of gi in more detail further below. (19) A man sori a buku gi wan tra man. def.sg man show def.sg book give? one other man ‘The man showed the book to another man.’



Creole in transition: Contact with Dutch and typological change in Sranan 

 77

The frequencies of SVCs and DOCs with the most common transfer and communication verbs in the corpus are summarized in Tab. 6 below. Monoverbal DOCs featuring gi ‘give’ (first line of Tab. 6) constitute the most frequent ditransitive construction, cf. (20). They cater for 47% of all occurrences of ditransitive constructions in the data. (20) Wan yonku-man e gi wan dame tu pata. one young-agn ipfv give? one lady two sports.shoe ‘A youngster is giving a lady two sports shoes.’ Tab. 6: Frequencies of ditransitive constructions. Event type

Verb

Gloss

“SVC”

DOC

Transfer

gi langa fringi tya(ri) teki sori seni kari taygi ferteri

‘give’ ‘hand (to)’ ‘fling’ ‘carry’ ‘take’ ‘show’ ‘send’ ‘call’ ‘tell’ ‘tell’

0 (0%) 32 (23%) 11 (8%) 4 (3%) 3 (2%) 1 (1%) 0 (0%) 0 (0%) 0 (0%) 0 (0%)

66 (47%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 2 (1%) 1 (1%) 8 (6%) 7 (5%) 6 (4%)

66 (47%) 32 (23%) 11 (8%) 4 (3%) 3 (2%) 3 (2%) 1 (1%) 8 (6%) 7 (5%) 6 (4%)

51 (36%)

90 (64%)

141 (100%)

Communication

Total

Total

The second most frequent construction in Tab. 6 is an “SVC” featuring gi as a marker of the Recipient argument. Example (21) is an alternative rendition of (20), provided by a different speaker in response to the same visual stimulus. (21) A boy e langa a susu gi def.sg boy ipfv hand.over def.sg shoe give? a uma. def.sg woman ‘The boy is handing (over) the shoe to the woman.’ Tab. 6 contains a large number of ditransitive constructions involving the verbs langa ‘hand over’ and fringi ‘fling’ as opposed to others like sori ‘show’ or seni ‘send’. This is partly an artifact of the elicitation, which involved a large number of throwing and handing-over events. It is nevertheless interesting to explore why “SVCs” of the langa–gi type (see (21)) are the second most frequent ditransitive construction (23%) after monoverbal DOCs featuring the use of gi

78 

 Kofi Yakpo

alone (as in (20)). This is somewhat unexpected since the literature on Sranan treats teki–gi ‘take–give’ sequences as a common structure for marking a Recipient in a ditransitive transfer event in the language (e.g. Sebba 1987: 74). The number of teki–gi sequences is, however, very low in the corpus (2% of total ditransitive constructions). When the sequence does occur, teki also receives a literal reading. In (22), for example, the Agent ‘goes, takes and gives’ the chair to the Recipient. (22) A heer go teki wan sturu gi a dame. def.sg gentleman go take one chair give def.sg lady ‘The gentleman went to take a chair to the lady.’ For the speakers in our corpus, langa-gi is therefore the preferred structure for expressing a literal transfer (of an object). The langa–gi “SVC” has gone unnoticed in works on Sranan and may be indicative of a change in progress in lexical preference that has been caused by contact with Dutch.4 The following two examples stem from our corpus of Surinamese Dutch (cf. Muysken, this volume). Both feature a prepositional object construction (POC), in which the Recipient is marked by the preposition aan ‘to’. The semantics and structure of the langa–gi string reiterate the most common equivalent Surinamese Dutch structures in the data, namely geven–aan ‘give to’ and overhandigen–aan ‘hand over to’. The Sranan and Surinamese Dutch structures are also semantically similar in that they both depict literal transfer events. Compare the correspondence in constituent order between the Dutch preposition aan ‘to’ in example (23) and the Sranan ‘verb’ gi in (21) above. (23) Een meneer overhandigt een schoen aan hands.over a shoe to a gentleman de vrouw. the woman ‘A gentleman is handing (over) a shoe to a woman.’ (Surinamese Dutch)

4 Supporting evidence comes, again, from Eastern Maroon Creole (EMC), the more conservative sister language(s) of Sranan. I quote a personal communication with Bettina Migge on the lexically more specialized uses of langa in EMC: “langa is not very common in natural discourse. If transfer involves movement of the person, people use tya ‘carry’. If the Theme is lying next to a person, people usually use gi and langa when reiterating a request. It seems to me that langa also often has the overtone of ‘hand over in order to rid oneself of something’ for example after a quarrel.”



Creole in transition: Contact with Dutch and typological change in Sranan 

 79

The most common alternative to the use of overhandigen in a POC by speakers of Surinamese Dutch in the corpus is a POC featuring the general verb of giving geven and a Recipient marked by the preposition aan, as in (24). (24) Een meneer geeft een tas aan de andere. a gentleman gives a bag to the other.one ‘A gentleman is giving a bag to the other one.’ (Surinamese Dutch) An alternative construction is also found in the data, in which the Recipient is coded as a (primary) object via word order in a DOC: (25) Hij geeft de man een pan. he gives the man a pot ‘He is giving the man a pot.’ (Surinamese Dutch) This kind of construction is the norm in Dutch in metaphorical (non-literal) transfer events, involving emotional interaction for example, e.g. in light verb constructions like (give an) embrace, (give a) kiss, as in (26). I will show in due course that the same holds for Sranan: (26) De man geeft de vrouw een zoen op de wang. the man gives the woman a kiss on the cheek ‘The man is giving the woman a kiss on the cheek.’ (Surinamese Dutch) In sum, the Sranan “SVC” involving langa–gi is very common in literal transfer events, where it immediately follows the use of a monoverbal DOC featuring gi in frequency. The langa–gi construction is structurally equivalent to two corresponding Dutch constructions namely the POCs overhandigen–aan and geven–aan. This raises the question of the syntactic status of gi ‘give’ in Sranan “SVCs” like (19) and (21) above. I suggest that gi is a preposition in such contexts and that the structures I have so far termed “SVCs” are, in fact, POCs, and I will henceforth continue referring to them as such. The element gi manifests a multiplicity of functions and syntactic behaviors that have led authors to assign it the status of verb in some instances and that of preposition in others (Voorhoeve 1975; Jansen, Koopman and Muysken 1978; Sebba 1987). Sebba (1987: 75) assumes that gi is a verb when it is used as the second verb in a verb series (the V2) in constructions like (19) and (21) above. However, I do not think that gi is a priori interpreted as a verb by Sranan speakers in these examples. The element gi shows enough preposition-like behavior in other contexts to allow speakers the interlingual identification of gi with the Dutch preposition aan in “SVCs” like (19) and (21)

80 

 Kofi Yakpo

above. These structures are characterized by a high degree of “linear equivalence” (Muysken 2000) between Dutch and Sranan.5 The only context in which gi is unequivocally verbal in Sranan is when it occurs as the only verb of a DOC like (20) above. Tab. 6 above shows that DOCs like (20), which feature gi in DOCs as the predicate, are twice as common as those featuring langa as the predicate in POCs, as in (21) above. The same speakers often use gi DOCs and langa-gi POCs interchangeably in literal events of giving as shown in (20) and (21). The larger number of gi DOCs in proportion to langa POCs is a result of the use of gi DOCs in metaphorical or non-literal transfer events in addition to use in literal transfer events, while the langa-gi POC can only depict the latter type. The following example features such a metaphorical transfer event with a gi DOC, the collocation gi lala ‘give nonsense, provoke’: (27) Mi kan de a wroko yu gi mi yu 1sg can cop loc work 2sg give 1sg 2sg lala, mi no meki trobi nanga yu. nonsense 1sg neg make trouble with 2sg ‘I could be at work, you’d give me your nonsense (and) I (still) wouldn’t pick a fight with you.’ Other idiomatic, metaphorical uses of gi in the data are collocations like gi odi ‘give greetings, greet’, gi grani ‘give honor, pay respect’, gi skoro ‘give school/ classes’, gi busu ‘(give a) kiss’, gi prisiri ‘give pleasure/enjoyment’, gi anu ‘give/ shake hands’. The parallels of some of these constructions with their Dutch equivalents are obvious and they are likely to have entered Sranan via calquing. Compare the Dutch collocations les geven ‘give classes’, zoen geven ‘(give a) kiss’, plezier geven ‘give enjoyment’. Communication verbs, the other class of ditransitive construction covered in Tab. 6, also participate in non-literal transfer events by their very nature. They therefore also exclusively occur in DOCs, compare (28) and (29). (28) Dan fa yu kan kari yu-srefi wan then how 2sg can call 2sg-self one bus-kondre pikin? forest-country child ‘Then how can you call yourself a person from the interior?’

5 See Sebba 1987 (73–75) for a brief treatment of “prepositional” uses of gi. Our corpus also contains many uses of gi marking a broader range of animate and inanimate participants beyond Recipient, i.e. Beneficiary (mi e kieze gi mi eygi taal ‘I opt for my own language’) Goal (yu e poti specerijen gi en ‘you put spices into it’), Stimulus (yu musu luku tu gi bigisma ‘you must also look at/watch out for the adults’), and Experiencer (a nyan switi gi mi ‘the food is tasty to me’).



Creole in transition: Contact with Dutch and typological change in Sranan 

(29) No dan te yu taygi den yu prijs neg then temp 2sg tell 3pl 2sg price ‘No when you tell them your price (…).’

 81

(…)

In sum, the speakers in our corpus employ langa (in POCs) and gi (in DOCs) as transfer verbs in literal transfer events. DOCs featuring gi are also employed in descriptions of metaphorical transfer events. The strong presence of langa in POCs is seen as a result of calquing of a corresponding Dutch POC featuring geven or overhandigen. The element gi functions as a preposition in these constructions multifunctional in Sranan. The status of gi as a preposition in POCs is also a result of contact with Dutch.

