Language Contact in the Arctic: Northern Pidgins and Contact Languages 9783110813302, 9783110143355

Table of contents :
Preface
Introduction
Northern pidgins
Dual-source pidgins and reverse creoloids: Northern perspectives on language contact
The special case of Arctic pidgins
Siberia
Language contact in northeastern Siberia (Chukotka and Kamchatka)
Chukchi, English, and Eskimo: A survey of jargons in the Chukotka area
A case of nongenetic development in the Arctic area: The contribution of Aleut and Russian to the formation of Copper Island Aleut
The Taimyr Peninsula Russian-based pidgin
Northwestern Russia and Scandinavia
Solombala-English in Archangel
The Vardø merchants’ reduced Russian
On the pidgin status of Russenorsk
Aspect marking and grammaticalization in Russenorsk compared with Immigrant Swedish
Greenland
Eskimo pidgin in West Greenland
North America
Language contact and pidginization in Davis Strait, Hudson Strait, and the Gulf of Saint Lawrence (northeast Canada)
An Inuit pidgin around Belle-Isle Strait (research note)
Broken Slavey and Jargon Loucheux: A first exploration
Arctic origin and domestic development of Chinook Jargon
Index

Citation preview

Language Contact in the Arctic

W G DE

Trends in Linguistics Studies and Monographs 88

Editor

Werner Winter

Mouton de Gruyter Berlin · New York

Language Contact in the Arctic Northern Pidgins and Contact Languages edited by

Ernst Häkon Jahr Ingvild Broch

Mouton de Gruyter Berlin · New York

1996

Mouton de Gruyter (formerly Mouton, The Hague) is a Division of Walter de Gruyter & Co., Berlin.

© Printed on acid-free paper which falls within the guidelines of the ANSI to ensure permanence and durability.

Library of Congress Cataloging-in-Publication-Data Language contact in the Arctic ; northern pidgins and contact languages / edited by Ernst Hakon Jahr, Ingvild Broch. p. cm. - (Trends in linguistics. Studies and monographs ; 88). Chiefly papers presented at the 9th International Troms0 Symposium on Language; Arctic Pidgins, which was held June 4 - 6 , 1992, University of Tromse. Includes bibliographical references and index. ISBN 3-11-014335-6 (alk. paper) 1. Pidgin languages - Arctic regions - Congresses. 2. Languages in contact — Arctic regions — Congresses. I. Jahr, Ernst Häkon, 1948. II. Broch, Ingvild. III. International Troms0 Symposium on Language (9th ; 1992 ; University of Tromso) IV. Series, PM7807.A73L36 1996 417'.22'091632-dc20 95-26183 CIP

Die Deutsche Bibliothek — Cataloging-in-Publication-Data Language contact in the Arctic : northern pidgins and contact languages / ed. by Ernst Hakon Jahr ; Ingvild Broch. - Berlin ; New York : Mouton de Gruyter, 1996 (Trends in linguistics : Studies and monographs ; 88) ISBN 3-11-014335-6 NE: Jahr, Ernst Häkon [Hrsg.]; Trends in linguistics / Studies and monographs

© Copyright 1996 by Walter de Gruyter & Co., D-10785 Berlin All rights reserved, including those of translation into foreign languages. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording or any information storage and retrieval system, without permission in writing from the publisher. Typesetting and printing: Arthur Collignon GmbH, Berlin. Binding: Lüderitz & Bauer, Berlin. Printed in Germany.

Preface

The majority of the papers included in this volume were first presented and discussed at a symposium held at the University of Tromsa, 4—6 June 1992: "The Ninth International Tromso Symposium on Language: Arctic Pidgins". (The papers by Louis-Jacques Dorais and Stephen A. Wurm were not presented in Tromso.) The symposium was organized by Ernst Häkon Jahr and Ingvild Broch, and supported financially by the University of Troms0, the School of Languages and Literature (University of Troms0), and the Norwegian Research Council. Tromso, 27 June 1994

Ernst Häkon Jahr and Ingvild Broch

Contents

Preface

ν

Introduction

1

Northern pidgins Peter Trudgill Dual-source pidgins and reverse creoloids: Northern perspectives on language contact Ian Hancock The special case of Arctic pidgins

5

15

Siberia Bernard Comrie Language contact Kamchatka)

in

northeastern

Siberia

(Chukotka

and 33

Willem J. de Reuse Chukchi, English, and Eskimo: A survey of jargons in the Chukotka area

47

Evgenij V. Golovko A case of nongenetic development in the Arctic area: The contribution of Aleut and Russian to the formation of Copper Island Aleut

63

Stephen A. Wurm The Taimyr Peninsula Russian-based pidgin

79

Northwestern Russia and Scandinavia Ingvild Broch Solombala-English in Archangel

93

Siri Sverdrup Lunden The Vardo merchants' reduced Russian

99

viii

Contents

Ernst Häkon Jahr On the pidgin status of Russenorsk

107

Ulla-Britt Kotsinas Aspect marking and grammaticalization in Russenorsk compared with Immigrant Swedish 123 Greenland Hein van der Voort Eskimo pidgin in West Greenland

157

North America Peter Bakker Language contact and pidginization in Davis Strait, Hudson Strait, and the Gulf of Saint Lawrence (northeast Canada) 261 Louis-Jacques Dorais An Inuit pidgin around Belle-Isle Strait (research note)

311

Peter Bakker Broken Slavey and Jargon Loucheux: A first exploration

317

William J. Samarin Arctic origin and domestic development of Chinook Jargon

321

Index

341

Introduction

This collection of papers aims at drawing the attention of scholars in pidgin and creole linguistics and in language development to the special conditions and features exhibited by the pidgins and contact languages of the Arctic and northern regions of the globe. Till now, most attention in creolistics has been directed towards the pidgin and Creole languages we find more or less close to the equator. From the Arctic region, only the pidgin Russenorsk has so far been taken into account in the more general discussion. Thus, most of the theoretical insights derived from pidgin and creole studies have been based on pidgins and Creoles found in southern regions. However, by widening the geographical area of study to include pidgins and contact languages of the far north it is possible to gain valuable new insights into the different mechanisms involved in pidgin origin and development. The papers presented here, most of which were first presented and discussed at a conference in Tromso in 1992, show that this opening up of the geographical scope has already provided a lot of new data as well as novel theoretical knowledge. In a summing-up statement at the Troms0 conference, Ian Hancock underlined this aspect: I feel the sensation that one gets after sitting for some time in a carriage of a train that has come to a stop somewhere along the line which suddenly jerks back into motion as it begins to move again. I truly feel that we have covered a lot of new ground in the past three days, particularly in the presentation of new data. We have also been forced to reexamine the theoretical criteria for selecting these languages for special attention, and it has been made clear that our existing terminology is inadequate and is in serious need of revision.

The papers included in the present volume discuss theoretical questions as well as give empirical descriptions of important pidgins and contact languages of the northern regions. Together, these papers cover the whole vast circumpolar area. Peter Trudgill discusses theoretical issues especially connected with the important question of dual-source pidgins. Ian Hancock gives an overview over all known pidgins and contact languages of the northern area. The Siberian far east is covered by Bernard Comrie and by Willem J. de Reuse who describe the Chukotka area. The special case of Copper Island Aleut is discussed by Evgenij V. Golovko. Stephen A. Wurm gives an account of Taimyr pidgin. Northwest Russia

2

Introduction

and north Scandinavia are represented by Solombala-English, which was used in Archangel in the nineteenth century (Ingvild Broch); by Vard0 reduced Russian, also a nineteenth-century phenomenon (Siri Sverdrup Lunden); and by Russenorsk in north Norway (Ernst Häkon Jahr, UllaBritt Kotsinas). Hein van der Voort gives an extensive description of Eskimo pidgin in Greenland, while Peter Bakker, Louis-Jacques Dorais, and William J. Samarin cover various contact languages of the far north of the American continent. It is the hope of the editors that this volume will stimulate more extensive studies into the fascinating features of the pidgins and contact languages of the Arctic and northern regions. We are confident that such investigations will yield results which will lead to important modifications in, and necessary redefinitions of, the theoretical models employed in pidgin and Creole studies.

Northern pidgins

Dual-source pidgins and reverse creoloids: Northern perspectives on language contact Peter

Trudgill

Most often, language contact does not lead to the development of new (pidgin or Creole) languages. In those cases where new varieties do form, however, there appear to be two fundamental mechanisms which are instrumental in their formation: 1. the inability of post-adolescent humans to learn new languages perfectly (see Trudgill 1989a); and 2. the process of focusing (see Le Page - Tabouret-Keller 1985), which may occur in certain social and linguistic circumstances. What we might perhaps refer to as the classical model of pidgin and creole formation, which many creolists, particularly those who have worked with the Atlantic Creoles, appear to subscribe to, although none of them, as far as I know, have actually formulated it in precisely these terms, can be presented as follows. (Naturally, none of the complexities and subtleties associated with what actually happens in real-life language contact situations can be accurately portrayed in any such model - see Hancock 1986).

Pidginization Whenever adults and post-adolescents learn a new language, pidginization takes place (Trudgill 1989b). Pidginization consists of three related but distinct processes: reduction, admixture, and simplification. Reduction, or impoverishment as it is sometimes, less happily, called, refers to the fact that, in pidginized form, there is simply less of a language as compared to the form in which it is spoken by native speakers: the vocabulary is smaller, and there are fewer syntactic structures, a narrower range of styles, and so on. Admixture refers to interference - the transfer of features of pronunciation and grammatical and semantic structure from the native language to the new language, an obvious feature of adult second-language

6

Peter

Trudgill

acquisition. Simplification, as is well known (see Mühlhäusler 1977), is a rather complex phenomenon, but it refers crucially to regularization of irregularities, to loss of redundancy (such as grammatical gender), and to an increase in analytic structures and transparent forms. Reduction can be considered as being due to incomplete learning and restriction in sociolinguistic function, while admixture and simplification are the result of imperfect learning.

Pidgins and pre-pidgins In some cases, where exposure to the new language is hiinimal, such pidginization may be extreme, and remain extreme. In certain cases, moreover (see Whinnom 1971), such extremely pidginized forms of language may, in the absence of native speakers of the original language, become important as a lingua franca, a means of communication between two or more groups who have acquired the pidginized forms and who have no native or other language in common. In these cases, focusing may well occur: the pidginized forms of the original language may acquire stability, with widely shared norms of usage, and a new language variety, a pidgin, will have come into being. Typically, then, a pidgin is a stable language, without native speakers, which is the outcome of reduction, admixture, and simplification of some source language, and where, also typically, pidginization has occurred to such a degree that mutual intelligibility with the source language is no longer possible. We know of at least one such variety from northern latitudes (see below). The language associated with the chronological stage that occurs before focusing leads to the achievement of stability and the development of shared norms, and where the pidginized forms are still relatively diffuse, can be referred to as a "pre-pidgin".

Creolization and Creoles Again according to this model, a further chronological stage may occur. In some circumstances, a pidgin language, as a lingua franca, may become the most important or indeed only viable shared language of a particular community. The pidgin will therefore be subject to expansion,

Dual-source pidgins and reverse creoloids: Northern perspectives

on language contact

7

in some cases rapidly, in other cases more slowly, so that it can be used in an increasingly wide range of functions, and come to meet the linguistic needs of native speakers. The result is a "creole" language. A creole, then, is indeed a pidgin which has acquired native speakers, but most crucially it is a pidgin which has undergone non-contact-induced expansion, where the expansion process (as Bickerton (1981) has pointed out, one of the most fascinating forms of change) "repairs" the results of the reduction process which occurred during pidginization. Non-contact-induced expansion is known as "creolization", which is a term which should not be used in a haphazard way for just any form of language mixture (see also Hancock, this volume). Nor, indeed, should it be used for just any kind of expansion: if a pidgin comes into renewed or closer contact with its original source language before creolization occurs, "depidginization" may take place. In this case, however, any expansion which occurs will be contact-induced and will lead in the direction of the source language, rather than being, as with creolization, internally generated. A creole language is thus a language which, relative to its source, is simplified (i. e., more analytic and regular) and mixed, but not in any way reduced: Creoles are perfectly normal languages with an unusual history.

Decreolization and post-creoles The next possible chronological stage that can be experienced by Creoles is decreolization. Like depidginization, decreolization is contact-induced. While depidginization is clearly the reverse of pidginization, decreolization is not the reverse of creolization, and for that reason it may be that we should develop another term for this process. Hancock (cf. Hancock 1988) favors the term "metropolitanization". If a creole comes into renewed or intensified contact with its source language, it may happen, if the sociolinguistic conditions are right, that it will begin to change in the direction of this source language. Clearly, changes in the direction of the source language will involve processes which reverse the effects of admixture and simplification. We may call these processes "purification" (the removal of words and forms which are not derived from the source language) and "complication" (the reintroduction of irregularity, etc.). Decreolization may eventually go to completion, so that the creole may become, or become perceived as, a variety of the source language,

8

Peter

Trudgill

as may well have happened with United States Black Vernacular English. Such a variety may be termed a "vestigial post-creole". Various intermediate stages are also of course possible. Such intermediate varieties can be called "post-creoles". A well-known phenomenon involving intermediate stages is the "post-creole continuum", such as that which exists in Jamaica, where a society demonstrates a cline of varieties ranging from a variety of the source language (Jamaican English) at the top of the social scale, to increasingly un-decreolized varieties of Jamaican Creole at the bottom. Post-creoles, or partially decreolized Creoles, will therefore demonstrate, relative to the source language, different degrees of simplification and admixture, although of course less than a totally un-decreolized Creole such as Sranan.

Creoloids and non-native creoloids It now becomes necessary to look beyond the traditional pidgin and Creole life-cycle model. We notice, first, that, interestingly, there are many varieties of language in the world which look like post-creoles but which actually are not. Such varieties demonstrate relatively undramatic admixture and simplification relative to some source language, but are known to have no pidgin history behind them. Such languages, as I have suggested elsewhere (Trudgill 1983: 102), can be called "creoloids", and the process which leads to their formation "creoloidization". The process of creoloidization thus consists of admixture and simplification. Unlike Creoles, however, creoloids have not experienced a history of reduction followed or "repaired" by expansion. Creoloids were never reduced in the first place. The difference between a creoloid and a partially decreolized Creole is thus a historical one. It is not apparent from synchronic inspection. Creoloidization is of course the result of the influence of imperfect learning by relatively large numbers of non-native adult speakers. Creoloids, however, are varieties which have never been reduced because they have maintained a continual native-speaker tradition. A good example of a creoloid is Afrikaans, which is clearly a creoloid relative to Dutch. Creoloids proper can be distinguished from non-native creoloids, such as Singaporean English (see Piatt - Weber 1980). A non-native creoloid may develop when, as in pidgin-formation, a pidginized variety of a source language becomes focused and acquires stability as a result of

Dual-source pidgins and reverse creoloids: Northern perspectives

on language contact

9

being employed as a lingua franca by two or more language groups who are not native speakers of the source language, and who have no other language in common. The difference between a non-native creoloid and a pidgin lies in the degree of pidginization which it has undergone. For example, Singaporean English is a recognizable and rather stable secondlanguage form of English which can be distinguished typologically from foreign-language forms of English such as, say, Japanese English by its institutionalization. Relative to metropolitan forms of English, Singaporean English is somewhat mixed and simplified and, because it is not spoken natively, also somewhat reduced. It is nevertheless still clearly a variety of English: the role of English in, for example, the education system and in Singaporean society generally, has meant that simplification and admixture have never been extreme, and its use as a primary (as opposed to first) language by many speakers in Singapore means that the reduction is also relatively slight. Unlike Afrikaans, however, it has no native speakers and therefore no native-speaker tradition to maintain.

Dual-source pidgins As we said above, the traditional pidgin/creole life-cycle model has been developed, if not always fully articulated, in connection with colonial, usually tropical, language contact, especially in the Pacific, and even more especially in the Atlantic Ocean areas. As our discussion of creoloids shows, however, this model needs amending if it is to give us a full and useful typological account of mixed and simplified languages. This requirement has become particularly clear as a result of our study of the less well-known Arctic contact varieties. Our examination of languagecontact situations, and their outcomes in the Arctic and other northern areas, indicates that the model needs to be supplemented in a number of ways. In particular, as is illustrated by the case of the best-known Arctic pidgin, Russenorsk, we need to take account of contact varieties that arise from the pidginization not of one source language but of two. As the terms "language contact" and "admixture" indicate, all pidgin formation obviously involves more than one language. However, most Atlantic and Pacific pidgin, creoloid, and post-creole languages have a single main source or lexifier language, so that we have no hesitation at all about

10

Peter

Trudgill

saying that, say, Sranan and Tok Pisin are both English-based varieties, in spite of the considerable minority input of other languages, e. g.; eight percent of Sranan lexis is African in origin. Creoloids like Afrikaans and non-native creoloids are also single-source varieties. Russenorsk, on the other hand, appears to be a pidginized form of Russian and Norwegian in about equal proportions. I suggest here that we need to distinguish such varieties typologically from other pidgins by labelling them "dual-source pidgins". These dual-source pidgins are of course linguistically different because their social genesis was different. Jahr (this volume) shows convincingly that Russenorsk was a stable, focused variety that had norms of usage which had to be learnt. Unlike in the case of those pidgins catered for in the traditional "Atlantic" model, however, it is clear that this focusing did not take place according to the Whinnomian scenario, i. e., in the absence of source-language speakers. Russenorsk, although it was also used by native speakers of Finnish and Sami, was mainly spoken by Russian and Norwegian speakers, and must have undergone focusing as a result of interaction between them. Importantly, also, Russenorsk was formed as a result of interaction between two groups of European trading partners rather than in a colonial or precolonial situation. (This does not mean to say, of course, that traditional pidgins cannot be found in the Arctic: Taimyr Pidgin Russian, spoken in an area of northern Russia colonized relatively late by Russian speakers, seems to be of this type.) There must, of course, have been a period, perhaps quite a considerable period, before Russenorsk acquired stability and became a relatively focused, named variety. I suggest here that we refer to this diffuse stage of a dual-source pidgin's development as a "jargon". The term "jargon" has had different uses in the pidgin literature (see the discussion in Mühlhäusler 1986), and is often used as being synonymous with "prepidgin". If, however, we are making a typological distinction, linguistically and socially, between pidgins proper and dual-source pidgins, it would be useful to be able to distinguish also between their precursors in the same way. A pidgin is therefore preceded chronologically by a prepidgin, a dual-source pidgin by a jargon. The Arctic area which gave rise to Russenorsk, perhaps unsurprisingly, seems also to have given rise to a number of relatively unfocused jargons of this type. For example, Solombala-English (Ingvild Broch, this volume), may have been such a jargon, as may have been the Basque-Algonquian "pidgin" (Bakker, this volume), and the Icelandic-Breton "pidgin" (Hancock, this volume).

Dual-source pidgins and reverse creoloids: Northern perspectives

on language contact

11

Dual-source Creoles In the case of single-source varieties, we saw that there was a potential chronological development of the form: pre-pidgin —• pidgin —• Creole —• post-creole —• vestigial post-creole In our discussion of dual-source varieties, we have so far noted only the following language types: jargon —• dual-source pidgin This raises the question of whether we can find further parallels: are there examples of the creolization and perhaps subsequent "decreolization" of dual-source pidgins? There seem to be none in the Arctic, and the typical social setting for dual-source pidgin formation — trading between equal partners - would seem to suggest that this would be unlikely: it is difficult to conceive of social situations where such a language would become the only viable language of a community. However, social situations, albeit highly unusual ones, can arise in which dual-source Creoles can develop. One language that clearly merits the label "dual-source Creole" is Pitcairnese (Ross - Moverley 1964). Pitcairnese is the native language of the Pitcairn islanders, and does not therefore demonstrate reduction. It is, however, a considerably simplified and mixed form of English and Tahitian. Moreover, its close relative, Norfolk, which has been more heavily influenced by English since the arrival of its speakers on Norfolk Island, can be regarded as a dualsource post-creole. (Naturally, decreolization of a dual-source Creole has to be in the direction of one of the source languages or the other, not both!)

Dual-source creoloids Our examination of Arctic contact varieties also shows yet another parallel between single-source and dual-source varieties. Copper Island Aleut (Golovko, this volume) is a language which is clearly the historical outcome of a mixture of Russian and Aleut with, for example, Russian verbal inflection and Aleut nominal inflection. There is, however, no reduction - the language is spoken natively. Unlike Pitcairnese, moreover, Copper Island Aleut demonstrates relatively little simplification. Indeed, one could argue that this is an example of language contact involving

12

Peter

Trudgill

child rather than adult bilingualism which has therefore in some respects led to complication (cf. Trudgill 1989b). The origins of Copper Island Aleut are a matter for dispute and conjecture. It has been argued, for example (see Golovko — Vakhtin 1990), that it was derived from some earlier pidgin. This seems unlikely, however, in view of the vast amounts of synthetic morphology which the language has retained. One possible scenario, therefore, is that there was no prior dual-source pidgin, and that rather the language represents, as it were, a mixture of, and compromise between, two native-speaker traditions. Two communitites, in long and intimate contact (see Samarin, this volume), gradually merged to form a single ethnic group, neither abandoning their native language but approximating it to that of the other. There are, that is, some parallels with the language-maintenance tradition of, say, Afrikaans, but of course it was two separate native-speaker traditions that were (in part) retained. It may be legitimate, therefore, to refer to Copper Island Aleut, and other similar languages such as Metsif, as "dual-source creoloids".

Reverse creoloids Dual-source pidgins and dual-source Creoles are the result of a break in native-speaker tradition followed by new-language formation. Dualsource creoloids, on the other hand, represent a particular kind of result of language contact combined with language maintenance. There is also, however, a third type of dual-source scenario that we have to consider in this context. There is a type of mixed language, exemplified in northern latitudes but also found elsewhere, which is a particular result of language contact accompanied by language shift. For example, Shetland Island Scots is clearly a variety of Scots, but one which shows considerable amounts of Scandinavian (Norn) influence, particularly in lexis. Unlike cases such as Afrikaans, where a language maintains its native-speaker tradition but is subject to considerable influence from non-native speakers, Shetland represents the opposite process, in which a community abandons its native language, but takes along with it, as it were, in the process of language shift, a considerable amount of influence from its original pre-shift language. A legitimate term for such varieties might therefore be "reverse Creoles". Other similar languages are Yiddish and Ladino. It is also possible that we could accurately refer to originally secondlanguage varieties such as Irish English as "vestigial reverse Creoles."

Dual-source pidgins and reverse creoloids: Northern perspectives

on language contact

13

Ethnolects such as Scandoromani (see Hancock, this volume) and Angloromani represent a special case of reverse creoloids, in that, while they do result from language shift, they are specialized codes rather than native varieties which are used for all purposes from childhood.

Conclusion The non-tropical contact varieties discussed in this volume demonstrate the importance of contact varieties formed out of interaction between two languages only, as opposed to the three-or-more language contact associated with the traditional Whinnomian Atlantic pidgins and Creoles. We have cited instances of jargons (as defined above), dual-source pidgins, dual-source creoloids, and reverse creoloids, all of which seem to have arisen in bilingual rather than multilingual situations. (We also noted the non-Arctic dual-source Creole, Pitcairnese.) A typology of these varieties is given in Table 1. Table 1. Single and dual-source contact varieties New-language formation jargon Icelandic-Breton pidgin pidgin dual-source pidgin Russenorsk creole dual-source creole Pitcairn post-creole dual-source post-creole Norfolk

Language shift

Language maintenance

reverse creoloid Shetland

creoloid Afrikaans

pre-pidgin

dual-source creoloid Copper Island Aleut

We can argue that the social situations in which these ^northern varieties were formed were often significantly different from those which obtained in the more tropical regions more frequently investigated by pidgin and creole scholars. We can also suggest, as always with sociolinguistic work, that it is dangerous to draw linguistic generalizations from only one type of linguistic community. Our further understanding of the range of possible outcomes of language contact can only be enhanced by studies of areas such as the Arctic where such contact has so far been relatively less thoroughly investigated.

14

Peter Trudgill

References Bickerton, Derek 1981 Roots of language. Ann Arbor: Karoma. Breivik, Leiv Egil - Ernst Häkon Jahr (eds.) 1989 Language change: Contributions to the study of its causes. Berlin-New York: Mouton de Gruyter. Golovko, Eugeni - Nikolai Vakhtin 1990 "Aleut in contact: The CIA enigma", Acta Linguistica Hafniensia 22: 9 7 125. Hancock, Ian 1986 "The domestic hypothesis, diffusion and componentiality: An account of Atlantic anglophone Creole origins", in: Pieter Muysken - Norval Smith (eds.), 71-102. 1988 "Componentiality and the origins of Gullah", in: James Peacock - James Sabella (eds.), 13-24. Hymes, Dell H. (ed.) 1971 Pidginisation and creolisation of languages. Cambridge: Cambridge University Press. Le Page, Robert B. - Andree Tabouret-Keller 1985 Acts of identity. Cambridge: Cambridge University Press. Mühlhäusler, Peter 1977 Pidginisation and simplification of language. Canberra: Pacific Linguistics. 1986 Pidgin and creole linguistics. Oxford: Blackwell. Muysken, Peter - Norval Smith 1986 Substrata vs. universals in creole genesis. Amsterdam: Benjamins. Peacock, James - James Sabella (eds.) 1988 Sea and land: Cultural and biological adaptation in the southern coastal plain. Athens, GA: University of Georgia Press. Piatt, John - Heidi Weber 1980 English in Singapore and Malaysia. Oxford: Oxford University Press. Ross, Alan S. C. - A. W. Moverley 1964 The Pitcairnese language. London: Andre Deutsch. Trudgill, Peter 1983 On dialect: Social and geographical perspectives. Oxford: Blackwell. 1989a "Contact and isolation in linguistic change", in: Leiv Egil Breivik - Ernst Häkon Jahr (eds.), 227-237. 1989b "Language contact and simplification", Nordlyd 15: 113-121. Whinnom, Keith 1971 "Linguistic hybridisation and the 'special case' of pidgins and Creoles", in: Dell H. Hymes (ed.), 91-115.

The special case of Arctic pidgins Ian Hancock

I am going to break the rules a bit and change the map, since the title of our conference, "Arctic pidgins", would strictly only allow me to talk about Russenorsk and Pidgin Inuit. 1 In this chapter, then, I have pushed the limit southwards to the sixtieth latitude, which touches the tip of Greenland and cuts through southern Norway and the southern coast of Alaska, and includes all of Iceland. Before selecting the languages I am going to discuss below, I will give a brief survey of the northern pidgins, some of which fall outside of the sixtieth parallel. All we know about most of these is simply a reference in this or that literary source; their geographical location is indicated on the map on page 16. 1. Chinook Jargon or Wawa was based lexically upon Chinukan, Nootka, Salish, Kwakiutl, and (later) French and English, with smaller contributions from Hawaiian, Chinese, and other languages. There is some evidence that it existed prior to European contact, although William Samarin (this volume) argues against this. It was originally spoken along the northwestern Pacific coast between the Columbia River and Vancouver Island; during the nineteenth and early twentieth centuries its area of use moved gradually northwards, eventually reaching the southern and western coasts of Alaska (Johnson 1978, Shaw 1909, Silverstein 1972, Thomas 1935, Thomason 1983). 2. There is some indication that a contact language based on Haida was also spoken at one time, in the Queen Charlotte Islands, though this remains undocumented. 3. Copper Island Aleut, spoken as a first language by a small population on Bering Island in the Commander Island group between Siberia and Alaska, is a contact language which derives from both Aleut and Russian. Discussed by Golovko (this volume). 4. Herschel Island Pidgin Inuit (Eskimo Trade Jargon), together with other Inuit-based pidgins used between Yupik and Inupiat Inuit and European seamen, spoken in Kotzebue, Point Hope, Point Barrow, and Marble Island along the northern coast of North America, are discussed by Stefänsson (1909).

16

Ian Hancock

Distribution of the Northern Pidgins

Figure 1. Distribution of the Northern Pidgins and Contact Vernaculars

The special

case of Arctic pidgins

17

5. Stefänsson also refers to a "more highly developed" contact language spoken between Inuit (Inupiat) and Athabascan-speaking Loucheux Indians in Mackenzie at Fort Arctic, Fort Macpherson, and the Red River Settlement. 6. An Athabascan (ΞΙανέ) pidgin once spoken on the Yukon and Tanana Rivers in central Alaska is reported by Dall (1870) and Whymper ("a broken Slavee", 1868: 226). Discussed by Bakker (this volume). Numbers of contact languages of different European bases emerged on the Atlantic coast of Canada during the early colonial period. Peter Bakker (1988) has listed the following (7-10): 7. A Portuguese-Algonkian from the Gaspe Peninsula. 8. A seventeenth-century Basque pidgin from Nova Scotia and Tadoussac. 9. A sixteenth-century Inuit-French pidgin from the Straits of Belle Isle. 10. A Newfoundland Pidgin English. 11. A variety of Greenland Contact Inuit is discussed by Van der Voort in this volume. 12. Van der Voort also discusses a Greenland Pidgin Danish, mentioned by Nielsen (n. d.: 79). 13. A French-Icelandic pidgin used by fishermen is reported in the 1911 edition of the Encyclopaedia Britannica; see also Bakker (1989). 14. Hualde (1991) discusses an Icelandic-Basque pidgin once employed by fishermen in the north Atlantic. 15. Dr. S. Sanderson of the University of Leeds reports (in personal communication) an Icelandic-Breton pidgin in use in former times by fishermen in the northern Atlantic. See also Sizaire (1976). 16. He also reports the earlier use of an Icelandic-English pidgin. 17. Russenorsk (Russonorsk, Moja pa Tvoja), deriving mainly from Russian and Norwegian, and used between speakers of those languages as well as others (Sami, Finnish), was in extensive use in northern Scandinavia and the Barents Sea in the nineteenth century, and survived into the 1920s (Broch - Jahr 1984a, 1984b). 18. A Baltic Maritime pidgin, called "Scandinavian" by its speakers, is still in use, according to Jan Fürst, a Canadian marine consultant familiar with it. It is spoken by Estonian and other sailors, and contains elements from Scandinavian, English, German, and other sources (contact with Mr. Fürst thanks to William Samarin). 19. A reduced variety of Russian was documented in Varde in the middle of the nineteenth century, and is discussed by Lunden (this volume).

18

Ian Hancock

20. An English-Russian pidgin was used in the port of Solombala at Archangel on the White Sea during the second half of the eighteenth century, and is discussed by Ingvild Broch (this volume). 21. Lunden (1978) mentions the possible existence of a Russian-English pidgin spoken in Moscow during the seventeenth century, also mentioned in Broch - Jahr (1984a: 30). 22. Jahr (this volume) also mentions a possible Norwegian-Sami pidgin. 23. Broch and Jahr mention a Swedish-Sami pidgin (1984b: 70). 24. Iversen (1950) has described a Norwegian-Romani cryptolect. Varieties of Romani-lexifier varieties of Scandinavian are also dealt with by Hancock (1992a). 25. A Pidgin Samoyed is spoken in the southwestern part of the Taymyr Peninsula, according to Evgenij Xelimskij (1987). 26. A Chinese-Russian pidgin spoken in the Mongolian Russian border region around Kjaxta (Kyakhta) from the 1720s is described by Neumann (1966). 27. A similar Chinese-Russian pidgin, which also contains elements from Polish, spoken in northeastern China around Harbin and Tang-Pei, has been described by Jabkmska (1957), and was still in use at the time of her writing. 28. A Chukchi-Russian pidgin spoken in the Chukotka Peninsula at the end of the nineteenth century has been discussed by De Reuse (this volume) and by Comrie (this volume). 29. Comrie (this volume) refers to a Yupik-Chukchi contact variety spoken in the Chukchi Peninsula. 30. Comrie also mentions a (Siberian) Yupik-Russian contact variety from the same area. 31. He also mentions an Aleut-Russian contact variety from the same area. 32. De Reuse (this volume) and Comrie (this volume) report on an Inuit pidgin. 33. Comrie also reports the existence of an English pidgin spoken in the same area. 34. Bogoras (1922) writes of a Kamchadal-Russian pidgin from the Kamchatka Peninsula in eastern Siberia. A "Russian Kamchadal Jargon" spoken in the village of Sedanka on the Tighil River is also mentioned by Jochelson (1928: 49). Comrie (this volume) also discusses this. 35. Bogoras (1922) also reports on a Koryak-Russian pidgin from the same area.

The special case of Arctic pidgins

19

Of the thirty-five contact languages listed here, only a few are actually spoken above the sixtieth parallel, and we cannot be sure which of them were, or are, stable linguistic systems, or simply reported xenolectal ("foreigner-talk") varieties. Bakker rightly states that "[i]t is often hard to distinguish pidgins, which should involve larger numbers of people in contact, and an individual's imperfect learning of a second language" (Bakker 1988: 11). It would be useful to incorporate into the discussion the terms "jargon" and "pidgin" as distinguished by Samarin (1986: 23), viz. A jargon is an unsystematic form of speech that characterizes either a given speaker or a group of speakers, or both; a pidgin is a stabilized form of speech, the consequence of pidginization in a language contact situation. A pidgin can therefore be the object or goal of one's attempt at learning to communicate; a jargon, by contrast, does not provide such a grammatical model.

There is a tendency, especially among nonspecialists, to use the terms "pidgin" and "creole" rather loosely to refer to any mixed or contact variety; there is, for example, Gonzalez' 1967 article on a southwestern US Spanish sociolect which he entitled "Pachuco: The birth of a creole language", and contact varieties of Romani are increasingly being referred to as "creolized Romani" (although the term "Pararomani" has been more recently introduced: Bakker — Cortiade 1991). Even the term "jargon" is used differently by the linguist; Chinook Jargon for example, does not meet Samarin's definition - an argument perhaps for instituting its alternative name, Wawa. The term "creole" when applied to people is similarly vague; in New Orleans, for example, some white Franco-Louisianans and some Afro-Louisianans both claim the label, while denying the legitimacy of the others' use of it. The Copper Island population in the northern Pacific refers to itself as "creole", while the mixed GermanHungarian population in central Europe is also called "creoles", according to Särosi (1971: 81). A second widespread misassumption also, though not exclusively, attributable to the nonspecialist is that the linguistic result of the contact of two languages will produce a pidgin. Thus Schultze (1933: 418) regarded pidgins as being "composed of corrupt fragments of at least, as a rule, two languages", and Black English specialist Geneva Smitherman (1980: 32) stated that "a pidgin is a mixture of two languages". In these definitions we find a third and a fourth misassumption, viz., that pidgins are "corruptions" of languages, and that they are "mixtures" of languages. Languages can certainly be corrupted, and they can certainly be mixed;

20

Ian

Hancock

a sample of the latter, an English-Spanish mixture, is provided as an appendix by way of example. But neither process is essential to the formation of a pidgin. "Corruption", of course, is a subjective term, meaning different things to different people, while language mixing is a universal characteristic of language contact, and therefore has no value as a criterion of distinctiveness. The lexicon of English is extensively "mixed", so much so in fact, that it retains little more than a quarter of its native word stock. Such "mixed" systems as "Spanglish", "Yinglish", "Finglish", "Lunfardo", "Cocoliche", and so on do not count as "pidgins" in the conventional sense; nor, probably, do most of those included on the accompanying map. Neither are such diglossic systems as Shelta or Angloromani or Scandoromani pidgins or Creoles. In the case of Russenorsk, however, Arnbjörndottir and Smith (1985) call that language the result of "the straightforward process of two-language pidginization" and conclude that "just two languages can suffice for the formation of a pidgin". Obviously, statements of this type cry out for a more rigorous definition of pidgin and pidginization. To start with, and social factors for the moment aside, fundamental to pidginization are, following Humboldt (1836) and Hymes (1971), the processes of reduction of inner form (phonology, morphology) and the expansion of outer form (syntax, lexicon), innovation and restructuring, and structural stabilization. The Bloomfieldian definition, variations of which are still the most widely cited, states that the nativization of a pidgin is a process called creolization, and that a nativized pidgin is a Creole; but it has been argued, by, e. g., Valdman (1977), that a Creole need not pass through a pidgin stage, and it might be argued that stabilization is equally as legitimate a factor in defining a Creole as is nativization. The notion that any kind of language contact qualifies a speech variety as a pidgin is traceable to nineteenth-century scholars such as Schuchardt, Coelho, and others who dealt with contact phenomena in a very broad way. Schuchardt did come gradually to acknowledge that there are different kinds of contact situation and that they yielded different linguistic outcomes; but it was Reinecke, in his 1937 doctoral dissertation, who identified and listed them. Of his ten different kinds of "contact" or "marginal"' languages (summarized in Hancock 1990), however, only the first, his "plantation Creoles", would qualify today as pidgins or Creoles by the criteria being discussed here. While all of the categories Reinecke established involved reduction, i. e., the loss, to a greater or lesser extent, of surface rules and narrower

The special case of Arctic pidgins

21

lexical distinctions (and compensationally, by the expansion of syntactic function and the semantic range of individual lexical items), only his "plantation Creole" category involves actual innovation and the creation of "new" or restructured grammars. I refer to a "new" grammar as that which has no model in any of the source languages. For example, it can be argued that both the past and the future constructions in, e. g., Sierra Leone Krio or in Sranan are wholly English in their derivational histories: (1)

English: They are going to walk Krio: Dsn go waka Sranan: Den 'o waka ('ο historically < go)

In English, both 6e-support ("are") and -ing are derived from "go" by intermediate transformations; likewise, the infinitivizer "to" is derived by a series of transformations from "walk". Similarly, a case transformation is necessary to generate subject "they". At the deepest level, the subject is *them go them walk. None of these English transformations was transferred with the lexical items into (or is even possible in) Krio or Sranan grammar, although all three languages share the identical noun phrase deletion transformation, resulting in their respective surface forms. (2)

THEM they BE they are they are

GO -ING going going

THEM they they 0

WALK walk walk to walk

The past constructions (Krio dm bin waka, Sranan den ben waka) can likewise be similarly derived from English "they have been walking". The fact that the semantic functions of the English sentences do not match those of the Krio and Sranan sentences does not argue against this; shift of semantic function is a fundamental characteristic distinguishing Creoles from their metropolitan congeners. Similar possibilities are evident in Gombo (Louisiana Creole French), cf. y'a mase (< eux va marcher) 'they will walk', ye te mase (< eux itait [en train de] marcher) 'they walked'. To recapitulate, the argument is being presented that forms such as these do not involve restructuring or innovation but merely reduction, a property common to all types of contact language, differing only in de-

22

Ian Hancock

gree from the way in which "he go", for example, differs as a reduced development from "he goes". We are dealing with the nontransferrence of transformational rules, whether just one or several. On the other hand, both Krio and Sranan, as well as Gombo, form their past-conditional constructions by combining the past and future markers with the verb, thus Krio dm bin go waka, Sranan den ben 'o waka, Gombo ye t'a mase (t'a < te va). This now is an example of an innovative generative rule, since it has no exterior model in English or French, nor is it a caique upon any other language. It has been generated solely by using the internal resources of Creole grammar, and is therefore an example of true creolization. Similar examples of true creolization would include for example the development of the w/z-adverbs as nounbased complexes, e. g., "what place" for where, "what time" for when, "what person" for who, "what fashion" for how and so on. English provides only the morphemes, not the words (lexemes). Also to be considered as innovative pidginization or creolization is the stabilization of syntactic patterns not part of the grammars of the source languages, thus in contact language number 27, Chinese Russian Pidgin tvaya shenma chifan iu? 'what are you eating' (lit. 'you what eat' + Q) corresponds neither to Russian nor Chinese syntax, cf. Russian cto vy kusaete 'what you eat', and Chinese ni ch'ih shen-me? 'you eat what'. Regarding innovation as one fundamental criterion for definition, we should then examine such systems as Russenorsk, Chinese Russian Pidgin, Chinook Wawa, Pidgin Inuit, etc., to determine the extent to which their grammatical rules are the result of the reduction, or freezing, of rules or structures in any of the source languages, and the extent to which they are structural innovations. On this basis, it should be possible to set up a typologically determined categorization of contact languages. Essential also to the processes of pidginization and creolization are the social factors, and these should be acknowledged in determining this typology. Whether, for instance, the emerging contact language acquires an independent, stabilizing existence, and becomes a target distinct from any of the other languages present in the matrix, or whether it remains, in Samarin's terms, simply a jargon. The structural dimension of this approach incorporates elements of the componential hypothesis, which I have discussed elsewhere (see in particular Hancock 1986, 1992b, 1992c). Briefly, it attempts to account for differences among related Creoles not solely by different rates of metropolitanization, but by different componential ratios in the formative matrix of the individual Creole. The components in the case of the Atlantic anglophone Creoles, for examples,

The spec ial case of Arctic pidgins

23

were varieties of English, African languages, and the dialects of Guinea Coast Creole English. These were differently represented in different places, and influenced each other under different circumstances. In the case of Chinook Wawa, the components were Chinukan, Salish, Nootka, Kwakiutl, French, English, and one or two other languages (see, e. g., Drechsel - Makuakäne 1982); in the case of Russeneorsk, they were Russian, Norwegian, Low German, Dutch, Sami, and some others. The componential approach also examines the lexicosemantic, demographic, topographic, and sociohistorical aspects of the formation of the individual languages. It is these latter which are of more relevance in examining those languages being discussed in the present study. There have been several surveys codifying the salient features of pidgins and Creoles, beginning with Taylor (1971: 294—295, but see also Bickerton 1981, and especially Holm 1987: 14), and attempts have been made to set up scales which rank related Creoles along a continuum of "creoleness". With the Atlantic anglophone Creoles we are dealing with a large number of distinct, but historically related, languages, which is not the case with Chinook Wawa, Pidgin Inuit, Chinese Russian Pidgin, or Russenorsk. Nevertheless, the same principles apply, viz., do they possess more or fewer of the characteristics of pidginization, and at what point might they be better classified simply as jargonized ("interlanguage", "xenolectal", or "foreigner-talk") varieties of the metropolitan language. When surface morphology (the inner form) of any of the donor languages is lost in the pidgin, a large part of the internal grammatical relationship is lost also, and must be compensated for. This loss happens in at least two ways: transmission of the rules generating them can be blocked entirely, as in the examples given above from Krio, Sranan, and Gombo, or else the forms are transmitted but not the grammatical information they carry. This is exemplifed especially well in Stefänsson's examples from (contact language number 4) Herschel Island Pidgin Inuit where, for example, the word for "hungry" means in the source language "I am hungry" (Stefänsson 1909: 218), so that "he is hungry" in the pidgin sounds like "he I am hungry" in Inuit. Similarly, in language number 1, Chinook Wawa, the personal pronouns, which in that language remain the same whether subject, oblique, or possessive, derive in fact from emphatic relative forms in ethnic Chinook, so that maika can mean "I", "me", or "my" in Wawa, but only "it is I who ..." in the lexifier language. In language number 27, Chinese Russian Pidgin, xocu means only 'want', but the Russian source form means Ί want'. This would have to be maja xocu in Pidgin (Jabloriska 1957 [1969: 143]).

