III The Mouton World Atlas of Variation in English

Edited by Bernd Kortmann and Kerstin Lunkenheimer

Brought to you by | Australian National University Authenticated | 150.203.230.100 Download Date | 2/11/13 1:07 AM IV

ISBN 978-3-11-027988-7 e-ISBN 978-3-11-028012-8

Library of Congress Cataloging-in-Publication Data A CIP catalogue record for this book has been applied for at the Library of Congress.

Bibliograhic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the internet at http://dnb.dnb.de.

© 2012 Walter de Gruyter GmbH, Berlin/Boston Cover image: iStockphoto/Thinkstock Typesetting: Dörlemann Satz, Lemförde Cartography: Hans-Jörg Bibiko, Leipzig Network diagrams: Christoph Wolk, Freiburg Printing and binding: Beltz Bad Langensalza GmbH, Bad Langensalza ? Printed on acid free paper Printed in Germany www.degruyter.com

Brought to you by | Australian National University Authenticated | 150.203.230.100 Download Date | 2/11/13 1:07 AM Table of contents v

Table of contents

Abbreviations of the WAVE varieties fi ix List of maps fi xi List of maps sorted by features fi xv List of phenetic networks fi xxi

Bernd Kortmann and Kerstin Lunkenheimer Introduction fi 1

Part I: The British Isles

Gunnel Melchers Orkney and Shetland English fi 15 Jennifer Smith and varieties of Scots fi 21 Markku Filppula Irish English fi 30 Jennifer Kewley Draskau Manx English fi 48 Robert Penhallurick fi 58 Graeme Trousdale English dialects in the north of England fi 70 Susanne Wagner Southwest English dialects fi 78 Peter Trudgill East Anglia fi 88 Anna Rosen Channel Island English fi 98

Part II: North America

Susanne Wagner fi 109 Alexander Kautzsch Earlier African American Vernacular English fi 126 Salikoko S. Mufwene Gullah fi 141 Robert Bayley fi 156

Brought to you by | Australian National University Authenticated | 150.203.230.100 Download Date | 2/11/13 1:07 AM vi Table of contents

Part III: The and South America

Jeffrey Reaser and Benjamin Torbert English in fi 169 Stephanie Hackert fi 180 Stacy Denny and Korah Belgrave Barbadian Creole English (Bajan) fi 197 Andrea Sand fi 210 Peter L. Jamaican Creole fi 222 Angela Bartens San Andres-Providence Creole English fi 237 Geneviève Escure fi 255 Hubert Devonish and Dahlia Thompson (Creolese) fi 265 Bettina Migge The Eastern Maroon Creoles fi 279 Tonjes Veenstra Saramaccan fi 291 Donald Winford Sranan fi 302 Dagmar Deuber and Valerie Youssef fi 320 Paula Prescod fi 329

Part IV: Africa

Malcolm Awadajin Finney Krio fi 343 John Victor Singler Liberian Settler English fi 358 John Victor Singler Vernacular fi 369 Magnus Huber fi 382 Magnus Huber Ghanaian Pidgin English fi 394 Rotimi Taiwo fi 410 Nicholas Faraclas fi 417

Brought to you by | Australian National University Authenticated | 150.203.230.100 Download Date | 2/11/13 1:07 AM Table of contents vii

Augustin Simo Bobda Cameroon English fi 433 Anne Schröder Cameroon Pidgin fi 441 Josef Schmied Tanzanian English fi 454 Alfred Buregeya fi 466 Jude Ssempuuma fi 475 Susan Fitzmaurice White fi 483 Rajend Mesthrie Black fi 493 Rajend Mesthrie Indian South African English fi 501 Sean Bowerman White South African English fi 511

Part V: South and Southeast Asia

Devyani Sharma fi 523 Ahmar Mahboob fi 531 Michael Meyler fi 540 May L-Y Wong fi 548 Verena Schröter Colloquial Singaporean English () fi 562 Stefanie Pillai Colloquial fi 573

Part VI: Australasia and the Pacific

Pam Peters and Peter Collins Colloquial fi 585 Peter Mühlhäusler Aboriginal English and associated varieties: shared and unshared features fi 596 Peter Mühlhäusler Norfolk Island/Pitcairn English fi 620 Rachel Hendery and Sabine Ehrhart Palmerston Island English fi 628 Kent Sakoda and Jeff Siegel Hawai’i Creole fi 643