4.2 Conclusion The preceding section has shown that Sranan ditransitive constructions have converged considerably with Dutch ones with respect to their morphosyntax and semantics. Some parallels between Sranan and Dutch are due to typological similarities and not likely to result from contact. For example, both languages have a single generic verb of giving, rather than a composite expression. A further typological overlap between Sranan and Dutch not likely to be contact-induced is Recipient-Theme constituent order in DOCs, rather than vice versa. I have analyzed other similarities between Sranan and Dutch as contact-induced, specifically the following ones: a) the use of gi as a preposition in ditransitive constructions, rather than a V2 in serial verb constructions (SVCs), mirroring the use of the preposition aan in Dutch prepositional object ditransitive constructions (POCs); b) the prolific use of langa in POCs featuring the Recipient as an object of the preposition gi in literal transfer events, calquing the use of geven (aan) ‘give (to)’ and overhandigen (aan) ‘hand over (to)’ in Dutch; c) a tendency to use DOCs featuring gi as the only verb in events that do not involve literal giving, covering idiomatic uses in particular, many of which are calqued from Dutch; d) a conspicuous absence of ditransitive constructions featuring teki-gi ‘takegive’ SVCs. I conclude that Dutch influence is equally pervasive in this domain of participant marking as it is in the expression of spatial relations, and this holds for both structural and semantic aspects. Dutch influence has also led to “deserialization” (Hajek 2006), the replacement of SVCs by POCs.

82 

 Kofi Yakpo

5 Processes and outcomes of contact-induced change in Sranan I have argued that contact with Dutch has made a considerable impact on the (a) grammar of space and (b) the way participant relations are expressed. I have shown that Sranan is undergoing a substantial typological shift with respect to (a), characterized by the demise of postpositional structures and the concurrent rise of Path-incorporating prepositions. This has led to the development of a GoalSource asymmetry in Sranan of the same kind as found in Dutch, in which Place and Goal can be marked by the same means (by static locative elements) and Source is marked differently (by Path-incorporating prepositions). The former pattern was symmetrical in Sranan with Place, Goal and Source Grounds uniformly marked by na. Changes were also recorded with respect to (b) in the syntax and semantics of ditransitive constructions. Sranan ditransitive constructions have converged considerably with corresponding Dutch ones. Here the principal change is that in transfer events, the Sranan verb gi ‘give’ can no longer be seen to function as a second verb in a “give” serial verb construction. Instead, the syntactic properties and semantics of gi have been calqued on those of the Dutch preposition aan ‘to’ in constructions that pattern closely with Dutch prepositional object constructions (POCs). Notably, serial verb constructions of the take-give type, mentioned in the twentieth century literature on Sranan, are exceedingly rare in the data. Double object constructions (DOCs), in which gi ‘give’ functions as a main verb are far more common with idiomatic uses than in literal transfer events, this too reflecting a pattern found in Dutch. I conclude that the alignment of Sranan patterns with Dutch ones has progressed quite far. In the domains studied here, Sranan is a creole in transition from a typology characterized by Kwa-like postpositional structures, complex locative constructions, a general locative preposition, and participant-marking SVCs, to a more Germanic type characterized by portmanteau prepositions, DOCs and POCs in line with its superstrate Dutch. The processes of contact-induced change involve, for one part, a small degree of matter borrowing (e.g. Sakel 2007), i.e. that of the Dutch preposition uit ‘out of’. Lexical material is not transferred ex nihilo and therefore not only involves the transfer of local meaning (e.g. ‘evacuation’ in the case of uit). We have seen that the non-native uit also patterns with the constituent order and the distribution of the native item fu ‘from’. Accordingly, uit is exclusively used as a preposition in Sranan, contrary to a native item like tapu, for which



Creole in transition: Contact with Dutch and typological change in Sranan 

 83

I give evidence of one-time postpositional uses. Equally, uit does not co-occur with the general locative preposition na, which is, again, possible with native locative elements other than fu. Ultimately, this means that Sranan speakers match semantic and morphosyntactic features when transferring and nativizing Dutch elements, and that matter and pattern borrowing (calquing) are inextricably interwoven. All other changes described in this chapter constitute instances of the transfer of lexical and morphosyntactic properties, rather than lexical material per se. This has been referred to in the literature, with varying degrees of overlap in meaning, by terms like “calquing” (Haugen 1950), “pattern transfer” (Heath 1984), “rule borrowing” (Boretzky 1985), “metatypy” (Ross 1996), “grammatical replication” (Heine and Kuteva 2003) and “pattern replication” or “pattern borrowing” (Sakel 2007; Matras 2009) among others. Pattern borrowing involves the carry-over of combinatorial possibilities, that is relational features, extending to constituent order, distributional potential and dependency relations (Yakpo and Bruyn 2015: 170–172). The outcomes of contact-induced change in Sranan are summarized below: Altered frequency and altered distribution: Elements and structures that were once more constrained in their distribution, relatively infrequent or marginal have become more frequent or central in Sranan or have acquired a wider distributional potential in alignment with Dutch patterns. At the same time, previously more frequent patterns have seen a decrease in frequency. Examples are (a) the typological shift from postpositional to prepositional structures; (b) the decrease in frequency of complex locative constructions and the corresponding increase of simplex locative constructions; (c) the continuing use of the general locative preposition na in specific types of Goal-oriented events, next to its demise in Sourceoriented events and a shrinkage of use in certain types of Place-oriented ones; (d) the use of Dutch-style POCs and DOCs rather than SVCs. Innovation: Elements fulfill new, previously unattested functions by exhibiting changes in their lexical meanings, and/or (any combination of) their distributional potential, constituent order, and dependency relations, following Dutch patterns. Instances are (a) the use of erstwhile locative nouns as portmanteau prepositions; (b) the use of tapu ‘on’ for expressing non-specific location and with idiomatic uses; (c) the shift from verb to preposition of the element gi ‘give’ in ditransitive constructions. The replication of Dutch patterns in Sranan therefore produces two interlinked types of outcomes in Sranan in the functional domains described in this chapter. Some of these are innovations, but the greater part consists of alterations of pre-existing patterns.

84 

 Kofi Yakpo

6 Conclusion The changes documented in this chapter constitute a typical case of linguistic convergence due to intense contact between the source language Dutch and the recipient language Sranan. Intensity is manifest in the duration of contact  – the two languages have been spoken alongside each other for about three and a half centuries – and in the extent of societal multilingualism. Dutch is no longer reserved for the socio-economic elite, as was the case during colonial rule. Multilingualism inclusive of Dutch has increased greatly for ordinary Surinamese (see e.g. Léglise and Migge 2015) due to socio-economic transformations since independence (Yakpo et al. 2015; Yakpo 2015). It is perhaps too early to ascertain whether the structural changes occurring in Sranan are taking part within a scenario of stable multilingualism and language maintenance, or whether convergence with Dutch is symptomatic of an ongoing language shift towards Dutch. In surveys, Sranan boasts a high degree of vitality and is claimed as a home or vernacular language by a large percentage of the population of Suriname (cf. data on language use in SIC 213–2005/02; Léglise and Migge 2015). The vitality of Sranan therefore seems to contrast with that of Surinamese Javanese, where many contact-induced changes seem to be symptoms of language loss and a shift to Sranan and Dutch as primary languages of the Surinamese-Javanese population (cf. Villerius, this volume). This also contrasts with the situation of Sarnami, a language that is generally seen as less threatened by loss and shift than Javanese, but nevertheless seems to be undergoing a shrinkage in speaker numbers in the urban agglomeration of Paramaribo (Marhe, p.c.; Yakpo and Muysken 2014; Hoeblal 2015). It can be concluded, however, that essentially, Sranan, like the other languages of Suriname described in this volume, is becoming more Dutch-like. Further investigations would probably confirm the general drift of Sranan towards Dutch in additional domains.

Abbreviations agn agentive suffix comp complementizer compl completive aspect cop locative-existential copula def definite article dist distal demonstrative doc double object construction



Creole in transition: Contact with Dutch and typological change in Sranan 

foc focus marker/identity copula fut future tense independent/object pronoun indp intj interjection ipfv imperfective aspect loc general locative preposition neg negator pl plural poc prepositional object construction prep general associative preposition pst past tense red reduplicant sbj subject sg singular serial verb construction svc

 85

Robert Borges

The Maroon creoles of the Guianas: Expansion, contact, and hybridization1 1 Introduction In Borges (this volume), I have already provided a brief overview of the origins of African slaves imported to Suriname in different time periods along with some information about the process of marronage these Africans faced once they decided to leave slavery behind. Before going into further detail, however, perhaps it is worth repeating that there are six widely recognized groups of Maroons in existence today who chiefly occupy areas along the river systems of central and eastern Suriname and adjoining areas in French Guiana as well as urban centers in both countries. The six groups can be further classified into two groups, the Central Maroons (Saramaccans and Matawai) and the Eastern Maroons (Ndyuka, Aluku, Pamaka, and Kwinti), and although all the Maroon groups are relatively homogenous in culture (Thoden van Velzen and van Weter­ ing 2004: 31ff), the language varieties associated with the Central Maroons on the one hand, and the Eastern Maroons on the other are reportedly not mutually intelligible. This is usually attributed to the high degree of lexemes of Iberian origin in the former group, along with a few phonological and morphosyntactic differences among them. Still, it has been posited that in modern Suriname, extensive demographic changes have caused a partial realignment of social structures among Maroons, whereby a pan-Maroon identity takes precedence over more traditional clan and ethnic group alignment among individuals, leading to a leveling of differences among Maroon language varieties (e.g. Migge and Léglise 2011). In this chapter, I will discuss the conditions leading to marronage and the divergence of the Maroon languages from each other and the plantation creole predecessor of Sranan, followed by the factors that fostered the development