24

(3)

Ian

Hancock

ChWmaika Ί , me, my', ECh 'it is I who...' CRP xocu 'want', RUS Ί want' CRP maja xocu Ί want', tvaja xocu 'you want', etc. ChW maika iätawa 'you go', maika na iätawa? 'are you going? CRP tvaja xazi 'you go', tvaja xazi iu 'are you going? RN tvoja spaserom po stova 'you go home' R N tvoja spaserom po stova li 'are you going home?' RN mo ja po skaffom Ί am eating' R N moja po moja stova Ί am at my house' ChW naika miiait kaba haus Ί am at home' ChW naika miiait makmak Ί am eating'

Compensation for this loss also occurs in two ways. Either new syntactic rules are generated, or new functions are assigned to old morphemes. Usually, a pidgin makes use of both. For example, one widespread innovative feature in pidgin languages is the insertion of an interrogative marker, thus in Chinook Wawa maika iätawa 'you go', maika na iätawa? 'are you going?', in Chinese Russian Pidgin tvaja xazi 'you go', tvaja xazi iu? 'are you going?', Russenorsk tvoja spaserom po stova 'you go home', tvoja spaserom po stova li? 'are you going home?'. Aspect to indicate progressive or ongoing action can be innovatively incorporated into pidgin grammar, as with the Russenorsk use of po (moja po skaffom Ί am eating'), which as a universal feature of the pidginization process appears to be related to the locative construction. An example from Russenorsk is moja po moja stova 'I'm at my house', with which the Chinook Wawa equivalents with miiait may be compared: naika miiait kaba haus 'I'm at home', naika miiait makmak 'I'm eating'. Another universal feature of pidgin languages is the innovative expansion of the lexicon. There are many processes accomplishing this (discussed in Hancock 1980), and while they are not unique to pidgins and Creoles, they are much more commonly found in those languages. Incoined forms (i. e., the creation of new words from already existing morphemes in innovative combinations) include from Russenorsk kua-skjorta 'hide', lit. 'cow' + 'shirt' (cf. Russian skura, Norwegian dyrehud, both 'hide'); from Chinese Russian Pidgin cuska-mjasa 'pork', lit. 'pig' + 'meat' (cf. Russian svinina, Chinese chu-jou, both 'pork'), or tolkai-tolkai 'sled', lit. 'push-push' (cf. Russian sani, Chinese shw-ii, both 'sled'). From Chinook Wawa are yuiqat-kwalan 'rabbit', lit. 'long' + 'ears', hdm-oputs 'skunk', lit. 'stink' + 'tail', and hayas-hulhul 'rat', lit. 'big' + 'mouse' (in ethnic Chinook, these are ske'epxoa, opanpan and qalapas respectively).

The special case of Arctic pidgins

25

From Pidgin Inuit we have anuni-anyanini 'storm', lit. 'wind' + 'big', igni-ravik 'oven', lit. Tire' + 'place', and ilwane-kamik 'sock', lit. 'inside' + 'boots' (in Inuit, these words are perksertok, igak, and pinnerak respectively). Lastly, a comment bears making upon the oft-repeated statement that speakers of pidgin languages are frequently under the impression that they are speaking the other person's language or, put another way, that the pidgin is itself the actual metropolitan language. Thus for Russenorsk, Brun (1878) claimed that "the interesting thing about such conversation [i. e., between Russians and Norwegians in Russenorsk] is that both partners believe they are speaking each other's language", while Stefänsson (1909: 217) writes of a government publication "of a book said to be on the Eskimo language, but which is in reality a study in ship's trade jargon". Such statements may hold true for some individuals, or from one side of the contact community, but it is hard to imagine that Native Americans believed Chinook Wawa to be the Europeans' mother tongue, for example, and it must have been clear to all involved in whatever situation, that members of the other community spoke differently, and unintelligibly, to each other when compared with how they spoke to their opposite numbers. To return to William Samarin's distinction, viz., that a pidgin can be a discrete linguistic goal in itself while a jargon is perceived by its speakers to be an imperfect attempt to speak an established, existing metropolitan language, it might be suggested that this awareness on the part of the speakers is also an important factor in the social definition of a pidgin. To conclude, one should accept that not all of the languages listed in the present survey qualify as pidgins, either by linguistic or by social criteria, and that not all of those which meet any such criteria do so equally. It is a mistake to typologize pidgins too rigidly, or to view them as constituting a monolithic category; rather, each contact language should be regarded componentially, in terms of its having incorporated more or fewer features of pidginization. The process of pidginization involves reduction through loss, in different degrees, of lexicon, phonology, and morphology in the metropolitan or source languages, and the compensational generation of innovative rules, especially syntactic and semantic rules, which tend to stabilize in the contact situation and which themselves come to constitute the learner's model or target, and which new system is then perceived to be such a target by its learners in the contact situation.

26

Ian Hancock

Appendix Sample of a mixed

language

The following is a lexically mixed (English and Spanish) register, consciously illustrating the speech of English-dominant Hispanic Americans particularly in Texas, New Mexico, Arizona, and California. Such varieties have been referred to as pidgins or Creoles (cf., for example, Gonzalez 1967), but are representative of neither. 'Tis the night before Christmas, and all through the casa Not a creature is stirring; Caramba? Que pasa? The stockings are hanging con mucho cuidado In hopes that Saint Nicholas will feel obligado To leave a few cosas aqui and alii For chico y chica (y something por mil) Los ninos are snuggled all safe in their camas Some in camisas y some in pajamas Their little cabezas are full of good things, Todos esperan que Santa will bring! Santa is down at the corner saloon (Muy borracho since mid-afternoon!) Mama is sitting beside the ventana Shining her rolling-pin para manana When Santa will come en un manner extrano Lit up like the star on the mountain, cantando Y mama lo manda to bed with a right Merry Christmas a todos Y a todos good night!

'house' 'what's happening?' 'with much care' 'obliged' 'things here; there' 'boy and girl; and for me' 'the children; beds' 'shirts; and' 'heads' 'all waiting for what' 'very drunk' 'window' 'for tomorrow' 'in a; strange' 'singing' 'and mama will send him' 'to all' 'and to all'

Note 1. This paper was originally presented at the Ninth International Troms0 Symposium on Language: Arctic Pidgins, Troms0, 4 - 6 June 1992.

The special case of Arctic pidgins

27

References Arnbjörnsdottir, Birna - Woody Smith 1985 Russenorsk. [Unpublished term paper, graduate creole seminar, University of Texas at Austin.] Bakker, Peter 1988 "Trade languages in the Straits of Belle Isle." Paper read at the Labrador Conference on Crosscultural Contact in the Strait of Belle Isle. 1989 "A French-Icelandic nautical pidgin", Journal of Pidgin and Creole Languages 4: 287-289. Bakker, Peter - Marcel Cortiade (eds.) 1991 In the margin of Romani: Gypsy languages in contact. (Studies in Language Contact I. University of Amsterdam Institute for General Linguistics Publication 58.) Amsterdam: University of Amsterdam. Bickerton, Derek 1981 Roots of language. Ann Arbor: Karoma. Bogoras, Waldemar 1922 "Chuckchee", in: Handbook of American Indian languages. Vol. 2. Washington: Bureau of American Ethnology, 631-903. Broch, Ingvild - Ernst Häkon Jahr 1984a "Russenorsk: A new look at the Russo-Norwegian pidgin in Northern Norway", in: Ureland - Clarkson (eds.), 21-65. 1984b Russenorsk - et pidginspräk i Norge [Russenorsk - a pidgin language in Norway] (Tromso Studies in Linguistics III). Second revised edition. Oslo: Novus Press. Brown, Arthur (ed.) 1980 The Ann Arbor Black English case. Gainsville: John Dewey. Brun, A. W. S. 1878 "Skildringer fra den norske Kyst" [Portrayals from the Norwegian coast]. Nor og Syd No. 11. Trondheim. Bure, Kristjan n. d. Greenland. Ringkjobing: The Royal Danish Ministry for Foreign Affairs. Dali, W. H. 1870 Alaska and its resources. Boston. Drechsel, Emanuel - H. Makuakäne 1982 "Hawaiian loanwords in two native American pidgins", International Journal of American Linguistics 48: 460-467. Gonzalez, R. J. 1967 "Pachuco: The birth of a Creole language", Arizona Quarterly 33: 343-356. Hancock, Ian 1980 "Lexical expansion in Creole languages", in: Valdman - Highfield (eds.), 6 3 88.

1986 1990 1992a

"The domestic hypothesis, diffusion and componentiality", in: Muysken Smith (eds.), 71-102. "Creolization and language change", in: Polome (ed.), 507-525. "The social and linguistic development of Scandoromani', in: Jahr (ed.), 3 7 52.

28

Ian Hancock 1992b 1992c

"The African component in Creole genesis", in: Mufwene (ed.), 84-102. "Creole language provenance and the African component", in: Mufwene (ed.), 182-191.

Holm, John 1987 "African substratal influence on the Atlantic Creole languages", in: Maurer - Stolz (eds.), 11-26. Hualde, Jose Ignacio 1991 "Icelandic Basque pidgin", Journal of Basque Studies in America 5: 4 1 - 5 9 . Humboldt, Wilhelm von 1836 Über die Verschiedenheit des menschlichen Sprachbaues und ihren Einfluss auf die geistige Entwicklung des Menschengeschlechts. Berlin: Königliche Akademie der Wissenschaften. Hymes, Dell (ed.) 1971 Pidginization and creolization of languages. Cambridge: Cambridge University Press. Iversen, Ragnvald 1950 The secret languages in Norway. Vol. 3. Oslo: Dybwad. Jabkmska, Antonina 1957 "J^zyk mieszany chmsko-rosyjski w Mandzurii", Przeglqd Orientalistyczny 2: 157-168. [1969] [translated by Anatole Lyovin as "The Sino-Russian mixed language of Manchuria", Working Papers in Linguistics 3: 135 — 164, University of Hawaii.] Jahr, Ernst Hakon (ed.) 1992 Language contact. Theoretical and empirical studies. Berlin: Mouton de Gruyter. Jochelson, Waldemar 1928 Peoples of Asiatic Russia. Washington: American Museum of Natural History. Johnson, Samuel V. 1978 Chinook Jargon: A computer-assisted analysis of variation in an American Indian pidgin. [Unpublished Ph.D. dissertation, University of Kansas, Lawrence.] Lunden, Siri Sverdrup 1978 "Tracing the ancestry of Russenorsk", Slavia Orientalis 27: 213—217. Maurer, Philippe - Thomas Stolz (eds.) 1987 Varia creolica. Bochum: Brockmeyer Verlag. Meisel, Jürgen (ed.) 1977 Langues en contact. Tübingen: Gunter Narr. Mufwene, Salikoko (ed.) 1992 Africanisms in Afro-American language varieties. Athens: University of Georgia Press. Muysken, Peter — Norval Smith (eds.) 1986 Substrata vs. universals in Creole genesis. Amsterdam: John Benjamins. Neumann, Günther 1966 "Zur chinesisch-russischen Behelfssprache von Kjachta", Die Sprache 12: 237-251. Nielsen, Frederik n. d. "Greenland culture", in: Bure (ed.), 67-88. Polome, Edgar C. (ed.) 1990 Research guide on language change. Berlin: Mouton de Gruyter.

The special case of Arctic pidgins

29

Reinecke, John E. 1937 Marginal languages: A sociological survey of the Creole languages and trade jargons. [Unpublished Ph.D. dissertation, Yale University.] Samarin, William J. 1986 "Chinook Jargon and pidgin historiography", Canadian Journal of Anthropology 5: 2 3 - 3 4 . Särosi, Bälint 1971 Gypsy music. Budapest: Corvina Press. Schultze, Ernst 1933 "Die Sklaven- und Dienersprachen", Sociologus 9: 377-418. Shaw, George C. 1909

The Chinook Jargon and how to use it: A complete and exhaustive lexicon of the oldest trade language of the American continent. Seattle: Rainier.

Silverstein, Michael 1972 "Chinook Jargon: Language contact and the problem of multi-level generative systems", Language 48: 378-406, 596-625. Sizaire, Pierre 1976 "Le parier maritimo-breton", in: Le parier matelot, Paris: Editions Maritimes et d'Outre-Mer, 121-126. Smitherman, Geneva 1980 "Talkin' and testifyin' on Ann Arbor's Green Road", in: Brown (ed.), 2 6 51. Stefänsson, Vilhjälmur 1909 "The Eskimo trade jargon of Herschel Island", American Anthropologist 11: 217-232. Taylor, Douglas 1971 "Grammatical and lexical affinities of Creoles", in: Hymes (ed.), 293—298. Thomas, Edward Harper 1935

Chinook: A history and dictionary

of the northwest coast trade jargon.

Port-

land: Binfords and Mort. Thomason, Sara Grey 1983 "Chinook Jargon in areal and historical context", Language 59: 820-870. Ureland, P. Sture - Iain Clarkson (eds.) 1984 Scandinavian language contacts. Cambridge: Cambridge University Press. Valdman, Albert 1977 "Creolisation sans pidgin; le systeme des determinants du nom dans les parlers franco-creoles", in: Meisel (ed.), 105-136. Valdman, Albert - Arnold Highfield (eds.) 1980

Theoretical orientations in Creole studies. New York: Academic Press.

VarduP, I. F. - Vladimir Belikov I. 1987

Vozniknovenie

i funkcionirovanie

kontaktnyx

jazykov:

materialy

rabocego

sovescanija [Emergence and functioning of contact languages: Material from a workshop], Moscow: Akademija Nauk SSSR, Institut vostokovedenija. Whymper, Frederick 1868

Travel and adventure in the territory of Alaska. London: J o h n Murray.

Xelimskij, Evgenij A. 1987 "Russkij govorka mesto kazat' budem (Tajmyrskij Pidzin)" [I shall say (it) in the Russian speech (Taymir Pidgin)], in: Vardul' — Belikov (eds.), 3 4 - 4 1 .

Siberia

Language contact in northeastern Siberia (Chukotka and Kamchatka) Bernard Comrie

Introduction The aim of this article is to investigate the kinds of language contact that have characterized northeastern Siberia, more specifically the Chukotka and Kamchatka peninsulas. 1 It is thus not directly concerned with Arctic pidgins — for pidgins in this area, reference should be made to the contribution to this volume by Willem de Reuse - but it does discuss the kinds of language contact that are a prerequisite to the development of pidgins. The main indigenous languages of Chukotka and Kamchatka are the Chukotko-Kamchatkan languages, generally reckoned to be five in number: Chukchi, Koryak, Kerek, Alutor, and Kamchadal (Itelmen). Of these, the first four are very close to one another, while Kamchadal is highly divergent. "Kamchadal" originally covered a group of probably three related languages or highly divergent dialects, but of these only Western Kamchadal has survived to be the subject of intensive linguistic investigation. Of the languages of the Chukotko-Kamchatkan family, discussion will be confined, for present purposes, to Chukchi and Western Kamchadal; most of what is said of Chukchi would also apply to Koryak, probably also to Kerek and Alutor, with the exception that these last two underwent no development as written languages in the Soviet period. In addition, a further indigenous language (or, more accurately, group of languages) will play an important role in what follows, namely Siberian Yupik, a member of the Eskimo branch of the Eskimo-Aleut language family. Siberian Yupik represents an immigration of Yupik Eskimo into Siberia, and the language is currently spoken both on the Russian side of the Bering Strait and on St. Lawrence Island, administratively part of the state of Alaska in the USA. 2 The main nonindigenous language relevant to language contact in this area is Russian, although English also plays a certain role, even in speech varieties of areas under Russian administration.

34

Bernard Comrie

Three contact situations will be discussed below: Chukchi and Russian (and also English), Central Siberian Yupik and Chukchi, and Western Kamchadal and Russian. It will be observed that two of these contact situations involve a nonindigenous language, while one involves two indigenous languages. This distinction, however, is not the most salient classificatory feature of these language-contact situations. Rather, the Chukchi-Russian (-English) contacts stand out as an instance of superficial lexical borrowing, in the traditional sense of this term, while those involving Central Siberian Yupik-Chukchi and Western Kamchadal Russian attest to a more intimate relationship between the two languages involved, more akin to language mixing or language shift. These terms - borrowing, language shift, language mixing - are used essentially in the senses in which they are introduced in Thomason and Kaufman (1988). Borrowing in its narrow sense refers to a situation where one language borrows, at least initially, only lexical items from another, primarily nonbasic lexical items, indeed in the first instance usually only lexical items to name new objects or concepts borrowed from the other language's speakers. Borrowing can be recognized by the prevalence of nouns among the borrowed elements, in the most superficial instances of borrowing to the exclusion of all other loans. Language mixing, by contrast, presupposes a fairly detailed knowledge of both languages by those who initially carry out the mixing, with the result that much less salient parts of one language's structure are incorporated into the other. As such, language mixing presupposes a closer social contact between speakers of the two languages. Language shift occurs where speakers of a language abandon their language in favor of another language, carrying over certain features of their original language into the new language; since these speakers are by definition familiar with the first language, they can carry over features that would require an intimate knowledge of that language.

1. Early Chukchi contacts with Russian (and English) By "early" contact I mean contact that took place before the development of Chukchi as a written language, which took place primarily in the period from 1930. As part of the process of developing Chukchi as a written language, a substantial volume of vocabulary was borrowed from Russian, written in exactly the same way as it is written in Russian (even to the extent of using cyrillic letters that form part of the Russian alphabet but are not used in indigenous Chukchi words). For the early period,

Language contact in northeastern

Siberia (Chukotka

and Kamchatka)

35

an invaluable source is Bogoras (1922), based on the author's own life among the Chukchi (and other Chukotko-Kamchatkan peoples) in the late nineteenth and early twentieth centuries, in particular during expeditions in 1895-1897 and 1900-1901. By this time, there were already well developed contacts between (at least some) Chukchi and local Russian seafarers and traders, so that Chukchi had had to come to terms with naming various foreign concepts. Bogoras notes that the usual Chukchi response at this period was to create new lexical items from indigenous lexical resources. Chukchi kelikeli-kel keli-t?ul kale-tko-ra-n tin-hle-t tin-uqqem tin-puvtdn rirje-nerj

translation 'write' 'drawing; writing; letter, book' 'paper' carve-piece carve-HAB-house-ABS 'school' '(eye) glasses' ice-eye-PL 'bottle' ice-deep:vessel ice-jar 'glass jar' fly-thing 'airplane'

gloss carve carving

The verb stem keli- in Chukchi originally meant (and can still be used to mean) 'carve'. Its semantic range was extended, under contact with literate Russians, to mean 'write'. 3 The next three items in (1) are derivatives of keli-. The same stem keli- can also be used as a nominal stem, referring to the product of the action denoted by the verb stem keli-, for this particular nominal formation, the absolutive singular form of the noun, also used as its citation form, is formed by reduplicating the initial CVC of the stem, a morphophonemic process that is quite widespread in Chukchi. The suffix -t?ul is widely used in Chukchi to mean 'piece, concrete object', so that keli-t?ul has the meaning 'paper', in some varieties of Chukchi also 'money'. Both the extension of the meaning of keli- and these derivatives with these senses are noted by Bogoras and are still current in Chukchi; they are listed, for instance, in Moll - Inenlikej 1957. The next item in (1) shows the use of Chukchi linguistic material to provide equivalents of words reflecting new social conditions, rather than just material objects; it is presumably subsequent to Bogoras's time, at least not being mentioned by him, but is listed by Moll - Inenlikej. The stem keli- appears in its vowel harmony alternant kale-·,4 to this is attached a suffix -tku (vowel harmony variant -tko), which indicates habit-

36

Bernard

Comrie

ual action; to this is attached the element -ra, which appears in independent form as the noun yaraip (stem yam-) 'house'; to which is attached finally the suffix -n, a lexically determined indicator of absolutive case in the singular. This item can thus be glossed more literally as 'house in which one habitually writes', a not unreasonable expression for 'school'. The Chukchi lexical item tin-, literally 'ice', is used in a number of neologisms to mean 'glass', and here we find it combined with ble-t 'eyes' (where -t is the plural absolutive suffix); the sense is thus literally 'ice (or glass) eyes', i. e., '(eye) glasses'. This item is cited both by Bogoras and by Moll and Inenlikej. In addition, Bogoras cites tin-uqqem 'bottle', while Moll and Inenlikej cite tin-puvtan 'glass jar'. The last item is post-Bogoras, though still current in Chukchi: the Chukchi suffix -nerj, attached to a verb stem, indicates an instrument for performing the action denoted by that verb, so that when attached to the stem riqe- 'fly', the result is an instrument for flying, or an airplane. Though rarer, Bogoras notes that Chukchi also borrowed some words directly from Russian. Some examples are given in (2). Chukchi taaq caqar cay (PL cag-td) kone-kon col toroma

Russian tabak saxar caj kon' sol' zdorovo

English 'tobacco' 'sugar' 'tea' 'horse' 'salt' 'hello' 5

In nearly all such instances, however, one finds adaptation of the loan word to the phonological and morphological patterns of Chukchi. Thus, Russian sounds that are nonexistent in Chukchi are replaced by the closest Chukchi sound, as when /s/ is replaced by the palatal affricate, here transcribed c, in caqar and col, or /x/ is replaced by the uvular plosive /q/ in caqar, or the distinctive palatalization of the word-final nasal of Russian kon' is lost, or /b/ is simply dropped in taaq. The noun cay follows the pattern of other Chukchi nouns with final y in selecting the -te variant of the absolutive plural suffix and changing y to g (a voiced velar fricative) before this suffix (as before t more generally). The word for 'horse' has the stem kone-, and its absolutive singular is formed by means of the same kind of initial CVC reduplication noted in (1) for keli-kel, to give kone-kon. Nearly all of the examples cited by Bogoras are nouns, the only exception being the greeting toroma, again a kind of item likely to

Language contact in northeastern Siberia (Chukotka and Kamchatka)

37

be easily assimilated even in superficial contact. Finally, Bogoras also notes items that combine use of native material and borrowing, as in taa2koyrpn 'pipe (for tobacco)'. The first component is the already noted loan taaq, with the regular shift of /q/ to [2] before a consonant. The second component, koyrpn, has the original meaning 'cup' in Chukchi. In addition to the contact with Russian, Chukchi also had direct contacts with English, resulting from direct contacts between speakers of Chukchi and American seafarers, with the result that there are even a few items of English origin, surviving in use to the present day, such as mane-man 'money' (and derived from English money, with the same kind of reduplication as in the case of kone-kon), and krscm-dn, which in Chukchi means simply 'holiday', but has its origin in English Christmas.6 To sum up the contents of this section: Chukchi lexical innovation under foreign influence at this period reflects primarily the creation of terminology to refer to new objects and concepts. The primary means was creation of new lexical items using indigenous patterns of derivational morphology, although there were also some direct loans from Russian (and English), all undergoing necessary nativization in phonology and morphology. It will also have been noted that virtually all the items cited are nouns. All of this is characteristic of borrowing in its narrow sense, as a result of relatively superficial contact between languages.

2. Central Siberian Yupik contacts with Chukchi Contact has been intense between Central Siberian Yupik and Chukchi, with the direction of influence being almost exclusively from Chukchi to Central Siberian Yupik, no doubt reflecting the numerical predominance of speakers of Chukchi. 7 The linguistic results of this contact have been noted by investigators of both the Central Siberian Yupik of Russia (e. g., Menovscikov 1968) and that of St. Lawrence Island (e. g., Jacobson 1977). The presence of such Chukchi loans in the Central Siberian Yupik of St. Lawrence is particularly interesting, since direct contact between St. Lawrence Island and the Siberian mainland was virtually nonexistent during the period of the Cold War (1948-1989), so Chukchi items showing up in the speech of St. Lawrence Islanders are presumably well ingrained in the language. Central Siberian Yupik items are cited below in the phonemic orthography used by Jacobson (1977), except that vowel length, not shown by Menovscikov, is indicated only for items also cited by Jacobson. Some of the lexical items of Chukchi origin found in

38

Bernard

Comrie

Central Siberian Yupik are nouns with concrete reference, such as yaywaali, from Chukchi yeyvel 'orphan', and guygu, from Chukchi guygu-n 'summer dwelling' (in the latter Chukchi item, -n is an absolutive singular suffix). However, Jacobson (1977: 90) notes: The loans from Chukchi are presumably fairly ancient borrowings. They are largely, though not entirely, uninflectable words such as adverbs, conjunctions, interjections, etc. 8

In (3), the first four items are cited by Jacobson (and Menovscikov), while the others are taken from Menovscikov; Badten et al. (1987) confirm that all items are found in the speech of St. Lawrence Islanders: (3)

Chukchi venhgi rdpet lureq enmec viin evdr dtiraq

Central Siberian Yupik wanlegi repall luuraq enmis wini iiwen enraq

English 'all the same' 'even' 'probably' 'already' 'aha' 'if' 'but'

An extended list of over 50 such items of Chukchi origin is given by Menovscikov (1968). Most of them are adverbs or particles and, in the absence of more detailed research, the English glosses should be taken with a certain pinch of salt: translations given by Jacobson do not always correspond to those given (in Russian) by Menovscikov, though this is surely in large part due to the difficulty of translating such idiomatic particles from any language into another (and in the case of Menovscikov as source, the fact that there is not only Menovscikov's translation from Central Siberian Yupik to Russian but also whatever further problems are introduced by my translation from Russian to English). The glosses of Central Siberian Yupik items, moreover, do not always correspond exactly to the glosses given to the corresponding Chukchi items by Moll - Inenlikej (1957), but this could reflect semantic shifts that occurred when the items moved from one language to the other, or that have occurred in one or the other language in the period since these words entered Central Siberian Yupik. Phonetically, all such items are adapted to the phonology of Central Siberian Yupik. While in some cases this involves little or no change, there are other instances of clear adaptation, perhaps noticeably to the

Language contact in northeastern Siberia (Chukotka and Kamchatka)

39

Central Siberian Yupik three-vowel system of i, u, a, thus losing Chukchi e and o. (In the orthography used on St. Lawrence Island, e represents shwa.) It should be noted that, prior to the appearance of these items of Chukchi, Central Siberian Yupik presumably, like most other varieties of Eskimo, was poor in uninflected words of this kind. In all varieties of Eskimo the primary devices corresponding to particles and conjunctions are verbal affixes or clitics, so that the absorption of these words from Chukchi has given rise to a typological innovation, or at least a statistical shift in type, in this variety of Eskimo. The social conditions that gave rise to this kind of borrowing - primarily restricted to particles - must have been very different from that characterizing the contact of Chukchi and Russian. In the case of Chukchi—Russian contact, one society (Chukchi) was essentially borrowing a range of objects and concepts from another society and forced into the position of having to name them. In the case of Central Siberian Yupik and Chukchi, the two cultures were presumably technologically at very similar levels, though with Chukchi clearly dominant numerically and possibly characterized by greater "prestige" in this encounter. The kinds of lexical items that were introduced from Chukchi into Central Siberian Yupik suggest a body of people familiar with both languages, and indeed, in the absence of direct historical evidence, a number of possibilities exist for how these Chukchi items found their way into Central Siberian Yupik; possibly some combination of these possibilities was at play: i. Speakers of Central Siberian Yupik who acquired a good command of Chukchi may simply have liked these newfangled forms they encountered in Chukchi, perhaps influenced by greater prestige of Chukchi, and have started using them in their own language, their usage subsequently spreading to the community as a whole, even to those speakers of Central Siberian Yupik who did not speak Chukchi. ii. Speakers of Chukchi may have learned Central Siberian Yupik but have retained these items from their native Chukchi, their usage subsequently being imitated by the Central Siberian Yupik community as a whole (again, with the prestige of Chukchi perhaps playing a role); the fact that the closest Central Siberian Yupik functional equivalents would usually have been affixes and clitics, more difficult for the foreign learner to isolate, may have been a factor favoring this development.

40

Bernard Comrie

iii. True bilinguals, perhaps the children of one Central Siberian Yupik and one Chukchi parent, may have created this "mixed" language (the quotes being used to indicate that the result is clearly basically Central Siberian Yupik, rather than a roughly equal mixture of the two linguistic systems), once again their usage subsequently spreading to the community as a whole.

3. Western Kamchadal (Western Itelmen) contacts with Russian Kamchadal, even in the narrower sense of Western Kamchadal, is now a virtually extinct language, which has been heavily eroded by Russian, not only in the sense of Western Kamchadal speakers shifting to Russian but also in the wholesale incorporation of Russian elements into Western Kamchadal. In Bogoras's time, however, what struck him was not, as in the case of Chukchi, the existence of nouns borrowed from Russian, but rather the fact that "most of the Kamchadal conjunctions have been replaced by the Russian (local) forms" (Bogoras 1922: 881).9 The items listed in (4) are among those identified by Bogoras in this connection; in the lefthand column, Bogoras's orthography is retained. (4)

Western Kamchadal Russian English i i 'and' dai dai 'and' je ze 'but' tolko tol'ko 'only' dotopera do topera 'so far, till now' potom potom 'then, subsequently'

It should be noted that what surprised Bogoras was not so much the fact that Western Kamchadal had borrowed conjunctions from Russian, but rather that this appeared to him to be the most salient kind of borrowing. Indeed, Western Kamchadal had not only borrowed Russian conjunctions, but had in the process virtually lost its native inventory of conjunctions. (The other Chukotko-Kamchatkan languages have a well defined set of coordinating and subordinating conjunctions, and there is no reason to suppose that Western Kamchadal was markedly different before its contact with Russian; a few indigenous Western Kamchadal conjunctions were retained.)

Language contact in northeastern

Siberia (Chukotka

and Kamchatka)

41

More recent studies of Western Kamchadal, in particular Volodin (1976), confirm the basis of Bogoras's surprise. The situation is particularly clear in traditional stories told by older speakers of Western Kamchadal. Their texts are virtually free of borrowed Russian nouns (in part, of course, because they are dealing with a traditional subject matter where such lexical items would be unlikely to occur), but are replete with Russian conjunctions and other particles, to the exclusion of Western Kamchadal equivalents. Example (5) is the beginning of a traditional story, from the full text given in Volodin (1976: 395—403). The narrator is Varvara Iosifovna Ponomareva (1909—1973), described by Volodin as the best of his informants and as someone who spoke only Western Kamchadal into her teens. Thus both the speaker and the topic are such as to suggest the lowest incidence of Russian loans. Yet the first 45 words of the text contain no fewer than seven Russian conjunctions and particles. This ratio is by no means atypical for the text as a whole. In (5), these items of Russian origin are indicated by underlining. 10 (5)

ksunlqzukne?n ememqut i smaqewt. theyrlived Ememqut and Sinyangewt. 'Once upon a time, there lived Ememqut and Sinyangewt.' lem ksunlqzukne?n sislxan lilixl?in si?rim, also they:lived Sislkhan sister Sirim 'There also lived Sislkhan and his sister Sirim,' qa?m wetatkenka?n not working 'who didn't work,' a si?rim qunix qamzatantalqzuzen qa?m k'enk laq. but Sirim also wanted:marry not anyone took 'but Sirim also wanted to get married but no one would take her.' ememqut klelexlcom klawülknen ememqut kyeniknen Ememqut with:sister they:sat Ememqut said 'Ememqut was sitting with his sister and Ememqut said:' kazza kwarjikas a kdmma kalxe?n tskqzayalcen. you sew and I arrows will:make 'You sew and I will make arrows.' tolko zaq kdmmanke dlckukaq only not at:me look 'Only don't look at me'

42

Bernard

Comrie

i kdmma knanke xe?Ac milckukicen. and I at:you not will:look 'and I won't look at you.' ememqut potom kdmck'elqkuknen Ememqut then got:bored 'Then Ememqut bot bored' i lelexlanke k'iretl klaxlknen. and at:sister sidewards he:looked 'and he glanced sidewards at his sister.' The precise incidence of Russian conjunctions does vary somewhat from text to text, although the basis for this variation is not clear to me. Of the two main dialects of Western Kamchadal, namely Sedanka and Khayryuzovo (called Napana by Volodin [1976: 16-19]), Ponomareva is a speaker of the Khayryuzovo dialect, the one least affected by Russian. As pointed out to me by Willem de Reuse, in the texts collected by Waldemar Jochelson (Vladimir Il'ic Ioxel'son) in 1910-1911 (Worth 1961), Russian conjunctions are absent from the Khayryuzovo texts but present in the Sedanka texts. In the approximately sixty years separating Jochelson's texts from Volodin's, the Khayryuzovo dialect seems to have moved to a degree of Russianization roughly similar to that of the Sedanka dialect sixty years before. As with contact between Central Siberian Yupik and Chukchi, we need to enquire what kinds of social conditions could give rise to this kind of transference between Russian and Western Kamchadal. The Russian speakers with whom the Western Kamchadal would have come into first intensive contact would not have been occasional visitors (like the seafarers with whom the Chukchi had their earliest contacts), but rather settlers, and while in certain respects the Russian speakers would have been culturally dominant, the differences were probably not too great: both Russian speakers and Western Kamchadal speakers were faced with essentially the same problems of survival in a not particularly clement environment, and while the new settlers would have had some material advantages, the indigenous population would certainly have had the advantage of greater experience. Mingling of the two communities through common households was probably widespread. One nonlinguistic, or at least not exclusively linguistic, reflection of the difference between Russian-Western Kamchadal and Russian-Chukchi contacts can be seen in the fact that while the Western Kamchadal have Russian family names, the Chukchi have family names based on their own traditional names.

Language contact in northeastern Siberia (Chukotka and Kamchatka)

43

Linguistic mixing presumably went hand in hand with the other close relations between Western Kamchadal and Russian speakers, with no doubt all of the following processes being to some extent involved: Western Kamchadal speakers with a good knowledge of Russian being attracted by Russian conjunctions and particles and introducing them into their own speech; Russian speakers learning Western Kamchadal but retaining some of their own conjunctions and particles, perhaps because they could find no exact Western Kamchadal equivalents, with subsequent introduction of these particles into the speech of native speakers of Western Kamchadal; bilinguals (perhaps the children of Western Kamchadal and Russian parents) introducing elements of one language into the other, with this mixed language then being adopted by the community as a whole.11 It is interesting that, whatever the origin of the introduction of Russian conjunctions and particles into Western Kamchadal, they were sufficiently integrated into Western Kamchadal that they are so salient in the speech of someone like Varvara Iosifovna Ponomareva, who considered herself a monolingual speaker of Western Kamchadal into her teens.

4. Conclusions As a result in part of studies of pidgin and Creole languages, in part of more wide-ranging studies of language contact like Thomason — Kaufman 1988, our horizons on the nature of language contact have been vastly expanded. The three case studies presented in this article might seem contradictory from a traditional language-contact perspective: we find two cases where languages have borrowed conjunctions and particles widely but without corresponding widespread borrowing of nouns; moreover, the patterns of borrowing do not seem to correlate with traditional conceptions of culturally more and less dominant languages. These different types of language contact must, empirically, be recognized. And while we may never know the precise demographic processes that underlie these differences in the particular case of the languages of northeastern Siberia, such knowledge as we have, supplemented by intelligent speculation, suggests that these differences are by no means arbitrary, but reflect in particular the difference between superficial language contact and language contact involving intimate relations between the speakers of the languages involved.

44

Bernard Comrie

Notes 1. I have benefited not only from comments from other participants at the Tromso workshop but also from comments by Willem de Reuse on an earlier written version. 2. A divergent variety of Siberian Yupik, Sireniki, spoken in the like-named settlement on the Russian side of the Bering Strait, is virtually extinct and will not be further considered here; the same is true of the likewise virtually extinct Inupiaq dialect of Big Diomede island. The rest of Siberian Yupik is divided between two mutually unintelligible varieties, Chaplino/St. Lawrence and Naukan; material here relates to Chaplino/St. Lawrence, hereafter referred to as Central Siberian Yupik. The Tungusic language Even (Lamut) is also spoken in Chukotka and Kamchatka; I have not yet investigated linguistic contacts involving Even. The remarkable contact language involving Aleut and Russian that has developed on Copper (Mednyj) Island, administratively part of Kamchatka, is independent of the contact situations discussed in this article; for discussion of Copper Island Aleut, reference should be made to the contribution to this volume by Evgenij V. Golovko. 3. One might compare the development of Old English writan, originally 'scratch' (cf. German reißen) to mean 'write'; German, by contrast, simply borrowed Latin scrTbere to give Old High German scrTban, modern schreiben. And note that even Latin scrTbere originally meant 'engrave'. 4. In Chukchi, if one morpheme of a word contains one of the dominant vowels e2, o, a, then all recessive vowels (/', u, et) of the word must change to the corresponding dominant vowel. Some morphemes containing only the neutral vowel 3, or indeed containing no vowel at all, are also morphophonemically dominant and trigger this replacement. The vowels ei and e 2 are at least morphophonemically distinct. 5. Glossed by Bogoras as 'how is your health', though in fact the Russian and, presumably, the Chukchi are used as greetings. 6. For a fuller list, see de Reuse (1992). 7. For a more detailed list of loans and account of the social background, see de Reuse (1994). 8. Willem de Reuse points out to me that, from a strictly numerical viewpoint, there are actually more noun stems than particles borrowed from Chukchi into Central Siberian Yupik. However, the proportion of Central Siberian Yupik nouns that is of Chukchi origin is several magnitudes smaller than the proportion of particles of Chukchi origin. Moreover, many of the nouns are concentrated in one area, reindeer herding, presumably an instance of loan words accompanying a borrowed phenomenon. 9. By "local" Bogoras refers to the fact that some of these items, such as topera 'now', are not characteristic of standard Russian (which has teper'), but rather of the varieties spoken by Russian settlers in Kamchatka. 10. As the morphological structure of the indigenous Western Kamchadal words is not at issue here, I have given a word-by-word rather than a morpheme-by-morpheme gloss in the second line. 11. On the plausible assumption that Western Kamchadal, like the other ChukotkoKamchatkan languages, did have indigenous conjunctions, the second process would not exactly have paralleled the contact between Central Siberian Yupik and Chukchi, where Chukchi speakers learning Central Siberian Yupik may well have had difficulty identifying Central Siberian Yupik morphemes corresponding in function to their own

Language contact in northeastern Siberia (Chukotka and Kamchatka)

45

particles. However, conjunctions and particles rarely provide exact matches across languages, even closely related languages, so that Russian speakers may still have been tempted to continue using their own particles, to retain the expressive power of their own language.

References Badten, Linda Womkon - Kaneshiro, Vera Oovi - Jacobson, Steven A. 1987 A dictionary of the St. Lawrence Island!Siberian Yupik Eskimo language. Fairbanks: Alaska Native Language Center, University of Alaska. Bogoras, Waldemar (= Bogoraz, V. G.) 1922 "Chukchee", in: Franz Boas (ed.), Handbook of American Indian languages, Vol. 2. Washington DC: Government Printer, 631—903. Jacobson, Steven A. 1977 A grammatical sketch of Siberian Yupik Eskimo as spoken on St. Lawrence Island, Alaska. Fairbanks: Alaska Native Language Center, University of Alaska. Menovscikov, Georgij Alekseevic 1968 Grammatika jazyka aziatskix eskimosov, c. 2 [Grammar of the language of the Asiatic Eskimos, Part 2.] Leningrad: Nauka. Moll, Tat'jana A. - Petr I. Inenlikej 1957 Cukotsko-russkijslovar'[Chukchi-Russian dictionary.] Leningrad: Ucpedgiz. de Reuse, Willem 1992 English loanwords in the native languages of Chukotka peninsula, Commonwealth of Independent States. [Unpublished manuscript, University of Arizona, Tucson.] 1994 Siberian Yupik Eskimo: The language and its contacts with Chukchi. Salt Lake City: University of Utah Press. Thomason, Sarah Grey — Terrence Kaufman 1988 Language contact, creolization, and genetic linguistics. Berkeley: University of California Press. Volodin, Aleksandr Pavlovic 1976 Itel'menskijjazyk [The Itelmen language.] Leningrad: Nauka. Worth, Dean S. 1961 Kamchadal texts collected by W. Jochelson. The Hague: Mouton.