Brought to you by | Australian National University Authenticated | 150.203.230.100 Download Date | 2/11/13 1:07 AM 628 Rachel Hendery and Sabine Ehrhart

Rachel Hendery and Sabine Ehrhart Palmerston Island English*

1Introduction

Palmerston English is spoken by just over 50 people on a remote atoll of the Cook Islands in the South Pacific. In 1863, the Englishman William Ma(r)sters settled on Palmerston with several Polynesian women from the Northern Group of the Cook Islands. Marsters ruled the island in an autocratic style: officially only English was allowed to be spoken and it is recorded that the children from his various wives spoke English with a strong British accent. The perpetuation of the speech was more difficult for the women than for the men be- cause they did not have a direct female example from Britain to imitate. The three branches of the family are now in their 8th generation and their offspring live mainly in Rarotonga, Aitutaki, New Zealand and Australia (several thousand people, with a varying degree of competence in the , depending on the contact re- tained with Palmerston). Despite the small size of the Palmerston Island community, there is a great deal of linguistic variation, both among speakers and in the speech of individuals. Moreover, there is a perception of linguistic variation, as shown in the following conversation, where RH is the researcher:

AB Some people on the island say “fowl” and some say “chicken”. “Chickens”, my dad simply say. “Can you go feed the chickens now?” RH So who says “fowl”? AB Um, mainly the the bush people. We’re the beach fellas, and they’re [gesturing to the other side of the island] the bush people. There’s a few of other things they say too. […] Cos most of the people on the bush side, they speak proper English and the beach side, they just communicate by their language. So if I would say to them the “shubble”, they would get angry at me. They don’t say it, cos they’re Engli- kind of closer to English, like MM and, uh, PT and them there.

This perception is supported by an initial statistical analysis of data collected in 2009. This analysis was con- ducted on the presence or absence of five Palmerston Island English features (chosen for their ease of recog- nition and frequency of use): pro-drop (F42–44), bare nouns for semantic plurals (F57,F58), bare verb form for present tense third person singular, -s morpheme on non-third-person present tense verbs, and the use of was with they, we, or you. For each transcription of at least 100 consecutive words, a score was assigned for each of the five features: the number of times the feature was used divided by the number of times it could potentially have occurred. These scores showed a significant difference (p<0.05) for “bush people” vs “beach fellas” While the above discussion focusses on lexical and grammatical variation, variation is found at all lin- guistic levels. It is sometimes a matter of choice between local variants or variants, but sometimes a choice between multiple local variants, e.g. the use of the term islet or motu, both of which are in frequent use on Palmerston Island. Statistical tests show that some variation is governed by gender (e.g. pro- drop is much more heavily used by men than by women), and some by age (use of singular forms of the verb be with plural subjects, e.g. they is, we is, you is is more frequent with younger speakers than with older people). Social group membership also seems to play a role, and may in fact be ultimately responsible for the apparent gender-related variation. At least some of the variation found in the speech of a single speaker is

* This research was supported under a Grant for the Excel- number DP110103714) for Rachel Hendery. We would also like lence of Research by the Deutsche Forschungsgemeinschaft to acknowledge the assistance of Teura Mouraui-Pokoati for in 2007 for Sabine Ehrhart and under the Australian Research answering questions about Cook Island Ma¯ori. Council’s Discovery Projects funding scheme (project

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM Palmerston Island English 629 conditioned by register and accommodation. A large amount of the inter-speaker variation, however, is not so easy to account for, and may be personal-pattern variation: the phenomenon described for small communities of Gaelic speakers by Dorian (1994).

2Notable features

2.1 “Substrate features”