1 I acknowledge the ERC advanced grant 230310 funding to the Traces of Contact ­Project. Various individuals have contributed to this work in various ways – s­ upervision, ­advice, (email) discussions, introductions in the field, and more – I list them here in ­A lphabetical order: Peter Bakker, Margot van den Berg, Dirk van der Elst, Friederike Lüpke, Bettina Migge Pieter Muysken, Robbert van Sluijs, H. U. E. Thoden van Velzen, and W. van Wetering. DOI 10.1515/9781614514886-004

88 

 Robert Borges

of an “organic” multilingual setting,2 eventually landing these languages on a full-circle trajectory, moving towards convergence with each other and modern Sranan. The following section will deal explicitly with the origins of each group, the conditions leading to their divergent language varieties, the relationship among these varieties, and the linguistic consequences of early contact between Maroons and Amerindians. In Section 3, I will discuss a number of more recent social and demographic changes that have provided an impetus for linguistic convergence. Specific instances of this convergence are identified in Section 4. Before continuing, it should be mentioned that this chapter’s coverage of the Maroon languages is not comprehensive, in that some varieties are given pre­ ference in various parts of the discussion. This has to do with the simple fact that some groups are relatively well represented in the anthropological literature and their languages relatively well studied (e.g. Saramaccan and Ndyuka) while others are considerably less well covered. Matawai and Kwinti for example are almost completely unrepresented in the linguistics literature. A second point relating to the literature is that within the realm of linguistic studies on Maroon languages, convergence tends to be a peripheral topic (at best). Take the Saramaccan language, which is the subject of dozens, if not hundreds, of scholarly articles, for example; works tend to either be descriptive in nature, or an attempt to uncover some aspect of the language’s origin via creolization. Very rarely is there a discussion in the literature about contact between a Maroon lan­ guage in a more contemporary setting – with other Maroon languages, Sranan, or Dutch (however, see Yakpo et al. 2015). Lastly, in looking at convergence and leve­ ling among Maroon languages, I rely primarily on field data collected among the Ndyuka and Kwinti in conjunction with sparsely available literature (e.g. Migge and Léglise 2011; 2013), thus the picture is limited to the Eastern Maroon setting. Although there is certainly room for more research in this regard, this work pre­ sents a snap shot of convergence contextualized in the cycle of language change.

2 Divergence and early contact Although there are currently six Maroon groups, as mentioned above, there have been other groups of Maroons in the past. Smith (2002:141) identifies

2 Organic multilingualism is a concept stemming from the Crossroads-KPAAM-CAM work­ shop held at SOAS from 16. The workshop brought together team members of the Leverhulme Crossroads project (https://soascrossroads.org/) and of the KPAAM-CAM project (pigforpikin. org/our-research/) on multilingualism in southern Senegal and northwestern Cameroon res­ pectively. The concept designates multilingual configurations and being elaborated by Lüpke (in preparation); I will elaborate on the concept in §4.



The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 89

two such groups known as the Karboegers (or Muraato) and the Brosu. These groups are relevant to the discussion here because they delimit the time period in which runaways had already acquired and began to employ some form of the plantation creole spoken in the area that would subsequently diverge into the Maroon languages as we know them today. This is possible because the earliest group, the Karboegers, fled ca. 1660 and was completely integrated linguisti­ cally to a group of Cariban speaking Indians with whom they began cohabiting. Smith (2014: 35) takes this as an indication that no creole variety had taken hold as a language for communication among the plantation slaves by that time. Similarly, the latter group, Brosu, which Smith dates to ca. 1820, spoke a variety similar to the modern varieties of Sranan spoken in and around Paramaribo (2002: 141–142, 150). The Brosu ceased to exist as such, and thus it is impossible to know if their language would have diverged into a distinct variety, as their formation occurred relatively close to the official end of slavery, thus marro­ nage, then and later, would have unnecessarily stifled the period of relative iso­ lation partially necessary for the divergence of the Maroon language varieties. Other groups of are noted in the literature that have either disappeared or been assimilated to other groups, e.g. Krabbeholle (Hoogbergen 1993), however the focus of the rest of this section will be on the historical factors that led to the establishment and divergence of the Maroons and Maroon languages that exist today; all of these groups formed in between the Karboegers and the Brosu. A number of factors contributed to the divergence of the Maroon languages. As can be inferred from the preceding paragraph, Smith (2002: 142) suggests that the timing of marronage was the foremost factor in determining the resulting lin­ guistic outcome. He proposes four stages of marronage, whereby the first and fourth correspond to the periods of the Karboegers and Brosu, respectively. The marronage of the six groups under consideration here fall into Smith’s Stages Two (1690–1710) and Three (1710–1800). Another factor, which is related to the timing of marronage, is the general location of the plantations from which the slaves ran away. Generally speaking, Maroons associated with Smith’s (2002: 142) Stage Two originated from the plantations around the Jodensavanne ‘Jews Savannah’ (for more details see Borges, this volume), while stage three Maroons originated in the area around Paramaribo, the Comewijne, and Cottica Rivers. Geography may be seen as a cover for a number of other factors behind variation in the plantation varieties. Van den Berg (2007: 379ff) discusses such dimensions, including the variable backgrounds of individual speakers and how long they had been present in the colony, differences in urban vs. rural varieties, stylistic variation, and other social factors (making note of varieties associated with European speakers and African speakers). While the precise details on different varieties and the social dimensions behind them are not very well understood, the divergence of Maroon languages has to be conceptualized in this context of wide spread variation among the early

90 

 Robert Borges

plantation creole speakers. As Thoden van Velzen and Hoogbergen (2011: 73) put it, Maroon groups as we know them today were formed by a conglomeration of clans (lo), which played a central role in the process of marronage and often ori­ ginated as a group from a single plantation, taking the plantation’s name for the clan; thus idiosyncrasies unique to a small group of runaways provided the basis for new divergent varieties. A further complicating factor is that not all clans, or individuals, fled at the same time; there was a considerable time span following the establishment of the earliest clans when new escapees could join. As far as I am aware, however, there have been no studies that systematically investigate variation in contemporary Maroon languages based on clan membership. In the remainder of this section, I will provide further details about the sociohistorical factors leading divergence of each of the Maroon groups. I will then address the role of language contact between the Maroons and Amerindians and finally provide a summary discussion about the genetic relationships among the Maroon languages.

2.1 Central Maroons The Central Maroons, Saramaccans and Matawai, are the oldest of the six groups of Surinamese Maroons. Smith (2002: 131), for example, writes that the oldest clan of the Saramaccans, Matjáu, was established around 1690, a date which is frequently reproduced as “the beginning” of Saramaccan in studies on the language. However, the Indian Wars (1678–1684) appear to have been an important event that provided a major early opportunity for slaves to make their escape. Runaways fought the colonists alongside Amerindians during these insurrections and were initially har­ bored by them. This is supported by archival documents and oral histories collected among Saramaccan and Ndyuka Maroons (Dragtenstein 2002; Thoden van Velzen and Hoogbergen 2011; Price 1983: e.g. 45; 2012). Discrepancies in timing aside, the Central Maroon Languages are also perhaps the most divergent among the Surina­ mese Creoles, differing particularly in terms of phonology and vocabulary. Several factors are responsible for these differences, in particular, the emer­ gence of these groups from the plantations in the Jodensavanne. The majority of plantations in the Jodensavanne were owned by Sephardic Jews, as the name sug­ gests, of Iberian origin. It is controversial whether a portion of the planters had previously been in Brazil (Pernambuco) and expelled by the Portuguese, whether they brought any slaves with them, and whether a L ­ uso-creole was spoken among them and relocated to the Jodensavanne in the mid 1660’s (Fontaine 1980: 33; Rens 1980: 36; Arends 1999; Ladhams 1999; Smith 1999). Nevertheless, a variety of the English based plantation creole was infused with a rather high percentage



The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 91

of words originating from Portuguese, either by ­borrowing directly from that lan­ guage via the Jewish planters, or as Smith (2014: 34) maintains, via partial relexi­ fication of the plantation creole by Portuguese (creole) speaking slaves. Whatever the case may have been, a primarily Anglo creole variety heavily influenced by Portuguese developed in the Jodensavanne, known as Djutongo ‘Jew language’. As the Central Maroons began their escapes, Djutongo had become their means of communication. Another factor that contributes to the distinctiveness of the Central Maroon languages is the relatively high retention of lexemes of African origin, this perhaps as a result of the early formation of these groups with a high number of runaways born in Africa. Smith (2015a, b) provides lists of Saramaccan words of Gbe and Kikongo provenance. The early history of the Central Maroons is relatively well covered in the lite­ rature, for example, Price’s (1983) work that meticulously corroborates accounts of oral history with archival material. This section will therefore remain rather concise. In brief, small groups of runaways originating independently from the Jodensavanne occupied the area to the south, between the Suriname and Sara­ macca Rivers, even as far south as the Tafelberg Mountain according to de Beet and Sterman (1981: 11). This continuous influx and conglomeration of clans eventually came under the leadership of three macro groups. Following severe infighting in the first half of the eighteenth century, and difficulties maintaining the 1762 peace agreement with the colonial government, one such group led by Becu and Musinga occupying the upper reaches of the Saramacca River, formally separated from the other two groups by concluding a separate treaty in 1767 (de Beet and Sterman 1981: 10–11; Price 1983: 89ff). Those groups who adhered to the 1762 Saramacca Peace Treaty and had been living along the upper reaches of the Suriname River became known as the Saramaccan people, while the Becu / Musinga group is now known as Matawai. 2.1.1 Saramaccan According to Price’s (2013) projections, as of 2014, there should be around 90,000 Saramaccan people; 28,500 remain in the Surinamese “interior”, some 29,000 in Paramaribo and it’s surrounding, 25,000 can be found in the coastal areas of French Guiana, and another 7,500 reside in the Netherlands. The Sara­ maccan language is vital, with fluent speakers and intergenerational trans­ mission in the locations mentioned. Saramaccan is relatively well studied and has an incidental written tradition. The Glottolog entry on the language, for example, lists 135 references to Saramaccan, including descriptive materials, case studies on particular linguistic structures, focused discussions on creoli­ zation, language learning materials and literature (Hammarström et al. 2016).