Chukchi, English, and Eskimo: A survey of jargons in the Chukotka Peninsula area1 Willem J. de Reuse

On the Chukotka peninsula in the Russian Far East, four native languages are presently spoken. They are: three Yupik Eskimo languages Chaplinski or Central Siberia Yupik (CSY), also spoken on St. Lawrence Island, Alaska; Naukanskij (Nauk.); and Sirenikskij (Sir.) - and the Paleosiberian (more precisely Chukotko-Kamchatkan) language Chukchi (Ch.). There is historical and linguistic evidence that these four languages have been influencing one another for centuries. Furthermore, in the second half of the nineteenth century, explorers' ships as well as commercial whaling ships often spent winters on the Chukotka coast. The social and economic relationships between Chukchis, Eskimos, and the mostly English-speaking crews of whaling ships had resulted in the appearance of no less than three different types of trade jargons in Chukotka by the end of the nineteenth century: an Eskimo jargon, a Chukchi jargon, and an English jargon. In this paper, I will provide a survey of the historical (mostly anecdotal) and linguistic evidence for the existence of these three jargons, a survey of their linguistic characteristics and their respective functions, and an account of the socioeconomic and political reasons for their complete disappearance by the middle of this century. Cases of "simplified trade languages" based on Chukchi or Eskimo in the Chukotka area have never before been mentioned in the literature on pidgins and Creoles. This is due to the fact that the evidence for the existence of Chukchi or Eskimo jargons would have to be searched for on the Chukotka coast, and this is an area virtually unexplored by Soviet linguists interested in pidgins and Creoles, and presently (1990) not accessible to Western pidgin and Creole specialists. With the expression "simplified trade language", I am referring to an elementary semicodified linguistic system that is nobody's first language, and has often been called a "trade jargon". A more precise definition of a jargon is given by Silverstein (1972: 622-623), who points out that a difference must be made between a "language community", which implies a shared and essentially identical grammatical system, and a "speech

48

Willem J. de Reuse

community", which does not imply such a shared system, and where communication is made possible only by simplifying or avoiding some of the differences between two distinct native grammatical systems, using a set of universal linguistic principles. Thus, there is a jargon speech community, but there is no jargon language community. I prefer not to use the term "pidgin" since this is often used to refer to more codified and elaborate systems. Well-known examples of such trade languages include Russenorsk, based on Norwegian and Russian, and to a lesser extent on Low German, Sami, and Swedish (Fox 1983: 97-104, Broch - Jahr 1984), and Chinook Jargon, based on Chinook, other Northwest coast Indian languages, French, and English (Silverstein 1972: 378-406, 5 9 6 623). Let me start with a discussion of jargon Chukchi. There is historical evidence that the Chukchi had special language simplifying skills. The Chukchi, especially before the October Revolution, consistently preferred speaking tjieir own language to outsiders; therefore it would have been natural for them to simplify their language somewhat in order to make it easier for the outsiders to learn some of it. In an Arctic expedition (1848-1849), reported on by Hooper (1853: 33), the Chukchi were met in Plover Bay, and in attempting to communicate, "they [the Chukchi] did not learn English nearly so quickly as many of us acquired their tongue". As Chukchi is typologically very different from English, one can have serious doubts as to the nature of the Chukchi acquired so fast by English speakers. The Swedish explorer Nordenskiöld, reporting on the Vega expedition (1878-1879), is more perceptive, and notes that "the difficulty of studying the language was increased, to a not inconsiderable degree, by the Chukchi in their wish to cooperate with us in finding a common speech being so courteous as not to correct, but to adopt the mistakes, in the pronunciation or meaning of words that were made on the Vega" (1882: 369). In the course of time, the natives also learned some words of Swedish, and so an international lingua franca was gradually formed, in which some of the crew members became very proficient. It was very poor in words, and "all grammatical inflections were totally wanting" (Nordenskiöld 1882: 371, 386). Nordquist (1883: 211), who was the linguistic investigator for the same expedition, regrets that he could not obtain more morphological information on the language, since the Chukchi would use the words in the same way they heard them used by the members of the Vega's crew, thereby causing the formation of a jargon. This is all the documentation I have been able to find on jargonized Chukchi.

Chukchi, English, and Eskimo: A survey of jargons in the Chukotka Peninsula area

49

As far as Eskimo is concerned, we are a little bit better off: a form of jargon has been documented in a detailed fashion for North Alaska (Stefänsson 1909: 217-232). Eskimo jargons have also been reported on in an anecdotal fashion for Western Canadian Inuit (de Poncins 1965), for Aivilik (Eastern Canadian) Inuit (Gilder 1881: 304-305), for Labrador Inuit (Dorais 1980, Bakker 1991), and for Greenlandic (Peterson 1979: 113-114). The Eskimos of the Chukotka area certainly had language simplifying skills similar to those of the Chukchi, since the existence of Eskimo-based jargons is recorded for the whaling period (i. e., the second half of the nineteenth century); these jargons will be further discussed below. A preliminary question I would like to ask is whether jargons were used before Euro-American contact for communication between various groups of Eskimo speakers and the Chukchi. It is possible that the appearance of Eskimo-based jargons was directly or indirectly caused by white contact, since the economic advantage recently gained by Eskimos in trading with whalers might have prompted the Chukchi to learn, at least imperfectly, some Eskimo, instead of requiring the Eskimo to learn some Chukchi or at least Chukchi jargon, which seems to have been the traditional pattern. For details on the sociolinguistic and economic relationships between Eskimos and Chukchis in this area, reference should be made to Bernard Comrie's contribution to this volume, and to my fuller account (de Reuse 1994). But considering the prehistoric linguistic diversity and various types of multilingualism involving Chukchi, and the various Eskimo languages (Krupnik - Clenov 1979: 25-26), it is more likely that the above circumstances only contributed to the further diffusion of Eskimo jargons that already existed before white contact. The correct historical scenario is impossible to determine, since the earliest accounts date from white contact. I will now discuss the historical, albeit mainly anecdotal, evidence for the existence of Eskimo jargons in the Russian Far East. It is striking that the early accounts that deal with the Chukchi do not fail to note that Eskimo was also spoken in the area, but was mixed with Chukchi, or contained many Chukchi words (Hooper 1853: 35; Nordenskiöld 1882: 459). In present times, it is not at all obvious, at least to the untrained observer, that there is such a striking Chukchi influence in Central Siberian Yupik. The above observations could mean that in the first half of the nineteenth century, Eskimo of the Russian Far East simply contained a lot more Chukchi loans than it does now. It is also likely that beside their own languages, the Eskimos used an Eskimo-Chukchi jargon to

50

Willem J. de Reuse

converse with the Chukchi and other outsiders, and that it is this jargon that accounts refer to as Eskimo mixed with Chukchi. Dall's account of Alaska Territory is probably the first work to explicitly mention this possibility: "There is reason to believe that there was originally a jargon, containing words of both languages, in use in communication between the men of the two races, as is at present the case near Norton Sound, between the Innuit and the Indians" (1870: 378). Dall also speculated perceptively on the Eskimo language itself: "Communication is constantly kept up between the Chukchees and the Tuski [i. e., Siberian Eskimo], and our knowledge of the language of the latter is still too meager to enable us to say whether among themselves they yet speak a pure Orarian [i. e., Eskimo-Aleut] dialect, but it is probable that it has become, from constant intercourse, permanently mixed with Chukchi words" (1870: 378). We are now certain that the lexical influence of Chukchi on Russian Far East Eskimo languages has become permanent (see Bernard Comrie's contribution to this volume), and we are left with the questions of what this jargon was like and how and why it disappeared, as is implied by Dall's statement "there was originally a jargon". Actually, it is not even certain that all traces of an Eskimo-Chukchi jargon have disappeared. Krupnik and Clenov (1979: 22) note that the Qiwaaghmiit, a Central Siberian Yupik group from Kivak, has close relationships with the neighboring maritime Chukchi settlement of Chechen, and, as a result of intermarriage with them, spoke a mixed EskimoChukchi jargon in the beginning of the twentieth century. 2 The same authors (1979: 23) also report that the Central Siberian Yupik Nunaghmiit group, originally from Nunak, south of Naukan, which had merged by the end of the nineteenth century with the Naukanskij, also spoke a kind of Eskimo-Chukchi jargon. Unfortunately, sources and exemplification are not provided. For linguistic evidence for such jargons, one must turn to a careful scrutiny of the early wordlists. Since there are no connected texts, there is considerable uncertainty as to what they represent. According to Krauss (1975: 45), there are about fifteen wordlists collected on the Siberian mainland before 1900. Of these, I have been able to examine in some detail two, both written by Americans: Dall (1870: 549) and Wells Kelly (1982: 67-72), originally published in 1890. As with all early wordlists, the translation of many words is inaccurate, the spelling is inconsistent and difficult to interpret, and typographical errors are very common. 3

Chukchi, English, and Eskimo: A survey of jargons in the Chukotka

Peninsula area

51

The authority given for Dall's list is "Hall (part)". This must mean that the list is part of a longer vocabulary collected by another author, but I have been unable to find a complete reference to this; Hall is probably Professor Asaph Hall of the 1869 US Eclipse expedition, who according to Dall (1870: 554) collected a Chukchi vocabulary at Plover Bay. Dall's list appears to be from the "Chü'klükmüt", probably the Siklughmiit, a group of Central Siberian Yupik speakers of Yttygran Island, in the Russian Far East. It contains thirty words, twenty-five of which are clearly recognizable Central Siberian Yupik nouns and numerals. The remaining five are clearly from Chukchi: they are the independent personal pronouns goom T , from Ch. gdm (Moll — Inenlikej 1957: 33); goot 'you', from Ch. gat (Moll - Inenlikej (1957: 35); en'kau 'he, she, or it', from Ch. anqen (Moll - Inenlikej 1957: 48) - this is probably a typo for en'kan, the spelling occurring in Hall's Chukchi vocabulary (Dall 1870: 552); and the particles ee 'yes', from Ch. ii (Moll - Inenlikej 1957: 39), and al'eah 'no', from Ch. eth (Moll - Inenlikej 1957: 158). It is remarkable that the nouns and numerals are from Central Siberian Yupik and the pronouns and particles from Chukchi, but of course in such a short list this peculiarity might be due to coincidence. The second, longer list will help elucidate this matter to some degree. Dall (1870: 410) and Stefänsson (1909: 217) note that vocabularies of whalers' and traders' Eskimo jargon have been published as studies of genuine Eskimo. Unfortunately, none of them give references. There is good evidence that Wells and Kelly's English—Eskimo, Eskimo—English vocabulary, first published by the US government in 1890, is one of these works. This booklet actually contains two separate vocabularies, both with English-Eskimo and Eskimo-English sections; the first vocabulary (1982 [1890]: 29—66) is basically North Alaska Inupiaq, and contains approximately 2260 words. It appears to contain some phonologically and morphologically simplified words, but also many inflected forms. On the whole, the words are not nearly as simplified as in Stefänsson's (1909) vocabulary. This is probably due to the fact that Stefänsson's intent was to describe a jargon, whereas Wells and Kelly's was to give a list of actual Eskimo words, so one would expect them to leave out all foreign-looking words. They did leave in some rather suspicious words, such as ab'abah 'to speak' [1982 [1890]: 42), of unclear origin but possibly Hawaiian according to Stefänsson (1909: 22); kok'o 'hard bread' (1982 [1890]: 30), from Hawaiian Pidgin English, further discussed below; koone'a 'wife' (1982 [1890]: 45), from Danish via Greenlandic (Stefänsson 1909: 226). There are also two unassimilated Russian loanwords: chy 'tea' (1982

52

Willem J. de Reuse

[1890]: 40), from Russian caj (but Imaklikskij has the assimilated loan sayuq (Menovscikov 1980: 294), like Alaskan Inupiaq saayu [Larry Kaplan, personal communication); and do'bra 'sufficient' (1982 [1890]: 49), from Russian dobro 'good'. There also appears to be one word of Chukchi origin, tin'oop 'silver-black fox' (1982 [1890]: 34), also occurring in Central Siberian Yupik and Sirenikskij. If the first vocabulary of Wells and Kelly is, at least partially, a jargon, it is reasonable to assume that the second vocabulary [1982 [1890]: 67— 72) entitled "Eskimos in Siberia, from Cape Behring to East Cape", contains a similar type of Eskimo. This second vocabulary, much shorter than the Inupiaq one, contains 354 entries in the English-Eskimo section, to which must be added eleven entries from the Eskimo-English section which are not found in the English-Eskimo section. Of this total of 365 words, I have been able to identify the origin of 281 with some amount of certainty; 84 thus remain obscure, but will probably turn out to be from Chukchi. Of the words with a determined origin, 185 are from Central Siberian Yupik, 76 are from Chukchi, of which 16 are also Chukchi loans into Central Siberian Yupik, ten are from Sirenikskij, and six are probably from Inupiaq. One might wonder whether there are also Naukanskij words in this list. Due to the fact that Naukanskij is generally phonetically close to Central Siberian Yupik, and to the atrocious spelling of the words, it is often impossible to tell whether a word is Central Siberian Yupik or Naukanskij. In most cases of potential ambiguity, such as stoke 'fingernail' (1982 [1890]: 68), compare Central Siberian Yupik estuk (Badten et al. 1987: 60) and Nauk. situk (Menovscikov 1975: 474), the spelling points to a Central Siberian Yupik word. Words that are probably Naukanskij are ahmeta'too (1982 [1890]: 67), ahme'ta (1982 [1980]: 69) 'ermine', Nauk. amitatuk (Menovscikov 1975: 394), but it could also be Sirenikskij amitatugg (Menovscikov 1964: 209) - Central Siberian Yupik has amikluk (Badten etal. 1987: 31); imp'net 'cliff' (1982 [1890]: 67), Nauk. dpnaq (Menovscikov 1975: 506), rather than Central Siberian Yupik pennaq (Badten etal. 1987: 156); im'uk 'water' (1982 [1890]: 69), Nauk. smaq (Menovscikov 1975: 503), rather than Central Siberian Yupik meq (Badten et al. 1987: 115); mun'tuk 'black skin' (1982 [1890]: 67), 'whale meat' (1982 [1890]: 71), Nauk. mantaq (Menovscikov 1975: 445), rather than Central Siberian Yupik mangtak 'edible whale skin' (Badten etal. 1987: 111); and toon'too '[rein]deer' (1982 [1890]: 72), Nauk, tuntu (Menovscikov 1975: 484), rather than Central Siberian Yupik tungtu (Badten et al. 1987: 235). One should also keep in mind that some of the above words could also be Central Alaskan Yupik from a phonetic

Chukchi, English, and Eskimo: A survey of jargons in the Chukotka

Peninsula area

53

point of view, so that if we assumed that Central Alaskan Yupik was included in the vocabulary, the evidence for Naukanskij would be reduced to three words, those for 'cliff', 'whale skin', and 'deer'. One has here an interesting mix of Chukchi and various Eskimo languages. It is legitimate to ask whether the compiler was aware that there are several languages in his vocabulary, and whether the mixture was the responsibility of one and the same informant. There are in this vocabulary many examples of the same English word with translations in the various languages, but usually involving genetically unrelated and therefore phonetically dissimilar forms. I will give some examples, providing the origin of the word, but, for the sake of brevity, without giving the native language equivalents in the modern spelling.

(1)

[From Wells - Kelly 1982 [1890]: 67] Bear, white room'ka (< Chukchi) [i. e., polar bear] naw'ook (< Central Siberian Yupik) Bear, brown pahkin'ok (< Sirenikskij) kainga (< Chukchi) nemain'kin (< Chukchi) Big ong'are (< Central Siberian Yupik) Coat erin (< Chukchi) at'kook (< Central Siberian Yupik) Deer toomtoo or toon'too (< Central Siberian [i. e., reindeer Yupik or Naukanskij) or caribou] toom'ta (< Sirenikskij) korong'a (< Chukchi) ilwil'loo (< Chukchi) Eat ningum'eta (< Chukchi) neg'a (< Central Siberian Yupik)

(2)

[From Wells - Kelly 1982 [1890]: 68] kitswe'a (< Sirenikskij) Hair (human) noo'yok (< Central Siberian Yupik) noo'ya (< Central Siberian Yupik, literally: 'his/her hair') yoke (< Central Siberian Yupik) Man, native kol'loun ('woman' (p. 70)) (< Chukchi) seek'eek (< Central Siberian Yupik) Marmot anuk'keo (< Sirenikskij) [i. e., ground e'li (< Chukchi) squirrel]

54

(3)

Willem J. de Reuse

[From Wells Raven Walrus

Kelly 1982 [1890]: 69] kwil'wit (< Chukchi) muttuk'look (Central Siberian Yupik) i'wok (< Central Siberian Yupik) chit'chu (< Chukchi, female pronunciation [Comrie 1981: 244])

One can be certain that several informants were involved. It would also be hard to imagine that the compiler would not notice that different, and even genetically unrelated languages are involved here; apparently he did not care. I would like to argue that this lack of concern for the precise number of languages spoken in an area is characteristic for a trade jargon in an unstabilized stage. The trader, (or in the case of this vocabulary, the collector), will memorize (or write down) several different words with the same meaning if that appears useful for trade over a large area. The jargon speaker will of course not bother to distinguish (or write down) phonetically similar cognates, as these will be considered to be the same language. Thus it is so hard to distinguish Naukanskij from Central Siberian Yupik in this vocabulary, because the compiler never noticed or cared to notice the difference. It is clear that the number of Chukchi loanwords in this vocabulary is much larger than the number found in any contemporary Eskimo language; as in Dall's vocabulary, there are also qualitative differences as to the words that are not Eskimo. So, in Wells and Kelly's vocabulary, the particles are also from Chukchi, even though they are not the same as in Dall: ki'wa 'by-and-by'(1982 [1890]: 67), Ch. qeywe 'it is true, really', also in Central Siberian Yupik as qaywa 'wait!; consider before proceeding!' (Badten etal. 1987: 179); wing'a 'no', Ch. uyqe 'no; there is no', also reported in jargon Chukchi as uinga, ouinga (Nordenskiöld 1882: 387, 411, 488). The pronouns in Wells and Kelly are rather problematic: there is oowung'a 'me' (1982 [1890]: 68), apparently from Inupiaq uvarja (MacLean 1980: 74); pung'a 'mine' (1982 [1890]: 68), of unknown origin; hum'neen 'mine' (1982 [1890]: 68), from Ch. gamnin 'my' (Moll - Inenlikej 1957: 33); oomungh'ham 'we', of unclear origin, perhaps from Sir. menga Τ (Menovscikov 1964: 53); shupa 'you' (1982 [1890]: 69), probably from Sir. elpi (Menovscikov 1964: 53), rather than from Central Siberian Yupik elpek (Badten et al. 1987: 56); hurt 'you', and hun'yun 'yours' (1982 [1890]: 69) respectively from Ch. gdn 'you', and Ch. gdnin 'your' (Moll - Inenlikej 1957: 33).

Chukchi, English, and Eskimo:

A survey of jargons

in the Chukotka

Peninsula

area

55

The fact that the particles are from Chukchi in both vocabularies, and that the pronouns are only from Chukchi in Dall, and from Chukchi and non-Central Siberian Yupik Eskimo languages in Wells and Kelly, is very intriguing, and could hardly be due to chance. The best explanation is that both vocabularies are in an Eskimo-Chukchi jargon, and that in this jargon, the particles were taken from Chukchi. Since independent pronouns are more commonly used in Chukchi than in Eskimo, it would also make sense for a jargon, which is more analytic than Eskimo and therefore needs independent pronouns, to borrow them from Chukchi. It is less clear why there are also some independent pronouns from Eskimo in Wells and Kelly. One might conjecture that the Inupiaq form was borrowed from an Inupiaq jargon from Alaska, and that Sirenikskij, like Chukchi, uses its independent pronouns more often than in other Eskimo languages, and as a result these were also taken over by the jargon. Another piece of evidence that Wells and Kelly contains a jargon is the fact that a few Eskimo word combinations are not ordered in the usual way. In Central Siberian Yupik, a modifying noun most often follows the modified noun, but in this vocabulary, as in Chukchi and English, the sequence is modifying noun plus noun. The examples occurring in the wordlist are the following: 4 (4)

(5)

(6)

sukalu'uke 'bad man' (1982 [1890]: 67);

tow'illery-yoke 'black man' (1982 [1890]: 68);

kot'ilery-yoke 'white man' (1982 [1890]: 68);

Central Siberian Yupik

Central Siberian Yupik

Central Siberian Yupik

*seghleq seghlegh-0 bad-ABs

yuuk yug-0 person-ABs

l* tagnelghii tagnegh-lghii-0

yuuk yug-0

be.black-iNP-ABs

person-ABs

* qatelghii qategh-lghii-0 be.white-iNP-ABs

yuuk yug-0 person-ABs

The preferred Central Siberian Yupik forms would be yuuk seghleq, yuuk tagnelghii, and yuuk qatelghii. The expressions yuuk tagnelghii, and yuuk qatelghii are also suspicious since they look like caiques of the English ethnic designations black man and white man. The correct Central Siberian Yupik ethnic terms are Pureki, from English Portagee, referring to the fact that Africans on the ships were often Cape-Verdean speakers of Portuguese, and Laluramka, a compound from Chukchi literally meaning 'bearded clan', and these terms do not contain a reference to skin color.

56

Willem J. de Reuse

I will now turn to English-based jargons. During the whaling period, the Eskimo and coastal Chukchi came in contact with the crews of US or European whalers' ships. 5 On some of these ships various pidgin varieties were spoken. Gerland (1883: 207) mentions an ungrammatical jargon based on Chukchi and Eskimo, mixed with Russian, Hawaiian, and English, which was used as a lingua franca during that period. There is linguistic evidence that English-speaking whalers must have taught a variety of pidgin English to the Chukotka natives. Whymper (1868), in a generally reliable travel account of Alaska and Chukotka in 1865-1866, gives the following words and phrases of "broken English" uttered by Chukchi natives: (7)

a. [about the engine of the ship] Too muchee wheel, makee man too muchee think!

(1868: 91)

b. [referring to objects as well as human beings] What's that fellow?

(1868: 91)

c. donkey engine 'ship's engine'

(1868: 91)

d. [explanation on the telegraph] S'pose lope fixy, well - one Melican Plower Bay, make talky all the same San Flancisco Melican.

(1868: 92)

e. [on eating meat with pepper sauce] Me sabe good deal, but me no sabe white man eat fire on meat. (1868: 121) f. too muchee 'too expensive'

(1868: 122)

The phrases given above show features typical of many varieties of pidgin English. For example, the following also occur in (the now extinct) Chinese Pidgin English: postconsonantal word-final /i/ in muchee, makee, fixy, talky, III for Irl, as in lope, Melican, San Flancisco (Hall 1944). The following feature occurs in Melanesian Pidgin English or Tok Pisin: fellow as an affix marking attributive adjectives [Mühlhäusler 1986: 153— 154). Other features are more widespread and they occur in Chinese Pidgin English as well as in Tok· Pisin: make as some type of all-purpose auxiliary, as in make talky; all (the) same 'like'; s'pose 'if'; and sabe 'to know'. Concerning this last word, another geographical account reports that the "corrupt Spanish" expressions mi savi Ί know'; mi no savi Ί don't know' are generally known around Plover Bay (Das Ausland 1882:

Chukchi, English, and Eskimo: A survey of jargons in the Chukotka

Peninsula area

57

904). Schuchardt (1883: 15) points out that they are actually pidgin Portuguese picked up by the whalers somewhere in the South Seas, and used also in English-based pidgins. It is thus likely that the above phrases are examples of a ship's jargon that originated farther south, and that it is an ancestor of present-day Pacific varieties of Pidgin English. Two English words recorded only in nineteenth-century Chukchi are probably also from whaler's pidgin English, and refer to goods commonly obtained from whaling ships. The word ram, or lum, from "rum", but referring to any type of liquor, is often quoted as well-known to the Chukchi (Whymper 1868: 115, Das Ausland 1882: 905, Nordenskiöld 1882: 367, 402, 403, 412). Nordenskiöld (1882: 471) also notes that the Chukchi used calico to refer to a shirt of seal-gut or cotton cloth. This word is attested in Inupiaq trade jargon: ka'-li-kö 'any kind of knitted or woven stuff' (Stefänsson 1909: 225); it apparently also entered North Alaskan Inupiaq itself as kaliku 'cloth' (MacLean 1980: 19). Travelers' accounts also often mention the presence of Hawaiian words in such whalers' jargons. In the period during which whaling ships stopped in Hawaii (1819-1880, with a peak in activity between 1840 and 1860), Hawaiians often served as crew menbers on the whaling ships (Drechsel - Makuakäne 1982: 464, Holm 1986: 296). Hawaiian words specifically noted for the Russian Far East coast were kaukau 'food', and pau, in Hawaiian 'ready, finished', but said to mean the same as uinga 'nothing, there is no' among the Chukchi (Nordenskiöld 1882: 388, 412, Das Ausland 1882: 904, Birgham 1883: 20). Kaukau, actually Hawaiian Pidgin English for 'to eat', is originally from Chinese, and was already considered a genuine Chukchi word by Nordquist (1883: 208); it has become so completely assimilated in Chukchi that it occurs in the Soviet dictionaries: kawkaw, kewkew 'ship's biscuit, rusk, small loaf, bread' (Moll - Inenlikej 1957: 50, 59). Ol'ga Tunnetuvge, a native speaker of Chukchi from Anadyr', has confirmed to me its existence in contemporary Chukchi (personal communication, 1991). The Chukchi scholar Inenlikej (1978: 77) notes that this word is more widespread in coastal Chukotka, as one would expect, but incorrectly assumes that its origin is onomatopoeic. Surprisingly, kaukau is not recorded in any of the Russian Far East Eskimo languages, even though it is well attested in the Inupiaq trade jargon (Wells - Kelly 1982 [1890]: 30, Murdoch 1892: 55; Stefänsson 1909: 226).6 The English-based jargons used by whalers disappeared with the end of the whaling period, about 1900, and the Eskimo or Chukchi jargons probably disappeared a little later, with the loosening of trade relation-

58

Willem J. de Reuse

ships between Chukchi and Eskimo, and the establishment of Russian schools in the Chukotka area, and English schools on St. Lawrence Island. Nowadays, at least on St. Lawrence Island, the existence of Eskimo or Chukchi jargons is completely forgotten. Even though all jargons have now apparently disappeared, and even though none of the present-day Eskimo languages of Chukotka can be considered pidginized, this does not mean that there has been no influence of Chukchi jargons on the present-day Eskimo languages of the area. There exist some unusual structural features in the three Chukotka Eskimo languages, such as an uninflected particle superstructure entirely borrowed from Chukchi, which occurs in virtually identical form in all three Eskimo languages of the Russian Far East. It is possible that this extensive (more than 100), and easy to use, set of particles, which included adverbs, conjunctions, and interjections, was borrowed from an Eskimo-Chukchi jargon. It is clear that some of these particles of Chukchi origin became well integrated into the Eskimo morphosyntax, but many are not well integrated. These are considered suspicious, or "Siberian", "reindeer herders' words", or just "slang" by Central Siberian Yupik speakers from St. Lawrence Island, whereas the same speakers will be convinced that the wellintegrated particles are genuine Eskimo. The low status of the not-wellintegrated particles is of course what one would expect if they came from a jargon. For a full discussion of this hypothesis, I refer to chapter 6 of de Reuse (1994). To conclude this paper, I will posit that in the course of history, there have been at least three unstable trade jargons in use on the western shore of the Bering Sea: (1) an Eskimo-based jargon, used for communication with the Chukchi and other Eskimo groups, containing many Chukchi loans, in particular particles and personal pronouns - this jargon might have been the origin of set of the Chukchi particles found in the Eskimo languages of the Russian Far East; (2) a Chukchi-based jargon, mostly simplified Chukchi, used for communication with outsiders; and (3) a whaler's jargon, probably English-based, with Hawaiian and Portuguese words, used by and with the ships' crews only. This is of course an idealized classification; in reality, these three types of jargon were not kept distinct, and there were probably intermediaries between them.

Chukchi, English, and Eskimo: A survey of jargons in the Chukotka Peninsula area

59

Notes 1. This paper was read at the Ninth International Troms0 Symposium on Language on June 5th, 1992, by Peter Trudgill, since I was unable to attend. I am most grateful to Peter for his skillful reading of my paper. I thank the session participants, in particular Peter Bakker and William Samarin, for their comments. 2. Peter Bakker (personal communication, 1992) points out to me that there are no cases of pidginization occurring as a result of mixed marriages; either one spouse learns the language of the other, or they both keep speaking their own language, or they speak a third language. I agree. I assume that the mixed Eskimo-Chukchi jargon spoken by the Qiwaaghmiit are actually not the result of mixed marriages, but that the mixed EskimoChukchi couples might have been more inclined to use a pre-existing Eskimo-Chukchi jargon, and thereby contributed to its spread. 3. Several participants at the Symposium pointed out that wordlists collected by Western explorers, missionaries, or traders in little-known parts of the world often look very strange. This is due to that fact that the native might misunderstand the purpose of the outsider, and provides him or her with a number of words from neighboring languages, either because of word taboo, or because the native thinks that the outsider has some utilitarian motive in mind, and assumes that what is desired is a list of words in use in this area. The result would then be a mixture of words from different languages, even in the absence of a specifically mixed jargon being used in that area. I agree that one must treat mixed wordlists with extreme caution, and that my arguments, developed below, for the existence of an Eskimo jargon in the Dall and Wells and Kelly wordlists do not constitute proof that these wordlists must represent a jargon. 4. Under the hypothetical forms of the modern Central Siberian Yupik spelling, I use the following abbreviations for the morphological analysis: ABS Absolutive singular case; iNP Intransitive participial mood inflection. 5. These trading contacts between English-speaking whalers and Siberian natives also resulted in the permanent adoption of over thirty English words into the native languages of the Russian Far East. These English loanwords designated foods (such as "beans", "butter", "candy", "flour", "onion", "pepper", "rice"), animals ("cow", "elephant", "monkey", "pussy(cat)"), festivals ("Christmas"), professions and nationalities ("doctor", "Japanese"), and items of material culture ("bomb-gun", "clock", "coal", "engine", "lock", "money", "sandpaper", "skate", "soap", "towel", "watch") which were unknown to the natives. A full discussion and list of these loanwords is provided in de Reuse (1994a). 6. An exhaustive list of Eskimo jargon words with Hawaiian etymologies is given in Drechsel - Makuakäne (1982: 462 -464).

References Badten, Linda W. - Vera O. Kaneshiro - Marie Oovi - Steven A. Jacobson 1987 A dictionary of the St. Lawrence Island I Siberian Yupik Eskimo language. Fairbanks: Alaska Native Language Center, University of Alaska.

60

Willem J. de Reuse

Bakker, Peter 1991 "Trade languages in the strait of Belle Isle", Journal of the Atlantic Provinces Linguistic Association 13: 1 — 19. Birgham, Francis 1883 "Fremdwörter bei den Tschuktschen", Das Ausland 56: 20. Stuttgart-Munich. Broch, Ingvild - Ernst Hakon Jahr 1984 "Russenorsk: A new look at the Russo-Norwegian pidgin in northern Norway", in: R Sture Ureland - Iain Clarkson (eds.), Scandinavian language contacts. Cambridge: Cambridge University Press, 21-65. Comrie, Bernard 1981 The languages of the Soviet Union. Cambridge: Cambridge University Press. Dall, William H. 1870 Alaska and its resources. Boston: Lee and Shepard. Das Ausland 1882 "Ueber die Expedition der Bremer Geographischen Gesellschaft nach der Tschuktschen-Halbinsel und Alaska", Das Ausland 55: 901-906. Stuttgart. Dorais, Louis-Jacques 1980 "Les Inuit du Labrador meridional: Donnees linguistiques", Etudes!Inuitl Studies 4: 167-174. Drechsel, Emanuel J. - T. Haunani Makuakäne 1982 "Hawaiian loanwords in two native American Pidgins", International Journal of American Linguistics 48: 460 —467. Fox, James Α. 1983 "Simplified input and negotiation in Russenorsk", in: Roger W. Andersen (ed.), Pidginization and creolization as language acquisition. Rowley, MA: Newbury House, 94-108. Gerland, G. 1883 "Zur Ethnographie des äussersten Nordostens von Asien", Zeitschrift der Gesellschaft für Erdkunde zu Berlin 18: 194—222. Gilder, William H. 1881 Schwatka's search. Sledging in the Arctic in quest of the Franklin records. New York: Charles Scribner's Sons. Hall, Robert Α., Jr. 1944 "Chinese Pidgin English: Grammar and texts", Journal of the American Oriental Society 64: 95-113. Holm, John 1986 "Substrate diffusion", in: Pieter Muysken - Norval Smith (eds.), Substrata versus universals in Creole genesis. Amsterdam: John Benjamins, 259-278. Hooper, William H. 1853 Ten months among the tents of the Tuski, with incidents of an Arctic boat expedition in search of Sir John Franklin, as far as the Mackenzie River, and Cape Bathurst. London: John Murray. Inenlikej, Piotr I. 1978 "K periodizacii razvitija leksiki cukotskogo jazyka (po dannym areal'nyx nabljudenij)" [On the periodization of the development of the Chukchi lexicon (on the basis of areal observations)], in: M. A. Borodina (ed.), Narody i jazyki Sibiri [Peoples and languages of Siberia]. Moscow: Iszatel'stvo Nauka, 74-76.

Chukchi, English, and Eskimo: A survey of jargons in the Chukotka Peninsula area

61

Krauss, Michael E. 1975 "St. Lawrence Island Eskimo phonology and orthography", Linguistics 152: 39-72. Krupnik, Igor I. - Mixail A. Clenov 1979 "Dinamika etnolingvisticeskoj situacii u aziatskix eskimosov" [The dynamics of the ethnolingistic situation with the Asiatic Eskimo], Sovetskaja Etnografija 2: 19-29. MacLean, Edna A. 1980 Inupiallu Tanrjillu Uqalwjisa Ilanich. Abridged Inupiaq and English Dictionary. Fairbanks: Alaska Native Language Center, University of Alaska, Barrow: Inupiat Language Commission. Menovscikov, Georgij A. 1964 Jazyk sirenikskix eskimosov [The language of the Sirenik Eskimo]. Leningrad: Izdatel'stvo Nauka. 1975 Jazyk naukanskix eskimosov [The language of the Naukan Eskimo]. Leningrad: Izdatel'stvo Nauka. 1980 Jazyk eskimosov Beringova proliva [The language of the Bering Strait Eskimo], Leningrad: Izdatel'stvo Nauka. Moll, Tat'jana A. - Inenlikej, Piotr I. 1957 Cukotsko-russkij slovar' [Chukchi-Russian dictionary]. Leningrad: Gosudarstvennoe ucebno-pedagogiceskoe izdatel'stvo ministerstva prosvescenija RSFSR. Mühlhäusler, Peter 1986 Pidgin and creole linguistics. Oxford: Blackwell. Murdoch, John 1892 "Ethnological results of the Point Barrow Expedition", in: John W. Powell (ed.), Ninth Annual Report of the Bureau of Ethnology to the Secretary of the Smithsonian Institution 1887—1888. Washington, DC: Government Printing Office. Nordenskiöld, Adolf E. 1882 The voyage of the Vega around Asia and Europe, with a historical review of previous journeys along the north coast of the Old World. New York: MacMillan and Co. Nordquist, Oscar 1883 "Tschuktschisches Wörterverzeichnis", in: Adolf E. Nordenskiöld (ed.), Die wissenschaftlichen Ergebnisse der Vega-Expedition. Vol.1. Leipzig: F. A. Brockhaus, 206-225. Petersen, Robert 1979 "Danish influence on Greenlandic syntax", in: Bjarne Basse - Kirsten Jensen (eds.), Eskimo languages, their present-day conditions, majority language influence on Eskimo minority languages. Aarhus: Arkona, 113—121. de Poncins, Gontran 1965 Kabloona. Alexandria, VA: Time-Life Books. de Reuse, Willem J. 1994 Siberian Yupik Eskimo. The language and Its Contacts with Chukchi. Salt Lake City: University of Utah Press. 1994a English Loanwords in the Native Languages of the Chukotka Peninsula, Anthropological Linguistics 36: 56—68.

62

Willem J. de Reuse

Schuchardt, Hugo 1883 "Melanesian English", edited and translated by Glenn G. Gilbert, Pidgin and creole languages. London: Cambridge University Press, 15-23. Silverstein, Michael 1972 "Chinook Jargon: Language contact and the problem of multi-level generative systems", Language 48: 378-406, 596-625. Stefänsson, Vilhjälmur 1909 "The Eskimo trade jargon of Herschel Island", American Anthropologist 11: 217-232. Wells, Roger, Jr. - John W. Kelly 1982 English—Eskimo and Eskimo—English vocabularies . Rutland, VT: Charles E. Tuttle. [1890] Original edition. Washington, DC: Government Printing Office. Whymper, Frederick 1868 Travel and adventure in the Territory of Alaska. London: John Murray.

A case of nongenetic development in the Arctic area: The contribution of Aleut and Russian to the formation of Copper Island Aleut1 Evgenij

V. Golovko

Very few pidgin and Creole languages have been identified in the Arctic area to date, although the search for them is far from finished. Recent discoveries (see below) prompt one to look for new evidence on the colonization of the North, especially Siberia. Russian colonization of this vast area, populated by numerous tribes which spoke various languages, inevitably must have led to the formation of pidgins and Creoles. Among the few language systems that have been identified, one can find varieties which, in spite of being referred to as pidgins and Creoles, are very different both in origin and structure. The following three illustrate this diversity:

1. Taimyr pidgin Although this language is not widely known, it is, in my opinion, the most typically pidgin-like of all the Arctic candidate languages. Several Samoedic grammars mentioned its continued existence in the Taimyr Peninsula, and it was finally investigated by Evgenij Xelimskij during his fieldwork in Taimyr in 1986-1987 (see his short grammar survey and appended text in Xelimskij 1987; see also de Reuse and Wurm in this volume). It is actually a pidginized form of Russian which was formerly spoken and is still partly preserved by the natives of the Taimyr Peninsula. Its major characteristic is its systematic adaptation of Uralic-Altaic grammatical structures.

2.

Russenorsk This is the best-known Arctic pidgin, and it has been thoroughly analyzed. One can debate whether or not it can be called a pidgin in the proper sense of the word (see Jahr, this volume).

3. Copper Island Aleut This language system was discovered by Georgij Menovscikov in 1963. It is still spoken by ten to twelve elderly people who are former inhabitants of Mednij (Copper) Island; they now live in a village on

64

Evgenij V. Golovko

Bering Island — one of the two Commander Islands - together with five to seven speakers of Bering Aleut and about 1500 Russian speakers. These Copper Island Aleut speakers represent the third (and definitely the last) generation of speakers. Copper Island Aleut is based on two unrelated languages - Aleut and Russian - and it presents, in a sense, a unique example of language interference, since it has borrowed grammatical elements from both languages. Copper Island Aleut differs in many respects from pidgins and Creoles. In the conclusion, I shall hazard some conjectures concerning its origin and offer possible parallels to this language system, but first I shall present a short survey of its main grammatical features. (A more detailed study of the grammar of Copper Island Aleut may be found in Golovko — Vaxtin 1990.) In order to introduce new material on this relatively unstudied language system, most examples used here are taken from my later fieldwork. (For more examples, see the texts in Golovko 1988 and 1989.) Copper Island Aleut is the product of very profound linguistic interference and is a peculiar amalgam of two languages, although its components seem to have preserved their shape after merging and can be easily extracted from the grammatical system through linguistic analysis. I shall follow the method used in Golovko - Vaxtin (1990) here and compare its constituent parts with the corresponding grammatical structures of the source languages. This will allow me to show, on the one hand, what this newly formed language is made of and, on the other, to demonstrate how these heterogeneous parts work together as an integrated whole.

The Aleut contribution Lexicon The most important and unexpected aspect of the Aleut heritage in Copper Island Aleut is that it forms the basis of the lexicon. (The vocabulary of most pidgins and Creoles comes from the language of the dominant group, which would be Russian in this case.) There are, of course, certain Russian loanwords in Copper Island Aleut, but they are not more numerous than in the "pure" Bering Island Aleut dialect which now coexists with it in the same village. The majority of these Russian-based words are nouns pertaining to imported objects - household articles, tools,

A case of nongenetic development:

The contribution

of Aleut and Russian

65

food, and clothing. Copper Island Aleut, as well as other Aleut dialects - including those spoken on the American side of the Bering Strait also borrowed some Russian verbs, mostly denoting new household activities or other activities which were not familiar to the Aleuts before the Russian invasion, for example:

(1)

Copper Island Aleut kraasil cidil struuzal muucil gulal misaajal

Russian krasit' cedit' strogat', strugat' mucit' guljat' mesat'

'paint' 'strain, filter' 'plane' 'torture' 'go for a walk' 'disturb'

Phonology Phonologically, Copper Island Aleut has preserved the main Aleut features, though some have altered under Russian influence: the velar/uvular opposition has lost its obligatory status, and the original Aleut bilabial stop [v] has been split into the bilabial stops [p] and [b] (for details, see Golovko - Vaxtin 1987).

Morphology The original system of Aleut morphology remains intact. Copper Island Aleut has preserved a system of two cases using the absolutive marker -x and the relative marker -m. In all Aleut dialects (but with some exceptions), the relative marker encodes the possessor as well as the actor when it does coincide with the topic. (The verb always agrees with a third person topic. There are two paradigms for this: one when the actor does not coincide with the topic, and one when it does; for details, see Bergsland 1989.) Copper Island Aleut has preserved only the first function of -m, for example: (2)

tajagu-m ulaa

'man's house'

(3)

ukina-m qicigii

'sharp knife'

66

(4)

Evgenij V. Golovko

kusugoora-m ilagaan 'from the hill; down the hill' (cf. Russ. kosogor 'slope of a hill')

Copper Island Aleut has preserved the full paradigm for personal possessive markers on nouns, for example: (5)

ula-ng ula-an ula-a

'my house' 'your (singular) house' 'his/her house', etc.

It has also preserved the entire system of nominal and verbal derivational suffixes. The nominal suffixes include those with the following meanings: 'big', 'small', 'good', 'bad', 'doer of the action', 'means of action', and 'place of action'. The verbal suffixes include those meaning: 'have', 'want', 'do well', 'begin', 'stop', 'state resulting from the action', and several expressing plurality of actions (for examples, see Golovko Vaxtin 1990). While personal agreement on the verb in Copper Island Aleut is completely Russianized (see below), the syntactic derivational suffixes that change the argument structure (valency) of the verb are surprisingly Aleut-like. The derived verbal forms change their argument structure in complete accordance with Aleut syntactic rules (the valency of the basic, nonderived verb is, naturally, Aleut as well). For example: (6)

-ca-\ factitive causation a. CIA: 2 una-l suupa-x tagaju-ca-l cook-coNj Soup-SG salty-CAUS-3 SG PAST Russ.: ja gotovi-l i sup peresoli-l I cook-PAST and soup put much Salt-PAST Ί cooked and put too much salt into the soup.'

cf.: b. CIA:

suupa-x

tagaju-t Soup-SG salty-3 SG PRES Russ.: sup solenyj soup salty [no verb] 'The soup is salty.'