The case of Palmerston Island English is not a prototypical “substratum-superstratum” situation. There was no resident population on the island when the first settlers arrived. Within the settlement group, however, English was certainly dominant in terms of power, as it is said that Marsters forbade the use of Cook Island M¯aori, the language of his wives, and there seems to be no record of other language use among the group, des- pite the possibility that one or more Portuguese speakers were present. In this sense, then, we can speak of a Cook Island M¯aori “substratum”. It is difficult to distinguish, however, between features that might have arisen in the learner English of this first generation and those that may have reached the language later as the result of contact with the surrounding region. The extreme isolation of Palmerston Island for many years does mean that we would expect fewer features to be the result of such later contact. The most obvious influence from Cook Island M¯aori is lexical borrowing. M¯aori words in Palmerston Is- land English are quite prevalent in the semantic domain of flora and fauna, although there is frequently an alternative English word in occasional use. There seems to be more differentiation of species in the M¯aori vocabulary than in the English “equivalents”: for example, various fish that Palmerston Islanders call by distinct names originating from Cook Island M¯aori can be referred to collectively as snapper. There are also some words of M¯aori origin in non-flora or fauna domains, for example umu ‘ground oven’, motu ‘islet’, ’uapo ‘meeting for prayer and song’, kaikai ‘feast’, orometua ‘missionary’. These are typically concepts and items associated strongly with the Cook Island culture and region. Grammatically there are both morphological and syntactic influences from Cook Island M¯aori, as well as some features that are common to learner English varieties (which was, after all, the situation for all speakers in the first generation except Marsters himself). We will return to the discussion of L2 features in section 2.3. The pronoun system is one feature that has been strongly influenced by Cook Island M¯aori. Cook Island M¯aori pronouns differ from English in having both a dual paradigm (F37) and an inclusive/exclusive distinc- tion (F36). Palmerston Island English has adapted the English pronouns to include both these features. In Table 1 we compare the pronoun paradigm of Rarotongan (the variety of Cook Island M¯aori spoken on the main island, Rarotonga) to that of Palmerston Island English. person singular dual plural

1 incl au taua tatou 1 excl maua matou 2 koe korua kotou 3 aia raua ratou 1 incl I, me we two, yami, two of us, we, us we (all), us (lot), we, us 1 excl two of us, we, us us lot, we, us 2 you you two, you you lot, yous, you 3 he, she, it, him, her himshe, them two, they, them they, them, big lot of them

Table 1: Comparison of Rarotongan and Palmerston Island English pronoun paradigms

It can be seen that there are multiple alternatives in some of the cells. Two of us does not seem to be restricted to either inclusive or exclusive use, while yami and we two are only used in the inclusive sense. Yami seems to be infrequently used, but the older people say that it used to be more common. Himshe, used for a mixed-

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM 630 Rachel Hendery and Sabine Ehrhart gender pair, seems to be a new development: Hendery only observed it in use among the youngest generation (up to the age of 18); Ehrhart, visiting 15 years earlier, did not observe it at all. Yous is believed by the older people to be a recent innovation, and is possibly imported from , perhaps via Rarotonga. In the examples in (1) we show a few of the Palmerston Island pronoun innovations in use.

(1) a. Yami go there tomorrow. b. What cup? The two of us cup. c. Just do you two job.

Another morphological feature of Palmerston Island English that can be explained as influence from Cook Is- land M¯aori is the optionality of morphological plural marking on nouns. In Cook Island M¯aori, number is marked on a separate lexeme preceding the noun, and this is a matter of suppletion, not inflection, as shown in (2).

(2)a. te puaka ‘the pig’ b. te nga puaka ‘the two pigs’ c. te au puaka ‘the pigs’ (examples from Buse 1960: 132)

In Palmerston Island English, semantically plural nouns do not take obligatory inflection (F57,F58). The op- tionality of this is shown in example (3b). This latter example could, of course, be interpreted as self-correc- tion, in which case the lack of plural marking was “preferred”, at least in this context. It is possible that since this occurred in an elicitation task, the speaker was consciously or subconsciously maximally distinguishing his English from Standard English, and self-corrected towards the more traditional Palmerston variant.

(3) a. One of the closest boat was a Fijian boat. b. While he was counting his baskets – while he was counting his basket […]

When plural marking does occur on the noun, the /s/ allomorph is generalised. There is, in any case, no voic- ing distinction in the dialect, and irregular plurals are frequently double-marked (F48), as illustrated in (4).

(4) Well, I’m not too sure how many childrens he had.

Plural forms of the demonstrative can occur with or without formally plural nouns, and singular forms of the demonstrative occasionally occur with nouns that are inflected for plural (F71), as shown in (5).

(5) a. Did all those work. (= ‘I did all that work’.) b. That trees over there. c. It’s more easier to have time for family these days.

Moving on to syntax, Palmerston Island English has a flexibility of word order that allows the formation of clauses that resemble the equivalent M¯aori constructions. Topics can be postposed so that the sentence be- gins with a verb (or in the case of copula deletion, which is frequent, with the material that would have fol- lowed the verb in the standard English sentence). Examples of these (cf. F224) are shown in (6), with the M¯aori equivalent to (6a) given as a comparison in (7).

(6) a. Is hot the sun. b. Is yours, the cup on the table? c. Too small the table. d. After Mama Shirley you.