92 

 Robert Borges

To acquaint the reader, the Saramaccan language is exemplified in (1) and (2), from McWhorter and Good (2012: 332–333). (1) Mi bi tá poobá u dá kái únu. Mi bi 1sg pst ipfv try nf give call 2pl 1sg pst ké u hii únu tú kó aki. want nf all 2pl two come here ‘I was trying to call you. I wanted both of you to come here.’ (2)

Mi ó tá, nee, mɛ́ ó gó alá, 1sg fut talk no 1sg.neg fut go yonder faádi mbéi me gó a New York? for.what make 1sg go loc New York ‘I would say, “No, I’m not going there, why would I go to New York?”’

Traditional political organization and cultural practices among the Saramaccans appear to be alive and well, with some aspects making their way to Paramaribo and beyond along with migrations (Bijnaar 2007; Gossens 2007; Price 2008). 2.1.2 Matawai The situation for Matawai, however, is much more serious. By Price’s (2013) esti­ mate, there are approximately 7,000 Matawai, just 1,300 of which can be found in the Surinamese “interior”. The rest are primarily found in Paramaribo. Villages on the upper Saramacca River are particularly depopulated, the majority of “resi­ dents” only returning from Paramaribo during holidays (Migge 2014: p.c.). The densest populations of Matawai in the interior can be found downriver towards the area where the JFK Highway crosses the Saramacca River. As a result of these (circular) migrations, a phenomenon already noted in the 1970s by anthropolo­ gists (de Beet and Sterman 1981: 415ff), along with other factors associated with life in the city such as interethnic networks and the position of Dutch in Surina­ mese society, many Matawai display preference for the use of Sranan or Dutch, with heavy code mixing of these three varieties (and others) in in-group contexts. The transmission of Matawai to children appears to be unsystematic at best, if not interrupted outside the rural context. To date there are virtually no published linguistic studies or other materials available on the Matawai language. According to the Ethnologue, Matawai is clas­ sified as a dialect of Saramaccan. However, this classification was established based on a few scant sentences contained in Hancock (1987). New research sug­ gests, however, that Matawai and Saramaccan are phonologically and morpho­ syntactically distinct, the former consisting of three varieties recognized by the



The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 93

Matawai speakers themselves. For example, Migge and Winford (2015) present data that demonstrates variation in future marking in Matawai. The future marker o, as found in all the other Surinamese Creoles is accompanied by an additional marker to in Matawai. It is clear that to is not a combination of the imperfective marker ta and future o since to and ta co-occur (3), as do o and ta (4) – examples reproduced from Migge and Winford (2015). (3) nɔɔ ade u to ta dɛ then there we fut ipfv there ‘Then there we will (generally) be.’ (4) komputer a o ta sei computer he fut ipfv sell ‘He will be selling computers.’ While to suggests some degree of divergence between Matawai and Saramaccan, the co-occurrence of future morphemes with imperfective markers is a shared characteristic of the two languages that sets them apart from Sranan and the Eastern Maroon Languages (Huttar and Huttar 1994: 507ff; McWhorter and Good 2012: 135). Future research will likely uncover answers to interesting questions surrounding to. What conditions the variation between this form and o? Where does this marker come from? It seems unlikely, without further explanation, that to and o derive from the same source, as with English go and o (van den Berg 2007; van den Berg and Arends 2004). Leaving aside the interrelated factors mentioned above, such as timing of marronage, the place/plantations of origin, and idiosyncrasies among the found­ ing clans, additional factors very likely contributed to the divergence of Saramac­ can and Matawai. Although de Beet and Sterman (1981: 11) refer to missionary documents that suggest the Matawai and Saramaccans stayed in regular contact following the treaties of the 1760s, the costly and difficult journey between the rivers was likely an infrequent occurrence. Additionally, from the 1860s, the Matawai were proselytized and largely converted to Christianity (de Beet and Sterman 1981: 173), driving a cultural wedge between the two groups.

2.2 Eastern Maroons Unlike the Central Maroons, the history of the Eastern Maroons makes this group ­considerably more difficult to describe as a unit. While the Eastern Maroon groups we recognize today – Ndyuka, Aluku, Kwinti, and Pamaka – share a great deal in terms of culture and speak highly intelligible language varieties, their origins lie in a number

94 

 Robert Borges

of separate events with groups of individuals escaping from different p ­ lantations. Further complicating the picture, only the Ndyuka Maroons concluded a peace treaty with the colonial government in 1760, which instilled animosity and mistrust between them and the other Eastern Maroons. Lenoir (1973: 15) reports that after this 1760 treaty security was tightened around the plantation area and the forests alike; non-“pacified” groups were stuck in this “no man’s land” and ­vulnerable to both colonists and the newly “pacified” Ndyuka. Despite this, slaves continued to escape the plantations (or to be captured by Maroons) to form or augment core groups of Maroons already established. Although the terms of the Ndyuka treaty stipulated that any post-treaty runaways be turned in to the colonial authorities, in practice this did not always occur. Thoden van Velzen and Hoogbergen (2011: 103, 259) note for example that a number of important Ndyuka ancestors joined the group after the 1760 treaty. Additionally, the Ndyuka were able to use this stipulation as leverage to extract labor resources from non-treaty Maroons (Lenoir 1973: 26). In this section, the Eastern Maroon groups are ordered by the ­chronological order that they have come into existence as reported in the literature on the dif­ ferent language varieties (e.g. Migge and Goury 2008). However, oral histories and works examining archival evidence suggest that our understanding of the ­temporal dimension of the formation of these groups could benefit from some revision. For example, the core founders of Ndyuka, described by Smith (2002: 142) as Stage Three (1710–1800), were established at the time of the Indian Wars (1678–1684), before the Stage Two groups (1690–1710), which were also ­established at this earlier time time. 2.2.1 Ndyuka The Ndyuka, also known as Okanisi, began as a number of clans that escaped inde­ pendently from plantations primarily along the Commewijne, Cottica, Tempati, and Perica Rivers, though there is some evidence that the Otoo, Misidyan, and Dyu clans originated from plantations on the Upper Suriname River and share some connection with the early Saramaccan Maroons (Price 1983: 110, 129–134; Thoden van Velzen and Hoogbergen 2011: 43–49, 62–63). The clans’ names are derived from the plantation name, or the plantation owner, from where the origi­ nal core of the clan escaped. The name Ndyuka originally only referred to a subgrouping of these clans, established in the early part of the eighteenth century. Thoden van Velzen and Hoogbergen (2011: 39–94) provide great detail about the origins, movements, and notable ancestors of each clan based on oral histories and supported with archival evidence where possible. Escapes are detailed and corroborated by the archives between 1706 and the 1750s, however they also note that the oral histories point to beginnings during the Indian Wars (1678–1686) (Thoden van Velzen and Hoogbergen 2011: 72).



The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 95

By the 1740s all the clans had apparently come together and settled in an area rich in small creaks known as the Mama Ndyuka, between what is now called The Lelygebergte Mountains and the Marowijne and Tapanohony Rivers. The clans were well organized and quickly established three groups of clans that rapidly expanded their collective territory, strictly allotting particular areas to these grou­ pings. Thoden van Velzen and Hoogbergen (2011: 103) write that “probably by around 1745 or a few years earlier” the Lukubun division had already established themselves in the Lonwataa ‘fast streaming water’ area of the Tapanohony, where there are numerous smaller sections of river with fast running water and frequent rapids. Lonwataa occupies the area from Slagboomvallen upriver to the village of Kisai. Migrations of the Miáfiya Bakaa division also began sometime before the 1760 treaty to the area of Aduwataa, a calmer wider stretch of the Tapanah­ ony upriver from Lonwataa. These migrations and establishment of new villages likely continued into the 1780s. The third group, the Ndyuka division remained in the Mama Ndyuka area, below Slagboomvallen, until the 1780s, then later also began migrating further up the river (Thoden van Velzen and Hoogbergen 2011: 103–113). These three groups of clans, sometimes referred to as “federations” con­ tinued to play a significant role in social organization, conflict management, ritual and religious responsibilities, and remain relevant until today. Accor­ ding to Price’s (2013) estimate, there are approximately 90,000 Ndyuka, 26,000 of which can be found in the Surinamese ‘interior’, 29,500 in Paramaribo, 5,500 in the French Guiana ‘interior’, 21,000 on the French Guiana ‘coast’, and 7,500 in the Netherlands. The language appears to be quite vital, with many speakers in all areas mentioned above, in-tact generational transmission, and a tendency for Ndyuka to serve as a target for convergence among speakers of other Maroon languages (see §4). Like Saramaccan, there are ample studies and materials available on the Ndyuka language, including descriptive mate­ rials, topic studies, and language learning materials. A syllabic orthography was proposed in the early twentieth century, which, as the story goes, came to one Ndyuka named Afaka from God in a dream around 1908 (Dubbelaar and Pakosie 1999: 17). The script was not supported by Granman Amakti (Huttar 1992), and today its use is severely limited, finding audience primarily in works of art. 2.2.2 Aluku The Aluku, or Boni, are a much smaller group than the Ndyuka. Price (2013) esti­ mates approximately 11,000 total Aluku, of which some 5,700 are to be found in the French Guiana ‘interior’, the rest on French Guiana ‘coast’, except for some 200 individuals in the Netherlands.