A case of nongenetic development:

(7)

The contribution

-ni-: distant causation a. CIA: ana-x aniqju-un

of Aleut and Russian

saga-ni-it

mother-SG baby-SG REFL sleep-CAUSE-3 SG PRES

Russ.: mat' svoego rebenka ukaciva-et mother her child swing-3 SG 'Mother makes her baby sleep.'

PRES

cf.: b. CIA:

aniqju-x saga-it baby-SG sleep-3 SG PRES Russ.: rebenok sp-it child sleep-3 SG PRES 'The baby is sleeping.'

(8)

-ixci-: permissive causation a. CIA: segodnja taanga-x bud-ut today

spirits-SG AUX FUT-3 PL

su-la-xci-t' take-MULT-CAUSE-INF

Russ.: segodnja spirtnoe bud-ut today

spirits

prodava-t'

AUX FUT-3 PL sell-INF

'Today they will sell spirits.' cf.: b. CIA:

oni taanga-x su-la-jut they spirits-SG take-MULT-PRES 3 SG Russ.: oni spirtnoe pokupa-jut they spirits buy-3 PL PRES 'They are buying spirits.'

(9)

-ja-: try to cause a. CIA: amu-n hagja-ja-it clothes-PL clean-CAUS-3 SG PRES

Russ.: on cist-it odezdu he clean-3 SG PRES clothes 'He is cleaning clothes.'

67

68

Evgenij V. Golovko

cf.: b. CIA:

amu-n

hagja-it

clothes-PL clean-PL PRES

Russ.: odezda cistaja clothes clean [no verb] 'The clothes are clean.' (10)

-Vsa-: transitivizer (where V = obligatory lengthening of the preceding vowel) a. CIA: on agitaaja-an aalu-usa-it he-NOM friend-3 SG REFL smile-TRANS-3 SG PRES Russ.: on svoemu drugu ulyba-et-sja he to his friend smile-3 SG PRES-REFL 'He is smiling at his friend.'

cf.: b. CIA:

aalu-it laugh-3 SG PRES Russ.: on ulyba-et-sja he smile-3 SG PRES-REFL 'He is smiling.'

(11)

-qagi-'· detransitivizer CIA: iglu-ng n'i tuta-qagi-it grandson-1 SG POSS NEG hear-DETRANS-3 SG PRES Russ.: moj vnuk ne slusa-et-sja my grandson not hear-3 SG PRES-REFL 'My grandson does not listen', i. e., 'My grandson does not obey.'

Copper Island Aleut has preserved some Aleut dependent (nonfinite) verbal forms: the conjunctive (see example [6a]), conditional, and intentional. Some other grammatical structures are also obviously of Aleut origin, e. g., the way of expressing an oblique object, the comitative, and comparison. All demonstratives, postpositions, and question words come from Aleut. One of the conspicuous features of Aleut is verb—topic number agreement, which is achieved through the coinciding of two normally sepa-

A case of nongenetic development:

The contribution

of Aleut and Russian

69

rated devices: anaphora and topic-prominence (for details, see Bergsland 1989 and Fortescue 1985). Below is an example from Bering Island Aleut: (12)

imli-ngis adu-ku-x hair-3 PL POSSD long-REAL-3 'Her hair is long.'

SG

The number of the possessor (which may be treated as the topic - it is omitted in [12] but is explicitly given in the preceding clause) is indicated at the end of the verb (-χ, third person singular). The following examples clearly show that this device is also at work in Copper Island Aleut: (13)

a. CIA:

kaju-ngi muscle-3 PL POSS Russ.: vse ego muskuly all his muscles 'All his muscles ache.'

huzu-ngi nana-it each-3 PL POSS ache-3 SG bolj-at ache-3 PL PRES

PRES

cf.: b. Russ.: ego muskul bol-it his muscle ache-3 'His muscle aches.' (14)

SG PRES

CIA:

cvetki-ning hula-l-a flowers-3 PL POSSD 1 SG POSS bloom-PAST-3 Russ.: moi cvetki rascve-l-i my flowers bloom-PAST-PL 'My flowers bloomed.'

SG FEM

Example [14] is even more striking than [13], since the verb agrees with the speaker/topic (a woman) and is thus marked with the Russian third person singular feminine morpheme. Russian gender morphemes are optional in Copper Island Aleut, and the feminine gender is marked in Russian only in the past singular, cf.: (15)

roza rascve-l-a rose bloom-PAST-FEM 'The rose bloomed.'

70

Evgenij V. Golovko

However, Russian-type agreement can occur in Copper Island Aleut. Example (14) can easily be transformed into (16), following the Russian agreement model: (16)

CIA:

cvetki-ning hula-l-i flowers-3 PL POSSD 1 SG POSS bloom-PAST-3 Russ.: moi cvetki rascve-l-i my flowers bloom-PAST-PL 'My flowers bloomed.'

PL

Both examples are considered "absolutely correct" by my informants.

The Russian contribution The most striking Russian feature in Copper Island Aleut is the use of its verbal paradigm. The verb is inflected for person, number, and tense (optionally for gender). Compare the conjugations below for the verb 'to stand': (17) 1 sg 2 sg 3 sg 1 pl 2 pl 3 pl

CIA Russian anqaxtastoanqaxta-ju sto-ju anqaxta-is sto-is anqaxta-it sto-it anqaxta-im sto-im anqaxta-iti sto-ite anqaxta-jut (-jat) sto-jat {-jutl-jat are variants used in different conjugation types)

The tense system in Copper Island Aleut is entirely Russian, as can be seen in the examples above: present [6b, 7, 8b, 9-13], past [6a, 14, 16], and future [8a]. The absence of person marking in the Russian past is compensated for by the use of Russian pronouns. (In Aleut there are no independent subject pronouns; see, for example, oni 'they' in [8b] and on 'he' in [10a].) Russian third person object (accusative) pronouns have been adopted by Copper Island Aleut and are used instead of the original set of endings, sometimes called the "objective conjugation" (see Bergsland Dirks 1981: 8-12):

A case of nongenetic development:

(18)

CIA:

ona she-NOM

hixta-it say- 3 SG

The contribution of Aleut and Russian

PRES

71

cto ona ego that she-NOM h e - A C C

ilaxta-it love-3 SG PRES Russ.: ona govor-it cto ona ego ljub-it she say-3 SG PRES that she him love-3 'She says that she loves him.'

SG PRES

cf. Bering Island Aleut (BIA): (19)

BIA: ilaxta-ku-u l o v e - R E A L - 3 SG OBJ 3 SG SUBJ

'She/he loves him/her/it.' The same set of pronouns (from Russian) is also used instead of the original Aleut object pronouns: (20)

CIA:

ty you Russ.: ty you

menja hamayaaxta-is me ask-2 SG PRES menja sprasiva-es' me ask-2 SG PRES

cf.: BIA: ting me

ahmayaaxta-ku-xt a s k - R E A L - 2 SG PRES

'You are asking me.' The set of original Aleut object pronouns (ting, txin, timis, etc.) has been preserved, but their usage is restricted to reflexive verbs: (21)

CIA:

halucagi-l txin sew-coNj 3 SG Russ.: ona si-l-a she

cuhni-l REFL prick-3 SG PAST i ukolo-l-a-s'

Sew-PAST-FEM a n d

prick-PAST-FEM-REFL

'She sewed and pricked herself.' The original Aleut analytical future is replaced in Copper Island Aleut by the Russian analytical future: the auxiliary bud- (often shortened to bu-) plus the infinitive in -/' (see example [8a]). Russian also has a syn-

72

Evgenij V. Golovko

thetic future tense connected with an aspectual meaning of completed action, which is expressed by a prefix. The person markers for this form do not differ from those for the present tense. There are no prefixes in Aleut dialects; all ways of expressing the future are thus analytical and consist of two parts - a verb in the intentional or conjunctive plus an auxiliary (Bergsland - Dirks 1981: 6 2 - 6 4 ; Golovko 1982), i. e., the order of constituents is the reverse of the Russian future. Copper Island Aleut chose the less ambiguous of the Russian ways of expressing the future (the analytical) and has completely ignored the aspect. Both Aleut negative suffixes, -laka(g)- 'negation of actual action' and -{g)ula-{x) 'negation of nonactual action', are replaced in Copper Island Aleut by the Russian prefix ne- (pronounced [ni]) - (as mentioned above, there are no prefixes in Aleut). Compare the following example from Bering Island Aleut with that from Copper Island Aleut given in example [11] and repeated here: (22)

CIA: iglu-ng n'i tuta-qagi-it grandson-1 SG POSS NEG hear-DETRANS-3 SG PRES BIA: iglu-ng tuta-qagi-laka-x grandson-1 SG POSS listen-DETRANS-NEG-3 SG PRES 'My grandson does not listen', i. e., 'My grandson does not obey,'

Copper Island Aleut has adopted some Russian modal words (e. g., nado 'ought to', dolzen 'must', moc' 'can', xotet' 'want') and the predicate negator net(u). Words with these meanings, as well as conjunctions and some adverbs (denoting time, degree, quality, etc.) are normally used a great deal in language contact and are borrowed at an early stage. Such Russian loanwords are found in Bering Island Aleut as well as in many other languages of Siberia. As can be seen from the examples above, the syntactic structure of the Copper Island Aleut clause is taken from Russian. The rigid SOV word order which is the core of Aleut syntax has been abandoned; phrases are produced according to Russian syntactic models. Copper Island Aleut shows a rather high degree of variability, especially in syntax. Compare the following two synonymous Copper Island Aleut sentences: (23)

a. ja segodnja cxuugi-n inka-ca-l I today linen-PL hang-CAUS-3

qaka-ca-anga SG PAST dry-CAUS-iNT

A case of nongenetic development:

The contribution

b. ja segodnja cxuugi-n inka-ca-l I today linen-PL hang-CAUS-3 qaka-ca-t'

of Aleut and Russian

SG PAST

73

ctoby ego in order it

dry-CAUS-iNF

Today I hung linen in order to dry it.' cf. Russian and "pure" Bering Island Aleut: (24)

a. Russ.: ja segodnja bel'e povesi-l-a {ctoby ego) I today linen hang-PAST-FEM (in order to it) vysusi-t' dry-INF

[The words in parentheses are normally omitted in colloquial Russian.] b. BIA: wan angalix cxuugis qaka-t-iingan this day linen dry-CAUS-INTR inka-t-na-x hang-CAUS-PAST-3 SG

'Today I hung linen in order to dry it.' In (23b), the original intentional form qaka-ca-anga from (23a) is replaced by the Russian infinitive construction (-t'). Both sentences are asserted by informants to be "perfectly correct". To summarize, two layers may be identified in Copper Island Aleut: one is of Aleut origin, the other is Russian. The first comprises the vocabulary, derivational morphology, nominal inflection, and certain Aleut structures. The second consists of verbal inflection, negation, infinitive forms, and syntax. Phonetically, Copper Island Aleut is a compromise between Aleut and Russian. Linguistic facts from Copper Island Aleut clearly show that it differs sharply from most language systems that can be considered to be a result of language interference. It is an independent language though it shows a high degree of syntactic variability. It obviously has nothing in common with pidgins, but, at the same time, it does not look like a Creole language, either historically or structurally. There is no convincing evidence that it could have developed from a pidgin, as there seem to be no people for whom such a pidgin could have served as the only possible means of communication (see Golovko - Vaxtin 1990: 113-118). Even if such groups could be found, most probably they would have used Russian or

74

Evgenij V. Golovko

a Russian-based pidgin, and not a pidgin with an Aleut lexicon. It is hard to imagine relexification leading to such a total change in the lexicon. In addition, Copper Island Aleut does not show real simplification in its grammatical structure. It has replaced certain Aleut grammatical structures very precisely with Russian ones, admitting at the same time a rather high degree of variability.

Possible parallels When a reasonable explanation for a phenomenon cannot be found immediately, it is sometimes helpful to look for similar phenomena which may shed some light on the matter. However unique Copper Island Aleut might seem, we know of at least one other language that bears a striking resemblance to it in its grammatical structure (there are actually more, but they have not been studied very thoroughly). Romani dialects spoken in Spain, Portugal, England, Armenia, Sweden, and Norway have almost completely lost their original grammatical structures, have borrowed sets of grammatical affixes, and have switched to the grammar of the dominant languages of those countries, while at the same time having preserved the original Romani lexicon. I would like to illustrate this linguistic situation with some examples analyzed by Ian Hancock (1976: 104). The following examples are taken from the Romani dialects of Great Britain, Spain, and "pure" Romani as spoken in Europe: (25)

a. Mandi' bor's chavvies are kellin' dre the poov. b. Es chabores de mindös rocamblo fican andre ο tarique. c. Mure amaleske savore khelen änd'e phuv. 'My friends' sons are playing in the field.'

One doesn't need a dictionary or Romani grammar to identify which country each of these Romani varieties comes from. As in the case of Copper Island Aleut, it seems highly unlikely that Romani speakers could switch to English (Spanish, etc.) and later (or at the same time) adopt the entire bulk of the Romani lexicon as loanwords. Rather than being a strange and inexplicable example of relexification, these cases represent instead a special linguistic situation. It is important to mention here that speakers of these Romani dialects identify themselves as Romani speakers (not English, Spanish, etc.): this

A case of nongenetic development:

The contribution

of Aleut and Russian

75

provides another parallel with Copper Island Aleut, which is considered by its speakers to be "the correct Aleut language". In Great Britian (mainly in Wales), there are groups of Gypsies who speak a variety which is very close to the Romani dialects of Europe. This variety is considered by the speakers to be an old, "deep" language, and Anglo-Romani is thought of as a new, "broken" language (Smart Crofton 1875: xi; Hancock 1977: 36). In their book on the dialect of English Gypsies, Smart and Crofton pointed out that, in the middle of the nineteenth century, most Gypsies spoke Anglo-Romani, but the "new" and the "old" languages were not taken to be two entirely different languages: the same speaker could use grammatical forms from both languages. My feeling is that the situation was, to a certain extent, analogous with the variability we see in Copper Island Aleut. The productivity of Aleut word-formation suffixes can be compared with the productivity (though restricted) of some Romany affixes in the Romani of Norway: according to Iversen (1944: 245-252), new words can be derived on the basis of Norwegian roots by means of Romani affixes. Hancock (1977) rightly points out that Anglo-Romani (as well as other Romani dialects of this type) is a kind of contact language which looks, in many respects, like a Creole. This similarity to Creole languages is clear, though it seems to be more like a mirror-image. Unlike pidgins and Creoles which go through relexification, these Gypsy dialects experience a process which can instead be called "regrammaticalization" (Eloeva Rusakov 1990: 38): (some) grammatical structures of the language are replaced, but the lexicon is not radically affected. This regrammaticalization can take place more than once: the speakers of Spanish Romani replaced their Spanish grammatical structures with Portuguese ones after Portuguese became the dominant language (Duff 1950, Bourgeois 1911, cited in Eloeva - Rusakov 1990: 38). Eloeva and Rusakov (1990: 42) assert that the major reason for the peculiar development of many Romany varieties was the need to preserve them as cryptolects which were inaccessible to the surrounding population. We have no evidence that Copper Island Aleut followed its peculiar path for this reason. However, the cases mentioned above demonstrate that such a development, occurring when all speakers of a language are involved in permanent code-switching and code-mixing, is not exceptional and displays certain universal features. Most probably, this process is also universal in geographical terms: we see that languages can evolve

76

Evgenij V. Golovko

in this peculiar way in many differents parts of the world, including places close to the Arctic Circle - such as Norway and the Commander Islands - and who knows where else?

Notes 1. I am grateful to Alexandr Rusakov, with whom I have discussed the problems considered in this paper many times. 2. Copper Island Aleut is abbreviated as CIA in the examples, Bering Island Aleut as BIA.

References Bergsland, Knut 1989 "Comparative aspects of Aleut syntax", Aikakauskirja Journal 82: 7 - 8 0 . Bergsland, Knut - Moses Dirks 1981 Atkan Aleut school grammar. Anchorage: University of Alaska. Eloeva, Fatima - Aleksandr Ju. Rusakov 1990 Problemy jazykovoj interferencii [Problems of language interference]. Leningrad: Leningradskij universitet. Fortescue, Michael 1985 "Anaphoric agreement in Aleut", in: A. Machtelt Bolkenstein - Casper de Groot - S. Lachlan Mackenzie (eds.), Predicates and terms in Functional Grammar. Dordrecht: Foris, 105—126. Golovko, Evgenij V. 1982 "Sistemy vremen ν aleutskom jazyke" [The tense system in Aleut], in: Nikolaj D. Andreev (ed.), Lingvisticeskie issledovanija. Jazykovye edinicy ν sinxronii i diaxronii. Moscow: Institut jazykoznanija AN SSSR, 34-41. 1988 "Materialy dlja izucenija jazyka mednovskix aleutov 1." [Materials for studying the language of Copper Island Aleut 1.], in: Nikolaj D. Andreev (ed.), Lingvisticeskie issledovanija 1988. Problematika vzaimodejstvija jazykovyx urovnej. Leningrad: Institut jazykoznanija, 73-78. 1989 "Materialy dlja izucenija jazyka mednovskix aleutov 2." [Materials for studying the language of Copper Island Aleuts 2.], in: Nikolaj D. Andreev (ed.), Lingvisticeskie issledovanija 1989. Struktura jazyka i ego evoljucija. Moscow: Institut jazykoznanija, 67—74. Golovko, Evgenij V. - Nikolaj B. Vaxtin 1987 "The convergence of contacting phonological systems", Proceedings of the XI International Congress of Phonetic Sciences 6. Tallinn: 172-174. 1990 "Aleut in contact: the CIA enigma", Acta Linguistica Hafniensia 22: 97 — 125. Hancock, Ian 1976 "Patterns of English lexical adoption in an American dialect of Romanes", Orbis 25.1. 1977 "The social and linguistic development of Anglo-Romani", Working Papers in Sociolinguistics 38, 83—104.

A case of nongenetic development: The contribution of Aleut and Russian

77

Iversen, Ragnvald 1944 The Romany language in Norway. Oslo: Jacob Dybvad. Menovscikov, Georgij A. 1964 "K. voprosu ο pronicaemosti grammaticeskogo stroja jazyka" [On the penetrability of the grammatical structure of a language], Voprosy jazykoznanija 5: 100-106. Smart, B. C. - Η. T. Crofton 1875 The dialect of the English Gypsies. London: Asher. Xelimskij, E. A. 1987 "Russkij govorka mesto kazat' budem (Tajmyrskij pidzin)" [I shall say it in the Russian speech (Tajmyr pidgin)], in: Igor F. Vardul' - Vladimir I. Belikov (eds.), Vozniknovenie i funkcionirovanie kontaktnyx jazykov: Materialy rabocego sovescanija [Emergence and functioning of contact languages: Materials from a workshop], Moscow: Akademija Nauk SSSR, Institut vostokovedenija, 8 1 - 8 3 .

The Taimyr Peninsula Russian-based pidgin1 Stephen A. Wurm

It has been known for quite some time that pidgin languages with a vocabulary derived predominantly from Russian and a syntactic base reflecting Uralian and Altaic features functioned as languages of intercultural and interethnic communication in parts of northern Siberia from before the turn of the century until beyond the 1950s, in some regions even up to the present day. One such variety is reported by Ubrjatova (1985: 67 — 68), who mentions a peculiar form of Russian used by descendants of the so-called "Transtundra Christians" in the Noril'sk region of the Taimyr Peninsula area. It had no sibilants, nor / or v, but had vowels as found in Dolgan, a sublanguage (or dialect) of the Turkic Yakut language spoken further east, displayed Dolgan syntactic features, and contained a very large number of Dolgan words. It appears that it might have become creolized, but today its speakers have become completely assimilated by the Dolgans. The first description of one of these Russian-based pidgins in northern Siberia has been provided by Khelimskij (1987), who studied a variety used by speakers of the Samoyedic Nganasan language in the Taimyr Peninsula area. It was, and to some extent still is, employed for communication between Nganasan and Russian speakers, and also as a lingua franca in situations involving speakers of Nganasan and of the Samoyedic Nenets, the Turkic Dolgan, and the Tungusic Evenki and Even (Lamut) languages. At present, this pidgin language is spoken predominantly by the middle-aged and older generations, and with many of its speakers it is beginning to show influences from colloquial Russian, especially when used in speaking to Russians - it is, according to Khelimskij, beginning to move into a post-pidgin continuum stage. The salient characteristics of Taimyr Peninsula pidgin are outlined in the following seven sections, based on material from Khelimskij's (1987) study. Before going into more detail and giving examples of these seven points, some remarks about the translations are necessary. Pidgin examples are given in transliteration, followed by a literal translation reflecting the Russian meaning (e. g., ljudi means 'human, man' in pidgin, but 'people' in Russian; in the pidgin literal translation, it is given as 'people').

80

Stephen Α. Wurm

This is followed by the Russian equivalent in transliteration, also with a literal translation. 2 This may help readers unfamiliar with Russian to understand and appreciate the features of this Taimyr Peninsula pidgin and the differences between it and Russian.

1. Phonetics The Taimyr pidgin has a phonetic system which reflects Nganasan and other indigenous features. Russian c is pronounced with a weakened t, or the t is missing altogether. Russian shibilants change to s in a number of lexical items. Initial consonant clusters are generally reduced, e. g., kol'ko (Russian skol'ko) 'how much'; se (Russ. vse) 'all of them'; cera (Russ. vcera) 'yesterday'; kazat' (Russ. skazat' 'to say') 'to say, to speak'. The vowel system corresponds largely to that of standard Russian, with some exceptions, but the intonation pattern is indistinguishable from that of Nganasan.

2. Vocabulary The pidgin's vocabulary is derived predominantly from colloquial Russian, but with many of the pidgin lexical items differing in meaning and phonetic shape from the Russian source words, and with a large number of elements derived from Nganasan and other northern Siberian languages. The vocabulary of the language is rather restricted, and descriptive lexical items are common; e. g., kori sanka (lit. 'horse sledge') 'cart' (Russ. telega). Many basic verbs have broader meanings than their Russian source words or have multiple meanings. Many of the pidgin lexical items have meanings which differ from those of the Russian source words, and/or have changed their shape to a greater or lesser extent (see sanka 'sledge' above, from Russ. sani or sanki). Other examples include: (1)

a. pojmat' 'to receive'; from Russian 'to catch' (Russ. polucif 'to receive') b. prijti 'to occur, be found', from Russian 'to come' (Russ. popast'sja 'to occur')

The Taimyr Peninsula Russian-based

pidgin

81

c. rezat' 'to cut, to pass time'; from Russian rezat' 'to cut' (Russ. provesti 'to pass time') d. lomat' 'to spoil'; from Russian 'to break, to smash' (Russ. [isjportit' 'to spoil') e. umet' 'to have, to be able'; from Russian 'to be able' (Russ. imet' 'to have') f. govorka 'word, speech, language'; from Russian go vor 'talk, speech' (Russ. slovo 'word', jazyk 'language'; govorka is slang for 'speaking') g. seredka 'half of something, a piece o f ; from Russian 'middle, middle part' (Russ. cast' 'a piece of').

3. Gender and Number The Taimyr Peninsula pidgin lacks the distinction between genders and number for nouns, adjuncts, and past tense verbs which is typical of Russian. The following examples illustrate the lack of gender distinction in adjuncts and past tense forms; masculine forms are used exclusively: (2)

a. russkij govorka 'Russian speech, language'; instead of russkaja govorka (Russ. russkij jazyk) b. takoj priduman'e 'such thinking'; instead of takoe priduman'e (Russ. takaja ideja) c. devka pel 'the girl sang'; instead of devka pela (Russ. devuska pela)

Also note this lack of adjunct and past tense gender concordance in the following sentences:2 (3)

P: cimi devka-to pesna pel Cimi girl-that song sang R: devuska cimi pela pesnju girl Cimi sang song 'The girl Cimi sang a song for

(4)

P: baba sovsem kityryj. woman entirely cunning

menja sestra mesto. me sister place moej sestre. to-my sister my sister.'

82

Stephen Α. Wurm

R: nekotorye zensciny ocen' xitry. some women very cunning (5)

Ρ: menja takoj priduman'e me such thinking R: u menja est' takaja at me there-is such Ί have such an idea.'

est'. there-is ideja. idea

Examples of pidgin nouns appearing in the Russian plural form with singular meaning are given below and illustrated in the following sentences: (6)

a. glaza 'eye'; in Russian 'eyes' (Russ. glaz 'eye') b. rebjata 'child'; in Russian 'children' (Russ. rebenok 'child') c. ljudi 'human, man'; in Russian 'people' (Russ. celovek 'human, man')

(7)

P: baba-to est' i rebjata odin. old-woman-that there-is and children one [The numeral odin 'one' often follows the noun it determines.] R: zena est' i odin rebenok. woman there-is and one child

(8)

P: tebja kakoj ljudi. you-ACC which people R: cto ty za celovek? what you for person? 'What sort of person are you?'

Because number is not indicated on nouns in the Taimyr pidgin, they remain unchanged after numerals, unlike in Russian - e. g., sem god 'seven years', sto kon' '(one) hundred horses' - except in certain set phrases, such as dva rublja 'two rubles' (Russ. rublja is the genitive of rubl' 'ruble'). If the indication of plurality is essential, auxiliary words are employed, e. g., djaringa se 'lads' (lit. 'lads they all'); mnogo drova 'trees' (lit. 'much [fire]wood').

The Taimyr Peninsula Russian-based pidgin

83

4. Pronouns Taimyr pidgin has a singular—dual—plural pronoun system embodying inclusive and exclusive forms in the nonsingular first person, with the first, second, and (in part) third person pronouns reflecting the accusative (or object) forms of the Russian personal pronouns. It must be stressed that the pronominal system of the pidgin is quite different from Russian. The forms menja 'me' and tebja 'you' (singular accusative) are used for Ί/my' and 'you (singular)/your' in all possible syntactic relationships, e. g.: (9)

P: menja tebja drova rubil. me you-ACC (fire)wood chopped R: ja tebe drova rubil. I for-you (fire)wood chopped Ί chopped (fire)wood for you.'

Very rarely, other pronominal forms occur (e. g., nasa 'our' [feminine]), taken from Russian, very likely recently. Khelimskij states that the materials collected make it possible, with some hypothesizing, to suggest the presence in Taimyr Peninsula pidgin of the forms for the dual and plural personal (and possessive) pronouns for the first and second persons shown in Table 1. Table 1. Hypothesized personal and possessive pronouns

First person inclusive First person exclusive Second person a b

Dual

Plural

menja tebja menja obaa tebja oba

menja tebja se b menja se tebja se

oba 'both' se 'they all' (Russ. vse)

For the third person personal pronoun, irrespective of number, two forms are used: oni and ego (lit. 'him'). The first of these appears in the usual syntactic position for nouns, e. g.: (10)

P: oni dva rublja rabotal. they two rubles worked

84

Stephen Α. Wurm

R: on zarabotal dva rublja. he earned two rubles 'He earned two rubles.' The second form, ego, appears immediately after the predicate and indicates that the third person subject is singular. Similarly, ego oba in that position indicates that it is dual, whereas for the plural, only ego seems to be attested: (11)

P: menja saman kostjum odel ego. me shaman costume dressed (someone) him R: on nadel moj samanskij kostjum. he put-on my shaman costume 'He put on my shaman costume.'

(12)

P: gorod mesto guljat' xodili, vecerom town place to-walk they-went in-the-evening prixodili ego. they came-back him R: on/oni posellposli guljat' ν gorod, vecerom he/they went to-walk into town in-the-evening vern ulsjal vern ulis'. he/they-returned 'He/they went for a walk to the town; in the evening he/they came back.'

(13)

P: voda taskat' ne xocet ego oba. water drag not he-wants him both R: oni (dvoe) ne xotjat nosit' vodu. they (two) not they-want to-carry water 'They two do not want to carry water.'

5. Verbs The pidgin is distinguished by the use of only very few categories and forms for the verb. The present tense is indicated by the use of the third person singular ending -t for all persons, e. g.:

The Taimyr Peninsula Russian-based pidgin

(14)

85

P: menja nicevo ne znaet. me nothing not he-knows R: ja nicego ne znaju. I nothing not I-know Ί know nothing.'

The past tense exhibits the undifferentiated use of the Russian masculine forms (singular) -/ or (plural) -//, as can be seen in example (12) above and in other examples in this paper. The future is denoted by use of the infinitive with budet 'it will be', e. g.: (15)

P: tebja

kogda letat'

you-ACC when

budet?

to-fly it-will-be

R: ty kogda polet is'? you when you-will-fly 'When will you fly?' Instead of adding budet to the infinitive to indicate the future tense, budet can also be added to the imperative or the past tense form, although both of these possibilities occur only rarely. The previous example could therefore also be rendered as tebja kogda letaj (or letal) budet? The subjunctive is signalled through by following the verb, e. g.: (16)

P: menja sto kon' kusal by. me hundred horses ate would R: ja by sto konej s'ela... [spoken by a woman] I would hundred horses I-ate Ί would have eaten a hundred horses ...

The imperative is formed using the Russian second person singular form for all persons, e. g.: (17)

P: menja delaj... me make R: sdelaju-ka ja... I-will surely-make I 'Let me make...'

86

Stephen Α. Wurm

In Taimyr pidgin, verbs can use the Russian reflexive forms with the suffix -sja or lack such an ending, irrespective of their meanings, as exemplified below: (18)

a. parit' 'to marry someone to someone else; to get married'; Russ. sparit' 'to make a pair' (Russ. zenit' 'to marry someone to someone else' and zenit'sja 'to get married') b. odevat' 'to get dressed'; Russ. 'to dress someone' (Russ. odet'sja 'to get dressed') c. zabyt'sja 'to forget'; Russ. 'to forget oneself; to fall asleep' (Russ. zabyt' 'to forget').

(19)

P: rodnja rannij god nel'zja parit'. relative early year impossible to-make-a-pair R: rodstvennikam ν starinu nel'zja to-relatives in old-time impossible zenit'sja drug na druge. was-to-get-married to each other 'In the past it was impossible for relatives to marry each other.'

(20)

P: kusat' koncal-to, Ivan-to odeval. to-eat finished-that Ivan-that dressed someone R: Ivan koncil kusat', odelsja. Ivan finished to-eat, got-dressed 'Ivan finished eating, (and) got dressed.'

6. Syntactic features from Uralian and Altaic Word order mirrors Uralian and Altaic characteristics, with SOV order predominating, e. g.: (21)

P: dva serenij rebjata tut oxota toze xoroso dikij vidit. two middle children there hunt also well wild sees

The Taimyr Peninsula Russian-based

pidgin

87

R: dva srednix syna na oxote toze xoroso vyslezivajut two middle sons on hunt also well track-out dikogo olenja. wild reindeer 'The two middle sons when hunting also track the wild reindeer well.' The logical subject usually stands at the beginning of an utterance, before the object and the predicate, e. g.: (22)

P: dva serenij rebjata prisel. two middle children came R: prisli dva srednix syna. came two middle sons Two of the middle sons (i. e., between the oldest and the youngest) came.'

For emphasis, however, an inversion of the usual SOV word order may take place, and the word celovek, Russ. 'human, man, person', is postposed after the predicate to indicate indefiniteness, e. g.: (23)

P: nasa zemlja prisel celovek, zacem vorovat' budet? our land came someone, why to-steal will-be R: esli kto-to priexal k nam, zacem emu if someone came (travelling) to us, why him vorovat'? rob 'If somebody came to our land, why rob him?'

When nouns are juxtaposed, the modifying noun precedes the modified one, indicating the genitive, e. g.: (24)

P: menja sestra devka muzik me sister girl fellow R: muz doceri moej sestry husband of-the-daughter of-my sister 'my sister's daughter's husband'

88

Stephen Α. Wurm

This prepositioning also applies to whole adjunct phrases, e. g.: (25)

P: eto this-(is) R: eto this-(is) "This is the

glaza bol'noj celovek. eyes hurting someone celovek s bol'nymi glazami the-person with hurting eyes someone whose eyes are hurting.'

(26)

P: kotoryj eto rebjata pojmal ljudi. who this children caught people R: celovek, kotoryj pojmal etogo mal'cika. the-man who caught this boy 'the man who caught this boy'

Other features under this heading include the absence of prepositions and their replacement by postpositions, and lack of marking of categories determined by the syntactic and semantic context.

7. The function word mesto The function word mesto (also appearing as meste) 'place' is used as a postposition in a number of diverse syntactic functions, i. e., with instrumental, comitative, dative, destination, and locative-directional uses; in addition, in comparisons it follows the compared noun. Some examples are given below to illustrate its various functions: (27)

P: tebja

koliti nozik mesto. hit small-knife place R: ty koli nozom. you split-(it) with-a-knife '(you) split (it) with a knife!' you-ACC

(28)

P: kon' meste horse place R: ex a I he-travelled 'He travelled on

xodil. he-went verxom on-horseback horseback.'

The Taimyr Peninsula Russian-based pidgin

(29)

P: dva devka meste menja tarn sidel. two girl place me there sat R: ja sidel tam s dvumja devuskami. I sat there with two girls Ί sat there with two girls.'

(30)

P: tarik mesto takan nalil. old-man place glass poured R: stariku nalil stakan. for-the-old-man (I) poured the-glass Ί poured the glass for the old man.'

(31)

P: gorod mesto usel obratno. town place went back R: usel obratno ν gorod. went back in town 'He went back to the town.'

(32)

P: karman meste den'gi pixal tuda. pocket place money pushed there R: zasunul den'gi ν karman. he-put the-money into pocket 'He put the money into his pocket.'

(33)

P: takoj-to baba mesto bol'soj golova. such old-woman place big head R: takaja golova, bolse cem u baby. such head bigger than at (old)-woman 'such a head, bigger than that of the (old) woman'

89

Notes 1. This article constitutes an altered version of a section of the author's (1992) contribution "Some contact languages and pidgin and Creole languages in the Siberian region", published in Language Sciences 14(3): 1 - 3 7 . Permission from Pergamon Press to publish this modified version is hereby gratefully acknowledged. 2. In the examples, Ρ indicates pidgin, R Russian.

90

Stephen Α. Wurm

References Ubrjatova, Ekaterina I. 1985 Jazyk noril'skix dolgati [The language of the Noril' Dolgans]. Novosibirsk: Nauka. Khelimskij, Evgenij A. 1987 "Russkij govorka mesto kazat' budem (Tajmyrskij pidzin)" [I shall say it in the Russian speech (Tajmyr pidgin)], in: Igor F. VarduF - Vladimir I. Belikov (eds.), Vozniknovenie i funkcionirovanie kontaktnyx jazykov: Materialy rabocego sovescanija [Emergence and functioning of contact languages: Materials from a workshop]. Moscow: Akademija Nauk SSSR, Institut vostokovedenija.

Northwestern Russia and Scandinavia

Solombala-English in Archangel Ingvild

Broch

Almost all the existing Russenorsk material has been recorded in Norway. Also it is reasonable to presume that Russenorsk was mainly used in northern Norway. The bartering connected with Russenorsk took place here, not in Russia. During our investigation of Russenorsk, we have been wondering about the possibility of other jargons or pidgins in the area, as well as about the origin of the verbal marker -om in Russenorsk. The existence of so many mutually unintelligible languages could provide a good reason for developing jargons. From northern Sweden we know about the Swedish—Sami jargon Borgarmälet (Högström 1747 [1980], Broch Jahr 1984: 69—71); we do not, however, know much about the language itself. Fox (1973) suggested that Russenorsk could be a grammaticalization of an international North European base vocabulary which had spread throughout the military garrisons in Scandinavia. There are, however, no indications that a base vocabulary connected to soldiers ever existed. We want to draw attention to another early jargon in the northern area, which in our view proves more vital to the discussion of the origin and development of Russenorsk than Borgarmälet, namely an English— Rüssian pidgin in Archangel. Little is known about this pidgin which developed in the port Solombala in the neighborhood of Archangel (hereafter called Solombala-English). The city of Archangel, which was founded in 1693 by Peter the Great, was also the meeting place of many nations. The so-called "English period" in the history of Archangel lasted about a hundred years, from around the middle of the eighteenth century till the middle of the nineteenth century. This indicates that Solombala-English probably developed in the second half of the eighteenth century. That is exactly the same period as suggested by Broch and Jahr (1984: 50) for the origin and the development of Russenorsk. If this is correct, then Russenorsk and Solombala-English developed simultaneously in northern Norway and in Archangel. Aleksandr Davydov, however, indicates that SolombalaEnglish was used in the second half of the nineteenth century (Davydov (1988: 14).

94

Ingvild Broch

The ports of Archangel were visited by a lot of foreign sailors in the summer season when the White Sea was navigable. In the summer of 1805, 12,000 foreign sailors were reported to have come to Archangel, in addition to five or six thousand other foreign visitors (Richter 1857). In 1827, in the Norwegian town of Hammerfest (which claims to be the northernmost town of the world), Keilhau (1831: 101) in one single day counted 60 English ships on their way to Archangel. More than half of the foreign sailors in Archangel are reported to have been English. In Archangel's early history, foreign ships were not allowed into the main harbor of Archangel, but had to stop in nearby Solombala. In Solombala there was an Anglican church situated in the harbor area and the bar "London" was a popular meeting place (Davydov 1988: 14). The Russians - shopkeepers, coachmen, dock workers, streetwalkers, and young boys wanting to barter with the sailors and trying to attract the attention of foreign sailors, did not use Russian, but Solombala's own jargon — "Solombala-English". So far only two short texts are known, the first in V. Verescagin's Ocerki Arxangel'skoj gubernii [Sketches from the district of Archangel] from 1849, the second in an Archangel journal, Arxangel'skie vedomosti, from 1867.1 In 1849, Verescagin reported the sentences shown as examples (1) to (3).

(1)

Vat ju vanted, asej! What you want sailor 'What do you want, sailor!

According to Verescagin (1849: 406) the foreigners were addressed as asej because of the English habit of exclaiming I say so often. The foreigners, on the other hand, would address the dock workers with slysty '[now] listen you' (Russian). With the exception of the word asej, this phrase is reduced English, probably with a pronunciation adapted to Russian. (2)

Baem buc, sus, asej! Kom sjudal Veri gud buc, Buy shoes shoes sailor Come here Very good shoes sus. shoes 'Buy shoes, shoes, sailor! Come here! Very good shoes.'

Of special interest here is the verb baem, obviously with the same verbal marker -om that is so typical for Russenorsk. Buc reminds the Russian

Solombala-English

in Archangel

95

speaker of North Russian bucni, a special kind of shoe, and the English of English boots. Then the speaker repeats the word shoes, to make sure that the sailor will understand. While (2) contains words both of English and of Russian origin, sjuda 'here' is the only standard Russian word. (3)

Asej, asej! Daj my kopejki! Sailor sailor Give me copecks 'Sailor, sailor! Give me some money!'

Here, daj is from the Russian daj (imperative singular); kopejki is from the Russian nominative/accusative plural; and my [mi] is from the English me. The spelling indicates a pronunciation as in the Russian first person plural personal pronoun. In 1867, Prusakevic noted the following examples of Solombala-English sentences (examples [4] to [6]): (4)

Asej? Kom milek drinkom. Sailor Come milk drink 'Sailor? Come and drink some milk.'

The verb drinkom is documented in Russenorsk too. (5)

Ol Uez! Boise dobra macka. Oh Yes very good much O h ! Yes! Very, very good.'

Russenorsk has bolsa and dobra. The macka is interesting, with mac from the English much combined with the Russian suffix -ka. (6)

Asej, asej, smotrom, bolse dobra sund.uk, vervvel' Sailor sailor look very good chest very-good skripim, gut verstom. lock good key 'Sailor, sailor, look, it's a very good chest, a very good lock, a good key.'

Here we have further examples of positive adjectives: vervvel' from the English very well, and gut, which could be German/Low German gut or even English good, with a voiced consonant in final position pronounced unvoiced according to Russian pronunciation rules. Smotrom is well

96

Ingvild

Broch

known from Russenorsk, with the Russian root smotr- as in the infinitive smotret' plus the verbal marker -om. The nouns skripim and verstom are translated for the Russian readers of the journal, it is not easy to see their origin. The word asej is used in most of the few phrases that are documented. Five of the six seem to be directed to the sailors. Sentence (5) is probably an example of how the Englishmen answered. Although they are few, these sentences nevertheless provide important clues to a better understanding of certain questions, concerning the origin of Russenorsk, since the verbs drinkom and smotrom are identical in Solombala-English and Russenorsk. Also, we note the verb baem ([bajom] < buy + om) where the English root is combined with the verbal marker -om. This particular verb is not attested in Russenorsk. But the very fact that both Solombala-English and Russenorsk exhibit the suffix -om as a verbal marker, and have at least two frequently used verbs in common, suggests a probable connection between these two languages. Still, it is difficult to ascertain which of the following three possible developments is the correct one: (i) -om may have developed first in Solombala-English and been copied at a later date in Russenorsk; (ii) -om could have developed first in Russenorsk and subsequently been introduced into Solombala-English; finally, (iii) -om may have developed simultaneously in the two pidgins. Until the precise timing of the development of SolombalaEnglish has been established it is hardly possible to determine which of these three alternatives is the correct one. These few sentences may, however, be taken to indicate that the verbal marker in Russenorsk has not only been pronounced [-um] as is claimed by Broch (1927). Because of the presumed pronunciation he rejects a possible connection between the Russian verbal suffix -om (first person plural present tense, in, e. g., pojdem [pyd'om] 'let's go'). The cyrillic spelling of Ivan Jakovlevic's Russenorsk text (Broch - Jahr 1984: 125—126) and of the Solombala-English texts does not indicate that the pronunciation should be [um]. On the contrary, the spelling indicates that it should be [om]. In North Russian dialects the pronunciation of smotrom, drinkom would be [smotrom], [drinkom]. We would also like to mention that a verbal suffix -um, although it is not used in standard Russian, does occur in North Russian dialects. The Russenorsk verb lavirum would be pronounced [laviru:m] in North Russian, while the standard Russian pronunciation of a verb of the same conjugation class would end in [-ujem], Lunden (1978: 15—16) discusses the -om suffix in Russenorsk verbs of Russian origin (although she does

Solombala-English in Archangel

97

not include this particular verb in her discussion), and I agree with Lunden that the pronunciation of the verbal suffix has varied. None of the suggested origins of the -om suffix, including imaginative suggestions about influence from the Swedish hortative (as in sjungom 'let us sing'), the Norwegian preposition om (as in Russenorsk väg om väg 'weight for weight', or trokk om trokk 'goods for goods', i. e., 'let us barter'), or even Latin (Broch - Jahr 1984, Perexval'skaja 1987), are convincing. Both the -um and the -om suffix can be found in North Russian, and although the most frequent Russenorsk verbs in the material would not have an -om ending in North Russian, a verb like podjom (cf. Russian pojdem, 'let's go') has certainly been frequent in speech.