96 

 Robert Borges

The Aluku are well known for the violent uprisings, known as the Boni Wars they staged against the colonists in Suriname, and as a result, they settled far south on the Lawa River, many of them on the French side. The Aluku concluded a treaty with the French authorities in 1860 and are the only group to perma­ nently settle on the French side; today they comprise a recognized component of French Guianese society (Bilby 1991; Migge and Léglise 2013: 84). In the 1960’s, the French Guianese government introduced a French style commune territorial organization that displaced some aspects of traditional social organization and caused changes in Aluku demographics. Aluku people began migrating away from traditional villages along the Lawa, such that by 1980s, only half of the Aluku population lived in villages along the Lawa River (Bilby 1991: 49). Much like the Ndyuka, the Aluku as a group formed via a conglomeration of independently fleeing clans that took their names from their plantation of origin, their numbers subsequently enlarged by new runaways or slaves captured in regular raids the group staged on plantations. The first reported leader of the Aluku was Asikan Silvester, who fled to the bush in the Cottica area in the chaos caused by the French invasion led by Cassard in 1712 (Hoogbergen 1989: 184). The names Aluku and Boni come from subsequent leaders of the group, who ruled simultaneously, but with a division of responsibilities (Hoogbergen 1989: 182). Aluku, known also as Askaan van Marceille (from the plantation Marceille along the Cottica river) was ­responsible for village life, ensuring the safety of the woman and children, while Boni, said to have been born in the bush, led raiding parties of men outside their camps to the plantation area where the sought supplies and additional recruits. Following the tightened security brought about by the 1760 treaty with the Ndyuka, the Aluku increased their attacks on the plantations, not only to source supplies and increase their numbers, but also in an attempt to intimidate the colonial authorities into concluding a treaty with them, in what Buddingh (1995: 133) called a “counterproductive” strategy. This period, from the mid 1760s to the latter half of the 1770s is known as the first Boni War, after which Boni and his followers retreated to French territory (Hoogbergen 1989: 175). Following the first Boni War, the Ndyuka and Aluku formed an alliance, which led to regular visits of the groups to each other’s villages, religious and economic exchange, as well as intermarriage (Thoden van Velzen and Hoogbergen 2011: 230–233). The Second Boni War began when Boni and his followers resumed their insurrections on the Surinamese plantations in 1789. This decision ended tragically when many of the Maroons were killed, Boni himself in 1793, and the remainder of the group, now less than a quarter of their numbers in the 1760s, retreated to the Abbatis Cottica area of the Lawa River under the leadership of Boni’s son Agosu (Hoogbergen 1989: 175; Bilby 1991). Great detail about this period of Aluku history can be found in the works of de Beet (1984) Hoogbergen (1985, 1989, 1990), and Buddingh (1995: 133–159).



The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 97

2.2.3 Kwinti The Kwinti are the smallest and least studied group of Maroons in Suriname. Price (2013) estimates that there are around 1,000 Kwinti and the majority of them –650– are to be found in Paramaribo, some 50 individuals in the ­Netherlands and the rest in the Suriname “interior”. Despite being linguisti­ cally affiliated with the Eastern Maroon group the Kwinti found in the ‘interior’ of Suriname occupy the westernmost Maroon villages on the Coppename River (Bitagron and Kaimanston) and villages among those of the Matawai on the ­ akapaka, middle reaches of the Saramacca River (Kwata Ede, Pakapaka, Pikin P and Makajapingo). The two Kwinti groups of the interior share a rather deta­ ched sentiment toward each other and it may be that over the past hundred or so years these two varieties have diverged to some degree; however there has been, to date, no linguistic data collected from the Kwinti living along the Saramacca River. The details of the early history of the Kwinti remain somewhat unclear (cf.  Wekker 1985). In short, the Kwinti appear to have marooned in the ­eighteenth century, settled later along the Saramacca River with the Matawai in the mid nineteenth century. Following tensions between the two groups, a portion of the Kwinti settled along the Coppename River later in that century (Wekker 1985: 83). Anthropologist Dirk van der Elst, who worked with the Coppename Kwinti in the early 1970’s, presents the following somewhat conflicting accounts of Kwinti origins from oral histories he collected (1975: 10–12): (1) The first possibility is that the Kwinti originated as escaped slaves from Berbice. They traveled along the Corentijn River (a.k.a. Kwinti Liba ‘Kwinti River’), then up the Nickerie River. They stayed along the Nickerie River for some time, but fearful of patrols, went further inland. There they built a village called ­ arriageable Pisii ‘pleasure’. They later joined the Matawai due to a lack of m women. Van der Elst notes that this account is unlikely, but cites the existence of the Karboegers as support for this possibility. Hoogbergen later presents archi­ val evidence that suggests the Karboegers were employed on patrols to catch the Kwinti making it rather unlikely that these were actually a single group (1992: 45). De Beet and Sterman (1980: 6) also note that there are no physical ­ merindians. characteristics suggesting the Kwinti ever did any mixing with A Additionally, there appears to be no linguistic evidence supporting the Kwinti’s origin anywhere other than Suriname’s plantation area (Smith and Huttar 1983; Borges 2016b). (2) A second option is that a group including Kwinti ancestors escaped westward from the Paramaribo area, avoiding all people, settling east of the ­Saramacca River. After bouts of strife, the group split into three: the Kwinti,

98 

 Robert Borges

who went west; the Pamaka, who went east, and the Duda clan who returned to their masters. Following the split, the Kwinti built Pisii. The story indica­ tes that the escapes happened after the 1760 treaties; otherwise there would have  been no point in avoiding other people, since the “pacified” Maroon groups would not have turned in other runaways before then. This account also seems ­unlikely except for the fact that Pamaka oral history also claims that the Pamaka lived in a village with the Kwinti in their early history (van der Elst cites Lenoir 1973: p.c.). De Beet and Sterman (1980: 3) relay another version of this story, told to evangelist C.S.E. Helstone in 1947 in Pakapaka. While in this version there is no mention of Pamaka, the Kwinti and the Duda settled in two villages deep in the forest between the Saramacca and the Coppename rivers after fleeing the plan­ tations in the Para area. They would raid the plantations as well as the Matawai settlements for women, children and tools. After some time, the two groups failed to get along, so the Kwinti took the opportunity to make peace with and settle among the Matawai. (3) Lastly, and perhaps the most likely of the three accounts, da Kofi, the Kwinti founder, led a band of Maroons in the south of Paramaribo’s plantation area for some time before convincing his younger brother Boni3 (see §2.2.2) to run away. In the ensuing uprisings,4 the groups were separated. Kofi’s group went westward and Boni’s group passed through Ndyuka territory finally settling on the Lawa River. Green (1974: 58–59) provides additional details to this account collected on the Saramacca river. Kofi’s and Boni’s groups escaped from a planta­ tion called Bunumike and became hostile to each other, each blaming the other for their discovery by the planters, though they traveled within shouting distance of each other and eventually reconciled before going in different directions. Green notes that some informants insist that the ancestors of the Pamaka Maroons later left the Kwinti group, or that they later fled the same plantation.5

3 At least half-brother, perhaps classificatory brother. Note that in the Aluku Oral histories, Boni is said to have been born in the bush. 4 For details see Hoogbergen (1985). 5 Bunumike, according to the account told to Green, was near two other plantations, Poesoegro­ enoe and Hamborg. Neither Poesoegroenoe nor Bunumike appear on a plantation map, though the map covers a later era (Bakhuis et al. 2003), nor are they mentioned in van Stripriaan (1993), which covers the years 1750–1863. However, the map shows two plantations called Hamborg, one on the Saramacca river and one on the Cottica river. These two are also mentioned in van Stripri­ aan (1993:460). The Pamaka are from plantations on the Cottica river, while in this account, the Kwinti should have fled plantations in the Para area.



The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 99

Van der Elst (1975: 12) notes that it is possible that there is truth to some ­elements in all the histories he presents; rather than conflicting with each other, they suggest that the founders of the Kwinti did not all come from the same place, which is certainly consistent with the formation of the other Maroon groups. It is interesting to consider these oral histories with reference to the development of the Suriname creoles, particularly the idea that the Kwinti have at the very least “crossed paths” with the Pamaka and Aluku in their early history (de Beet and Sterman 1980: 5). Accounts (2) and (3), as well as linguistic evidence, point to a close relationship among the Kwinti, Pamaka and Aluku, all having fled the plantations in the latter half of the eighteenth century, as opposed to the Ndyuka, Saramacca and Matawai, who were already relatively well establis­ hed by the beginning of the eighteenth century. However, Hoogbergen’s (1992) ­examination of archival evidence suggests that the Kwinti were well “settled” and self-­sufficient before 1760 (based on their number of settlements, provision grounds, ability to quickly escape patrols and effectively hide). This adds additi­ onal complications to the oral histories, particularly (2), which suggests that the first Kwinti runaways escaped after the “pacification” of Matawai, Ndyuka and Saramacca, and were poorly adept at bush craft (i.e. plantation born). Irrespective of their origins, it is clear from the archival evidence that the eighteenth century was a turbulent period of marronage for the Kwinti. Following the peace treaties organized between the colonial government and the Matawai, Ndyuka, and Saramacca groups (1760’s), the Kwinti regularly faced hostilities from Amerindians and slaves, nor did they enjoy solidarity with other Maroons who were often compelled to turn in runaways and non-pacified Maroons (for a ransom). Additionally, burgherpatrouilles, bands of planters, slaves, and free blacks who hunted runaways, were regularly destroying villages and provision grounds of the Maroons.6 While the Kwinti were never officially “pacified”, by the nineteenth century no more patrols were sent after the Kwinti and the govern­ ment in Paramaribo lost interest in them (Hoogbergen 1992: 51).7 During the time of the Matawai granman ‘paramount chief’ Josua (1835– 1867), the Kwinti first made contact with the Matawai (Green 1974: 59). Green’s account, collected from informants on the Saramacca River, states that the main impetus for taking up contact with the Matawai was to find marriageable men for the group’s disproportionate number of girls. One man called Tata Djafu first trekked to the Saramacca river, and once the Matawai had established that he wasn’t there to steal women, but rather to “become free”, Djafu led a group of

6 For a detailed account of the early history of the Kwinti, see Hoogbergen (1992). 7 The Kwinti were only ‘recognized’ by the government in 1887.

100 

 Robert Borges

Matawai back into the bush where they informed the rest of the Kwinti that the war was over and they no longer had to hide from the whites. The Kwinti were all persuaded to relocate to the Saramacca River8 where they concluded a peace oath (bebe soi ‘drink show’ comparable to Ndyuka diingi sweli ‘drink swear’) with the Matawai (1974: 60). Relations did not remain peaceful between the Kwinti and the Matawai for long because the ‘Kwintis were aggressive with Matawai women’ (Green 1974: 62). After a fight for which several Kwinti were punished excessively during the time of granman Noah Adrai (1867–1893), a number of the Kwinti left, lead by Kapiten Aketemoni and Alamo, to resettle along the Coppename River. Permission to relocate was granted by governor Tonckens in 1883 (van der Elst 1975: 12). These Kwinti first settled at a site called Coppencrisie as suggested by a group of Ndyuka lumberjacks working in the area. Not long after the site was settled, a mission was sent to the Coppename Kwinti in 1889 led by Christian Kraag (de Beet and Sterman 1980: 1). Internal strife in the village lead to people abandoning it for the current Kwinti villages of Kaaimansiton and Witagron9 in the beginning of the twentieth century (van der Elst 1975: 12–13). Aside from their official status under the Matawai Granman, the Coppename Kwinti are neither economically nor culturally dependent on any other Maroon group. 2.2.4 Pamaka The Pamaka or Paramaccans is another group that formed in the “no man’s land” around the plantations of the Commewijne and Cottica Rivers and the territory of the “pacified” Ndyuka to the south. Lenoir (1973: 21) dates the formation of the group to the late eighteenth century, when clans from the Hantros (Antoisie) and Hazard (Asaiti), along with another group of unclear origin, Molo, established permanent settlements between the Commewijne and Marowijne Rivers. Like the other groups, they were dependent on maintaining what Lenoir (1973: 21) calls “clandestine contact” with the slave populations on their respective plantations in order to source metal goods. Unlike the Aluku, the Pamaka tended to avoid military conflicts, which according to Lenoir (1973: 18) is perhaps why Pamaka history has received such little attention from scholars. However, Thoden van Velzen and Hoogbergen (2011: 222–223) tell us that a group of Maroons, known in the literature as Tesisi (see also Hoogbergen 1990) but called Paamaka by the

8 With the exception, apparently, of one woman who was too fat to go anywhere. She remained in the bush. 9 Kaimanston is probably older.



The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 101

Ndyuka, attacked the plantation of Jukemombo, killing the owner’s wife and kidnapping the remaining slaves in 1765. As a result, the Posthouders increased pressure on Ndyuka to patrol the Tesisi area and in 1771 they captured and killed several people from Tesisi. But a year later, the Ndyuka concluded a peace agree­ ment with the Pamaka from Tesisi and took 36 individuals under their care/in their territory. In the 1780’s the group’s area was cordoned by the planters and they fled further south to the Tempatie Creek area. At this time Lenoir (1973: 22) suggests that a possible split of the group under military pressure, forming the Pamaka and Kwinti as distinct groups. In the Tempatie area, they were discovered by the Ndyuka and fled again to the Nassau Mountains around the headwaters of Paramacca Creek (Lenoir 1973: 24). From the 1810s, the Pamaka population gra­ dually migrated downstream towards Marowijne. They reestablished economic contact with Ndyuka around 1830 (Lenoir 1973: 25); the Ndyuka provided metal goods and obliged the Pamaka to cut timber in exchange, which would be sold by the Ndyuka to plantations. By the 1890s, the Pamaka had spread out along the Marowijne (Lenoir 1973: 26). According to Price’s (2013) estimates, there are recently around 1,100 Pamaka in total, 4,300 of them to be found in the Suriname “interior”, another 1,100 in greater Paramaribo and about 700 in the Netherlands. Price estimates 1,000 in French Guiana’s “interior”, and another 3,900 along the French Guiana “coast”.

2.3 Maroon-Amerindian contact Immediately following the Dutch takeover of the colony, the colonists instiga­ ted a peace agreement among Suriname’s Amerindian populations. The newly united Indians, however, found that they had a common set of grievances against the colonists and were able to stage a series of severe uprisings, known as the Indian Wars (1678–1684). During these insurrections the colony was thrown into chaos – plantations were burned and planters and slaves were murdered (Dragtenstein 2002: 39). As the initial Carib-Arawak alliance unraveled, African slaves seized the opportunity to escape and to fight alongside the Amerindians. The oral histories of the Saramacca, Ndyuka, and Aluku, along with sup­ porting archival documents, suggest that initially the Maroons were harbored by the Amerindians, and later took Amerindian wives (at times non-consensually), such that virtually all clans had “Indian Mothers” (Dragtenstein 2002; Hoog­ bergen 1989; Price 2012; Thoden van Velzen and Hoogbergen 2011). Price notes (2012: 4) that, strikingly, according to oral history these women were quickly inte­ grated into Maroon society.

102 

 Robert Borges

The reflex of these unions is not only evident in, for example, techniques for subsistence farming, hunting and fishing, and environmental knowledge (Price 2012: 5), but in the Amerindian contribution to the lexicon of the newly diver­ gent Maroon Creoles as well. The Amerindian lexical contribution to the Maroon Creoles is significant within the realm of lexemes related to the local knowledge that was passed on by the Amerindians. In Good’s (2009) sample of 1,000 Sara­ macca words, he found 15 lexemes with Carib etymology. These were, categori­ cally: animal names, food, drink and cooking utensils, clothing, personal adorn­ ment and care, warfare and hunting. Borges (2015) examines the SIL dictionary of Ndyuka (Shanks et al. 1994) and finds 47 lexemes of Amerindian origin listed. Just as in the Saramaccan case, the words primarily describe items pertinent to the local environment, i.e. plants and fruit, 11 words; birds, 9 words; land animals, 8 words; aquatic animals, 4 words; cooking utensils, 3 words; insects, 3 words; ailments, 2 words; miscellaneous, 7 words. Although the Indian Wars provided an impetus for cooperation between the Maroons and the Amerindians, and for the subsequent sharing of local practices and related lexemes, these partnerships did not last. Following the unraveling of the Carib-Arawak alliance ca. 1680, the colonial government re-engineered ani­ mosity between the two groups. Then, following a 1684 treaty stipulating that all Amerindians in the colony were free and could not enslaved, the colonial govern­ ment employed various Amerindians as forest guides and hunters in their quest to capture and punish runaway slaves. As a result, interactions between Maroons and Amerindians became rather limited in the colony’s early history. It is difficult to see any influence from Amerindian languages in the develop­ ment of Maroon languages aside from these lexical items. However, following their southward movements on the Tapanahony in the years surrounding the treaty, Ndyuka established a trade monopoly over their southern Amerindian neighbors, which resulted in the development of a trade pidgin (see Meira and Muysken, this volume). This trade pidgin is apparently no longer spoken and today, the majority of Amerindians living along the river systems of Eastern Suriname and bordering western French Guiana speak some form of Eastern Maroon Creole or Sranan.

2.4 On the divergence of Maroon languages An examination of the historical development of the Surinamese Maroons tells us that the earliest runaways that contributed to the formation of the six groups in existence today left the plantations at least as early as the late 1670s. Typically small groups would flee from a single plantation, forming the core of a clan and taking the plantation name as a clan name. For a time, these groups would maintain some



The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 103

form of contact with the plantation, either in covert alliance with slaves or by out­ right incursion, in order to source metal and other goods unavailable in the forest, and to increase their numbers by recruiting and/or kidnapping plantation slaves. These clans necessarily formed alliances with each other, with Amerindians, and even with agents of the colonial government, some of which were quickly dissolved while others remain relevant to this day. Although the histories of the Maroons are highly intertwined, they are also independent enough that the language varieties have developed independently of one another.10 Smith’s work (1987, 2002, 2015c) places primary importance of these divergences on the timing of Maroonage. Excluding the (still controversial) possibility that a lusocreole played some role in the formation of the Central Maroon languages, the divergence of the languages could be represented as in Fig. 1, adapted from Smith (1987, 2002).

ProtoSurinamese Creole

Saramaccan-Matawai Ndyuka-Pamaka-Aluku Kwinti Modern Sranan

Fig. 1: Classification of the Surinamese creoles by Smith (1987: 4)

The earliest groups (Smith’s Stage Two) are dated to 1690, however, we have seen that the Indian Wars beginning more than a decade earlier provided the impetus for the escape of core groups of early Saramaccan and Ndyuka Maroons, the latter considered part of Stage Three (from 1710) and that several of the Ndyuka clans’ point of origin was in the Jodensavanne. Additionally, in Smith’s work and repea­ ted throughout the literature, 1712, the year of French Admiral Cassard’s assault on the colony is cited as a large impetus for mass escapes of those who would later form part of the Ndyuka. Interestingly though, Thoden van Velzen and ­Hoogbergen (2011) make no mention of this event in their extensive description of the Lowéten ‘runaway time’. This is not to say that no slaves joined the ranks

10 It should be noted that, while from the perspective of a non-Maroon or a linguist, it might ap­ pear that some of these varieties would better be considered dialects, e.g. the Eastern Maroon Lan­ guages, due to their high degree of lexical and structural similarity, the differences among them are socially salient enough that the speakers themselves consider them distinct varieties. This also holds for the Central Maroon languages; Although Matawai is considered a dialect of Saramaccan in the ethnologue, Migge and Winford (2015) describe distinct morphosyntactic characteristics and further note that the Matawai speakers themselves recognize three varieties of that language.