Note 1. I thank Aleksandr N. Davydov of the Ecological Centre of the Ural branch of the Russian Academy of Sciences for drawing my attention to this material.

References Broch, Ingvild - Ernst Häkon Jahr 1984 Russenorsk — et pidginspräk i Norge [Russenorsk - a pidgin language in Norway], 2nd edition. (Tromso studier i sprakvitenskap 3.) Oslo: Novus. Broch, Olaf 1927 "Russenorsk", Archiv for slavische Philologie 41: 209-262. Davydov, Aleksandr N. 1988 Arxangel'sk vo vtoroj polovine χ ix — nacale xx veka: Naselenie — gorodskaja sreda — obscestvennyj byt [Archangel in the second half of the nineteenth century and the beginning of the twentieth century: population - town atmosphere - social behavior], Leningrad: Akademija Nauk SSSR. Fox, James A. 1973 Russenorsk: A study in language adaptivity. [Unpublished term paper, University of Chicago.] Högström, Pehr 1747 Beskrifning öfwer de til Sweriges Krona lydande Lapmarker [Description of the Lap areas belonging to the Swedish Crown], Stockholm: Lars Salvius. [1980] [Umeä: Tvä Förläggare Bokförlag (Norrländska Skrifter n. 3.)] Keilhau, Balthasar M. 1831 Reise i 0st-og Vest-Finnmarken samt til Beeren-Eiland og Spitsbergen i Aarene 1827 og 1828 [Travelling in East and West Finnmark and to Bear Island and Spitzbergen in the years 1827 and 1828], Christiania. Lunden, Siri Sverdrup 1978 Russenorsk revisited. Meddelelser 15. Oslo: Slavic-Baltic Institute, University of Oslo.

98

Ingvild Broch

Perexval'skaja, Elena V. 1987 "Russenorsk kak primer nacal'nogo etapa formirovanija pidzina" [Russenorsk as an example of the first stage in the development of a pidgin], in: Igor F. Vardul' - Vladimir I. Belikov (eds.), Vozniknovenie i funkcionirovanie kontaktnyx jazykov, materialy rabocego sovescanija [Emergence and functioning of contact languages: Materials from a workshop]. Moscow: Nauka. Prusakevic, Ivan 1867 "Solombala zimoju a letom" [Solombala in winter and summer], Arxangel'skie gubernskie vedomosti [Archangel district news] 85. (Arxangel'sk). Richter, Theodor F. Μ. 1857 Reisen zu Wasser und zu Lande. Leipzig: Arnoldische Buchhandlung. Verescagin, Vasilij 1849 Ocerki Arxangel'skoj gubernii [Sketches from the district of Archangel]. Saint Petersburg: Jakov Trej.

The Vard0 merchants' reduced Russian Siri Sverdrup Lunden

In 1862, a little book called Norsk-Russisk ordbog og Parleur [Norwegian—Russian vocabulary and phrasebook] was published in Bergen. It was unknown to Russian scholars until it was discovered in the University of Oslo library by cand. philol. Helene Strand Johansen, then a student of mine. At first glance, the book's continued anonymous existence might seem warranted. The Russian words listed in it - about 770 - are rendered in the Latin alphabet, are not in standard Russian, and are often quite unrecognizable. However, for two reasons I found that this book deserved attention: 1. it is the earliest attempt at compiling a manual of Russian for Norwegians; 2. in the preface, the author recommends his book for the use of "merchants and fishermen trading in Finnmark as well as other people wishing to become proficient in the Russian language" (my translation). I therefore undertook a study of the book in the hope of finding new material on Russenorsk. It turned out that the Russian of the vocabulary has little in common with that pidgin. Only in a single instance do we find the typical first person singular pronoun moia; in all other cases it is the normal Russian ja. But we note that this is the earliest documented example of moia. Two other words, the verbs stanum and paajcedom remind us of the Russenorsk forms with an -om ending, but they may simply be due to North Russian ekan'e (unstressed e pronounced as o), and thus the verbal ending -em would be pronounced -om, e. g., Russian stanem, North Russian stanem. Before proceeding with analyzing the material, I tried to find some information on the author. According to the title page, the book was "published by A. Hansen". Now, Hansen is a very common surname in Norway, but I am sure that I have found the right man: the wine merchant Andreas Nicolai Hansen of Vardo in eastern Finnmark, born in Kongsberg, southern Norway, in 1819. A wine merchant was just the

100

Siri Sverdrup

Lunden

sort of person who could be expected to have dealings with the Russians during the trading season. This phrasebook in fact contains a passage on buying liquor, and there are numerous passages in both Russian and Norwegian sources on the "cheap spirits" bought by the Russians in Finnmark. The case for the author being this particular A. Hansen is strengthened by an advertisement in the Troms0 newspaper Tromso Stiftstidende (no. 64, 1862) signed by A. Hansen of Yardo. He was seeking a suitable position for his "sixteen-year-old son, who has i. a. spent two years in Russia in order to learn the language". So the Vardo merchant Andreas Nicolai Hansen is regarded as the person who compiled the dictionary and had it published. His son is regarded as one of his informants. Hansen Sr. must have studied some Russian, or at least had some formal grammatical education, for he offers information on Russian grammar, though of a very reduced kind. That his practice is not in accordance with his theory will be apparent from a few examples.

Nouns Hansen writes: "Nouns are listed in the singular; in the plural they usually terminate in /". Most nouns are actually listed in the nominative singular, though a plural (or genitive?) sometimes occurs: Bomagi 'paper', Djefki 'girl'. The entry form may also be an oblique case, representing the form heard most frequently in context. In several cases, the word occurs in just this form in the conversation section, so probably it has been transferred from this part to the word list, indicating that the conversations were written first. Examples include: a. Berigo 'coast'

- cf. Russ. na beregu

b. Vremini 'time'

- cf. Russ. skol'ko vremeni

c. Vcesnoi 'spring'

- cf. Russ. vesnoj

d. Raamo 'rum'

- cf. Russ. kupil romu

e. Kitaaf 'whale'

- cf. Russ. videl kitov

f. Lcetom 'summer' - cf. Russ. letom

The Vardß merchants' reduced Russian

101

The rule is that a noun has one form only. Exceptionally, a case form is recognized: Hansen lists Daam 'house' and writes nad Daamaam 'over the house'. Gaad in the word list is translated 'year', but in the expresssion 'how many 20, 40 years', the form Gaadda is used.

Adjectives Hansen gives this rule for adjectives: "The Russian adjectives in the positive degree usually end in -oi, comparative in -e, and superlatives in -cets, e. g., kort 'short' — kaaraatkoi — kaaraatie — kaaraatces." Gender and declension are not mentioned, but agreement with a feminine noun is found in the following word groups, which are certainly very frequent: (2)

a. Bcelamoka 'white flour'; Russ. bela(ja) muka b. Aafsarna Moka 'oatmeal'; Russ. ovsjana(ja) muka c. saalaana Reba 'salted fish'; Russ. solenaija) ryba

(Contraction of the 'long' adjective ending, a widespread phenomenon in the Pomor dialect, is assumed in these cases.)

Verbs Though modern Norwegian has only one general form in the present tense, Hansen recognizes the three persons in the Russian present tense, but his grammatical description does not extend to singular/plural: Verbs in Russian in the first person singular usually end in jo, in the preterit in -/, and in the future tense in -m, e. g., veed 'know' - snajo, vidste 'knew' - snal, skal vide 'will know' - snam. In the second and third person they end in -t. The

infinitive

Practically all the Russian word forms corresponding to a Norwegian infinitive (about 125) end in though they may be based on other gram-

102

Siri Sverdrup

Lunden

matical forms. Twenty-five "infinitives" are based on a present tense, e. g., maaset, omraat, paajadaat, platset, sabudet, vrot (mozet, umret, poedet, placet, zabudet, vret). In some cases we seem to have the imperative plus -t: naleit, prenisit, paadit, skarsit (nalej, prinesi, podi, skazi) — the last form reminds us of the frequent Russenorsk expression moia ska si Ί say, I tell you'. It could represent the standard Norwegian ska{l) si, but I think the Russian imperative skazi may have played a role.

Vocabulary As mentioned before, the vocabulary is limited; it is confined to the words needed in a trading situation. The geographical horizon is restricted to eastern Finnmark and the adjacent areas of Russia. There are some topographical terms and a quite comprehensive list of the points of the compass, winds, and sea conditions. Naturally, it also includes names for types of vessels and crew members, and words used for fishing and fishing tackle, demonstrating what we already presumed — that Hansen must have had frequent dealings with Russian fishermen. There are 16 different fish names and two words for fish liver, a valuable delicacy. The list of goods brought by the Russian traders is lengthy, but it corresponds to what we know from other sources: flour, groats, wheat flour, fishing lines, longlines, hemp, linen, twine, birch bark, tarred rope, iron, peas, milk, butter, poles, logs, sail cloth, canvas, thread, and planks. There is little, of course, on fauna and flora. People and professions are included. Women, of course, do not figure prominently in Hansen's vocabulary list: Djefki 'girl', Devitsa 'girl', Tsaanka 'a married woman' and Kasasika 'a female servant'. Male professions are well represented, though they are confined to trading, fishing, customs officers, etc.

Conclusion The language found in Hansen's book is not a pidgin; it is a drastically reduced form of Russian. In contrast to Russenorsk, the vocabulary is

The Vardo merchants' reduced Russian

103

entirely Russian. It is limited to the circumstances in which the two parties met - fishing, trading, and in the case of Hansen Jr., apprenticeship. An interesting aspect of the vocabulary is that it is typical North Russian, with a number of dialect words not found in ordinary dictionaries. In spite of the rudimentary spellings, we can clearly discern North Russian phonetic features — okan'e, ekan'e, etc. But is it more than an individual reduced Russian spoken by the Hansens? I think so, considering the conditions of learning. Hansen Sr., as a wine merchant, had dealings with the Russians who came to Finnmark to fish and trade during the summer season, as did many other Finnmark merchants. His son learnt his Russian under conditions of apprenticeship, as did many other young Norwegians. Examples from these situations are given in the appendix. Obviously, the Russians speaking to these Norwegians used a "simplified" Russian, but not the Russenorsk that the less "educated" people used. It is unthinkable that A. Hansen would have printed this book at his own expense unless he was sure that there was a market for it. Several other Finnmark merchants must have learnt their simplified Russian from this book.

References Broch, Ingvild - Ernst Hakon Jahr 1984 Russenorsk - et pidginspräk i Norge [Russenorsk - a pidgin language in Norway]. 2nd edition. (Tromso-studier i sprakvitenskap 3). Oslo: Novus. Broch, Olaf 1927a "Russenorsk", Archiv for slavische Philologie 41: 209-262. 1927b "Russenorsk", Maal og Minne 8 1 - 1 3 0 . 1930 "Russenorsk tekstmateriale" [Russenorsk text materials], Maal og Minne 114-140. Hansen, A[ndreas Nicolai] 1862 Norsk—Russisk Ordbog og Parleur [Norwegian-Russian vocabulary and phrasebook], Bergen: Η. I. Geelmuydens Enkes Officin.

Appendix — Hansen 1862: 27—28 Hvad skylder jeg dig nu? 'How much do I owe you now?' 5000 Rubel. '5000 rubles.'

Staa ja daalsnoi tebe tepere? (Russ. Skol'ko ja tebe dolzen?) Pcettitsetsa Rublie. (Russ. Pjat' tysjac rublej.)

104

Siri Sverdrup

Lunden

Har du skrevet det? 'Have you written it down?' Det er skrevet. 'It is written.' Jeg skal imorgen betale det. Ί shall pay it tomorrow.' Du er atter her? 'You are here again?' Ja, jeg erindrer, hvad jeg skylder dig. 'Yes, I remember what I owe you.' Vil du gjere Afregning? 'Will you square your account?' Jeg har kort Tid, i Aften reiser jeg. Ί have not much time, I am leaving tonight.' Nu er Veiret imod. 'Now the wind is contrary.' I Aften bliver det godt Veir. 'Tonight the weather will be good.' Hvoraf veed du det? 'How do you know?' See du hvorledes Skyerne flyver i Luften. 'Do you see how the clouds fly in the air?' Staa op din lade Hund, hvorfor sover du saaleenge? 'Get up, you lazy dog, why do you sleep so long?' Hvormange er Klokken nu? 'What time is it now?' Solen er i Syd. 'The sun is in the south.' Jeg har arbeidet meget inat. Ί worked much in the night.'

Te cetaat sapiscenna? (Russ. Ty eto zapisal?) Sapisanna. (Russ. Zapisano.) Ja buda saftra sapletit cetaat. (Russ. Ja zaplacu zavtra.) Te opcet stes? (Russ. Ty opjat' zdes'?) Jes, ja paamino, staa ja daalsnoi tebe. (Russ. Da, ja pomnju, skol'ko ja tebe dolzen.) Te nadaat raasetaimse stanum? (Russ. Ty xoces' rascitat'sja?) Omina kaaraatkos Vremina, ja Vcetseri paajcedom. (Russ. U menja malo vremini, ja uezzaju vecerom.) Tepere stret paadvoeter. (Russ. Veter teper' vstrecnyj.) Vcetsero kaaraasaa Vceter budet. (Russ. Vecerom pogoda budet xorosaja.) Lcetsaagaa te snat cetaat? (Russ. Otkuda ty eto znaes?) Smaatri, kuda Oblakaa Ice tit Djaam. (Russ. Posmotri, kak letjat ν vozduxe oblaka.) Stavai leenivoi Subaka satsam te daalge spiit? (Russ. Vstavaj, lenivaja sobaka, zacem tak dolgo spis'?) Kaalke Tsasa teperi? (Russ. Kotoryj teper' cas?) Saansa Icetniki. (Russ. Solnce uze na juge.) Ja Naatso mnaagaa robaatat. (Russ. Ja noc'ju mnogo rabotal.)

The Vardß merchants' reduced Russian

Hos hvem? 'With whom?' Hos Smeden. 'With the blacksmith.' Staa op, tag Pakhusnogelen og gaa at veie Fisk. 'Get up, take the key to the storehouse and go weigh fish.'

105

Okaagaa? (Russ. U kogo?) Ο Kutsnitsi. (Russ. U kuzneca.) Stavai, vaasmi Klüts na Ambare i paadi vesi Reba. (Russ. Vstavaj, voz'mi kljuc ot ambara i idi vesat' rybu.)

On the pidgin status of Russenorsk Ernst Häkon Jahr

1. "Whinnomian" pidgin development and Russenorsk In situations around the world where pidgins have developed, it is most often the case that those who have created the pidgin have had a colonial and, therefore, socially superior target language from which to draw the main bulk of vocabulary, and their own substrate languages (two or more) from which they develop the pidgin grammar. When speakers from the different substrate languages use the pidgin for communication between themselves, we have a situation of "tertiary hybridization", in Keith Whinnom's (1971) terminology. Whinnom claims that a pidgin proper has a stable norm of grammar and that it contains "certain linguistic items (structures, etc.) which are not immediately assignable to the native or to the target language" (1971: 105). The "tertiary hybridization" is, according to Whinnom, one of two ways of pidgin development to the attainment of such a stage ("relexification" being the other). For the dual-source idiom Russenorsk, however, the social setting was quite different. There was no social difference between the two parties, the Norwegians and the Russians. Therefore, no target language in Whinnom's terms existed. From the outset, Russian was the target language for the Norwegians; for the Russians it was Norwegian. Whinnom argues quite convincingly that a stable pidgin cannot develop between one target language and only one subordinate or substrate language. (His example is "Cocoliche", used extensively by Italian immigrants in Argentina, but not by Argentinians.) However, if it is possible to demonstrate that Russenorsk exhibits the Whinnomian criteria for a proper pidgin (a stable norm of grammar which includes certain innovations), it can be argued that a pidgin may come into being through a third process in addition to "tertiary hybridization" and "relexification". That is, longstanding contact between two socially equal languages may also yield a pidgin proper if the need for learning each other's language does not exist due to the limitation of the direct contact. The aim of this paper is to discuss Russenorsk in the context of these questions.

108

Ernst Häkon Jahr

2. The social history of Russenorsk Russenorsk was used for trade and barter by Norwegians and Russians in northern Norway (the counties Finnmark and Troms) for a period of approximately 150 years, from the end of the eighteenth century up until the second decade of the twentieth century. The language originated among Norwegians and Russians when bartering fish, flour, and grain in northern Norway. The contact was limited to a few summer months each year. The barter business was highly favorable for both parties and therefore developed in spite of being illegal in the beginning. The Russians had a grain surplus but lacked fish, the Norwegians had fish but lacked grain. The Russians bought the fish at a time of the year (summer) when it was difficult to sell in Norway, and they were not too fussy about the quality of the fish. The Russians came from different villages along the Murman Coast and from Kem' and Sumskojposad by the White Sea. They usually returned to the same places in northern Norway every summer. Their main destinations were Vardo, Vads0, Hammerfest, and Troms0. Once in a while they also went further south, to the Lofoten islands. This trade is usually referred to as the "Pomor trade". (The White Sea coast is called Pomor'e in Russian, and those who live there and along the Murman coast are accordingly called "Pomors".) In the beginning of the nineteenth century, Russenorsk spread to the most important trading centres in Finnmark and Troms. The area in which Russenorsk is reported in use up until 1842 stretches from Kola in Russia in the east to Tromso in Norway in the west. Around the middle of the nineteenth century, an interesting change in the social status of Russenorsk seems to have taken place (cf. Broch — Jahr 1983). Before 1850, Russenorsk was commonly used by Norwegians, both fishermen and merchants, in dealing with the Russians, and Russenorsk enjoyed a high status in society. After 1850, it was mostly fishermen who used Russenorsk, because the merchants learned more Russian by spending a year or two with colleagues in Russia and also by actively studying Russian in Archangel. According to Tsjudinov (1992: 29), in 1856 more than ten Norwegians studied Russian in the Archangel area. Subsequently, the merchants of Finnmark developed their own grammatically simplified variety of Russian (described in this volume by Siri Sverdrup Lunden). It is striking that the available reports and evaluations of Russenorsk on the whole are more positive before 1850 than after. The reason for this is that as long as the merchants used - and had to use - Russenorsk,

On the pidgin status of Russenorsk

109

it was socially accepted by the local upper classes as a language equal to other languages. But when the merchants started learning Russian, the social status of Russenorsk was devalued. This change in social status is, of course, not surprising. On the contrary, it is entirely what could be expected when the local upper classes stopped using Russenorsk and started using Russian proper, and Russenorsk as a result was limited to the common fishermen. Thus, around and after 1850, the social evaluation of Russenorsk was downgraded to the status which characterizes most pidgins in the world. In the second half of the nineteenth century, Russian trade reached a substantial level. Several sources claim that the Russian grain prevented starvation in northern Norway during the famine years in the beginning of the century, and Norwegian expansion and activity in Finnmark during the nineteenth century would have been unthinkable without the Russian trade. Around the turn of the century, trade with Russia changed in character from basically a form of barter, becoming increasingly cash-based. On both sides the big merchants often understood the language of their trading partner: hence the need for Russenorsk declined. The ability to speak Russenorsk presumably became less widespread as a result of the disappearance of the communicative situation in which it had been used for 150 years. The Russian revolution in 1917 put a definitive end to this trade and consequently to the use of Russenorsk.

3. Short survey of Russenorsk The Russenorsk texts which are available today consist of single words and isolated sentences, various word lists of different lengths, and conversations in the form of dialogues. 1 The text material comprises some 400 different words. However, about 50 percent of them appear only once. This means that the vocabulary of Russenorsk consisted of a core of 150-200 words. The characteristic features of Russenorsk can be summarized briefly as follows: i. Russenorsk phonology is based on Norwegian and Russian, but sounds and consonant clusters not found in both languages are avoided or simplified.

110

Ernst Häkon Jahr

ii. The most frequently used pronouns are moja ('I/me/my') and tvoja ('you/your'). iii. The verbs have no markers for tense, aspect, or person. The suffix -om is a general verbal marker, though it is not always used. (On aspect in Russenorsk, cf. Kotsinas, this volume.) iv. Nouns are not declined, and the suffix -a seems to have a nounmarking function, like -om for verbs. There is no copula, but liggel liggom 'lie down' or stannom 'stand' are used with existential meaning. v. pä is in principle the only preposition in Russenorsk. vi. Coordinating conjunctions are i,ja,jes. Kak functions as a subordinating conjunction (and as a frequently used question word). vii. The vocabulary is mostly of Norwegian or Russian origin, but some lexical items are derived from other languages (Low German/Dutch, English, French, Swedish, Sami). For many concepts both a Russianbased and a Norwegian-based item are attested. viii. The fundamental word order is SVO. Sentences are combined paratactically by means of juxtaposition or conjunctions; embedding and subordination are attested. Still, the syntactic possibilities are quite restricted. The largest variety seems to have developed in interrogative sentences, which is not unexpected considering that Russenorsk was used to ask questions about prices and barter for merchandise. It should be noticed that, although Russenorsk was not a difficult language to acquire, at least not for a Norwegian-speaking person, it seems that no one had direct and immediate access to it. This is reported in several sources (cf. Broch - Jahr 1981: 149, 1984: 58). Thus, like any other language, Russenorsk had to be learned.

4. 4.1.

The origin and development of Russenorsk "Grammaticalization"

Referring to Whinnom (1971) and to Silverstein's (1972) analysis of Chinook jargon, Fox concludes that Russenorsk is "not a true pidgin" (1973: 25). Instead he argues that Russenorsk originated as an international base vocabulary which was later "grammaticalized" to a certain degree. (His view is repeated and further elaborated in Fox 1983.) However, according to Fox, each Russenorsk speaker depended for the most part

On the pidgin status of Russenorsk

111

on his primary grammar - in a drastically reduced form. Accordingly, Russenorsk should be seen as reduced Norwegian and Russian, and Fox claims that the lexical variation evidenced in Russenorsk excludes it as a pidgin (1973: 22). Russenorsk never reached a stability "which might tempt one to write a 'grammar of Russenorsk' without reference to the primary languages" (1973: 41). Such a grammar was, however, suggested by Dan I. Slobin a few years after Fox made this claim (Slobin 1977, 1979; see below). Also, Fox's position is not compatible with the reported observation that no one had immediate access of Russenorsk, but had to learn it (cf. above). Arnbjornsdottir and Smith (1986) contribute to this discussion with an analysis of Fox's list of international words in Russenorsk. 2 Although Fox (1973) calls the number of international words "impressive" (1973: 27), it contains altogether only about 20 words (1973: 28), or about 10 percent of the core vocabulary in Russenorsk. In addition, Arnbjornsdottir and Smith question the status of most of these words as "international", pointing instead to possible Norwegian or Russian origin. I cannot follow Arnbjornsdottir and Smith to more than a certain extent in their analysis of these words. (Their etymologies, or suggested possible Norwegian etymologies concerning Russenorsk words like sprek, njem, vater, vat, better, slipom, are clearly unfounded.) Still, I agree with their main view that Fox has not been able to substantiate his view concerning the origin of Russenorsk from an already existing international contact vocabulary that later was, in Fox's terminology, "grammaticalized".

4.2.

"Relexification"

Arnbjornsdottir and Smith also object to what they see as Broch and Jahr's (1981) hypothesis about a possible Russenorsk relexification of a pidgin which existed in the area prior to Russenorsk, e. g., the SwedishSami Borgarmälet. I agree with their refutation of such a hypothesis connected only with Borgarmälet, but fail to see that Broch and Jahr have made such a suggestion as anything more than one possibility among others. We wanted to draw attention to Borgarmälet in our 1981 book (and also in Broch - Jahr 1980), since Fox (1973) in support of his own hypothesis about Russenorsk's origin explicitly stated that no such language existed in the area prior to Russenorsk. In a later paper, when we had published all the available details on Borgarmälet, Fox also reported the existence of this language (Fox 1983: 98). 3

112

Ernst Häkon

Jahr

In Norway, no Norwegian - S a m i pidgin has been reported, although Qvigstad (1899: 11) mentions " 'gavppe-daro' (das für den handel nötige norwegische)" (from Sami gavppe 'trade' and daro 'Norwegian') (also mentioned by Heiland 1906: 419). On the other hand, Brooke (1823: 175) reports that the merchants of Finnmark "knew" Sami. It is not unlikely that what Brooke observed about the merchants' use of Sami together with Qvigstad's "gavppe-daro" constitute a close parallel to the eighteenth-century Swedish-Sami Borgarmälet. Also, one may draw attention to still another early pidgin in the area, "Solombala- English". Little is known about this nineteenth-century (and perhaps even older) English—Russian pidgin in Archangel, but it will probably prove more important to the discussion of Russenorsk's origin and development than Borgarmälet (cf. Broch, this volume). Arnbjornsdottir and Smith (1986) suggest, after they have refuted both Fox's (1973) "grammaticalization" hypothesis and also a possible relexification of earlier pidgins in the area, that Russenorsk developed as a pidgin from the long-lasting contact between two socially equal languages. They do not, however, substantiate this suggestion with any argument based on Russenorsk data. In the following, I want to discuss some arguments based on available Russenorsk data to support the view that Russenorsk contained a stable grammatical core, and also that it exhibits some linguistic items which are not easily assignable to either of the two base languages - in other words, that Russenorsk shows a possible innovation quality.

5. A "grammar of Russenorsk" In order to use Russenorsk as an example of an extreme minimum language as regards grammatical constructions, Dan I. Slobin (1979: 4 3 46) lists a total of eight rules which in his view give an exhaustive description of the grammatical structure of Russenorsk (cf. also Slobin 1977). The grammatical rules given by Slobin are the following, and it should be noted that they are formulated without any reference to either Norwegian or Russian (my comments in brackets): 1. A word which refers to a person, place, thing, conceptual entity, etc., is a noun. [Here he could have mentioned that the suffix -a is often in use as a noun marker, however, not as clearly as the suffix -om for verbs, cf. 2.]

On the pidgin status of Russenorsk

113

2. A word which ends in -om is a verb. [But not all verbs exhibit this marker.] 3. The noun preceding a verb is the agent, and the noun following the verb is the patient of the action referred to by the verb. [This rule establishes the fundamental SVO structure of Russenorsk. Perexval'skaja (1987: 65) posits an SOV structure, but she is clearly mistaken, see Broch - Jahr 1981: 54-55, 1984: 41-42.] 4. In a sequence of two nouns, the first noun is the possessor and the second is the thing possessed. [In the entire Russenorsk corpus, this rule only applies to kuasalt ("cow salt", i. e., 'salted meat') and kuasjorta ("cow shirt", i. e., 'cowhide'), both of which should be analyzed as compound words. Instead, a possessive relationship in Russenorsk is expressed by a pronoun preceding the noun, or by a prepositional phrase: kor ju stova? 'where (is) your house?', ρά moja stova 'in (lit. 'on') my house', klokka ρά ju "watch on you", i. e., 'your watch'.] 5. Interpret a sequence of noun —ρά—noun as any semantically and contextually plausible combination, excluding possession. [Since rule 4 is inaccurate, it follows that the part of this fifth rule which concerns possession is likewise false.] 6. A sequence of words with a rising intonation is a question which can be answered by "yes" or "no". 7. A sequence of words beginning with a question word ("what", "when", "how many", etc.) is an informational question. [Kak is in principle the only wh-word in Russenorsk.] 8. A sequence of words ending with falling intonation is a declaration. This concludes Slobin's overview of Russenorsk grammar. As stated by Broch and Jahr (1981: 61), Slobin has obviously oversimplified the grammar of the language. For example, he posits (1977: 201) that Russenorsk exhibits no conjunctions and no embedded constructions. This is not the case, since Russenorsk had developed both coordinating conjunctions (e. g.,ja,jes\ see below for the origin of these conjunctions), and a subordinating one (kak, cf. Broch - Jahr 1984: 38, 41). 4 Thus, although Slobin's grammatical rules do capture and account for a large part of the known Russenorsk data, they still oversimplify the structure of Russenorsk. His "Russenorsk grammar" is not exhaustive and not completely accurate. However, the main point here is that Slobin has demonstrated that it is possible to write grammatical rules for Russenorsk without bringing Norwegian or Russian grammar into the description. Fox's (1973) claim

114

Ernst Häkon Jahr

that no one would even attempt to do so, is evidently refuted. Slobin's grammatical rules for Russenorsk show that a grammatical core can indeed be abstracted from the Russenorsk texts, and, with the necessary additions, these rules can account for "the grammar of Russenorsk". It is also important to notice that the Russenorsk texts - although some of them show considerable variation, especially in vocabulary - nevertheless for most of the more fundamental grammatical rules exhibit amazing consistency over a period of almost 150 years.

6. The question of grammatical innovations in Russenorsk An important defining feature of a "Whinnomian" pidgin is, as mentioned above, that it shows some grammatical innovation quality, i. e., features not derivable from any of the base languages. For Russenorsk it is possible to point to some features in morphology and syntax which seem to have originated within Russenorsk itself or in connection with other Arctic pidgin languages. 6.1.

Morphology

In morphology, the verbal marker -om is perhaps the best candidate for a Russenorsk trait not derivable from Norwegian or Russian grammar. Although the etymology of the marker -om has been much discussed and several suggestions have been advanced (cf. Broch — Jahr 1984: 35-36), 5 the use of -om as a general verb marker did in all probability develop in connection with the English-Russian Archangel pidgin mentioned above, which also has the om-ending. 6 The suffix -mann to designate nationality/ethnicity (e. g., russmann 'Russian', burmann 'Norwegian', filmann 'Sami') or occupation (e. g., kukmann 'merchant') is also an interesting Russenorsk feature, although this suffix appears in Norwegian as well. It is attested in Russenorsk as early as 1785 (Broch - Jahr 1984: 49) and is found also in the eighteenthcentury Swedish-Sami Borgarmälet (Broch - Jahr 1984: 52), as well as in a not too well documented pidgin in Iceland, Fäskrüösfjardarfranska (Noreen 1911, Bakker 1989). Finally, it is found in the word kukmann 'merchant' in an eighteenth-century Swedish-Russian word list and phrase book, probably used by the Pomors (Broch - Jahr 1984: 53-54). Although Norwegian does have this suffix, it is not improbable that the

On the pidgin status of Russenorsk

115

use of -mann in Russenorsk is in some way connected with the use of the same suffix in other Arctic contact languages. It should also be mentioned that the tendency in Russenorsk to mark nouns and adjectives with an α-suffix is paralleled in the seventeenthcentury (i. e., pre-Russenorsk) Icelandic-Basque pidgin attested in Iceland. In this pidgin, nouns and adjectives end in -a (Hualde 1984: 51, cf. Bakker 1987). 6.2.

Syntax

In Russenorsk syntax, two rules can be especially highlighted. The first is the tendency to have the verb in final position when the sentence contains an adverbial. This word order is marginal in Russian and impossible in Norwegian. Sentences without adverbials normally exhibit SVO ordering in Russenorsk: Sentences without adverbials: (1)

tvoja kopom oreka? you buy nut 'You buy nuts?'

(2)

moja kopom fiska. I buy fish Ί buy fish.'

(3)

tvoja kralom tros. you steal hawser 'You steal the hawser.'

Sentences with adverbials: (4)

moja tri vekkel stannom. I three week stand Ί stayed three weeks.'

(5)

moja ette ort perevoj ras pä Norge stannom. I this year first time on Norway stand Ί am for the first time in Norway.'

116

Ernst Häkon Jahr

(6)

moja tvoja pä vater kasstom. I you on water throw Ί shall throw you in the water.'

(7)

davaj pä moja skip kjai drinkom. please on my ship tea drink 'Please, come and have tea on my ship.'

The second rule states that the negator (i. e., Norwegian ikke or Russian njet) is restricted to second position, a rule not found either in Norwegian or in Russian: (8)

etta njet dobra. this not good 'This is not good.'

(9)

moja njet lygom. I not lie Ί don't lie.'

(10)

pä den dag ikke russefolk robotom. on that day not Russians work 'On that day, Russians do not work.'

There are, however, exceptions to both rules. Therefore, they represent tendencies more than absolute rules. The following sentence, for example, violates both rules: (11)

mangoli är moja njet smotrom tvoja. many year I not see you Ί have not seen you for many years.'

According to the two syntactic rules in question, this sentence "should" have been like this: (12)

mangoli är njet moja tvoja smotrom. many year not I you see

On the pidgin status of Russenorsk

117

7. On the Sami impact on Russenorsk Possible Sami impact on Russenorsk has been discussed earlier. It is doubtless the case that Sami speakers were extensive users of Russenorsk. Julius Aars, a priest in Finnmark as well as an MP, wrote in a newspaper article (Morgenbladet No. 205, p. 2) in 1849 that "In the Samis' trade with the Russians, the common medium of communication is a kind of Norwegian", i.e., Russenorsk (cf. also Fox 1973: 12-17, Broch - Jahr 1984: 51-53). 7 However, the unfortunate lack of texts from Sami Russenorsk speakers (this is also, by the way, true for the Finnish Russenorsk speakers) makes it difficult to determine the exact role of Sami in the development of Russenorsk. Olaf Broch has previously discussed (1927: 257; cf. also 1928) the possibility of an older, more original layer of Sami fish names in Russenorsk. Likewise, the suffix -a in nouns may partly be attributed to Sami influence (Broch - Jahr 1984: 34), although several other origins have also been suggested (Broch - Jahr 1984: 34). However, the use of ja as a coordinating conjunction in Russenorsk is perhaps the best candidate for a specific Sami contribution to Russenorsk grammar. In Russenorsk, the Russian da 'yes' is used in the affirmative meaning only, not as a conjunction with the meaning 'and', which it may have in Russian. Instead, the word ja is used in Russenorsk (together with jes < Eng. yes) meaning both 'yes' and 'and'. In Norwegian ja means only 'yes', but in Sami (and Finnish) ja means 'and'. In Russenorsk we see, then, that the word ja has a meaning derived from both Norwegian ('yes') and Sami ('and'), making a parallel to Russian da.8 The relationship between Russian da, Norwegian ja, Sami ja and Russenorsk jaljes can be illustrated as in Figure 1. Norwegian Russian Sami Figure 1. The relationship between Russ. da, Norw. ja, Sami ja, and Russenorsk jaljes

8. Lexical variation Lexical variation is a characteristic feature of pidgin languages in general, and also of Russenorsk. Fox (1973: 22), however, dismisses Russenorsk

118

Ernst Häkon Jahr

as a pidgin because of its great lexical variation and claims that "a pidgin would not be characterized by the sort of lexical variation which we find in Russenorsk". Fox does not elaborate on what sort of lexical variation could, in his view, be acceptable for a pidgin proper, but his claim cannot in any case be defended when we take into account the great variation we find in most pidgins. Even if the lexical variation in Russenorsk is considerable, it is still well within the variation we have to allow for in pidgins in general. Fox's opinion must therefore be dismissed. 9

9. Conclusion A cautious conclusion is that the dual-source idiom Russenorsk was a language very close to being a pidgin even in the "Whinnomian" sense (i. e., according to Whinnom 1971). To a stabilized grammatical and lexical core, a core that was transmitted from generation to generation, a great variety of lexical items were added when the situation called for it. In order for the parties to understand each other outside the realm of direct bartering, one had to rely on whatever knowledge one had of words from each other's language. In this respect, Russenorsk was a very restricted language. Nevertheless, it did have a core grammar of its own, constructed in accordance with pidgin grammars as they are found all over the world, and with certain features and rules not directly derivable from, or paralleled in, Norwegian or Russian. In addition, Russenorsk exhibited a core vocabulary oriented toward the barter business, including words for fish, prices, goods of different kinds, and words connected with a limited range of social interactions, like tea drinking. A few elements in Russenorsk can be shown to be possible contributions to the language from Sami speakers of Russenorsk. 10

Notes 1. All known Russenorsk texts up to 1981 are included in Broch - Jahr (1981: 107-147). A few new texts were added in the second edition (1984). Here, I use the opportunity to include the texts which have been found after 1984. A. Edle Hartmann 1936 [1987], Hartmann's examples of Russenorsk are from Tromse in the 1870s:

On the pidgin status of Russenorsk (i)

(ii)

(iii) (iv) (v)

(vi)

119

kak ju snakka? what you speak 'What do you say?' kanski den prinsipal pä stova? maybe that skipper on house 'Is the skipper present?' yes 'yes' njet 'no' fiska kopum. fish buy Ί want to buy fish.' grot slipum? big sleep 'Slept well?'

B. Helge Stangnes 1986: 7 8 - 8 0 . (i)

(ii)

(iii)

moja grot krank pa guano. I big ill on (fish-)head Ί have a big headache.' pä gonora skjcerom. on head cut 'Cut off the head (of the fish).' kanali norfar kodi boska? crazy Norwegian where bucket 'Crazy Norwegian, where is the bucket?'

The following "verse" is found on the back side of a bill from 1895: (iv)

kupi buy kupi buy

vina, jibi wine, fuck bolsa, jibi more, fuck

mina me dolsa longer

C. Sentences gathered by Helge Stangnes: (i)

(from an informant on the island of Senja) russmann, grot frott! Russian, big louse '(Hey), Russian, you have a lot of lice.' (ii) (from Harald Lind, Kiberg in Finnmark) stari gammel, snart pä kjcerka slipom. old old, soon on church sleep Ί am old, I shall die soon.' [said by an old Russian skipper] 2. I thank Ian F. Hancock for making Arnbjornsdottir and Smith's paper available to me. 3. Fox (1973: 36): "It seems reasonable to conclude that no continuous pidgin or jargon tradition can be established, nor is one likely, for this area prior to the development of Russenorsk". Fox (1983: 98): "In looking for conceivable antecedents of R N , I was

120

Ernst Häkon Jahr

astounded at the richness of exotic language varieties in the linguistic repertoire of Scandinavia. It includes a seventeenth-century Swedish trade pidgin (Borgarmäl), used in the southern Lappmarks of Sweden;...". 4. The following sentence shows examples of both coordination (using ja and jes) and subordination (using kak)\ (i)

5.

6.

7. 8.

9. 10.

kak ju vil skaffom ja drikke te, davaj pa sjib tvoja ligge ne if you will eat and drink tea, please on ship your lie down jes pä slipom. and on sleep 'If you want to eat and drink tea, please come on board your ship and lie down to sleep'

Bakker (1987: 19) suggests that p ä too can be used as a complementizer. He bases his view on this sentence: davaj pä skib kom, brat, pä tjei drikki (lit. 'please on ship come, brother, on tea drink'), where the second pä is interpreted by Bakker as a complementizer. It seems more plausible to view this use of pä as parallel to what we find in sentences like these (also quoted by Bakker): sprek pä moja (lit. 'talk on me'), moja pä tvoja kludi (lit. Ί on you beat'). Cf. also Belikov - Perekhvalskaya (1989: 288). Cf. Lunden (1978: 16), who suggests that -om could have developed within Russenorsk itself from the phrase trokk om trokk ('ware for ware') > trokkom trokk ('let us barter'), in view of the frequent occurrence of the expression trokk om trokk. Arnbjornsdottir and Smith's (1986: 9) reference to the Old Norse (dative plural) ending -om is clearly irrelevant. No northern Norwegian dialects north of the Lofoten islands kept the dative case or the dative plural om-ending till the eighteenth century. My translation from the Norwegian original. I thank Bente Martinussen for making me aware of this statement by Aars. According to Belikov and Perekhvalskaya (1989: 289), O. Broch (1927), in his translations, gives Russenorsk jes the meaning 'yes' only. Although it is the case that Broch obviously has given some incorrect translations, he nevertheless remarks (Broch 1927: 233) that jes may also be used for 'and', parallel to Russian da. Note that both Suzanne Romaine (1988: 124-130) and Jean Aitchison (1991: 181) in recent books have referred to Russenorsk as an example of a stable pidgin. Perexval'skaja (1987) views Russenorsk as a Norwegian - not a Russian - pidgin, and therefore as a Germanic and not a Slavic language. I do not see the direct importance of this distinction, especially since the notions "Germanic" and "Slavic" refer exclusively to the Stammbaum theory, and it is evident that this theory does not account for the development of pidgin and Creole languages. A bias towards Norwegian may rather be seen as a consequence of the fact that almost all the available Russenorsk texts are collected from Norwegian speakers of Russenorsk. Also important is the fact that Russenorsk was used almost exclusively in Norway, not in Russia.

References Aitchison, Jean 1991

Language change: Progress or decay? Second edition. Cambridge: Cambridge University Press.