104 

 Robert Borges

of the nascent Ndyuka in the wake of this event, but it is telling that it does not play a central role in the formation of any of the clans as reported by the work on this period. A reading of Thoden van Velzen and Hoogbergen (2011) also makes it clear that groups of escapees continued to join the ranks of the Ndyuka through­ out the eighteenth century, even after the treaty of 1760, which stipulated they could not accept any new runaways into their group. A similar temporal discre­ pancy arises in relation to the Aluku, who are usually considered a post-treaty group, i.e. forming in the second half of the eighteenth century. But as noted in Hoogbergen (1989) the core of the oldest Aluku clans can be traced to the 1712 invasion. Another issue that arises from Smith’s (1987) grouping is the actual ­positioning of the Eastern Maroon varieties in relation to each other; the ­Ndyuka-Pamaka-Aluku cluster is considered a dialect cluster rather than a group of languages, with Kwinti falling somewhere outside this cluster. This is addressed somewhat in his 2002 paper where he arranges the groups in order of their emergence: Saramaccan 1690–1710 Matawai c. 1700 Ndyuka 1712 Aluku before 1760 Kwinti before 1760 Pamaka c. 1800 It should be reiterated that these dates for emergence of the different groups should be made earlier in virtually all instances, although given the literature surveyed above, the order is accurate. Goury and Migge account for this with regard to some of the later groups.

Fig. 2: Development of the Surinamese creoles, based on Goury and Migge (2003: 18)



The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 105

A secondary consideration is the location of the plantations from which slaves escaped. In the literature, this has been described as the determining factor for the high degree of difference between the Central and Eastern Maroon ­languages. Since the majority of the slaves that would later comprise the former group fled plantations of the Jodensavanne, where the European component consisted ­primarily of Sephardic Jews of Iberian origin, their resulting languages would be characterized by a higher degree of Portuguese lexical admixture. Although understudied, the location of origin may have also played a more significant role in the divergence of the Eastern Maroon varieties as well. Given the widespread variation in early Sranan (see van den Berg 2007; van den Berg and Arends 2004) it is likely that a group of escapees from one plantation could take with them a significantly different variety of Plantation Creole than another group fleeing another plantation at the same time. Proving this would be parti­ cularly difficult as archival evidence for such a historical variationist approach is likely to be insufficient. Further muddling the picture is the fact that early clans frequently conglomerated and absorbed smaller groups in addition to kidnapping plantation slaves during incursions. Two possibilities might prove fruitful in this regard. Since the origin myths of most clans are well known (if not in the acade­ mic literature, then to the clan members themselves) and some corroborated by archival evidence, then a study of clan based variation may allow for the recons­ truction of some of the geographic and temporal variation of Plantation Creole. A second possibility would be the study of koo sama tongo ‘the language of the dead’ which is spoken when individuals are possessed by yooka ‘ancestral spirits’. From an emic perspective, the yooka takes possession of its host and speaks through it (presumably in the variety used when s/he was alive). While this may sound like lunacy to a western academic, from an etic perspective, given the centrality of ancestor worship in Maroon religion it is certainly reasonable that “language varieties” are passed through the generations just as other characteristic elements of “religious performance” (see Borges 2014: 50–51). Like the origin myths details of important ancestors (particularly those who recurrently possess the living) are often well known and sometimes corroborated in the archives.

3 Nineteenth century economic advances and the social and demographic upheaval of the late twentieth century Although the treaties between the Maroons and the colonial government stipula­ ted that runaways who were not among the treaty groups were to be turned over to the authorities or killed, van Wetering and Thoden van Velzen (2013) report

106 

 Robert Borges

that there was widespread cooperation among the Maroons in eastern Suriname. Very shortly after the Second Boni War, the Ndyuka and the Aluku concluded a peace agreement (van Wetering and Thoden van Velzen 2013: 66) and in the wake of the Netherlands’ incorporation into the French Republic, the English o ­ ccupied Suriname (1799–1802, 1804–1816) resulting in a decrease in (the intensity of) campaigns against the non-treaty Maroons (van Wetering and Thoden van Velzen 2013: 71). This allowed the Ndyuka, for example, to cooperate with, and some­ times exploit the labor of other groups to further their own economic interests, ­ etering notably the Pamaka and the Krabehollo in the logging industry (van W and Thoden van Velzen 2013: 117ff). While attempts were made to restrict the movement of the treaty Maroons to the plantation area and the coast via pass system,11 many Maroons simply didn’t bother to obtain such passes, and from the earliest years of the 19th century, posthouders lodged complaints that virtually all able-bodied men, incl. village captains, had left their villages to engage in wage labor (van Wetering and Thoden van Velzen 2013: 72). In the first half of the nineteenth century, Maroons from different groups primarily worked together in work camps cutting lumber on the outskirts of the plantation area, on the Cottica River, the Courmotibo area, and the Lower Saramacca. Logs were sold in Parama­ ribo or on plantations directly (van Wetering and Thoden van Velzen 2013: 71). A smaller number of Maroons were employed directly by plantations or saw mills, and some engaged in other activities, such as gold panning (some Ndyuka esta­ blished a settlement on Sara Creek around 1800 for this purpose (van Wetering and Thoden van Velzen 2013: 71), farming for market in the coastal areas, and following discovery of gold on the Lawa River in the 1870s, Maroons were able to make a great deal of money carrying people, equipment and goods through the difficult river passage. Over the course of the nineteenth century, Maroons were able to improve their economic standing and make increasingly significant contributions to the overall economy of the colony, securing their position as the agricultural sector tanked following the abolition of slavery (1863)12 and steep drops worldwide in the price of and demand for sugar (Hoefte 1990: xx). Through their economic acti­ vities, the Maroons were in ever increasing contact with other groups of Maroons and non-Maroons alike. Maroon-Maroon contact in this period is evidenced, for

11 Van Wetering and Thoden van Velzen (2013: 72) report for example that in 1837, there were a mere 40 of such passes available to the Ndyuka. 12 Although slavery was officially abolished in 1863, for fear that the economy would fail com­ pletely, slaves were required to continue working in the agricultural sector, on a plantation of their choosing, for a period of 10 years.



The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 107

example, by the spread of the Sweli Gadu cult13 in the late nineteenth century from the Ndyuka, first to the other Eastern Maroons, then to the Central Maroons (van Wetering and Thoden van Velzen 2013: 117ff). It appears that throughout the nineteenth century, Maroons who engaged in wage labor primarily did so on a temporary basis, returning to their (wives’) villages after a period of work – a practice that continues today among rural Maroons. Hoefte (1990: xii) claims that the abolition of slavery was a major impetus for urbanization of the Surinamese population, though it took some time to transform the society to a more urbanbased one. This, given the 10 year supervisory period wherein newly freed slaves were required to continue plantation work and the subsequent importation of the Asian labor force in an effort to maintain the prominence of the agricultu­ ral sector. This urbanization becomes remarkably noticeable in the first half of the twentieth century, where the population of Paramaribo more than doubled despite overall population decrease country-wide (de Bruijne 2001: 32). Although many Maroons continue to live a rural lifestyle, and circular migrations are still well practiced, events of the twentieth century, particularly the second half, sparked urbanization of Maroon populations en masse (de Bruijne 2007: 17). As the sugar economy tanked and other agricultural alternatives proved insufficient to prevent a complete economic collapse, the discovery of bauxite deposits in the country was an economic saving grace. In 1917, a major bauxite mining venture was established by an American firm Suralco – Suriname Alumi­ num Company, a subsidiary of Alcoa, Aluminum Company of America – which paid better wages and offered better fringe benefits than other opportunities in the country. Bauxite successfully replaced sugar as main export. In the economic boom brought about by WWII, there was much urbanization in Suriname directly related to this industry. Aside from local individuals who were required as ext­ raction laborers, American troops were stationed in Suriname to protect Bauxite mines and key transport routes. Suriname was a principle supplier of Bauxite to the Allied Forces at this time; Hoefte (1990: xxi) reports that approximately two-thirds originated in Suriname. Bauxite continues to be a main export of the Surinamese economy to this day. The take-off of Bauxite marked the beginning of

13 The Sweli ( −−−−−>

−−−−−> −−−−−> −−−−−>

Single – Dual-language context Dense Code switching + goal maintenance + suppression of interference – opportunistic planning competitive schemas

– goal maintenance – suppression of interference + opportunistic planning cooperative schemas

Muysken

alternational

“Temporal”

Synchronic variation −−−−−> −−−−−> −−−−−> −−−−−>

“structure”

Mixed Code

insertional

congruent lexicalization Diachronic change

Fig. 3: Progression of multilingual discourse and interactional contexts toward language change.



The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 115

conventionalized code switching. So while mixed languages and pidgins are not probable results of intense multilingualism, Auer’s term must rather refer to contact induced change resulting in a/the language(s) involved becoming more similar in grammar and lexicon – i.e. convergence (following Yakpo et al.’s (2015) account of convergence in Suriname). I will return to the question of convergent varieties’ “languagehood” towards the end of this section.

4.1 Congruent lexicalization While Migge (2007) describes a number of social and discourse related functions achieved in the Maroon context via language choice, it is also quite common that there are no marked discourse functions of the code mixing, what Myers-­Scotton (1993b) calls “code switching as the unmarked choice”. Given that the most common languages among the Surinamese Maroons, are (a) Maroon language(s), Sranan, and Dutch, and that the former two generally share a great deal in terms of both lexicon and morphosyntax, the most common type of code mixing can be described as congruent lexicalization, from a structural point of view (Muysken 2001). In other words lexemes are freely exchanged over shared grammatical structures. Like Auer’s (1998) code mixing and Green and Abutalebi’s (2013) dense code switching, codes are exchanged at great frequency with little impact on dis­ course where the grammars of the languages involved coincide. Heavy congruent lexicalization is most noticeable with exchanges between and among Maroon languages and Sranan, as illustrated in the following example from Migge and Léglise (2013: 273–274), where several men discuss plans for a hunting trip. The transcription conventions are as follows: Eastern Maroon: bold; Sranan: under­ lined; shared: italics; Dutch: underlined, italics. (5) Br: Mi nanga A. nanga ete wan tra man be o go. I with A. with yet one other man pst fut go ‘Me and A. and another guy were to go.’ Pe: No da a no bun. Ma mi kan rey u, no then it neg good but I can drive you(pl) mi e rey u. Da mi e teki u later, I ipfv drive you then I ipfv take you(pl) later snel, snel snel. quick quick quick ‘No, then that’s no good. But I can drive you, I’m going to drive you.’

116 

 Robert Borges

P: Winsi u go wan dri sma tok. even we go one three people right ‘Even if we go with three people, right.’ Br: No da luku dyaso den fisi di u o kisi, den no then look here det fish rel we fut get det fisi di u o kisi den tyaipi. fish rel we fut get they a.lot ‘No, look here, the fish that we’ll get, they’ll be a lot.’ Pe: Ok mi e ley tu rit. Efu i e teki wan ok I ipfv drive two trip if you ipfv take one taa wagi. other car ‘Ok, I’m going to make two trips. If I’m taking another car.’ Br: No, no fanya a taki. no neg mess.up det talk ‘No, don’t mess up the discussion.’ Pe: Fustan san mi e taki, mi no lobi oli understand what I ipfv say I neg love hol u w’woyo i sabi tok. you(pl) market you know right ‘Understand what I’m saying, I don’t like to be part of your unorganized talking, you know.’ Br: A no oli u w’woyo. it neg hold our market ‘It’s not random talk.’ P: Suku dri man da u dri man e go wan search three man then we three man ipfv go one trip tok. trip right ‘Look for three men, then us three are going to take one trip.’ Pe: luku mi de nanga wagi, san mi taki, mi sab san look I cop with car what I say I know what o go pasa dat se, ma mi nanga wantu fut go happen that side but I with one.two man set kba taki u o go dat se sonde man set already talk we fut go that side Sunday ‘Look, I have a car, what I said, I know what will happen there, but I and some others have made arrangements already that we’ll go there on Sunday.’



The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 117

Here, the question of whether (5) illustrates code mixing or borrowing must be left open for the time being (but see the discussion below), but given the high degree of shared elements/structures, what languages are the recipient and donor in case of borrowing or “matrix language” and “embedded language” (in the sense of MyersScotton 1993b) in the case of code mixing is very difficult to determine. However quantitative analysis of diverse diachronic data gives some clues about the trajectory of contact-induced language change. In the following sub-sections, I will present such analyses of a number of commonly found switched elements in a sample of Traces of Contact project’s Maroon corpora (see Borges 2014a, 2014b, 2015). 4.1.1 /r/ variation The /r/ is a very marked sound in Suriname in that it is indicative of Sranan or urban speech varieties. It is a phoneme in Sranan, though not in the Maroon languages of Suriname. Many lexemes have an /r/ and /r/-less variant, such as moro and moo ‘more’ whereby use of the /r/ variant in a Maroon language conveys an urban feel. A number of such pairs were investigated for comparison. Tab. 1 details the actual number of occur­ rences of each pair. Kwinti and urban Ndyuka show a variable alternation between /r/ and /r/-less tokens, while in rural Ndyuka variation does not exist in this realm. Tab. 1: /r/ tokens in comparison -r/+r

gloss

Kwinti

urban Ndyuka

rural Ndyuka

booko/broko daai/drai doo/doro foo/fowru fuu/furu kii/kiri kondee/kondre moo/moro nengee/nengre peesi/presi seefi/srefi siibi/sribi soutu/s(r)ou(r)tu taa/tra taanga/tranga wataa/watra wooko/wroko yee/yere

broken turn door, arrive, through bird full kill country, village more black person, person place self sleep sort, lock other strong water work hear

1/10 13/5 20/4 11/4 9/3 17/4 2/29 44/8 6/3 10/10 22/5 1/1 3/0 25/14 0/2 12/2 13/7 65/11

2/2 0/1 12/1 0/0 3/8 7/5 7/4 14/14 29/1 0/2 22/1 2/0 2/4 13/11 0/0 0/0 0/11 9/2

15/0 22/0 13/0 0/0 0/0 3/0 5/0 20/0 2/0 9/0 36/0 10/0 7/0 53/0 2/0 8/0 16/0 8/0

118 

 Robert Borges

In the varieties with variation, most /r/-less tokens in the pair are used with a higher frequency, though several pairs are used with approximately equal fre­ quency. Just two /r/ tokens, broko ‘broken’ and kondre ‘village’ occur more fre­ quently in Kwinti than their /r/-less variants. 4.1.2 Conjunctions Conjunctions also differ between EMC and Sranan. EMC uses e(n)ke ‘like’ and anga ‘and’, while Sranan uses lek(i) ‘like’ and nanga ‘and’. The (n)anga type is used for coordinating nominal elements (nouns, numerals, pronouns, and NPs) within a variety of clause types, as well as sentential coordination (Huttar and Huttar 1994: 16, 34–36, 194, 229, 240–243, 248, 532; van den Berg 2007: 134–13). Lek(i) and e(n)ke display a different distribution. They are mainly used as subordinators to manner adverb and equative adverb clauses (Huttar and Huttar 1994: 113, 121; van den Berg 2007: 363–66, 375) or as prepositions in equative phrase (Huttar and Huttar 1994: 293; van den Berg 2007: 153). Tab. 2 compares the variation in the use of e(n)ke vs. lek(i). Kwinti speakers vary these forms rather evenly. Interestingly, of the eight speakers who actually produced one of these conjunctions, three produced only one, three produced only the other, and the remaining two produced both, though each favored a dif­ ferent variant. Tab. 2: Coordinators compared

lek(i) e(n)ke

Kwinti

Urban Ndyuka

Rural Ndyuka

10 13

6 11

0 9

Urban Ndyuka speakers also displayed some variation, though they favored the Maroon variant. Rural Ndyuka speakers did not exhibit variation in this regard. Less dramatically, anga and nanga indicate a more consistent usage pattern in favor of the Maroon variant in all three language varieties, detailed in Tab. 3. Tab. 3: NP coordinators compared

anga nanga

Kwinti

Urban Ndyuka

Rural Ndyuka

69 5

53 11

87 1



The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 119

4.1.3 Negation The forms of Sranan and Maroon negations do not correspond. In all the Surina­ mese creoles, preverbal negation combines phonologically with a following imper­ fective marker e. The resulting merged forms also differ. Tab. 4 below details the different forms. The Maroon style preverbal negation (n)á(n) is used more frequently than Sranan no by Kwinti and rural Ndyuka speakers. Once imperfective marking is added, ne becomes the preferred strategy for the urban Ndyuka and Kwinti. Tab. 4: Sranan versus Ndyuka pre verbal negation

neg neg + ipfv

Sranan

Ndyuka

no ne

(n)á(n) (n)ái

Tab. 5 compares the strategies for preverbal negation in the three Maroon varie­ ties. While the form of preverbal negation varies in all three varieties, the rural Ndyuka variation is somewhat negligible, favoring Maroon type negators. Urban Ndyuka shows a slight preference for Sranan type negation and Kwinti speakers prefer the Maroon variant without imperfective marking. Curiously however, there is a strong tendency towards ne among Kwinti speakers. It is not clear though, if ne is a variable from Sranan or if it is Kwinti’s inherited form. In all other Maroon languages the neg + ipfv is realized as nái except for Aluku, one of Kwinti’s closest sister languages, which employs a phonetically similar form nee (Goury and Migge 2003: 18; Smith 2002: 132). Tab. 5: Negation strategies compared

(n)á(n) no nái ne

Kwinti

Urban Ndyuka

Rural Ndyuka

116 68 6 45

31 48 11 14

88 5 12 0

It is also possible that nái has made its way into Kwinti via the other Maroon vari­ eties encountered in Paramaribo. If the latter were to be proven, nái would be an excellent piece of evidence demonstrating the influence of other Maroon varieties on Kwinti. But it is perhaps just as likely that these two variables have existed in Kwinti since the beginning.

120 

 Robert Borges

4.2 Heavy insertional code mixing and morphosyntactic integration There are just a few areas where Maroon languages and Sranan do not corres­ pond in their grammatical structure, where insertional mixing would potentially lead to some sort of morphosyntactic consequence. No instances of this have been identified in the current data. Loosely related to this point would be the changes in modal marking in urban Ndyuka, following the pattern and forms of Sranan; the syntax is unchanged, but the categorial semantic organization of what is marked and how is altered (see Borges et al., this volume). It is relatively more probably to encounter instances necessitating morphosyntactic integra­ tion when Dutch elements are inserted. As with Maroon languages and Sranan, Dutch items are also potentially inserted in Maroon discourse at high frequency, as illustrated by the following example from urban Ndyuka. Dutch elements are shown in bold. a saanantongo dati mi taki tu, ya helemaal. ma a seefi buku a pe de’e poti a (6)  kwa ini olansi baka dus . . . eigenlijk . . . want den taki a nederlans ma so sani nái kon overeen . . . dus waarder. i abi fu taki okaans anga den ndyuka nengee. nou over het algemeen sama... de meeste sama nái si taki mi a ndyuka nengee want den nái schatten so furu. de’e taki gewoon taki meisje a nengee ma m’e taki nederlans over het algemeen dus drap . . . drape mi nái si echt wan probleem so. ‘I also speak Sranan. Yes, completely. But the same book there, they’re translating it back into Dutch, thus... actually... because they speak Dutch, but many things [about the languages] aren’t the same, there­ fore [they’re] valuable. You have to speak Ndyuka with the Aucanners. Now in general people . . . most people don’t see that I’m an Aucanner because they don’t judge so well. They’re just saying the girl is black, but I generally speak Dutch, so there . . . there I don’t really see any problem.’ Non-nativized functional and non-functional words are frequently attested in these instances of insertional codeswitching in the urban Ndyuka and Kwinti samples. Lexemes that appear time and again are detailed in (7). Some such insertions result in the mismatch of structures, necessitating some degree of mor­ phosyntactic integration. In (8) a Dutch adjective is inserted; property concepts in the creoles are generally expressed predicatively as verbal elements, but here the adjective requires a copula, de in this case. It introduces existential, stative, or locational predicates.



(7) (8)

The Maroon creoles of the Guianas: Expansion, contact, and hybridization 

 121

omdati