On the pidgin status of Russenorsk

121

Arnbjornsdottir, Birna - Constance S. Woody Smith 1986 Russenorsk. [Unpublished seminar paper, University of Texas at Austin.] Bakker, Peter 1987 "A Basque nautical pidgin: A missing link in the history of fu", Journal of Pidgin and Creole Languages 2: 1-30. 1989 "A French-Icelandic nautical pidgin", Journal of Pidgin and Creole Languages 4: 129-132. Belikov, Vladimir - Elena Perekhvalskaya [Perexval'skaja] 1989 "Russian Norwegian po reconsidered", Journal of Pidgin and Creole Languages 4: 287-289. Broch, Ingvild - Ernst Häkon Jahr 1980 "The problem of stability in Russenorsk", in: Even Hovdhaugen (ed.), The Nordic languages and modern linguistics, [4], Oslo - Bergen - Tromsa: Universitetsforlaget, 242-250. 1981 Russenorsk - et pidginsprak i Norge [Russenorsk - a pidgin in Norway], (Troms0 Studies in Linguistics 3.) Oslo: Novus. 2 ed. 1984. 1983 "Does a pidgin necessarily have a low social status? The case of Russenorsk", Nordlyd: Tromso University Working Papers on Language and Linguistics 7: 36-45. 1984 "Russenorsk: A new look at the Russo-Norwegian pidgin in northern Norway", in: P. Sture Ureland - Iain Clarkson (eds.), Scandinavian language contacts. Cambridge: Cambridge University Press, 21 - 6 5 . Broch, Olaf 1927 "Russenorsk", Archiv for slavische Philologie 41: 209-262. 1928 "Noen fiskenavn i 'russenorsk'" [Some fish names in Russenorsk], in: Det Norske Videnskaps-Akademi i Oslo. Arbok 1927 [The Norwegian Academy of Science and Letters in Oslo. Yearbook 1927], Commissioned by Jacob Dybwad. Oslo: Norske Videnskaps-Akademi, 10. Brooke, A. de Capell 1823 Travels through Sweden, Norway, and Finmark, to the North Cape, in the summer of 1820. London: Rodwell and Martin. Fox, James A. 1973 Russenorsk: A study in language adaptivity. [Unpublished term paper, University of Chicago.] 1983 "Simplified input and negotiation in Russenorsk", in: Roger W. Andersen (ed.), Pidginization and creolization as language acquisition. Rowley, MA: Newbury House, 94-108. Hartmann, Edle 1936 Kjentfolk gjennom ärene [People of renown over the years]. Oslo: Nikolai Olsens boktr. [1987] [reprinted in Nordlys, 26 September.] Heiland, Amund 1906 Topografisk-statistisk beskrivelse over Finmarkens amt. [Topografic-statistical description of the county of Finnmark] Vol. 2. Kristiania [Oslo]: Aschehoug. Hualde, Jose I. 1984 "Icelandic Basque pidgin", Journal of Basque Studies in America 5: 41-59. Lunden, Siri Sverdrup 1978 "Russenorsk revisited", Meddelelser 15. Oslo: Slavic-Baltic Institute, University of Oslo.

122

Ernst Häkon Jahr

Noreen, Adolf 1911 "Scandinavian languages", Encyclopedia Britannica, vol. 24: 293. Perexval'skaja, Elena V. 1987 "Russenorsk kak primer nacal'nogo etapa formirovanija pidzina" [Russenorsk as an example of a first step in pidgin formation], in: Igor F. VarduF — Vladimir I. Belikov (eds.), Vozniknovenie i funkcionirovanie kontaktnyx jazykov, materialy rabocego sovescanija [Emergence and functioning of contact languages: Materials from a workshop]. Moscow: Nauka, 63-67. Qvigstad, Knut Just 1899 "Über die Geschichte der lappischen Sprachforschung", Journal de la Societe Finno-Ougrienne 16(3): 11-29. Romaine, Suzanne 1988 Pidgin and creole languages. London—New York: Longman. Silverstein, Michael 1972 "Chinook jargon: Language contact and the problem of multi-level generative systems", Language 48: 378-406, 596-625. Slobin, Dan Isaac 1977 "Language change in childhood and in history", in: John Macnamara (ed.): Language learning and thought. New York: Academic Press, 185-214. 1979 Psycholinguistics. Second edition. Glenview, IL: Scott, Foresman and Company. Stangnes, Helge 1986 " - 'som han sa, russen' - Om restar av det russenorske spraket i NordNorge" [ - 'like he said, the Russian' - On remnants of Russenorsk in North Norway], Arbok for Senja 15: 78-80. Tsjudinov, Konstantin [Cudinov] 1992 "Norskundervisning i Nord-Russland i pomortida" [Teaching of Norwegian in northern Russia in the Pomor period], Ottar (University of Troms0) No. 192: 29-31. Whinnom, Keith 1971 "Linguistic hybridization and the 'special case' of pidgins and Creoles", in: Dell Hymes (ed.), Pidginization and creolization of languages. Cambridge: Cambridge University Press, 91-115.

Aspect marking and grammaticalization in Russenorsk compared with Immigrant Swedish UUa-Britt

Kotsinas

It is well known that aspect is a grammatical category which tends to be expressed early in child language (Antinucci - Miller 1976, Bronckart Sinclair 1973), and that grammatical markers of aspect are created at an early stage when pidgins creolize (Bickerton 1981, Traugott 1978, Mühlhäusler 1986). There also seems to be a diachronic tendency in languages for TMA systems (tense, modality, aspect) to change more frequently than other subsystems of language (Traugott 1978, Fleischmann 1982). This suggests that aspect is an essential category, which could be expected to be expressed in some way also in pidgins and at early stages of second-language acquisition. In the now extinct trade pidgin Russenorsk, however, neither tense nor aspect seems to have been expressed (Broch - Jahr 1981: 58), in spite of the fact that other tendencies in the preserved texts indicate that Russenorsk under certain circumstances could have developed functionally and grammatically into a Creole (Broch - Jahr 1981: 16). A comparison between Russenorsk and other pidginlike varieties, however, reveals similarities as regards the use of certain lexical items of a type which in the creolization process are often used to form aspect markers.

The marking of aspect Even if aspect marking is inclined to appear early in the processes of firstlanguage acquisition and creolization, this does not imply that aspect is always manifested in the same way in the emerging language as in the language that forms the model for the speaker, i. e., the target language for first- or second-language learners or the lexifying language of the pidgin/creole. When children first use tense forms, for example, they tend to attribute an aspectual, rather than a temporal, meaning to these forms (Bronckart - Sinclair 1973). When pidgins creolize, on the other hand, there seems to be a tendency not to use the existing morphological or periphrastic means of the base

124

Ulla-Britt

Kotsinas

language to mark aspect but rather to create new TMA markers out of certain lexical items. Very often these lexical items consist of verbs meaning 'stay', 'stop', 'go', 'come', 'finish', etc. In Hawaiian Creole English, for example, stei (stay) and go mark aspect and modality, respectively, according to Bickerton (1981: 26); in Tok Pisin stop has developed into a locative copula (Sankoff 1980: 148), and in many Creoles, for instance Tok Pisin, verbs meaning 'finish' are markers of perfective aspect. Also certain particles have developed into aspect or tense markers, for instance in Tok Pisin where baimbai has turned into a marker of future tense, bai (Sankoff 1977). The tendency to create aspect/tense markers from a highly constrained set of features is not restricted to pidgins and Creoles. On the contrary, verbs, adverbs, and prepositions of the above mentioned type very often underlie aspect markers in "natural" languages. Comrie (1978: 106) points out that directional expressions like 'come' and 'go' are used to mark prospective aspectual meaning in West African languages, and Traugott (1978: 389) and Bybee (1985: 11) observe that verbs for remaining, standing, sitting, or lying form progressives in many languages, such as Spanish (estar), Italian {stare), Latin (stare), Eastern Aramaic, Modern Amharic, Tunica, Ya Tue, Hindu-Urdu, Punjabi, and Mandarin Chinese, i. e., in languages many of which are unrelated to each other. Verbs meaning 'finish', which according to Traugott (1978: 391) are ablative in the underlying structure, occur in a large number of languages as markers of terminative/resultative aspect, such as Spanish, Japanese, Swahili, KukiChin, Turkish, Hungarian, Arabic, Amharic, Malay, Punjabi, Hindi, Urdu, and Burmese (Traugott 1978: 390, Bybee 1985: 11), and when particles such as 'on', 'up', 'down', etc., are added to the verb as prefixes, suffixes or free morphemes (burn up, burn down), they have an effect on its aspect (Bybee - Dahl 1989: 85-86). As demonstrated by Traugott (1973, 1978) and Bybee and Dahl (1989), aspect and tense markers are almost exclusively developed from originally locative verbs and particles, although future tense may also develop out of modals, such as desideratives meaning 'want' and 'desire' or verbs meaning 'to owe' or 'to be obliged'. Traugott (1978: 373) remarks that there seems to be a constant tendency for tenses and aspects to be revitalized in language acquisition, language change, and language contact, and that tense markers are often historically derived from aspectuals (Traugott 1975: 209). Markers of perfective, terminative aspect, for instance, tend to develop into markers of perfect tense, while inceptive or inchoative aspectuals, in particular verbs

Aspect marking and grammaticalization

in Russenorsk

with Immigrant

Swedish

125

of motion, tend to give rise to future tenses (Fleischmann 1982: 23). Bybee and Dahl (1989: 85) suggest that the aspectual distinction is more basic in language than the tense distinction. The grammaticalization process seems to be quite similar in creolization and language change. In both cases the development goes from analytic to synthetic expressions. Originally a word has a full lexical, concrete meaning. If the word is frequently used, the meaning may be generalized and become more abstract. Later on, the original semantic meaning may be almost completely lost and it becomes a function word, i. e., an auxiliary, which finally may be cliticized or even become inflectional at the same time as it is phonologically reduced (Bybee — Dahl 1989: 63—67). In Latin, French, and Spanish, for instance, future markers diachronically have been renewed again and again in a process where "lexical, i. e., full verbs, typically bleach out into modal auxiliaries, amd modals and locative expressions develop into aspect markers, which in turn develop into tense markers" (Fleischmann 1982: 128). Another example is the development of have into a temporal auxiliary in English (Bybee - Pagliuca 1985: 72-73). Similarly, in Hawaiian Pidgin English go was frequently used as a main verb and as a preverbal modifier of "extremely indeterminate meaning and wildly fluctuating distribution" (Bickerton 1981: 31), while in the creole it has turned into a grammatical marker (Bickerton 1981: 79). In Tok Pisin baimbai seems to have gone through a similar process, i. e., an increase in frequency, semantic extension, phonological reduction (bai), and changes of sentence position (SankofT 1977). Since no obvious markers of categories such as tense and aspect usually exist in pidgins, the development of grammatical markers in Creoles has traditionally been regarded as a result of children's creativity and sometimes also as evidence of an innate language-developmental program. Bickerton (1981) and Mühlhäusler (1986), however, both observe that the pidgin input plays an important role in creolization, and SankofT remarks (1980: 155): "nativization, that is, child acquisition, is not the only thing that can lead to grammatical expansion. Adult, second-language speakers already had done a pretty good job of it by the time native speakers of Tok Pisin came along". Mühlhäusler, however, rejects the opinion that the substrate languages have a decisive impact on the development of a creole, especially as regards syntactic innovation. Instead he points to universal principles of language development and pragmatic factors and remarks: "Dramatic growth can take place in the absence of creolizing children" (Mühlhäusler 1986: 182-184).

126

Ulla-Britt

Kotsinas

Evidently, aspect is a pragmatically important category in language, overlapping with tense and modality. Therefore, it seems reasonable to assume that the process of aspect marking is inclined to start already in restricted pidgins like Russenorsk or pidginized varieties of second-language acquisition, and furthermore that verbs and particles of the abovementioned type would be good candidates for being on their way to such markers. In the following, the possibility of the occurrence of early aspect markers in two pidginlike varieties, Russenorsk and Immigrant Swedish, will be discussed.

Russenorsk and Immigrant Swedish Russenorsk, described by Broch (1927) and Broch and Jahr (1981) was used as a lingua franca, mainly in communication between Russian merchants and Norwegian fishermen along the Arctic coast of northern Norway, during a period of at least 140 years until contacts were broken in 1917. Unlike most pidgins, Russenorsk was the result of contact between socially equal groups, a fact that is mirrored in the lexicon, which contains roughly equal proportions of Russian and Norwegian words, and in addition some German, English, Swedish, Finnish, and Sami words, not seldom synonyms, for instance Russ. davaj and Norw. vcersgo 'please', Russ. malenka, Norw. lite 'a little'. Broch and Jahr (1981) characterize Russenorsk as a restricted pidgin with certain rules, for example the addition of a verb-marking suffix -om to verbs of both Norwegian and Russian origin: betalom 'pay', robotom 'work'. Romaine (1988: 124) characterizes Russenorsk as a stable pidgin and suggests that creolization in Russenorsk was counteracted by repidginization, caused by seasonal fluctuations of trade. Clyne (1968), Meisel (1977a, 1977b), the Heidelberger Forschungsprojekt (1977), Mühlhäusler (1986), Holm (1989), Kotsinas (1985, 1989), and many others have pointed to the similarities between pidgins and the linguistic outcome of the immigration to western Europe and Scandinavia during the last few decades, the so-called Gastarbeiterdeutsch and similar varieties. In Sweden, in fact, the social and linguistic situation in which pidginized second-language varieties have emerged is remarkably similar to the one in plantation areas, where "classic" pidgin/creoles like Hawaiian Pidgin English and Tok Pisin once arose.

Aspect marking and grammaticalization

in Russenorsk

with Immigrant

Swedish

127

Since the mid-sixties, Sweden, like many other West European countries has received a huge wave of immigrants from agricultural areas in southern Europe and, more recently, refugees from all parts of the world. Today approximately 140 languages are represented in the country. Most of the immigrants live in suburbs on the outskirts of the big cities, where many of them have limited contacts with native Swedes, partly due to social barriers, partly due to the type of jobs available to them. In some of these areas, 50 to 70 percent of the inhabitants are of foreign origin, and 30 to 40, in some cases up to 100, languages are spoken. Since no ethnic group, not even the Swedish one, is in a majority in areas like these, Swedish has to be used as a lingua franca in shops, cafeterias, at post offices, etc. Although some immigrants to Sweden speak Swedish very well, many use pidginized varieties. So far, however, no stable pidgin seems to have crystallized, and tendencies in that direction are counteracted by the constant influx of new immigrant and refugee groups as well as by an ongoing moving-out to other areas of more established groups. The term Immigrant Swedish (IS) is in the following restricted to pidginized varieties, or rather idiolects, used by those immigrants and refugees who have lived in Sweden for several years. In fact, the term is used in an even more restricted way, in that almost all the following examples are taken from one single speaker, a Greek man who has spent nine years in Sweden. His variety of Swedish deviates markedly both from Swedish and from his mother tongue Greek and exhibits many typical pidgin features (Kotsinas 1983, 1984a, 1984b, 1985, 1989). Similar solutions, however, are easily found in other ethnic groups, and some examples are taken from speakers with Spanish, Kurdish, and Finnish mother tongues (ISSp, ISKu, ISFi). Finally, a few examples are taken from literature on non-European pidgins and Creoles, such as New Guinean Pidgin/Tok Pisin (NGP), Hawaiian Pidgin English (ΗΡΕ), and from Gastarbeiter-pidgins in Germany (IG) and America (IE). (A list of abbreviations can be found at the end of this paper.)

Grammaticalization As in most pidgins, the vocabularies of Russenorsk and Immigrant Swedish are very restricted, and compensatory strategies such as lexical overuse and semantic overextension, circumlocution and repetition are fre-

128

Ulla-Britt

Kotsinas

quent. Verbs and nouns are used in invariant forms, i. e., neither tense, aspect, nor number is morphologically expressed. Function words such as articles, copulas, subjunctions, and auxiliaries rarely occur. Subordinate clauses are rare and reported speech is replaced by direct quotations (as in examples [1] to [5]). (1)

RN: Kristus grot vrei, tvoja ljugom. Christ very angry you lie 'Christ is very angry, because you lie.'

(2)

IS: ja inte kommer, ja jobba I not come I work Ί will not come, because I am working.'

(3)

RN: moja ska si ju: ju grot lygom. I will tell you you much lie Ί tell you that you are a big liar.'

(4)

IS: a hart säjer: inte behöver göra hör. and he say not have to do here 'He said that I did not have to clean here.'

(5)

ΗΡΕ: stei tel mi: ο neks wik, hi kamin, kamin. stay tell me oh next week he coming coming 'He kept telling me that he would come the following week.'

The lack of grammatical markers, however, is at least partly compensated for by analytical marking of certain categories. Tense and sequencing are, for instance, as usual in pidgins, marked by time adverbials in cases where the context is not enough (as in examples [6] to [9]). (6)

RN: kanske morradag mera pris. maybe tomorrow more price 'Maybe the price will be higher tomorrow, later.'

(7)

IS: ikväll inte träffa. tonight not meet 'We are not going to meet tonight.'

Aspect marking and grammaticalization

in Russenorsk

with Immigrant

Swedish

129

(8)

RN: kor ju stann om paa gammel ras? where you stay on old time? 'Where were you last time?'

(9)

IS: nästa gammal är 53 timmar fel. next old year 53 hour wrong 'The following year I got the wrong [salary for] 53 hours [of work].'

There also seems to be a tendency for certain particles to be involved in a process of grammaticalization, for instance for interjections meaning 'please' to mark imperative and for adverbs meaning 'maybe' to mark epistemic modality and futurity. In both Russenorsk and Immigrant Swedish requests sometimes are marked by the direct mention of the addressee (examples [10] to [12]). Very often, though, davaj or vcersgo 'please' in Russenorsk and varsdgod 'please' in Immigrant Swedish, are placed sentence-initially (examples [12] and [13]). Normally, markers like these are used to indicate politeness, but both the frequency of the markers and the fact that the semantic content of them in some cases seems to be diluted suggest that these words have a purely grammatical function. Example (12), for instance, uttered by a captain, is rather an order to a subordinate than a polite invitation. (10)

RN: junga grebi moja po lan. boy row me on land 'Boy, row me ashore.'

(11)

IS: kom du tolk med mej. come you interpreter with me 'Come with me as an interpreter.'

(12)

RN: vaersgo ju pa moja skib vaskom. please you on my ship wash 'Clean my ship.'

(13)

IS: varsego, titta en fel. please look one error 'Look, I have made only one error.'

130

Ulla-Britt

Kotsinas

A similar development is described by kanske 'maybe' which, like davaj, etc., is usually placed sentence-initially. Kanske is in Russenorsk and Immigrant Swedish often used as a marker of questions (examples [14] to [18]), but it occurs so frequently in Russenorsk in other contexts, too, that Broch and Jahr (1981: 52-53) suggest that it approaches a general marker of politeness. Examples like (17) to (21), however, indicate that the word in certain contexts is used as a marker of epistemic modality, often with future reference. The irreality of future, marked by kanske, is especially evident in (21), where a Spanish woman talks about her future plans. In Immigrant Swedish and Immigrant German, but not in Russenorsk, finally, a few cases occur, where kanske and corresponding words seem to be used as subjunctions, marking a conditional clause (examples [22] and [23]). (14)

RN: kanske principal pa kantor? maybe captain on office? 'Is the captain in his office?'

(15)

RN: kanske tvoja rik man? maybe you rich man Ί believe, that you are rich?'

(16)

IS: den skriva Dagens Nyheter kanske? it write Dagens Nyheter maybe 'Does it say Dagens Nyheter? Could it be Dagens Nyheter [a Swedish daily newspaper]?'

(17)

RN: kanske Uta tjai drikkom? maybe some tea drink 'Shall we / Do you want to drink some tea?'

(18)

RN: kanske kapitan pa skib reisa? maybe captain on ship travel 'Will the captain go on board?'

(19)

IS: förstär inte i dag, kanske imorgon förstär. understand not today maybe tomorrow understand 'Maybe I will understand better later.'

Aspect marking and grammaticalization

in Russenorsk

with Immigrant

Swedish

131

(20)

IE: Maybe next time I understand more.

(21)

ISSp: ja kanske spara lite I maybe save some ä kanske ja gifta mej ά and maybe I marry and Ί may save some money and and marry. I don't know.'

(22)

IS: kanske spela du entusen kommer etthundra. maybe play you one-thousand come one-hundred 'If you bet 1000 crowns, you get 100.'

(23)

IG: vielleicht komme beim toilette... maybe come by toilet 'If you come/go to the lavatory

penga ä köpa lite hus money and buy little house kanske vet inte. maybe know not maybe I will buy a little house

Aspect According to Comrie (1978: 13), the verb in a sentence either expresses a static or a dynamic situation. As an hyperonymous term Comrie uses "situation". The dynamic situations in turn are of two types, processes that are "viewed in progress... (imperfectively)", and events, that are "viewed as a complete whole (perfectively)" (1978: 13). Figure 1 illustrates the use of certain verbs in Immigrant Swedish (here represented by the above-mentioned Greek speaker) and in Russenorsk. In Norwegian and Swedish, states are expressed by verbs meaning 'stay', 'live', etc., or by a copula, used in practically the same way as in English, i. e., attributive {she is nice), equative (he is my brother), and locative (he is in London). As it seems, no copula is used in Russenorsk (Broch - Jahr 1981: 46), (examples [24] and [25]). In Immigrant Swedish, the copula is normally omitted (26), although a copula-like morpheme accidentally occurs in certain expressions resulting from holophrastic learning, den e 'it is', e de 'is it', etc. (Kotsinas 1989).

132

Ulla-Britt

Kotsinas

Situations

ingressive copula attributive equative

locative

Immigrant Swedish:

0

0

Russenorsk: 0 0 or stannom slipom ligge ne

stannom leve

'eat' 'drink' 'write' etc.

komma present om kastom etc.

Figure 1. The use of certain verbs in Immigrant Swedish and Russenorsk

(24)

R N : russmann bra mann. Russian-man good man 'The Russian is a good man.'

(25)

R N : moja grot krank. I very ill Ί am very ill.'

egressive

Aspect marking and grammaticalization

(26)

in Russenorsk

with Immigrant

Swedish

133

IS: grammatik my eke t svärt. grammar very difficult 'Grammar is very difficult.'

In both Russenorsk and Immigrant Swedish, however, certain verbs seem to be involved in a process of semantic extension, similar to the one observed in many pidgins: In nearly all pidgins, the usage of the verb 'to be' differs in some way from its use in the standard. It may be lost entirely or be replaced by, or be in free variation with, another verb or copula with a meaning like 'to remain', 'to stay', 'to sit', 'to be temporarily' etc. (Naro 1978: 330)

In Russenorsk, for instance, there are examples where the verbs stannom 'stay/stop', liggene 'lie down', slipom 'sleep', and maybe also spasirom 'walk around' seem to have lost some of their original meaning and are used in a sense close to a locative copula (examples [27) to [32]), while they have retained their original lexical meaning in other examples. Similar use of the verbs stop and ste 'stay' is found in, for instance, New Guinean Pidgin and Hawaiian Pidgin, respectively (examples [33] to [36]). Although stanna 'stay' in Immigrant Swedish does not seem to have developed into a copula so far, it is used by the Greek Immigrant Swedish speaker in a very generalized sense, i. e., to replace other stative verbs meaning 'live', 'be alive', 'remain' etc. (example [37]) (Kotsinas 1989). (27)

R N : kor ju stann om ρά gammel ras? where you stay on old time 'Where were you last time?'

(28)

R N : kor ju ligga ned ρά gammel Dag? where you lie down on old day 'Where were you yesterday?'

(29)

R N : moja pä stova ρά Kristus sprek stannom. I on house on Christ speak stay Ί was in church [in the house where they speak about Christ].'

(30)

R N : altsamma ρά salt ligge ne. everything on salt lie down 'Everything [the fish] is salted down.'

134

Ulla-Britt

Kotsinas

(31)

RN: altsamma pä salt slipom. everything on salt sleep 'Everything [the fish] is salted down.'

(32)

RN: saika grot paa Gaf spasirom. coalfish much on sea walk 'There are many coalfish in the sea.'

(33)

NGP: hous bilong yu i stap we? 'Where is your house?'

(34)

NGP: hamas man i stap? 'How many men are there?'

(35)

NGP: some egg he stop? 'Are there any eggs?'

(36)

ΗΡΕ: hi ste maui. 'He is in Maui [temporarily].'

(37)

IS: den tjugo är stanna Joannina. it twenty year stay Joannina 'She lived in Joannina [a town] for 20 years.'

The Immigrant Swedish speaker, however, also uses a verb of movement, komma 'come', very frequently and in an even more generalized way, for instance whenever a verb with the meaning 'come', 'go', 'travel', 'give' or 'receive' would be adequate. Seemingly, komma is thus used to replace all words with the semantic feature [change of location] (examples [40], [42]). The same speaker also uses komma for words meaning 'begin', 'start', and other words with the semantic feature [change of time] (example [44]). Finally he uses komma to replace words meaning 'realize, understand', 'to get to know' etc., i. e., words which could be said to have the semantic feature [change of state] (example [45]). As a matter of fact he uses komma as if it had only the semantic feature [change], i. e., to denote all kinds of events where an ingressive verb is needed (Kotsinas 1989). In some cases the original semantic content of the word, thus, is totally lost and komma seems to have the function of an ingressive copula. Similar generalizations of verbs of movement may be found in other varieties of second-language acquisition as well as in many pidgins. It may also be

Aspect marking and grammaticalization

in Russenorsk

with Immigrant

Swedish

135

noted that words meaning 'come' in some natural languages have developed into ingressive copulas (Engl, become) or words meaning 'receive' (German bekommen). In Greek, though, the corresponding verb irchume is restricted mainly to the notion [change of location]. In Russenorsk, at least four verbs of movement are used: komma, spasirom 'walk', reisalreisom 'travel', marsjirom 'march', and in one example gä. Komma is not very frequently used, and most often the original meaning is retained (examples [38] and [39]). There are, however, two instances (examples [41], [46]) in which komma is used in ways which would be hard to explain, were it not for the similarities to the use of komma in Immigrant Swedish. Below, examples of the use of verbs with the original meaning 'come' in Russenorsk, Immigrant Swedish, Immigrant German and a few pidgins (New Guinean Pidgin and French Seychelle Pidgin) are ordered according to meaning and function (examples [38] to [52]). Verbs meaning 'come' in Russenorsk, Immigrant Swedish, and some pidgins: 1. Change of location a) intransitive, 'come', 'go' (38)

RN: vcersägo, burmann, pa skib komm. please fisherman on ship come 'Come [or go] aboard, fisherman.'

(39)

RN: ju sproek paa moja Kantor kom. you speak on my office come 'You said that you would come to my office.'

(40)

IS: kommer ρά exepetor, frägar kom. come on supervisor ask come Ί went to the supervisor and asked her to come.'

b) transitive, 'give', 'receive' (41)

RN: burman kom fiska. fisherman come fish 'Give me the fish.'

(42)

IS: en flicka den flicka vill kommer cigarrett. a girl that girl want come cigarette 'The girl who wanted [to have, to get] cigarettes.'

136

Ulla-Britt

(43)

Kotsinas

IG: ich spreche, Kognak bei dir komme, drei Botellas I say brandy to you come three bottles Kognak komme eine Schenke. brandy come a gift Ί said to him: I will give you three bottles of brandy as a gift·'

2. Change of time (44)

IS: klockan fem kommer, klockan nie slut. o'clock five come o'clock nine finish Ί start at five o'clock and finish at nine.' [I work between five and nine.]

3. Change of state a) 'understand', 'get to know' (45)

IS: kanske kommer fru var ja ringa. maybe come wife where I call 'Maybe my wife will understand to whom I call [if I call a Swedish girl].' [A joke.]

b) 'become', 'get', ingressive copula (46)

RN: nokka lite pjan kom. a little drunk came Ί got a bit drunk.'

(47)

IS: den kommer 20 är. it come 20 years 'She became 20 years old [adult].'

(48)

NGP: em i kamap tisa. 'He became a teacher.'

(49)

NGP: brukim nain long tri, na em i kamap tri. 'Nine divided by three is three.'

(50)

FS: telma kisa lager t'in ariv efreja. La guerre etait devenue si effrayante 'The war had become so frightening.'

Aspect marking and grammaticalization

in Russenorsk

with Immigrant

Swedish

(51)

IG: komme Sohn fümf Jahre kom in die Schule. come son five years come in the school 'When my son became five years old he began school.'

(52)

IG: Mädchen komme grosser, noch verheiratet gehen. girl come bigger married go 'The girls will become bigger and get married.'

137

In Immigrant Swedish, a few words, sluta 'finish' and färdig 'finished' still seem to be markers of aspect (examples [53] and [54]). Similar examples are found in Immigrant German (example [55]). In Russenorsk, however, there are no traces of such use. (53)

IS: de nie mänad slut. it nine month finish 'After nine months it [the pain] stopped.'

(54)

IS: jag alia toalett slut fardig. I all lavatory finish finished Ί had finished [cleaning] all the lavatories.'

(55)

IG: ich fertig arbeit. I finish work Ί had finished my work.'

As we have seen, words meaning 'come', 'go', 'stay', or 'finish' in pidgins tend to develop into aspectual or temporal auxiliaries after a period of semantic generalization (examples [56] to [62]). In Russenorsk and Immigrant Swedish (as represented by the Greek speaker), however, the corresponding words seem to have kept their status of main verbs in the sentence, with a few exceptions [example [63]). Similar examples are found in other varieties of Immigrant Swedish (examples [64] and [65]). (56)

FP: hi kam gro da paemili. he came grow the family 'The family was beginning to grow up.'

(57)

FGC: li ka ale asas. 'He is going to hunt.'

138

Ulla-Britt

Kotsinas

(58)

HCE: ai no kea hu stei hant insai de, ai gon hunt Ί don't care who's hunt/wg there, I'm going to hunt

(59)

CP: ζ' go kam. 'He will come.''

(60)

CP: mi a go go. Ί will go:

(61)

NGP: em i ritim dispela buk pinis. 'He has read this book.'

(62)

NGP: mi tokim ju pinis. Ί told you.'

(63)

IS: min fru kommer jobba klockan dtta. my wife come work o'clock eight 'My wife starts work at eight o'clock.'

(64)

ISFi: jag semester kom ä slut, jag kommer arbet. I vacation come and end I come work 'My vacation came to an end/ended, and I started work:

(65)

ISKu: sen ja komme vänta. then I come wait 'Then I began to wait:

The "odd" way of using komma, stanna, and sluta in Immigrant Swedish and to a smaller extent in Russenorsk, in fact seems very structured, and possibly may be seen as the first step in the development of an aspect system, following the same path as is usual in creolization and language change. The Immigrant Swedish speaker, from whom the examples of komma, stanna, slut are derived, has apparently found a way to express aspect by replacing all verbs with the features [state], [change (ingression or egression)] with stanna 'stay', komma 'come', and sluta 'finish' or färdig 'finished', Stanna is also used as a locative copula, and komma as an ingressive copula, 'become', while equative, attributive, and locative copulas are unmarked. By an extreme semantic extension of these words, he manages to use his very restricted Swedish vocabulary (approximately 400

Aspect marking and grammaticalization

in Russenorsk

with Immigrant

Swedish

139

words, Kotsinas 1985) to express his needs of a basic aspect system (although not always understandable to a native Swede). Only in the field of verbs denoting processes, i. e., concrete words meaning 'dance', 'eat', 'drink', 'read', 'write', etc., has his acquisition of Swedish resulted in a more varied vocabulary (approximately 30 words). It must be pointed out that this speaker's solutions to the problem of aspect marking do not derive from previous knowledge of any more or less stable pidgin outside Sweden, where similar solutions could have existed, nor are they the results of interference from his native language, especially as regards the use of komma and sluta, since the use of corresponding words in Greek is even more restricted than in Swedish (see Figure 1).

Prepositions and aspect The use of prepositions and adverbial particles derived from prepositions frequently has aspectual notions, and in some languages aspect is connected to case (inflections, prepositions, etc.) (Traugott 1973, 1978, Bybee - Dahl 1989). Like most pidgins, Russenorsk has only one preposition, pä (paa, po), derived from the Norwegian preposition pä and the Russian preposition/ affix polpo-. In Immigrant Swedish, as represented by the Greek speaker, four prepositions are used, ρά 'on', / 'in', med 'with', and till 'to'. The last three, however, are used almost exclusively in phrases resulting from holophrastic learning or, as regards med, in expressions where the homophonous and synonymous Greek preposition me would have been used. Pä, therefore, is the only "free" preposition, and far more frequent than any of the others. Both in Russenorsk and in many varieties of Immigrant Swedish, pä is used in a very generalized way, corresponding to "all purpose" prepositions in pidgins (Kotsinas 1984c). In both Russenorsk and Immigrant Swedish ρά is used as a marker of all kinds of locative expressions, i. e., position (examples [66] and [67]), direction (examples [68] and [69]), and origin (examples [71] and [72]), i. e., in expressions where in Norwegian and Swedish prepositions meaning 'on', 'in', 'to', 'from', etc., are used. In New Guinea Pidgin long is used in the same way (examples [70], [73]). (66)

RN: moja krank pä maga. I sick on stomach Ί have a stomach-ache.'

140

Ulla-Britt

Kotsinas

(67)

IS: sitter pä soli. sit on sun 'He is sitting in the sun.'

(68)

RN: ju sprcek pä moja kantor kom. you speak on my office come 'You said that you would come to my office.'

(69)

IS: kommer pä kontor. come on office Ί went to the office.'

(70)

NGP: em i kam long haus. 'He came to the house.'

(71)

RN: nogoli dag tväja reisa pä Archangel otsuda. some day you travel on Archangel away 'How long time did you travel, have you been on your way from Archangelsk?'

(72)

IS: komer en papper pä kyrka. come a paper on church Ά paper came from the church.'

(73)

NGP: Yu tekewe mani bilong mi long bokis. you take money belong me long box 'You took my money from the box.'

Temporal expressions are usually derived from locatives and closely connected to aspect (Traugott 1975, 1978). Expressions answering the question 'When?', for instance, mark a point on the time line, i. e., perfectivity, while expressions answering the question 'During how long a time?' mark an ongoing activity, imperfectivity. This aspectual distinction seems to be expressed in both Russenorsk and Immigrant Swedish. While punctual expressions are marked with ράΙρο (examples [74] and [75]), no preposition occurs in durative expressions (examples [76] and [77]). (74)

RN: moja paa morradag kom. I on tomorrow come Ί will come tomorrow.'

Aspect marking and grammaticalization

in Russenorsk

with Immigrant

Swedish

(75)

IS: kanske kommer pä päsk maybe come on Easter 'She will come at Easter.'

(76)

RN: ja pä madam Klerck tri daga ligge ne. I on Madame Klerck three days lie down Ί stayed at Madame Klerck's estate for three days.'

(77)

IS: fräga kanske tio timmar. ask maybe ten hour 'We were talking for ten hours / a long while.'

141

Pä is also used to mark case in Russenorsk almost without exception according to Broch — Jahr (1981: 58). In both Russenorsk and Immigrant Swedish the indirect object is always marked with a preposition (examples [78] to [83]), and there are also examples oipä marking possession (examples [85] to [87]). Similar examples occur in New Guinea Pidgin (example [84]). (78)

RN: moja paa ju presentom baanbaan. I on you give candy Ί will give you candy.'

(79)

IS: köpa pä barn... buy on child '[I bought clothes] for my child'

(80)

RN: mera better pä moja. more better on me 'It is better for me.'

(81)

IS: sämre pä oss. worse on us 'It is worse for us.'

(82)

RN: ju sprcek paa moia you speak on me ... 'You said to me...'

142

Ulla-Britt Kotsinas

(83)

IS: ringa pä flickor call on girls Ί will call the girls.'

(84)

NGP: Mi singout long you Ί called out to you.'

(85)

RN: pä tvoja kona? on you wife 'Do you have a wife / Are you married?'

(86)

RN: mangeli klokka pä ju? how-much clock on you 'How much is your clock / What time is it?'

(87)

IS: stan pä din mamma. town on your mother 'Your mother's village.'

As to the direct object, however, the situation seems more complicated in Russenorsk. In some examples (examples [88] to [90]) no preposition is used, while in others the direct object is marked by pä (example [94]). However, there are also examples of parallel expressions where the preposition seems optional (examples [90] and [91]). In Immigrant Swedish the direct object is only occasionally marked with a preposition (examples [92] and [93]). Whether this inconsistency has to do with for example the semantic content of the verb or aspect marking (cf. Bybee - Dahl 1989: 89 for the interaction between aspect and case-marking), or is just an example of the well-known variability in pidgins, is difficult to say. (88)

RN: tvoja fisk kopom? you fish buy 'Do you want to buy the fish?'

(89)

RN: tvoja vegom fiska. you weigh fish 'Weigh the fish.'

(90)

RN: vcersgo paa moja Skib kom tjai drikkom. please on my ship come tea drink 'Please come aboard (and) drink tea.'

Aspect marking and grammaticalization

in Russenorsk

with Immigrant

Swedish

(91)

RN: davaj paa Skip kom, Brat, paa Tjei driki. please on ship come brother on tea drink 'Please come aboard, brother [and?/to?] drink tea.'

(92)

ISFi: dricka pä kaffe. drink on coffe 'Drink coffee.'

(93)

IS: kontrollera pa pass. control on passport 'Control the passport.'

(94)

RN: vcersego ju paa moja skib vaskom. please you on my ship wash 'Clean my ship!'

143

Preverbal prepositions In Russenorsk there are 18 examples of pä in preverbal position (examples [95] to [111], and similar examples are found in Immigrant Swedish (examples [112] to [116]) and pidgins (example [117]). (95)

RN: vcersago, lita klceba pä presentom. please some bread on give 'Please give me some bread.'

(96)

RN: davaj pä skrivom. please on write 'Please write.'

(97)

RN: davaj pä skib tväja ligene jes pä slipom. please on ship you lie down and on sleep 'Please lie down and sleep on board.'

(98)

RN: davai paa Skip tvoia ligge ne, grot pä slipom. please on ship you lie down much on sleep 'Please lie down on board and sleep.'

144

Ulla-Brit t Ko tsinas

(99)

RN: davai päproberom. please on-try 'Please try.'

(100)

RN: davai päsmotrom pä skib. please on-see on ship 'Please look at the ship.'

(101)

RN: no tväja pä kastom visit treska. now you on throw weigh codfish 'Throw the codfish and weigh it.'

(102)

RN: russemann därlig mann, grot skade pä tvoja, Russian-man bad man much hurt on you päschlagom. on hit 'The Russian is a bad man, he causes you great harm, hit him.'

(103)

RN: podjom, 'Come here.'

(104)

RN: moja paslagom pä tvoja. I on-hit on you Ί will hit you.'

(105)

RN: Reise pä Rosland! Pä slagom/ Pä slagom/ Travel on Russia On hit On hit 'Go to Russia! Hit them! Hit them!'

(106)

RN: tvoja po-rejza? you on-travel 'Are you going to travel?'

(107)

RN: kanske litt pä skaffom? maybe little on eat 'Maybe you want something to eat?'

Aspect

marking

and grammaticalization

in Russenorsk

with Immigrant

Swedish

145

(108)

RN: nä gaima kanskje trovva pä presentom? now brother maybe wood on give 'Well my brother, maybe you will give me some wood.'

(109)

RN: principal pä slipom, eller njet? captain on sleep or no 'Is the captain asleep?'

(110)

RN: strasvi, junka. Principal ρά skib? hello, boy. Skipper on ship a. jes, pä skaffom. yes on eat b. jes, pä slipom. yes on sleep 'Good morning, boy. Is the captain aboard? Yes, he is eating. Yes, he is asleep.'

(111)

RN: gä pä slipom. go on sleep 'Go to sleep.'

(112)

IS: alia papper den gammal pä kasta. all paper it old on throw 'Throw away all those old papers.'

(113)

IS: fru: den e inte pä kasta, stanna. wife: it is not on throw stay 'My wife said that I should not throw away the papers but keep them.'

(114)

IS: den barn sex mänar kommer pä skriva pä kyrka. it child six months come on write on church 'The child had become six months old and we had to register him at church'

(115)

ISSp: stoppa dom tre fyra timmar taget pä undersöka stop they three four hours train on examine borne. child 'They stopped the train for some hours to examine the child.'

146

Ulla-Britt

Kotsinas

(116)

ISSp: och gä vi pä simma och äta. and go we on swim and eat 'And we went to swim and eat.'

(117)

NGP: em i go long kisim sut. 'He went to have a shot [an injection].'

The interpretation of examples like these in Russenorsk is difficult, and different explanations have been presented. Fox (1973, quoted in Broch - Jahr 1981: 59-60), for instance, suggests that (95) "appears to be perfective" and (110b) "limited durative", while some of the other examples are "purposive". Broch and Jahr (1981: 59—60) reject this interpretation, referring to the difficulty of finding a semantic element, common to all occurrences. They notice that ρά + V in most of the examples seems to mark tense, i. e., futurity, but since this explanation is not possible for examples (109) and (110a, b) they conclude that pä + V in Russenorsk is facultatively used. Bakker also (1987: 19—20) observes that ρά sometimes seems to be used as a future marker, for instance in (104). He suggests that pa is used as a complementizer, and notes that davaj is often used before the complement, for instance in example (91) {davaj paa Skip kom, Brat, paa Tjei driki), and sometimes as a polite imperative, as in example (97). He concludes that pä in preverbal position is used as "a kind of complementizer, a polite imperative, and a marker of obligation, all ... referring in some way to the future". Bakker's suggestions are rejected by Belikov and Perekhvalskaya (1989) with reference to the fact that many of the occurrences of pä may be explained by influence from Russian, where the preposition/particle po is used in ways very similar to those in Russenorsk, for instance in the following examples: (118)

a. RN: altsamma paa salt ligge ne. Russ.: Vse posoleno. 'All has been salted.' b. RN: davaj paa slipom. Russ.: Davajpospim. 'Let us sleep.'

Aspect marking and grammaticalization

in Russenorsk

with Immigrant

Swedish

147

c. RN: davaj poproberom. Russ.: Davaj poprobujem. 'Let us try.' As it seems, Broch and Jahr are right in the observation that no single explanation is possible for the interpretation of preverbal pä in Russenorsk. This, however, does not necessarily imply that pä is placed preverbally facultatively, neither that pä has no meaning or function. Polysemy (at least from the target language point of view) and multifunctionality are, as we have seen, typical not only of pä in Russenorsk but of prepositions, adverbs, certain verbs, etc., in most pidgins. As noted by Bakker, many of the examples in Russenorsk could be interpreted as imperatives. In six cases (examples [95] to [100]) the imperative markers davaj or vcersgo are placed sentence-initially, and at least one more example (101) is a request (cf. the very similar Immigrant Swedish examples [112] and [113]). Example (103) seems to be a transfer from Russian, while example (111) could be interpreted either as a request or an infinitive form. Another two examples, (104) and (105), seem to express the speaker's intentions, and example (106) is a question about the listener's intentions. Example (102) is either a request, directed to the listener, or expresses the speaker's intention. The notion of intention is also evident in the example from New Guinea Pidgin, (117), and the Immigrant Swedish examples, (114) to (116), where the preposition functions as a complementizer, marking embedded final, purposive clauses. According to Litteral (1969: 42) the "allpurpose" preposition long in New Guinea Pidgin in preverbal position has the meaning 'to' or 'in order to', and Mühlhäusler (1986: 187) remarks that complementizers were internally developed by means of reanalyzing the use of existing material, for instance the preposition long. Imperatives mark deontic modality, and like intentions they aim at a certain goal, i. e., some event in the near or distant future. As it seems, Russenorsk and possibly also the Immigrant Swedish speaker, have found a way to mark deontic modality, either by using words meaning 'please' or by adding pä preverbally, or both. As observed by Bybee and Dahl (1989: 58) markers of obligation and intention as well as desideratives diachronically tend to develop into futures. The development of prepositions into subjunctions is also well documented. Examples (107) to (109), on the other hand, are questions, two of them marked with the modality marker kanske, which, as discussed above, also may mark future.

148

Ulla-Britt

Kotsinas

Examples (109) and (110), though, are not possible to interpret as requests, intentions, or anything else that is goal-directed or has to do with the future. There is, however, another possible explanation for the use of preverbal ρά. As observed by Traugott (1978: 392-393) the addition of an originally locative particle to a verb results in a change of aspect. Usually the verb gets a terminative or resultative aspect of the verb itself (cf. drink and drink up, German essen and aufessen). Bybee and Dahl (1989: 85-86) have coined the term "bounder" for such markers, since they have the effect of making the process denoted by the verb "bounded" ("telic"). In fact, some of the occurrences of ρά in Russenorsk may be interpreted as bounders. Added to verbs with an inherent perfective aspect, such as presentom 'give' (examples [95], [108]) and kastom 'throw' (example [101], and the Immigrant Swedish examples [112] and [113]), ρά gives a resultative or telic aspect to the verb, 'give away', 'throw away', while päproberom 'try' (example [99]), and pdschlagom 'hit' (examples [104] and [105]), and in the verb of movement pa rejza 'travel' (example [106]), the preposition rather seems to mark ingressive or iterative aspect: 'start trying', 'beat up', 'go away'. When ρά is added to durative verbs, such as slipom (examples [97], [98], and [111]), and ρά skaffom 'eat' (example [107]), on the other hand, one interpretation would be that ρά in these cases, just like po- in Russian (Comrie 1978: 22) indicates a temporally restricted situation; 'sleep a while', 'eat something'. Another interpretation, especially for gäρά slipom 'go on sleep' (example [111]), davajpa skrivom 'please on write' (example [96]), and davai päsmotrom 'please on see' (example [100]), would be that the inherent imperfective aspect of the verb is turned into a perfective aspect, 'go to sleep', 'write down', 'look'. Neither of these explanations seem likely for examples (109) {principal ρά slipom, eller njet?) and (110) (jes, ρά skaffom, jes, ρά slipom). Locative particles have, however, also been used in many languages in the formation of progressive forms (Traugott 1978: 388-389). The English progressive form contained for instance originally the preposition on (Bybee - Dahl 1989: 79). This could also be a possible explanation for examples (109) and (110), which in that case would mean 'Is the captain asleep?' and 'Yes, he is eating', 'Yes, he is asleep' (cf. asleep derived out of on sleep). Particles with an original preverbal position tend to turn into affixes during the grammaticalization process, and interestingly enough, some of the occurrences of preverbal ρά in Russenorsk seem to be treated as such (Broch - Jahr 1981: 60).

Aspect marking and grammaticalization

in Russenorsk

with Immigrant

Swedish

149

It may be noted that pa never precedes any of the very frequent stative/ positional verbs stannom, ligge ne, or sitte ne, i. e., words with a distinct notion of imperfectivity which also tend to replace a copula. On the other hand, there are parallel examples of, for instance, presentom, slipom, and smotrom, where pä sometimes is added, sometimes not (examples [119] and [120]), possibly due to the variability in pidgins and maybe also to the fact that the Russenorsk texts were written down by different individuals over many years. There are, however, also examples that indicate semantic differences between verbs with and without pä (examples [121] and [122]). (119)

RN: vesagu fiska presentom please fish present 'Please give (me) fish.'

(120)

RN: veersägo, Uta klceba pä presentom please little bread on present 'Please give me some bread.'

(121)

RN: mo ja smotrom ju kralom I see you steal Ί saw you steal it.'

(122)

RN: davaj päsmotrom please on see 'Please look.'

Whether preverbal pä is interpreted as affecting the whole utterance or primarily the verb or both at the same time, all the occurrences seem to be markers of aspect, in one way or another. The fact that no single interpretation is possible for all occurrences is nothing peculiar to Russenorsk. In fact, this is typical of particles like this in natural languages as well as in pidgins, and the development of pä, both preverbal, and temporal and case-marking, seems to follow the same paths as in many languages. Although Russenorsk pä has a close match in Russian, po, there is, therefore, no need to suggest a direct interference from Russian. Rather, it may be that Russenorsk pä and Russian po have developed in the same direction, starting off as locatives. Mühlhäusler (1986), for example, argues that interference is of minor importance in the creation of

150

Ulla-Britt

Kotsinas

pidgins, especially as regards the development of syntax. Recent research on second-language acquisition points in the same direction. This, however, does not necessarily imply that substrate languages have no impact on language acquisition and pidgin development. Holm (1988: 73), for instance, remarks that apparently universal tendencies in pidgins may be reinforced by substrate influence in some cases. The mere fact that Norwegian and Russenorsk possess the near-homophonous particles pä and po, both of them multifunctional and used for a quite wide range of meanings, may have encouraged some of the developments of ρά in Russenorsk. The Greek speaker, for example, in whose native language no immediately recognizable equivalent exists, uses pä very frequently and in much the same ways as it is used in Russenorsk, with the exception that preverbal ρά is rare. Another Immigrant Swedish speaker, whose native language is Spanish (examples [115] and [116]), however, seems to use pä as a complementizer quite frequently, possibly influenced by the similar use of Spanish para.

Conclusion A comparison between Russenorsk and pidginized varieties of Immigrant Swedish reveals striking similarities as regards the development of aspect markers. In both Russenorsk and Immigrant Swedish aspect seems to be marked by certain, originally locative, lexical items, involved in a process of grammaticalization. There are also apparent similarities between the development of these markers and similar processes in other pidgins and diachronic changes in natural languages. The choice of lexical material is only to a minor extent the effect of interference or transfer from substrate languages. As could be expected, there is an interplay between markers of tense, modality, and aspect. Imperative is in both varieties marked by lexical elements meaning 'please', and epistemic modality and future tense with words meaning 'maybe'. Verbs meaning 'stay' and 'come' and the preposition pä are semantically extended to cover a wide range of meanings and functions. In Russenorsk, stannom 'stay' has developed into a locative copula, while in Immigrant Swedish, komma 'come' functions as an ingressive copula. The preposition pä has in both varieties aspectual functions in temporal expressions. In Russenorsk, pä in preverbal position seems to have different aspect-marking functions, for instance to mark deontic modality, future tense, and ingression.

Aspect marking and grammaticalization in Russenorsk with Immigrant Swedish

151

Abbreviations CP FGC FP FS HCE ΗΡΕ IE IG

Cameroon Pidgin (Todd 1974, 1979) French Guyana Creole (Saint-Jacques Faquenoy 1974) Filipino Pidgin (Bickerton 1977) French-based pidgin, Seychelles (Bollee 1977) Hawaiian Creole English (Bickerton 1981) Hawaiian Pidgin English (Bickerton 1981) Immigrant English (Schumann 1978) Immigrant German, Spanish or Italian immigrants (Klein - Dittmar 1979, Dittmar 1981, Keim 1978) IS Immigrant Swedish (Kotsinas 1984a, 1984b, 1984c, 1985, 1989) ISFi Finnish immigrant in Sweden ISSp Spanish immigrant in Sweden NGP New Guinea Pidgin/Tok Pisin (Sadler 1973, Schuchardt 1989 [1980], Litteral 1969, Romaine 1988) RN Russenorsk (Broch - Jahr 1981)

References Antinucci, Francesco - Ruth Miller 1976 "Guyanese: A French Creole", Journal of Child Language 3: 167-189. Bailey, Charles-James - Roger Shuy (eds.) 1973 New ways of analyzing variation in English. Washington: Georgetown University Press. Bakker, Peter 1987 "A Basque nautical pidgin: A missing link in the history of fu", Journal of Pidgin and Creole Languages 2: 1—30. Belikov, Vladimir - Elena Perekhvalskaya 1989 "Russian Norwegian po reconsidered", Journal of Pidgin and Creole Languages 4: 287-289. Bickerton, Derek 1977 "Pidginization and creolization: Language acquisition and language uniVersals", in: Albert Valdman (ed.), 49-69. 1981 Roots of language. Ann Arbor: Karoma Publishers. Blount, Ben G. - Mary Sanchez (eds.) 1977 Sociological dimensions of language change. New York-San Francisco - London: Academic Press. Bollee, Annegret 1977 "Two morpheme structure rules in an English proto-creole", in: David DeCamp - Ian Hancock (eds.), 118-129. Broch, Ingvild - Ernst Häkon Jahr 1981 Russenorsk. Oslo: Novus forlag. Broch, Olaf 1927 "Russenorsk", Maal og Minne: 81 -130. Bronckart, J. P. - H. Sinclair 1973 "Time, tense and aspect", Cognition 2: 107-129.

152

Ulla-Britt Kotsinas

Bybee, Joan 1985 On the nature of grammatical categories. A diachronic perspective. [Paper presented at the Second Eastern States Conference on Linguistics, Buffalo.] Bybee, Joan - Osten Dahl 1989 "The creation of tense and aspect systems in the languages of the world", Studies in Language 13: 51 — 103. Bybee, Joan - William Pagliuca 1985 "Cross-linguistic comparison and the development of grammatical meaning", in: Jacek Fisiak (ed.), 59-82. Clyne, Michael 1968 "Zum Pidgin-Deutsch der Gastarbeiter", Zeitschrift für Mundartforschung 35: 135-139. Comrie, Bernard 1978 Aspect. Cambridge: Cambridge University Press. Corder, Pit - Eddy Roulet (eds.) 1977 The notions of simplification, interlanguages, and pidgins and their relation to second language pedagogy. Geneve: Droz. DeCamp, David - Ian Hancock (eds.) 1977 Pidgins and Creoles: Current trends and prospects. Washington DC: Georgetown University Press. Dittmar, Norbert 1981 "On the verbal organization of L2 tense marking in an elicited translation task by Spanish immigrants in Germany", Studies in Second Language Acquisition 3: 136—163. Fisiak, Jacek (ed.) 1985 Historical semantics. Historical word-formation. Berlin: Mouton de Gruyter. Fleischmann, Suzanne 1982 The future in thought and language. Cambridge: Cambridge University Press. Fox, James A. 1973 Russenorsk: A study in language adaptivity. [Unpublished term paper, University of Chicago.] Gilbert, Glenn G. (ed.) 1980 Pidgin and Creole languages, selected essays by Hugo Schuchardt. London: Cambridge University Press. Greenberg, Joseph H. (ed.) 1978 Univiersals of human language: Word structure. Stanford: Stanford University Press. Heidelberger Forschungsprojekt 1977 Die ungesteuerte Erlernung des Deutschen durch spanische und italienische Arbeiter. Osnabrück: Osnabrücker Beiträge zur Sprachtheorie. Hyltenstam, Kenneth - Katrin Maandi (eds.) 1984 Nordens spräk som mälspräk [The Scandinavian languages as target languages], Stockholm: University of Stockholm, Department of Linguistics. Holm, John 1988 Pidgins and Creoles. Vol. 1. Cambridge: Cambridge University Press. 1989 Pidgins and Creoles. Vol. 2. Cambridge: Cambridge University Press. Keim, Inken 1978 Gastarbeiterdeutsch. Forschungsberichte des Instituts für deutsche Sprache 41, Tübingen: Günter Narr.

Aspect marking and grammaticalization in Russenorsk with Immigrant Swedish

153

Klein, Wolfgang - Norbert Dittmar 1979 Developing grammars. Berlin: Springer. Kotsinas, Ulla-Britt 1983 "Repetition in immigrant Swedish", Indian Journal of Applied Linguistics 9(1-2): 69-88. 1984a "Ask maybe ten hours. Semantic over-extension and lexical over-use", Scandinavian Working Papers on Bilingualism 2. Stockholm: University of Stockholm, Department of Bilingualism, 23-42. 1984b "On the acquisition of vocabulary in Immigrant Swedish", in: Hakan Ringbom (ed.), 75-100. 1984c "Om prepositionsanvändningen i nägra invandrares svenska talspräk" [On the use of prepositions in immigrants' spoken Swedish], in: Kenneth Hyltenstam - Katrin Maandi (eds.), 188-203. 1985 Invandrare talar svenska [Immigrants speak Swedish]. Malmö: Liber. 1989 "Come, stay, finish. On the development of aspect markers in interlanguage and pidgin/creole languages", in: Lars Gunnar Larsson (ed.), 33-48. Litterai, Robert 1969 A programmed course in New Guinea Pidgin. Hong Kong: Jacaranda Press. Larsson, Lars Gunnar (ed.) 1989 Proceedings of the second Scandinavian symposium on aspectology. (Studia Uralica et Altaica, Acta Universitatis Uppsaliensis 19.) Uppsala: Almqvist and Wiksell. Meisel, Jürgen 1977a "Linguistic simplification: A study of immigrant workers' speech and foreigner talk", in: Pit Corder - Eddy Roulet (eds.), 88-98. 1977b "The language of foreign workers in Germany", in: Carol Molony etal. (eds.), 184-212. Molony, Carol - Helmut Zobl - Wilfried Stölting (eds.) 1977 Deutsch im Kontakt mit anderen Sprachen. Kronberg/Ts.: Scriptor. Mühlhäusler, Peter 1986 Pidgin and Creole linguistics. Oxford: Basil Blackwell. Naro, Anthony 1978 "A study on the origins of pidginization", Language 54: 314-347. Ringbom, Hakan (ed.) 1984 Conference on psycholinguistics and foreign language learning. (Publications of the Research Institute of the Abo Akademi Foundation, 86.) Abo: Abo Akademi. Romaine, Suzanne 1988 Pidgin and Creole languages. New York: Longman. Sadler, Wesley 1973 Untangled New Guinea Pidgin. Madang: Kristen Pres. Saint-Jacques Fauquenoy, Marguerite 1974 "Guyanese: A French Creole", in: David DeCamp - Ian Hancock (eds.), 2 7 37. Sankoff, Gillian 1977 "Creolization and syntactic change in New Guinea Tok Pisin", in: Ben G. Blount - Mary Sanchez (eds.), 119-129.

154 1980

Ulla-Britt Kotsinas

"Variation, pidgins and Creoles", in: Albert Valdman - Arnold Highfield (eds.), 139-164. Schuchardt, Hugo 1889 "Beiträge zur Kenntnis des englischen Kreolisch II. Melaneso—Englisches". Englische Studien 13: 158-162. [1980) [Reprinted "Melanesian Pidgin", in: Glenn G. Gilbert (ed.), 23-29.] Schumann, John H. 1978 The pidginization process. Rowley, MA: Newbury House Publishers. Todd, Loreto 1974 Pidgins and Creoles. London: Routledge and Kegan. 1979 Some day been dey. London: Routledge and Kegan. Traugott, Elizabeth 1973 "Some thoughts on natural syntactic processes", in: Charles-James Bailey — Roger Shuy (eds.), 313-322. 1975 "Spatial expressions of tense and temporal sequencing: A contribution to the study of semantic fields", Semiotica 15(3): 207-230. 1978 "On the expression of spatio-temporal relations in language", in: Joseph H. Greenberg (ed.), 370-400. Valdman, Albert (ed.) 1977 Pidgin and creole linguistics. Bloomington: Indiana University Press. Valdman, Albert - Arnold Highfield (eds.) 1980 Theoretical orientations in creole studies. New York—London: Academic Press.

Greenland

Eskimo pidgin in West Greenland 1 Hein van der Voort

Introduction Because of "the absolute want of scientific information on the subject of northern climates", the Briton O'Reilly decided to become surgeon on the English whaling ship Thomas and sailed to Davis Strait (the strait between the west coast of Greenland and the east coast of Baffmland in northern Canada) in 1817. His observations and opinions that are laid down in his book of nearly 300 pages cover many subjects, for example, the Norse settlement, the Inuit of West Greenland, zoology, the Northwest Passage, etc. On pages 84 and 85, O'Reilly (1818) gives an Eskimo word list, and of the 27 entries, nine are not proper nineteenth-century Greenlandic. The full list is given in Appendix III, but I will mention a few remarkable words here, with their ultimate origin, O'Reilly's translations, and the Greenlandic equivalents of these translations (see Table 1). Table 1. Non-Greenlandic words from O'Reilly's (1818) Eskimo word list O'Reilly

Origin

Translation

Greenlandic

Uskee Cunä Brumik Picaninnee Canu Puma

Algonquian Old Norse Germanic Portuguese Arawak Eskimo

'land' 'woman' 'bread' 'child' 'boat' 'whale'

nuna arnaq iffiaq meeraq, qitornaq qajaq, umiaq arfeq

O-mu-as-a-wak

'Will you go on board the ship?', from Greenlandic umiarsuaq 'big ship' (umiaq); cf. Greenlandic ikaarnialerpit?

If we consider the ultimate origins of the words shown in Table 1, their divergence is striking. Uskee, Canu, and possibly also Picaninnee have probably entered the list through English. O'Reilly does not inform us about the situation in which most of these words were used, nor does he explicitly reveal anything about the way he came by other Eskimo words

158

Hein van der Voort

in the text. How is it that O'Reilly heard these words from West Greenlandic Inuit, who do not use them among themselves, and why did O'Reilly not know this? I will return to the question of the value of O'Reilly in section 3.1.1. In this paper, I want to argue for the former existence of a more or less stable contact variant of the Eskimo language that was used in communication between the Greenlandic Inuit and European whalers and traders from the seventeenth until the nineteenth century. In doing this, I will present unique material from a hitherto unknown source from the eighteenth century brought to light by whaling historian Jurjen Leinenga from the Arctic Centre of Groningen University that gives this hypothesis more solid ground. This paper is part of an overall study on Eskimo contact languages, the result of which will be published elsewhere. Here, I will only deal with the contact variety of Greenlandic in detail, instead of giving a wide, but necessarily superficial overview of all Eskimo pidgins. First, I will have to give a short sketch of the emergence of the historical setting in which jargons or pidgins could have developed in Greenland.

1. The early contacts It is known fact that the first recorded contacts between natives of the American continent and Europeans were those between Scandinavians and Indians and Inuit in the Middle Ages. The Norse (Icelanders and Norwegians) explored parts of Greenland and eastern Canada in the late tenth century A.D., and established settlements. These disappeared around 1450 (Gull0v 1982: 233; but according to Gad 1970 even later) in Greenland and much earlier in Canada. An important cause for the disappearance of the Norse in Greenland was their isolation due to the increasing interest of Scandinavia for the Hansa trade rather than trade with the West, and increasing lack of mobility of the Greenlandic Norse due to the scarcity of large logs of wood on which their tradition of shipbuilding depended. During the nearly 500 years of existence of the Greenlandic settlements, the Norse had occasional contacts with Indians and Inuit on their trips for timber and trade in Canada, and in the second half of this period, they more often met Inuit of the southwardly expanding Thüle culture in Greenland. From the old Norse sagas and Inuit oral traditions

Eskimo pidgin in West Greenland

159

about this latter period (Kleivan 1984), it appears that the Inuit and Norse lived in separate settlements with differing cultures, traditions, means of subsistence, and languages. From archeological evidence, however, it appears that there was frequent and increasing trade contact between both peoples, probably leading to close coexistence (McGhee 1987), and in this situation a trade language may have developed. Today, only a few possibly old Norse loanwords in Greenlandic remind us of the earlier contacts with the Norse. 2

2. Contacts and travel in Greenland According to Gad (1970), southwestern Europeans, presumably pirates and Basque whalers, were seen incidentally in Arctic waters and Greenland, probably before the disappearance of the last Norse settlements; but the first Europeans after the Norse to frequent the northwestern Atlantic on a commercial basis were Portuguese cod fishers (Morison 1971) around A.D. 1500.3 For a detailed account of the history and the nature of pidgins in use between the natives of eastern Canada and early Europeans, see Bakker (1989, 1991b, and this volume). While Basques and traders of other nations were commercially active in Labrador on a seasonal basis, the British explorers Martin Frobisher and John Davis sailed north in search of the northwestern passage to China, and when they landed in West Greenland in 1578 and 1586, they met the local Inuit and traded with them. 4 This encouraged the at that time Dano-Norwegian kingdom to finally come to grips with its longcherished plans to reclaim historical rights to tax, trade, and travel in the old colony, and, heeding the call of its Lutheran consciousness, to convert the Catholic or heathen-again Norse Greenlanders, should they still exist. After a few trips on which Inuit were kidnapped, renewed wars with Sweden put a halt to further imperial activity in the western Arctic. Meanwhile, in the early seventeenth century, aside from their extensive whaling activities in eastern Arctic waters, the Dutch and Frisians from the Netherlands had started regular trips to Davis Strait with the purpose of bartering with the Greenlandic Inuit (especially to obtain narwhal teeth and furs) and for some marginal whaling. Some of the earliest Dutch placenames in Greenland, Delftsche Haven and Brielsche Haven, date from 1624 (Gullov 1985). Still it was only after the "Little Ice A g e " had begun, in the second half of the seventeenth century, that whaling in the Davis Strait increased, pursued expecially by Frisian and Hamburg

160

Hein van der Voort

enterprise. This cold period, which lasted about a hundred years, yielded so much drift ice around Spitsbergen (Svalbard) that the whales were forced to abandon there feeding grounds for western Arctic waters (Vibe 1967, cited in GulLav 1985). By around 1719 (Haan 1720, Zorgdrager 1720, cited in Gull0v 1987) European whale exploitation in Davis Strait had acquired massive proportions. Before the turn of the seventeenth century, there was a trade system between South Greenlanders and Inuit along the west coast of Greenland as far north as Disko Bay. South Greenlanders travelled north to obtain baleen and soapstone, both of which were necessary parts of their fishing equipment, in exchange for caribou skins, fox furs, and driftwood from local Inuit. The trade between Disko and the south took place especially on offshore islands along the way (for details see Gullev 1987), where Greenlanders from different areas held their annual summer camps (aasiviit), and where marriages and other festivities also took place, and conflicts were solved in song contests. During this time, the Dutch, who were mainly interested in trading, anchored on the southwestern shores at places where fresh water could be obtained, and near the locations of the aasiviit, using them as marketplaces. Around 1700 the Dutch and Frisian traders' interest shifted to whaling and they also sailed further north to Disko bay (where whales could be found in abundance), but both the trade of the South Greenlanders with the local Inuit and the trade with the Dutch continued. At Zuyd Bay, on the island Ukiivik, near Holsteinsborg/Sisimiut, many Dutch anchored on their return journey, and a more or less regular market was established. This went on until about 1733—1734, when smallpox transmitted from Europe decimated the population of the Godthäb area, where the Norwegian (Danish) missionary Hans Egede had established a colony in 1721. Following this, only the South Greenlanders kept trading furs with the Dutch for iron, glass beads, kettles, etc., mainly around the Disko area. This came to an end when the Moravian missionaries (who had arrived with the smallpox in 1733) persuaded the travelling South Greenlanders to stay in their settlements. Meanwhile, the Danish crown had become increasingly serious about effectuating its claim to sovereignty over Greenland after 1728, and in 1740, trading with the Inuit became officially a monopoly of the Danes. The Dutch whalers continued to trade, however, even after the Dutch government had officially acknowledged the Danish monopoly in 1762.5 The Dutch (and implicitly all other nations present, such as the British, French, Germans, Spanish, etc.) were

Eskimo pidgin in West Greenland

161

only allowed to hunt whales, and not to land in order to trade with the Inuit, but these rules were now and then violated, until the severe winter of 1777 which destroyed a large part of the Dutch and Hamburg fleet as well as their whaling spirit. This made it an advantageous winter for the Danish and especially British competitors in whaling. The 150 year period of Dutch and Frisian activity in Davis Strait and their contacts with the Greenlanders is well documented in Danish, Dutch, and Moravian (Herrnhuter) mission sources, and amply described in, amongst others, Bobe (1917), Dekker (1979), Gad (1970, 1973) and Gullov (1985, 1987).

3. Contact varieties of Eskimo in Greenland 3.1

Sources

In spite of the fact that the history of European-Inuit contacts in the Atlantic is laid down in such a wealth of contemporary documents and publications, reference to the actual language that was used between its participants is rare, let alone that descriptions or even examples are given. The early eighteenth-century descriptions of West Greenlandic "proper" by Hans Egede were the first of a professional nature, but until this day non-native speakers are rare. Bilingualism was certainly also rare in the eighteenth century, and for a long time thereafter the reverse has also been true: very few Inuit knew Danish or any other European language. In the early contact period there were Inuit who had learnt English or later on Danish, and also Europeans who had learnt Eskimo, and they were used as interpreters. Still, people like the Greenlander Hans Zakaeus (or John Sakeouse) who accompanied John Ross on the expedition to the Polar Eskimos in 1818 (see, e. g., Gad 1984, Oswalt 1979), and the Dutch boatswain who assisted Egede for two weeks in 1721 (Bergsland - Rischel 1986: 11), were exceptions and most of the European-Inuit contacts took place without interpreters. Furthermore, these contacts were short, but rather frequent, as hundreds of ships landed each summer. As this annual trade on such a scale went on for at least a hundred years, contact varieties of Eskimo must have developed. Schultz-Lorentzen (1928a: 217) writes for example: For a long time before Hans Egede arrived in Greenland, the west coast of Greenland was regularly navigated by Dutch, German and English ves-

162

Hein van der Voort

sels, the crews of which traded with the Eskimos, so that the necessity arose for words, which both parties were able to understand.

Unfortunately, no examples are given. As regards a possible Danish Greenlandic Pidgin, Nielsen (n. d.: 79) writes in the 1950s: A curiosity of the early years of Danish colonization was a kind of "pidgin Danish", now out of fashion

but he does not say anything else about this subject either. Not even the decades of Dutch domination in the trade with the Greenlanders, that the Danes were so irritated about, had given rise to bilingualism to any notable extent, as may be inferred from the following passage from the Danish merchant Dalager's diary (1752, cited in Bobe 1915: 45), if we take the word forstaae 'understand' to pertain here to the understanding of language: If the Dutch would thoroughly understand the Greenlanders, and had the proper experience in the trade, it would not take very long until the Danish trading posts would have to pack their bags, (my translation)6

Apparently there was a long tradition of trade contact. Furthermore we know that the trade partners spoke mutually unintelligible languages and that there was hardly any bilingualism among them. These are circumstances of the type in which pidgins emerge. During the twentieth century, Europeans with whom the Greenlanders have had regular intercourse have mainly been Danes. The coexistence of Danes and Greenlanders in the larger settlements has necessitated the continued use of the pidgin which Schultz-Lorentzen (1928b) calls Kolonisprog 'colonial settlement language' and about which he writes: In Greenland it has happened that the main ingredient of the colonial settlement language is decidedly Greenlandic, but certainly a form of Greenlandic which has become influenced and reshaped, so that a fully uneducated strange Eskimo, indeed a Greenlander, would feel it as something strange and different from the language that is his own. (my translation)7

He suggests that it started to emerge even before the whaling era, in Norse times, and that there are also contemporary variants, like Kekkengronlandsk 'kitchen Greenlandic': This language plants its most remarkable flowers in the Danish kitchens, where the maid talks confidentially with her Danish mistress about all sorts of things in the language that is called kitchen Greenlandic and which is an interesting variant of the usual colonial settlement language, until finally both of them believe they are speaking correct Greenlandic. (my translation) 8

Eskimo pidgin in West Greenland

163

Today, the Greenlandic population is becoming increasingly bilingual as many learn Danish fluently. Danes, however, are less likely to learn Greenlandic (see Kleivan 1977), so there are still circumstances in which the necessity for a pidgin has in principle not diminished. As hardly any research has been done on present-day contact varieties of Greenlandic, we appear to have more information about past varieties. Until now, only a few sources have surfaced which contain forms that provide data on such varieties. They will be introduced in the following subsections. 3.1.1. O'Reilly

(1818)

In the introduction, we have seen some words from O'Reilly. It must be mentioned here that O'Reilly's work is not considered by Eskimologists to be a serious contribution to the study of the Arctic. O'Reilly was already severely criticized by contemporary reviewers, and is reputed for this in the Arctic Bibliography (Tremaine 1953,2: 1910). His critics (e. g., Anonymous 1818) falsified many of his observations, exposed his plagiarism of other works about Greenland and mock his Greenlandic vocabulary by ironically stressing its supposed "purity". 9 The value of this source for the purpose of this paper, however, lies in the fact that the etymology of the vocabulary points to pidginization. Already the (French-born) German romantic writer and traveller Adelbert von Chamisso (1781-1838) took notice of the unusual ("angeblich grönländischen" [supposedly Greenlandic]) character of O'Reilly's list, and the creolinguist Hugo Schuchardt (1883, cited in Gilbert 1980) used this observation in his reconstruction of the spreading of words like pickaninny 'little, child' from the whalers' jargon to and throughout the Pacific. There is some ground for suspecting O'Reilly of never himself having heard Greenlanders speak, and instead having got his information from other books and having made up the word list from what he heard the ship's crew say. It is strange, for example, that he writes about the word Karalit 'Greenlanders' in a well-informed manner, but that he does not mention this common Greenlandic word in the word list, and instead uses the English nickname JJskee there (see appendix). Still, O'Reilly's list contains some fascinating items confirming the continuity of a tradition of pidginized Greenlandic. For example Meyer's (1767; see section 3.1.3 below) word Promek 'bread' is found in O'Reilly as Brumik, and Boemek 'whale' as Puma. Fortunately, we came across a highly interesting remark on this latter word in Brown (1868: 5 3 4 553).10 About "whale" (Balaena Mysticetus), Brown writes, after having summed up all the established names for it in different languages:

164

Hein van der Voort

I have also heard both the Greenlanders and Western Eskimo call it puma, but I cannot learn what is the origin of this word, and suspect it to be whaler, - a corrupted jargon of Scotch, English, Danish and Eskimo, joined with some words which seem to belong to no language at all, but to have originated in a misconception on either side, and to have retained their place under the notion that each party was speaking the other's language, something of the nature of the Lingua Franca of the Mediterranean, the Pigeon English of China, and the Chinook jargon of north-west America.

As appears from Brown, there was an Eskimo pidgin called "Whaler", and it seems justified to suppose it was in some way related to Greenlandic Pidgin. 3.1.2 Resen (1687) Another source of Greenlandic pidgin is a German-Greenlandic word list in a manuscript from 1687 about Greenland by the Danish scholar Peder Hansen Resen (1625-1688), which was only recently published (Kisbye Melier 1987). This word list is in part based on words that were taken down from the mouths of the three Greenlandic women, Kabelau, Gunnelic, and Sigjo who were kidnapped (with a man, Ihiob, who died in Norway) and taken to Denmark by explorer David Dannel in 1654 (see Bobe 1916).11 The list contains some forms that can be seen as analytic expressions of notions that in normal Eskimo would be expressed synthetically. It concludes with some sentences and shorter phrases from Dannel's travel diaries which are characterized by Petersen and Rischel (1987) as "pidgin-language" because of their analytic traits. 3.1.3 Meyer (1767) The least known and most significant source of pidginized Greenlandic lies chronologically somewhere in between the previous two sources. This is the travel journal of a German surgeon from Ansbach, Johann Michael Meyer, which was published as a book in 1767.12 Meyer was a newcomer to the Arctic. His first journey as a ship's doctor was on a Hamburg whaler to Spitsbergen (Svalbard) in 1763. His second trip, in 1764, on another Hamburg whaler called the König Salomon (King Salomon), under the command of Wilhelm Heinrich de Hahn from the Frisian island of Borkum, brought him to Davis Strait and Greenland. The purpose of this journey was whaling and trading with the Inuit. Meyer's experiences in Greenland are laid down in the second part of his book, and cover 100 of the total of 163 pages.

Eskimo pidgin in West Greenland

165

The ship on which Meyer sailed in Greenland coastal waters was manned by a crew of 45 persons in total. It set off in Hamburg on February 25th. On the 27th of April South Greenland came into sight, but only after two weeks, on the 13th of May, did the wind allow the ship to enter Davis Strait. From the 21st of May they sailed in the waters around the island Disko, until the 11th of June, when they turned to the open sea again to hunt whales. On the 5th of July they saw the Canadian coast at latitude 70° North (the same latitude as Disko Island). On the 11th of July they commenced their return voyage to Hamburg, where they arrived on the 24th of August. The trading activities took place mainly around the Whale Islands in Disko Bay, just south of Disko Island, an area that was frequented by the European whalers of that time. During the three weeks between the 21st of May and the 11th of June the crew including Meyer went ashore on several of the islands to obtain fresh water, collect scurvy grass and other herbs, and hunt hares and geese. Also, there was much trade with the Inuit, but this was preferably done aboard the ship. Although the crew was rather afraid of the Inuit, they had brought the usual trading goods like beads, combs, mirrors, fishhooks, etc., and exchanged them for the usual skins, walrus teeth, etc. Meyer devotes several paragraphs to the Inuit, the way they make kayaks, clothes, houses, the way they hunt, and the way they trade. These paragraphs are not only interesting because of the eighteenth-century ethnographic observations they contain, but also because of the northern European attitude towards other peoples of that time which they reflect, and the way they at the same time reveal something of the European self-image of economic, scientific, cultural, moral, and religious superiority. Meyer furthermore describes several situations in which the Inuit and the crew, including himself, talk to each other, but he says nothing about the nature of the language that was used, nor the way or the amount of time in which this language was learnt. The last five pages of the work contain a list of the "most usual words that the inhabitants of the islands that lie in the Strait Davis avail themselves o f ' (Meyer 1767: 159). It is a list of 122 supposedly Greenlandic words, with their German equivalents. The interesting thing about this list is that it is shot through with non-Eskimo words, some of which have also been listed in other sources that contain pidgin words. Another reason for the significance of Meyer's book in relation to pidginized Eskimo is that quoted conversation in 'Greenlandic', with translations, can be found in the narrative. Most of these illustrative sentences show atypi-

166

Hein van der Voort 75

150

225

300 km =)

BAFFIN BAY

Whasen eyland / hare island waygatz / vajgat disko eyland bläfjeld / blauen land (?) wallfisch eylande rottgansz eyland hunde eyland nord bay / — nord bucht? rifkol / rebko (?) cap de walsingham sud bay / ukiivik -

0

.schwarze vogel bay / appat . klokkerhuk / alluttoq - rodebay / oqaatsut makkelijk out jacobshavn green islands / angisat christians häb süd ostenbucht aqqitsoq & kanissut

JAMES EYLAND / BAFFIN ISLAND

KALAALLIT NUNAAT/ GREENLAND noemey / nuuk

STRAIT DAVIS

Places visited or mentioned by Meyer 1767

staatenhuck / kap farvel / faruel •

Eskimo pidgin in West Greenland

167

cal Greenlandic features, reminiscent of a pidgin. Because Meyer's journal is the source I will mainly concentrate on, I will try to explain its rarity and give the existing bibliographical information in the following section. 3.1.4. Bibliographic information on Meyer (1767) The title on the first page of Meyer's journal is: J. M. Meyers /eines / deutschen Chirurgi / Beschreibung / seiner / auf den Wallfischfang / nach / Spitzbergen / gethanen Reise. / Mit Obrichkeitlicher Erlaubniß. — Straßburg - gedruckt und zu finden bey Joh. Heinr. Heitz, / Universitäts-Buchdr. 1767. [J. M. Meyer(')s (a German surgeon's) description of his journey to Spitsbergen on a whale hunt. With official permission. Strasbourg: printed by and available from Joh. Heinr. Heitz, University book printer. 1767.] (my translation, the slashes indicate a new line)

The first 60 pages of this book contain his travel diary of his trip to Spitsbergen. Pages 63 to 163 form the second part, in which he describes a journey to Greenland and Davis Strait. The title page of this part reads: J. M. Meyers / Reise / in die / Straße Davis. [J. M. Meyer(')s journey to Davis Strait.]

On the inside of the cover of the book the following is hand written, with the name "Meyers" in Roman script and the rest in Fraktur: Meyers / Erste Reise 1763. / 2te Reise 1764. [Meyer(')s First journey 1763. Second journey 1764.]

As far as the information of the present author goes, the book is mentioned only in Popst and Schöller (1983, Part 95: 387). There it is referred to as: "Meyer; Johann Michael. Beschreib, seiner auf d. Wallfischfange nach Spitzbergen gethanen Reise u. in die Straße Davis 8. Straßb. 767", mentioning both his journeys to Spitsbergen and Davis Strait. Because of the discrepancies between this reference and the actual information on the title page of Meyer, it is possible that Popst and Schöller got their information from a publishers' catalogue or another indirect source. The name "Meyers", however, as printed on the title page of the original, is most likely a genitive form of "Meyer". Popst and Schöller apparently assumed this to be the case. In their bibliography, the author's name is apparently printed in Roman type by the editors themselves, whereas the rest of the reference is a facsimile copy of an earlier reference printed in Fraktur. The abbreviation of 1767 to 767 is conventional. In Popst and

168

Hein van der Voort

Schöller (1983, Part 93: 542) we find Meyer as: "Mayer, Johann Michael, s. Meyer" [Mayer, Johann Michael, see Meyer]. There are a few reasons that Meyer's work is probably relatively unknown. One reason is that the title page only refers to Meyer's first journey and therefore explicitly mentions only the name Spitsbergen. Spitsbergen was in those days often called New Greenland, and what we now understand by Greenland was called Old Greenland, as it was thought that Spitsbergen was a part of Greenland. Spitsbergen also was often just referred to as Greenland, but because of the lack of such obvious ambiguity here, Meyer's book may have been ignored by Eskimologists. Another, maybe more important reason is that the work seems to be quite rare, in spite of its appearance in print. This book was brought to my attention by Peter Bakker and whaling historian Jurjen Leinenga, who encouraged me to work with the material. Leinenga obtained a photocopy from the owner, a book collector in Austria. His copy may very well be the only one extant. In the meantime, however, the book's owner has passed away and it is not known what his heirs have done with the collection, and with this book in particular. This makes publication of all the linguistic material it contains (Appendices I and II of the present paper) all the more important. A facsimile reproduction of the whole work is intended. Before I deal with Meyer's material, it is necessary to give a short description of the Greenlandic language, in order to place the kind of simplifications and adoptions in the pidgin which it reflects in a linguistic context.

3.2 The structure of West

Greenlandic

Greenlandic forms a part of a linguistic continuum that stretches from the northeastern tip of Siberia through Alaska and northern Canada to East Greenland. There is no clear evidence that Eskimo is historically related to any other language except Aleut spoken around the Bering Strait. Greenlandic syntax can be characterized as variable, but basic word order, in all moods, is SOV, the basic structure of the NP is "head modifier", and possessive constructions have "possessor-possessed" order. Greenlandic has only three lexical categories: nouns, verbs and socalled uninflectionable particles. There are no adjectives, articles, or pre-

Eskimo pidgin in West Greenland

169

positions. There are no gender distinctions in Greenlandic. There is no infinitive form of the verb. Verbs agree in person and number with both subject and (indirect or direct) object, and personal pronouns are only used for emphasis. Verbs can also be inflected for eight different "moods", four of which can form subordinate sentences. Nouns are inflected for one of eight possible cases. Six of them are so-called "local" or "oblique" cases for different semantic functions. There are two grammatical cases, the absolutive and the relative and the case-marking system is ergative. This means that the subject of an intransitive verb has the same case marking as the object of a transitive verb, here absolutive, whereas the subject of a transitive verb is marked differently, here relative. The instrumental case is often used for an antipassive relation. I will give some examples of transitive and intransitive verb agreement: (1)

anguti-p arnaq man-Rel women.Abs taku-aa see-he.him (either masculine, feminine, or inanimate) 'The man saw the woman.'

(2)

angut pisup-poq man.Abs walk-he 'The man walked.'

The following example shows absence of lexical pronoun and subordinate mood: (3)

taku-gakkit see-I.you.Ca 'Because I saw you.'

In possessive constructions, the possessor is marked with the relative case form, functioning as a genitive. The possessed is marked for number and person of the possessor, its own number, and for case depending on the function of the whole constituent in the sentence: (4)

{anguti-p) illu-a man-Rel house-his '(The man's) His house.'

170

(5)

Hein van der Voort

(anguti-t) illu-init. man-Rel.pl house-their.Abl 'From (the men's) their houses.'

Eskimo morphology is further characterized by a very extensive and productive derivational component, and in relation to this it is reputed to be a polysynthetic language. An Eskimo word has a lexical stem to which by recursive derivation an in principle endless string of suffixes can be attached. There are about 400 productive derivational suffixes, a number of which have category-changing effect, while there are only two prefixes. Composition, however, is (with a few marked exceptions) not possible in Eskimo. The basic Eskimo word structure contains only one lexical stem, as in: (6)

[stem- + (derivation[s]) + inflection[s] + (enclitic[s])]

An example: (7)

iminnger-naveersaar-tu-nngor-tussaa-vunga. [stem—derivation-der-der-der—inflection] be.inclined.to.drinking-try.not.to-one.who-become-shouldI.Ind Ί should become a teetotaller.'

Although West Greenlandic phonology is not representative for other dialects, especially not for the more archaic western dialects in Alaska, it has undergone little change since the eighteenth century. In this paper, I will touch upon phonology only occasionally. Some specificities, however, may not be left ummentioned. The phonological system has: i. (three) vowels: /a/, III, and lul (with the allophones: [e] and [o] before /q/ or Irl); ii. and consonants: I f , g, j, k, 1, m, n, q, p, q, r, s, t, v/. The former distinction between alveolar Is/ and post-alveolar Isl is now rapidly disappearing. The Iql is a voiceless dorso-uvular plosive. There are two phonologically distinctive quantities of vowels and consonants: long and short. A geminate (long) /!:/ is voiceless and fricative, as we know it from, for example, Welsh and Icelandic.

Eskimo pidgin in West Greenland

171

In the present paper, I have used the modern West Greenlandic orthography. The older standardized spelling was designed in the nineteenth century by the well-known Samuel Kleinschmidt, son of a Moravian missionary, and the first linguist of Greenlandic who attempted to approach the language in a way that was as unbiased as possible, i. e., not modelled on Latin grammar. His orthography, which officially was in use until 1973, differs from the modern one in several respects, of which I will mention the most important here. First, phoneme quantity was indicated by diacritic symbols, where today, the phoneme symbol is doubled. Secondly, several diphthongs have generally become long monophthongs in modern West Greenlandic pronunciation. This is now also reflected in the modern orthography by double vowels. Also, many consonant clusters were assimilated, resulting in geminate consonants, a fact which, with some exceptions, is reflected by double consonants. Finally the palatalized /s/, which used to be represented by ss is now spelled s. In the modern pronunciation the difference from /s/ is not expressed any more, nor is it any longer distinguished in the orthography. Because there are many assimilatory processes in modern West Greenlandic morphophonology, the underlying morphemes are less recognizable than they used to be. Where the older spelling used to be more transparent on the morphological level, the new spelling is phonologically more representative. In short, morphology is a very important component of Eskimo grammar. It is far more elaborate than that of any Indo-European language. Typologically, Eskimo and the European languages are nearly as far apart as languages can be. A detailed description of West Greenlandic, the Eskimo dialect on which Meyer's material is based, can be found in Fortescue (1984). 3.3 The pidgin

material

In this section, I will discuss the characteristics of the West Greenlandic pidgin in the following order: syntax, morphology, lexicon, phonology. My sources are Meyer (1767), O'Reilly (1818), and Resen (1687). All the data found in Meyer (1767) and O'Reilly (1818), including the original translations, have been reproduced in the appendices below. 3.3.1 Syntax I will compare the pidgin sentences with both Germanic (as a generic term for Danish, Dutch, Frisian, Low German, and High German, all of

172

Hein van der Voort

which may have played a role) and Greenlandic structure, especially where these differ. The syntax reflected in the sentences is variable. The original translations and their genuine West Greenlandic equivalents can be found in Appendix I, where a list of all sentences encountered in Meyer is found. The numbers in parentheses following some examples indicate, respectively, the page and phrase number in Meyer. The following sentences show OV order, as in West Greenlandic: (8)

Kiak Kinoka snipa. (106:8) kayak (of) captain lower-it/sloop Ο V 'Launch the (captain's) sloop.'

In Resen (1687) we find: (9)

Una Pile him/her let him/her Ο V 'Let him/her (get/have/do [?] it).' 13

More sentences, however, show a tendency towards VO order, which is more in line with Germanic syntax: (10)

Besingele Kakamia Bosamia. (107:9) see fox-skin (and) seal-skin V Ο 'See (these) fox skins (and) sealskin.'

(11)

Besingele pingele. (116:29) see child V Ο 'See the child.'

(12)

Koenet besingele tante. (114:25) girl see tent V Ο 'GirlQ see the tent.'

Unfortunately, most sentences in Meyer are interrogative or imperative expressions. In genuine West Greenlandic the difference between these moods and the indicative has no direct consequences for word order or

Eskimo pidgin in West Greenland

173

lexical expression of pronouns, however. In Resen (1687) we find a truly indicative sentence where a subject is expressed and SVO order can be observed: (13)

Uvanga I S Ί speak

Ocaluctung Maccua Invin. speak this/many Inuit V Ο (like/to) these/many persons.'(?) 14

Still, (S)VO order is not always the case, as follows from Meyer: (14)

Kynoka handla- Kakkamia Bosamia ajongelak besingele. basche (106:6) captain trade foxskin sealskin (it.is) good see a. S V S V V '(Does the) captain (want to) trade? (The) fox skins (and) sealskin is good. See!' b. S V Ο V V '(Does the) captain (want to) trade fox skins (and) sealskin? It is good. See!'

There are different ways to analyze this sentence. It is possible that Meyer noted down parts of conversation he remembered without paying attention to the fact that they may consist of several distinct sentences. In reading a. of this example, the first two words constitute a sentence, and the next three also, and the last one could have been a one-word utterance. In reading b. on the other hand, the first four words belong together as an SVO structure, leaving the fifth and sixth words as one-word sentences, with their argument represented by verb inflection instead. The form ajongelak is a nearly exact phonological and morphological match of the genuine Eskimo verb ajunngilaq 'it is good', 'fine', 'excellent'. Still, it is an often used one-word expression, and Europeans who are not acquainted with Eskimo proper will take it for an exclamatory adjective. The last word of the next utterance shows an absence of pronouns: (15)

Kynoka König Salomon ajongelak ishigaget. (106:5) commander King Salomon it.is.good I.saw.you NP V V 'Commander (of the) King Salomon? All right. I saw you (already before).'

174

Hein van der Voort

where ishigaget is a fully inflected transitive verb that can occur as an independent sentence: isigaakkit Ί saw you/looked at you'. It consists of the stem isigi- + first person singular subject and second person singular object indicative marking suffix -vakkit. This one-word sentence, with the absence of overt lexical arguments, corresponds fully to Eskimo "proper". 1 5 Such nonreduced grammatical forms are rare in Meyer, however. The sentences also show the loss of function words (grammatical morphemes or lexemes), characteristic for pidgins, like the lack of an interrogative element in the same example, (15), above: '(Are you) commander (of the) King Salomon?'. The same sentence may contain an understood possessive element, as the ship is called King Salomon. Conjunctional particles or Greenlandic enclitics are also lacking, as in example (10), where conjunction is expressed only by bare juxtaposition. The internal order of the possessive construction in (15) is not the canonical Greenlandic, but rather possessed-possessor. The same is the case in (8) where a possessive indication is also lacking: 'sloop (of the) ship/commander'. In short, function words are lacking, the order of possessed-possessor is Germanic rather than Greenlandic, as is the order verb-object. 3.3.2

Morphology

The forms in the sources are not characterized by systematic inflection. The few inflected verbal forms that were found have the most neutral usage in real Eskimo, like third person singular subject indicative for example in ajongelak (although that is used also interjectively in West Greenlandic). Some verbs have no inflection at all (an impossibility in genuine Greenlandic). Absolutive case is found on practically half of the roughly 90 nouns in the word list, and the other half (including most of the numerals) are in the instrumental case, which is the most unmarked oblique (nonobligatory argument) case. There is no specific "citation form" of nouns in genuine Greenlandic. Identification of items is possible by means of expressions that require an absolutive or an instrumental case form. The instrumental case in Greenlandic often has the value of an indefmiteness (or nonspecificity) marker of direct objects and is also used to refer to indefinite masses. 3.3.2.1 Nouns Of the 46 instrumental case entries, six are non-Eskimo words. Probably, the instrumental (the inflectional ending -mik) is interpreted in the pidgin

Eskimo pidgin in West Greenland

175

as the neutral citation form for nouns. In the first place, the instrumental is semantically one of the most neutral cases in Eskimo. In the sentence, it marks the oblique object and/or can have instrument/manner/means adverbial status. Furthermore, it is the object case under antipassivization of the verb, which occurs frequently. It may be the most frequently heard nominal ending, and the Danes sometimes poorly and half-jokingly stereotype West Greenlandic as "m/A>language". Ironizations like kaffemik 'hen party' (lit. 'coffee'-instrumental marker) and dansemik 'dance party' (lit. 'dance'-instrumental) have found their way into the Danish language, and they may originate from the Danish Greenlandic pidgin SchultzLorentzen (1928) writes about. None of them have found their way into West Greenlandic. The Dutch may also have noted the frequent ending -mik. In the pidgin, -mik clearly lost its original function. Thirdly, also two verbal stems seem to be provided with instrumental markers (see Kyatcemik 'is small' and Kyakkatiomik 'a little boat'). Locative case is only found in place names, and all place names are in the locative, apart from two unclear cases, Nolok and Karleproetramik. Inflections that could be fossilizations are found in composition-like kakamia 'foxskin' and bosamia 'sealskin' which seem to come from kakap amia and puisip amia. Here the final -a is a relict of a Greenlandic third person singular possessive inflection: (16)

kaka-p ami-a fox-Gen skin-his 'fox's skin, the skin of a fox'

(17)

puisi-p ami-a seal-Gen skin-his 'seal's skin, the skin of a seal'

3.3.2.2 Verbs Although some verbs have inflection markers, forms like those of the European-based verb handlabasche must be fossilizations, as this word appears in different examples with different person and number translations. In (14) it was used to mean 'you trade it with us', in (18) it means 'we trade it with you': (18)

App Handlabasche. (106:7)16 yes trade-I/we-you(pl) / -you(pl)-them Ί/we want to trade with you.' [Meyer's translation] 'You want them (the skins from us).'

176

Hein van der Voort

The stem handla- is Germanic (see Appendix II). The ending can be the reflex of the following Greenlandic forms: (19)

-vassi -vaasi -varsi -vast

'-l.you(pl) (Ind/Int)' '-he.you(pl) (Ind/Int)' '-you(pl).him' '-you(pl).them'

The pidgin word could have been derived from any of these. The Greenlandic ending probably has no function in the pidgin, however, since its meaning is clearly variable. The intricacies of Greenlandic anaphoric reference agreement and of phonological quantity were apparently disregarded. Another form that could be a fossilization is (20): (20)

Symene ishigaget. (161:79) where you.see.me? 'Where did you see me?' [Meyer's translation]

The West Greenlandic "proper" expression most closely resembling (20) has a somewhat different meaning: (21)

su-mi=mi isiga-akkit what-Loc=but see-I.you? 'But where did I see you?'

Meyer's translation of ishigaget in example (15) above is also correct from the perspective of West Greenlandic inflectional grammar: W G isigaakkit Ί saw/looked at you'. Viewing Meyer's translation of (20) from a West Greenlandic perspective, however, one would expect a different form, namely one with a second person subject and first person object inflection: WG sumimi isigaanga 'But where did you see me?'. This suggests that the verb form isigaakkit was used as the generic verb 'to see'. I will return to verbal morphology immediately below. 3.3.2.3 Analyticity The constructions in Meyer's material appear to be rather analytic. The tendency to simplify synthetic constructions through alternative analytic expression is a phenomenon of which we find rather clear examples in

Eskimo pidgin in West Greenland

177

Resen (1687) and in various sources from Canada and Alaska (among them Dorais 1979, Hanbury 1904, Stefänsson 1909, Tyrrell 1897 [1908]) and which can be seen both as a foreigner-talk strategy and as a pidgin feature. In example (13), from Resen, repeated here, the subject is expressed lexically with a pronoun: (22)

Uvanga Ocaluctung Maccua Invin I speak(he/I) this/many Inuit Ί speak (like/to) these/many persons.'(?)

Note that the subject is not expressed simultaneously by the inflection of the verb. In West Greenlandic arguments are always expressed by means of inflection, and simultaneous pronominal expression of the argument of a verb has strong focusing effect, but that is not the function of the pronoun in (22). In the pidgin the pronoun is the only way to express an argument, and inflection of the verb is a fossil. 17 In West Greenlandic, a possible equivalent of (22) is: (23)

Inut-tut oqallorip-punga. person-like speak.well-I Ί speak Greenlandic well.'

The synthetic structure of West Greenlandic has an analytic counterpart in Greenlandic pidgin. Analyticization is also very obvious in Resen, with the following juxtaposition of the bare concomitants in a "possessive" relation: (24)

a. Uvanga Nulia. I wife/his.wife 'My wife.' (lit. Ί (his-)wife') 18 b. compare WG nulia-ra. wife-my 'My wife.'

The same is also found in other Eskimo pidgins, e. g., in Stefänsson's (1909: 224) Herschel Island trade pidgin:

178

(25)

Hein van der Voort

a. awona kammik. I boot 'My boots.' (lit. Ί boot') b. compare WG kami-kka. boot-my.pl 'My boots.'

3.3.2.4 Derivation West Greenlandic derivational grammar is highly complex and in principle infinitely recursive. It enables the Inuit to express innumerable shades of semantic distinction in one word, something for which Europeans need separate words and complex syntax. The pidgin material does not show any sign of the derivational possibilities that West Greenlandic "proper" has. If the semantic distinctions that appear in Meyer's German translations of the pidgin (for example [15]) were translated into (modern) grammatical West Greenlandic, the words would turn out to be much longer: (26)

isige-qqamminngil-akkit. look.at-a.while.ago.lsg.2sg.Ind. 'I've seen/looked at you a while ago.'

(27)

isige-reer-pakkit. look.at-already-1 sg.2sg.Ind. 'I've already seen/looked at you.'

Sometimes Meyer's words do contain derivational morphemes, but they are hardly recognizable, if at all. One of the rare exceptions to the derivational intransparency of the forms in Meyer is ajongelak, from ajor'be bad' + the negating suffix -nngiC-, although this is a rather fixed expression: 'it's fine'. This derived word probably has not been transparent to Meyer at all, note his rendering of the underived word ajorpoq 'it is bad': eupok (see Appendix II). Sometimes, an unrecognized derivation has helped me in finding out about the meaning of the stem of a listed word, for example: Akesjamik 'is big' ["ist groß'] which may be aki 'price, payment, counterpart' + -(r)suaq 'big, bad one', which used to contain a palatalized III, < akersuaq

Eskimo pidgin in West Greenland

179

4- Ins < akersuamik 'high price'. The Greenlandic word for 'is big' is angivoq. In one case, something that looks like a foreign diminutive is found: Kornette 'young girl' ["Ein junges Mägdchen"], which consists of kone 'woman' and possibly -ette. Could this be the diminutive and feminine suffix -ette that originally comes from Latin and was spread by French throughout Europe? Maybe it was a productive suffix in Inuit French Jargon, but it could also have come from other European languages. This suffix is found for example in French loanwords in Middle Dutch (Cefas van Rossem, personal communication), but it is rather unlikely that it was incorporated into Greenlandic Pidgin. Although Eskimo has, with a few exceptions, no composition, some compounds are found in Meyer, e. g., kakamia 'fox skins' and Koenekiak 'women's boat', the latter of which is built up of kone 'woman' and qajaq 'kayak' and which must denote 'umiak', the traditional open skin-boat that is used to transport whole families and which is rowed by women, with single-bladed paddles. The Danish word for it is konebäd 'women's boat', and Dutch whalers also used koeneboot in referring to these boats. It is even reported that the Inuit used the word koeneboot (Kat 1818: 31). This is an indication of the early colonial Greenlandic Pidgin Danish referred to in Nielsen (n.d.: 79). In short, more or less analytic structure, frozen inflections, use of pronouns, nonproductive derivation and the use of compounds (rather than inflected possessive constructions) all are characteristic of this pidgin, and of pidgins in general. 3.3.3 Lexicon As regards the lexicon, we found roughly 140 different (pidgin) words in Meyer. Quite a few of these are apparently not used in the same way as in West Greenlandic. Above (in examples [15] and [20]) we have seen the Greenlandic stem isigi- 'look at, observe' used in the sense of 'see, meet'. In early as well as in modern genuine West Greenlandic, taku- is more often used in this sense. Some of Meyer's forms are both morphologically and semantically quite transparent and cause little difficulty in interpretation, whereas others are obscure in both respects, and require quite an interpretational effort in order to find an explanation in terms of a related form from the same semantic field. An example of the latter is:

180

(28)

Hein van der Voort

Oneto 'ribbon', and also 'small glass beads of various colors, that they hang on a string, to be able to sew them on their clothes, like our gold and silver braids (galloons)'. The root could be related to WG unger- 'tie (it) up with ribbon', ungerut 'tying rope' or to unngutit 'certain wrinkled part of a boot-sole'.

Sometimes, the relation is somewhat clearer, as in: (29)

Boemek 'whale' ['Ein Wallfisch"] - West Greenlandic verbal bare stem puumig- 'dip head under water'. May be related to Brown's (1868) and O'Reilly's (1818) pumä 'whale' (see section 3.1.1 and Appendices II and III).

In a few cases, etymological reconstruction faces the danger of becoming subject to the "cafeteria principle". That is the phenomenon where there may be so many etyma possibly involved in the emergence of a certain trait or a word that there is always a language or a form at hand that matches it, which reduces the work of reconstruction of its origin to a blind choice (as if from a cafeteria). Under such circumstances, a shot in the dark can hardly miss, but as an explanation it is weak, as in: (30)

Sinen Gademik 'a cutting knife' ["Ein Schnittmesser"] - WG savik 'knife' is found in Resen (1687), among other forms as Sivenbeck, but that does not explain all of Meyer. Maybe the second word has something to do with WG kaataq 'hammer' + Ins > kaatamik (to make a knife with), but on this level of speculation one could just as well propose a European etymology: Dutch gat 'hole' (that can be made with a knife), and maybe also for Sinen < dialectal Dutch snieen 'to cut'.

In some cases, the relation is more pragmatic than semantic. For example to relate the form tui 'shoulder' to the notion 'son-in-law' one has to consider the concrete situation that was at hand when Meyer tried to communicate with a Greenlander. (31)

Douwe 'son-in-law' ["Tochtermann"]

He may have put his hand on the shoulder of the Greenlander's son-inlaw. Or when asking what the word for 'floorboard' was, he may have

Eskimo pidgin in West Greenland

181

pounded on a floorboard with a loading rod for guns, which is called immiit in Greenlandic: (32)

Emietiemik 'floorboard' ["Eine Diehle"] - WG immiit 'ramrod' (which he pounded on the floorboard with), Ins > immiimmik / immiisimik.

Due to the lack of knowledge of languages of the other party, there must have been mutual misunderstandings as to what was intended. It is also well possible that Meyer or someone else has corrupted the manuscript when it was prepared for printing. Also, he may have had second-hand information in that he wrote from memory a little while later, or from what other Europeans told him. Such unverifiable possibilities will, however, not be taken into account here too often. Many of Meyer's entries are not words one would expect to find in the lexicon of a restricted trade jargon, and this explains why such words are often fully equivalent to contemporary Eskimo forms. But forms in Meyer's word list denoting objects exchanged in trade are also often ordinary Eskimo words. The pidgin character of the language variety as represented by Meyer's list or rather, by forms from this list, is only apparent with a number of forms (the full annotated word list is given in Appendix II): (33)

Abysie 'gun' ["Eine Flinte"] - maybe older Danish bysse or older Dutch bus(se) 'gun'.

(34)

Askepcese 'young boy' ["Ein junger Knab"] - Dutch Assepoester or Danish Askepot 'Cinderella', not WG assaq 'ashes'.

(35)

Bingele 'children' ["Die Kinder"] - maybe Portuguese pequenino 'little child', the pidginized form of which is also found in O'Reilly: picaninnee 'child', or Dutch or German Bengel 'cheeky/impudent child'.

(36)

Bliktemik 'tin' ["Ein blecherner Kessel"] - Dutch or Danish blik 'tin/white iron' + Ins.

(37)

Bogaklek 'whiting' ["Ein Weißfisch"] - Portuguese bacalhau 'cod'?

182

Hein van der Voort

(38)

Boemek 'whale' ["Ein Wallfisch"] - West Greenlandic bare verb stempuumig- 'dip head under water'. O'Reilly (1818) and Brown (1868) mention pumä 'whale'.

(39)

Dennemik 'sword' ["Ein Degen"] — probably from Danish degen 'sword' which is pronounced approximately as [de:j(a)n] + Ins.

(40)

Hageltimek 'shot' (shotgun charge) ["Schroot"] (109:11) from Dutch, Frisian, or German hagel, or Danish hagl 'shotgun charge, shot' + Ins.

(41)

Handlabasche: in context (106:7): App Handlabasche Ί want to trade with you.' ["Ich will mit euch handeln"] - Dutch handelen, German handeln, or Danish handle 'trade' + a West Greenland verbal inflection -vassi.

(42)

Hoekemik 'stockfish' (cod) ["Stockfisch"] - Dutch heek 'type of cod of which stockfish is made' + Ins, or WG uuaq 'tommy/fjord cod'. Also in Thevet (1558).

(43)

Kakkamia 'foxskin' ["Ein FuchsfelF] — WG kakap amia 'fox's skin'. The West Greenlandic word for 'fox', however, is terianniaq. Maybe pidgin based on WG kakkakkaartor- (v) 'to bark (especially of a fox)' like English bow-wow that is used to indicate a dog (e. g., in conversation with small children).

(44)

Kcene 'the son's wife' ["Des Sohns Frau"] — ΟΙ kone 'woman', found in Eskimo pidgins across the Arctic.

(45)

Koenekiak 'a small women's boat' ["Ein Frauen-Schifflein"] — Ol kone + WG qajaq. Periphrastic for WG umiaq.

(46)

Kralit 'a savage man' ["Ein wilder Mann"] — WG Kalaaleq 'Greenlander'. According to Paul Egede (1750: 68, cited in Petersen 1976: 194 and Thalbitzer 1904: 36), this is how Greenlanders referred to themselves when communicating with non-Inuit, claiming that the old Christians (Norse settlers) used this term. They call themselves inuit 'people'.

Eskimo pidgin in West Greenland

183

(47)

Kynoka 'commander' ["Ein Commandern"] — maybe Anderson's (1756: 202) loanword in West Greenlandic: kinneka 'child'; Dutch, Frisian, and Low German: kinneke{n)lkindeke '(little) child', or, maybe more likely, WG qinigaq 'elected/chosen one'.

(48)

Oesemik 'copper kettle' ["Ein kupferner Kesser] — Danish noun ose 'scoop' + Ins, or WG uussut 'something to cook in' + Ins < uussummik.

(49)

Promek 'bread' ["Brod"] (109:15) - Dutch brood, German Brot or Danish bred. O'Reilly: brumik 'bread' + Ins.

(50)

tante 'house, living' [" Wohnung"] - Dutch tente or Frisian tinte 'tent'.

Most of these words are non-Greenlandic (Meyer contains about 25 nonEskimo words or stems), and of some of them we can say that they must have belonged to an international nautical jargon, especially those we found confirmed by other pidgin or jargon lists (those of O'Reilly and Thevet), or those that are explicitly referred to as such (see Brown's remark in section 3.1.1 above). Some of them probably belonged to a colonial Danish-Greenlandic contact variety, and some words may even date back to Old Norse times. Paul Egede's remark about kalaaleq 'Greenlander' (see above, and the Appendix entry for Kralit) is particularly interesting: it is an explicit reference to something which points to a distinct foreigner-talk register that the Inuit developed. Historical data support such a view. A number of words from these times have been adopted into Greenlandic and naturalized (see Petersen 1976), like kalaaleq. There is one originally Dutch or Frisian word from the whaling period that has survived in modern West Greenlandic: malamuk 'fulmar/petrel (Fulmarus Glacialisy (Peter Bakker, Personal communication). This word ultimately comes from the Dutch compound mallemok lit. 'crazy gull', which has gone out of Dutch use this century. It may have entered the Greenlandic lexicon via the loanword in Danish: mallemuk 'fulmar', but it is not impossible that it was borrowed into Greenlandic before it entered into Danish. The word is attested in earlier sources than Cranz (1770) who, according to Salomonsen (1979), claims it is used by Greenlanders. It is found in almost every whaler's journal, and also in nearly all treatises

184

Hein van der Voort

from the eighteenth and nineteenth centuries on birdlife.in the Arctic, as being coined or used by Dutch whalers. It is also mentioned in O'Reilly (1818: 137) as "Procellaria Glacialis (fulmar petrel, or mallemuck)". One of the original West Greenlandic words for 'fulmar' is qaqulluk. One phenomenon that Mühlhäusler (1986: 1—3) associates with pidgin characteristics can also be observed in some of Meyer's entries: in cases where words have both coincidentially similar forms and meanings across languages in contact, they are highly likely to become part of the lexicon of an emerging pidgin through mutual reinforcement. According to Mühlhäusler, up to 50 percent of the early Tok Pisin lexicon can be traced back to more than one language, in this case Tolai and English. One of his examples of thie phenomenon is even from Stefänsson (1909: 227): Herschel Island Eskimo Pidgin miluk 'milk' could originate in Eskimo as well as in European. Another instance of this in Herschel Island trade jargon is pau-dlu 'gunpowder', and Stefänsson (1909) relates it to English powder. Herschel Island Eskimo Pidgin vocabulary indeed seems to have contained a large portion of English words. In a word list from nearby Point Barrow (Ray 1885), however, we find a similar form ραύ-rä 'powder' and elsewhere the word pau 'soot'. The Eskimo form pauq for 'soot, lampblack' is common in many of the geographically rather dispersed Eskimo dialects (see Jenness 1928: 97), and with the introduction of firearms its meaning seems to have been extended to embrace the sense of 'gunpowder'. English must have played a role here as well, but no lesser role was played by Eskimo. The semantic extension in Meyer's Baussamek 'powder', literally 'soot to be used' (Ins), which probably emerged independently from English, shows this. The influence of English on the eighteenthcentury West Greenlandic pidgin lexicon is minimal or absent. For certain cases in Meyer, like Kynoka, origins in either European or Eskimo may be equally unlikely. In some cases, however, both may be equally likely, as in Hcekemik above, or Bikewai and Oesemik (see Appendix II). The origins of some other forms like Askepcese seem less equally likely. Mühlhäusler also notes that pidgins sometimes develop compounds of dual origin. One instance of this in Meyer's word list has already been mentioned above (in section 3.3.2.4): Kcenekiak 'umiak, women's boat' from Scandinavian or pidgin kone 'wife, woman' and West Greenlandic qajaq 'kayak'. This is one of the processes responsible for innovative expansion of the lexicon, which is described by Hancock (this volume) and argued to be a universal feature of pidgin languages.

Eskimo pidgin in West Greenland

3.3.4 A note on

185

morphophonology

If the origin of Hoekemik 'stockfisch' is Dutch heek, the addition of an instrumental case inflection according to West Greenlandic morphophonological rules would result in hiimmik, as certain stem-final (prederivational or preinflectional) consonants become assimilated in West Greenlandic. As a loanword ending in such a consonant, however, heek must have got an epithetic vowel (see Fortescue 1984: 355) in the absolutive: heeki, hence the instrumental heekimik. Note that it is also possible that the form on which this entry is based ended in an -e\ older Dutch hecke. If the origin is West Greenlandic uuaq, a special West Greenlandic rule of morphophonology has determined the instrumental form: uukkamik. Therefore it can be considered possible that Hoekemik has a dual origin. The proposed origins of some words in Meyer's diary can only be maintained, however, if the possibility of foreigner-talk strategies in morphophonology is taken into account. The question here is, for example, how can a word like Blik take on the instrumental case suffix -mik and become Blik-te-mikl I think an explanation lies in the fact that the deletion, assimilation, etc., of sounds when morphemes are combined makes forms less transparent than processes like insertion, and therefore the latter may as a rule have become generalized in foreigner-talk. In a certain class of West Greenlandic words, namely those ending in -t, a recessive vowel (/i2f) may be inserted (or rather, may surface) in many cases; for example in angut 'man' + Ins.pl. -nik becomes angutinik. This rule may have been generalized to foreign forms not ending in -t(i2), to such an extent that even a It/ is inserted to make preinflectional vowel insertion possible, serving an analytical sort of morphophonological adaption. They may be the case in entries like Bliktemik from Dutch blik, Hageltimek from Dutch hagel and more, and maybe even in a word of West Greenlandic origin Engeltimek (see Appendix II). Remember that if blik were a West Greenlandic word, the instrumental would have been blimmik, and if it were only a loanword, canonically blikimik. I have not encountered any indications of foreigner-talk morphophonology outside of Meyer, however, and neither do I know of any systematic research on it yet. Some linguists, e. g., Samarin (1993), have done some research on the opposite: changes towards less transparency through, e. g., the condensation of pidgin morphophonology. 3.3.5

Phonology

As can be expected, those traits of Eskimo phonology that have no correspondences in the European (here Germanic) system, were in the main

186

Hein van der Voort

disregarded (in the variety described) by Meyer, for instance, quantity: Meyer uses no diacritics (like the early Danish sources of Eskimo "proper" do) which could have indicated that it sounded like accent to him. As the author in all likelihood used an alveolar trill for /r/ in his own native language, he reacted to the uvular /r/ of Greenlandic by transcribing it with a back stop: (51)

Oksok 'bacon' WG orsoq 'blubber' (whale or seal fat)

(52)

Beklemik 'net', 'tie' WG perlaamik 'twisted sinew thread' (Ins);

Consonant clusters with III may have been voiced or voiceless at the time (for a discussion on this matter see Bergsland - Rischel 1986: 31-33). The geminated voiceless [I:] is found in: (53)

altlemeck 'something else, other' WG allamik < alia '(an)other' + Ins: 'something else'; see also Osloeket in Appendix II, and Idelbmimik for voicedness

The uvular quality of the /q/ is not indicated either, and it is interpreted as a /k/, in examples (51), (52), and: (54)

Kyak 'ship' WG qajaq 'kayak'

although it appears from the following example that this /q/ did not pass wholly unnoticed by Meyer: (55)

Kralit 'run' WG qaagit! 'come here', second person subject imperative marker

It is not so strange that Meyer writes kra here because the combination /q/ + /a/ sounds exactly that way to foreign ears, and is still used in schoolbooks to illustrate this sound. He also noticed the different allophones of a phoneme like /p/, in Greenlandic and apparently in the pidgin, ranging from [b] to [p] in:

Eskimo pidgin in West Greenland

(56)

Badik 'paddle' WG paatik 'paddles'

(57)

Bingele / Pingele 'children' (see Appendix II)

(58)

Promek 'bread' Dutch brood

187

In part, foreigners' generalizations, like the interpretation of both [q] and [k] as the same phoneme /k/, may have been the basis of the phonological system in the pidgin or jargon. Other words and spellings point to an earlier stage in West Greenlandic. There used to be a distinction in West Greenlandic between /s/ and (palatalized) Is/. According to Bergsland and Rischel (1986: 31), this phoneme may originate from a more voiced sibilant that became voiceless during the seventeenth century. The disappearance of the distinction between /s/ and /s/ in West Greenlandic is only of recent date. In the following example, we find the /s/ in the derivational suffix -{r)suaq 'big one' represented by Meyer as sj: (59)

Akesjamik 'is big' WG akersuaq 'high price' + Ins > akersuarmik. If Akesjamik means 'expensive', then WG akisooq 'expensive', lit. 'one with a high price' + Ins > akisuumik may be a better solution. Then the sj would be less easy to explain, however.

Older West Greenlandic had many consonant clusters and diphthongs, and historical documents usually confirm Kleinschmidt's standardized nineteenth-century spelling in this. However, these clusters and diphthongs have been subject to assimilatory processes since the eighteenth century and today they have largely disappeared from West Greenlandic and became long geminate consonants and monophthongs. Not surprisingly, they can still be found in Meyer: (60)

Nieksik 'hook' WG nissik 'hook, boathook', in the old spelling nigsik.

(61)

Baussamek '(gun)powder' WG paassat 'gunpowder' + Ins > paassanik, old spelling paugsat\ plural from paoq 'sod, gunpowder'.

188

(62)

Hein van der Voort

Mauwe 'come here' WG maanga 'hereto'; Mauwe reflects the diphthong in the older form maunga 'hereto'. Egede has mäve 'here' (1727: 866) where the /u/ of the diphthong is reflected by the v.

Some words point to influence from the South Greenlandic dialect, which, like East Greenlandic, has /i/ in many places where West Greenlandic has /u/: 19 (63)

Kerniktik 'black' WG qernertoq 'black'. Note the i in the last syllable.

(64)

Agietemik 'an axe' WG agiut 'file'; cf. SG agiit + Ins > agiitimik

4. Discussion The main problem with research on Greenlandic Pidgin is the lack of conclusive data. For instance, we are not able to infer a complete and dependable syntactic picture as we lack among others firm indications about the (absence of) ordering of adjectives and nouns, and the place of negation. An additional problem with some of the data we do get lies in their sketchy roughness. Meyer was not a linguist, and the forms he noted down probably do not give a full picture, so we cannot always decide whether a certain form is Eskimo "proper" or Eskimo "pidgin", and in case of the latter, reconstruction is hardest. Furthermore, it is not perfectly certain that we fully understand what Meyer intended to convey with the spelling that he used. Remember for example the word Hcekemik; many facts speak for a European origin of this word, namely Dutch heek, but the orthographic symbol a nun biefe 3nfulaner· ju utt* an unfere fiftlbnun^er (Sommanbeurfam, fagtcii fie; Ohmakniblit? ba$ ift, 2Jtonn! wer fciji) tfjr? wief)eifietil)r? ber Sommanbeurfagte: Kinoka König Salomon, ba$ ift/ fcer (Som* tnanbeur beä 6φΐΑτii ber tfonig (Salomon genannt Söeil mm ber Sommanbeur Γφοη öftere mit liefern &t)iff l)ier gewefen, fa iemtetenifjn einige/ unb riefen: Kynoka Konig Salomon ajongelak ishigaget; ba£ ijl: fetjb ii)r ber (Sommanbeurtjon bem β φ ί ^ i>et Äonig 6älomott? es ijl gut, ίφ l)abe eucb (φοη ofterö gefeijett. 6 0 öalb ber (jomman* i>em:mjt3aantwortete, fagtentfe: Kynoka handlabafche Kakkamia Bofamia ajongelak befingele ; ba$ ift: ob er mit itynett i)anbeln wollte ? fte Ratten Γφ6ηε SBaaren iei) M / fφόne gelte-ton gucbfen unb @ee* frmbett, er follte fte nur befefjen, 2)arauf antwortete ber (Sommanbeur : App Handlabafche -beüngele Kakkamia Bofamia; btö ijl ? 2 Φ will mit euφ tobein, lagt ηήφ eure Söaaren fei)eru fuhren barauf weiter fbrt/ Kiak Kinoka ihipa, ba$ ift: S a f t bie ^ f c a l u p p e Pom © φ ί ^ herunter, bamitwir d>aluppeu l)crutttev j i r lajfeu; ba nun biv SÖilbm auf unferrn &btff waren / fprac&cn (te: BeGngeleKakamiaBo•famia ajongelak; baöift: 55cfei)et Mlfere gucb$>mib ^HouppenfcUe, fa fmb ffton. 17. S)iefe wilben beuten waren bc# iljrem £anbcl fo mtötrauifcb; baß jie iijr* SÖoaren b e f e b i g an einem ®nbe mit bert g a n t e n W i e n / nnb brn^auö niebt efter. öanjlict) fahren lieficn, alö fte baftir jatten^ watf 1\t wollten, au$ Surcbt, man mochte; fte iljiien ncljnwn SBiv aber tonnten un$ webt gauig bor iijnen in acbt!iel)inen/ bafljfe nicl;t etwas binweg nabinen. üjjre giuivi juni ftri)t«i erfuhr ϊφ felbft mit meinem eigenen (£cbabcn / inbemfie inirmeinSftefler famt beröabel biebifeber 2&ife eutwenbeten* SBir gaben nicbtäU/ bag fie mit eminent ©e* wcljr auf unfer βφί|τ'tarnen/ α μ φ lieffen wir feine größere 9(njaf)l auf einmal un$A als bie wir im gall* ber Sftotfj überwältigen ju tonnen glaubten; beim tl)t Umgang ijlaU jejeit πβΓάδΓΐίφ. hatten wir e$ bloä mit itftfnncrn $u tljun; id) werte beffer un* ten-aueb uufere ftanbelfcbaft mit iljreu S$ei* bern crjeblen, unb auf was für eine 9jrt wir biefe iOieiiije Scute wieber toon unferrn ;c Dberhanb Mommen, nnb/ ltd) auf ihr ^ufoer unb ©ewehr tKrlafictib, bie (Summer/ fo |ίφ Bep ihnen niebergelafieit, angreifen nnb auf· teibeit/ ja enblich gar allen Europäern beit £anbel mit ihnen ober ben SBaliftfcbfang Verwehren: aber mau hat jeberjeit herauf geantwortet: ba bie 23ilbett ttitfere SSaaren tt&thiger brauchten ate wir bie ii)rigeit/ ba jte unferer SBaarett burchauö ηΐφί entbehren ionnten, inbem ite folebe Weber felDer 511 »erfertigen ηοφ anberswoher ju eri)alien im (Stanbe fegen/ fo hatte man fein folebe* feinb* Unternehmen, bunt) welche^ (te |jcl> ja felOft ben großen 6cbaben jufwgeit wiir< fcett, t)on iftnen ju befürchten. Nebelt bem. Ober fo tonnten fte leicht aufieretanb gefef3t Serben, ben Europäern aufebaben, wenn fciefelben , fo halb fte ctwa^l berületcbcn jnerien würben, unterlceffen, iijnen Ruftet

199

200

Hein van der Voort

i n bie 6 t r r t j $ e 2D te ba§ t>ic © r o u l a n b e r m i t b e m (Scbicfeeweljr/ w e n n jte genugfamen 25or* rati) an Quitter R i t t e n , eben fb w o f ) l u m * iugcl)en wufjteii/ altf t i e (ginropaer; iljr g e n m a a d ift u o r t r e f i W . SBeim jte a b e r g u t fct^tefTen w o l l e n , le^eu lie (ίφ mit b e m 9tu* tfen a u f t>en 2 3 o b e n , fjeben ben ^ u ß in bte £>6t)e/ legen ben Sattftoornen b a r a u f / b e n jColben aber briitfeu jte fejt an ben S B a c f e n , faljreu eubliil)/ w e n n jie etwad g e w i ß t r e f * fen w o l l e n / mit bem allgemach in Me Jjboöe/ bruefen lotf/ unb feljle» feiten. . 2 ) Verlangen jie H a s e l t i m e k ober β φ ν ο ο ί . 3 ) Engeltimek, Keffer. 4) K o e g e l t i t u e k * © t e e f n a b e l n , ifjre jtleibev bamit j u f a m m e n juijeften. 0 M i k k e m i k , SWjeita&efa. bte SßeliEleiber bamit n a g e n m Wimen. 6) P r o m e k , ißrob. 7 ) Akkafchjomek, ® t Q & ftfebe. SSelcbe unter ifinen n u n m i t b e m iu> frieben w a r e n , w a s w i r ifjnenfttr e i n i g e i ö ^ rer SBaarcn geben w o l l t e n , j a g t e n : A p p . , j a / eö ijl g u t ; bie aber/ benen ber £ a u b e i nicijtgefiel, fagten: N ä h m e e u p o k A l t l e n i e k , taugt »id)t/ rt ijt iurt;t g u t / ten ifjre 2 3 a a r e t t .

uub U t y t U

§.. 18. wir. aber tiollifl mit beit3Jiau Ucbteit ttrtdjtfagen tonnen, hingegen κ φ Iangeten biefe wilbe (ScMnijeiten toon un$ i ) Sapannik, gldfertte 3£oraUeu fo grofj al$ Qrr&fen / Welche fie att einen gaben anfäffett/ um bie £aitbe binbett unb iljre größte ftren* be baratt fyaOen* 2) Tegaekfak, fleine Öpügei^ Umil)re worauf fie |ίφ ften fo biel / alä baö (Europdiftöe Stauen* jimmer, ein6ilben> bäriltnett &efcbaueit uttb i>«tvac!?tttt iu {fatten* yj Oneto, Heinere

201

202

Hein van der Voort

i n bte e t r a ß e S t e m s ,

m

glaferne Korallen m i allerlei Sar&ett/ u m jie a u eine Θ φ ι ι ι ι ν · (jangem unb a n ffire j f t e i b i m g / wie Deg un$ bie ® o l b > o b e r jJ'-ei/. bag jttwSBilbe mit iljrem-£ait> bei auf unferem berweilten/ $i£ wir, tow gewöhnlich,. 23ett(tunbe galten mugteni SD'ä nun biefe einfältigen Seute faijeu, wie ttnr.ünö ηαφ einer gewiffenDrbnung infam* meit jeftten r unfere £)jite unb jCappen bow tfppf traten,ftunben.jteboUer ^evwuube* rung ba unb Taljen einanber jtu&enb an; ba jie aber enblicb faljen,. wie wir alle ju einer geit.jurifere: SDUuler.auffperreteu uub mit wiüt(Stimm unfer ®efang anfangen/ üb err gel ffe eine ΓοΙφε Surest unb f r e i e n / bag iet eine in ber grogten Sil auf$ Serbecf lief am eeil hinunter in fein (Scbifflew tteg, ber anbere aber buitf) ben ßußen/ßa. iinauS in,ftineu$af)u fpwng; 23elc&£ luftige

f

203

204

Hein van der Voort

in bie 6 t r a ß e £(H>fe;

113

tufti0e Segefienfieit uns alle uitfcre 9(nfcact)t todljreiib tiefer Söettfluhö Oeilaljm. §. 20. Sftmi mill ίφ αηφ ifire ©etfaft be* ftörei&eii. £)ie 9J}aun£pa'ione» finb fteirfe kfdjte Scute/ l)aben fajl alle, wann tie au& Gemachten |Inb, einerlei; GMfie unb ftaare/ welche febwarj/ t>icif unb fo I)avt fmb, altf Wic^fcrbljaarc/ fäworje ^XuaOt'auiieu, U t U ite fch\vav3e Slujjeit, welche tönen etwirä tief im jfypf liege» , ein vtmbeä ©ejlcbt, feine Sfcafeiibeüter / wie wir Europäer / fonbern nur einen fntnipfen unb bieten 3tafenfpi$ / WeU&et fo tief im ©eftcbt liegt, bag er ncäjt Por bie Söacfeii Ijcvattf gel)t, feinen Söavt unb ra(V ten |1φ baljer αίιφ »iebt, groge breite S Ü t o lei·/ fcMue weiiie 3»4»ijw / aber feine ιινφ eine dui)erlic(;e ©ewalt eim öefcblrt^i'u werben. 3 0 r $111» ober Unter* Siefer ift ηίφί getpolbt, wiefcepunä, foit* beni nana platt.