(7) ‘e verateRa nom hot def Sun ‘The sun is hot’

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM Palmerston Island English 631

Noun phrases can occasionally take the order possessed-possessor, not only with of as in Standard English, but also with the clitic ’s or as a compound with no explicit possessive marking, as in (8).

(8) a. blood pig b. shit chicken c. Two barrel diesel Korinako’s

For comparison with (8), some examples of Cook Island M¯aori possessive constructions are given in the examples in (9). It can be seen that, like in English, both orderings of possessor and possessed are poss- ible.

(9) a. tutae moa faeces chicken ‘chicken faeces’ b. To te Ariki Pare POSS DEF chief Hat ‘the chief’s hat’ (Buse 1963: 402) c. te tama-iti aPuna DEF boy POSS Puna ‘Puna’s boy.’ (Buse 1963: 398)

The order of questions is something we will mention here for completeness, but such ordering seems to be a feature common across many L2 varieties of English, which Palmerston Island English originally was, so that we cannot be certain it is due simply to M¯aori influence. In Palmerston Island English there seems to be a strong, almost absolute preference for no word order change when a declarative is converted to a question (F228). An interrogative replaces the questioned material, and is optionally fronted. Do-support is generally absent in wh-questions. Examples of wh-questions are shown in (10).

(10) a. Where you was? b. How that happen? c. Where you two going? d. What Shirley was doing there when you went? e. What for you want that? [=‘What do you want that for’?] f. What for is sore? [=‘Why are you sore’?] g. Whose this? [=‘Whose is this’?] h. Why you keep knocking on bloody rocks?

Occasionally verb inversion occurs, most commonly in the idiom What’s your time? or What’s the time like?, both of which mean ‘What’s the time?’ or in other contexts where the abbreviation what’s is pos- sible. Embedded questions usually resemble those of Standard English in their word order. This means that in Palmerston Island English embedded questions and non-embedded questions can be identical. Two examples of embedded questions are given in (11).

(11) a. Do I have to tell him where the line went out? b. Do you want me to tell you what them? ‘Do you want me to tell you what they are?’

In Hendery’s corpus there are a few cases of verb inversion in embedded questions (F227). These were not all addressed to the researcher, so it is not possible to dismiss them all as hypercorrection. Example (12) shows a single sentence which contains both an inverted and a non-inverted embedded question.

(12) You describe where is the dog and where you go to the fish.

There is also at least one example where the verb appears both before and after the subject of the embedded question. There are no hesitations in this sentence, so it does not obviously seem to be a matter of self-correc-

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM 632 Rachel Hendery and Sabine Ehrhart tion on the fly; instead it could be a case of what’s being treated as a single fused unit, which would be con- sistent with the use of what’s in non-embedded questions, as mentioned above.

(13) You know what’s coconut is? [=‘Do you know what a coconut is’?]

A Palmerston Island English yes/no question can take one of several forms. A standard-style do-support con- struction exists, and this is sometimes also used in questions about the future, where Standard English would use will. An example of this appears in (14e).

(14) a. Do we know this was happening? b. Do you have? c. Does they eat it? d. Did we ever report ours? e. Do you help me to make a hat for brother?

The use of will is also possible, however, as shown in (15).

(15) Will you sharpen this one?

Another way to form a yes/no question is simply through intonation with declarative clause word order (F229). This is illustrated in the examples in (16).

(16) a. You lot playing ping-pong? b. You know who my grandmother is? c. You like the fish?

Since pro-drop is possible in Palmerston Island English, such declarative-style questions can still begin with a verb, as in (17).

(17) a. Gonna ‘ave a cup of coffee? b. Was a cyclone, eh?

The Standard-English-style verb inversion yes/no questions with auxiliaries and modals are also possible, as shown in (18).

(18) a. Are you foolish? b. The yacht, is ’e gonna full up? [= ‘Is it [the yacht] going to be full’?] c. Can you say it whole start again? [= ‘Can you say the whole thing again from the start’?]

A striking feature of Palmerston Island English is the flexibility of words with regard to word class. Many nouns have been extended to verbs, as in (19a-c). The use of another instead of again in (19c) suggests that broom is not functioning entirely the same way as a “normal” verb, yet its position in the rest of the sentence, and the lack of other verbs suggests that it is definitely being treated as verbal in most respects. Some adjectives have also become verbs (19d-f). It also seems relatively common to incorporate nouns into a verbal complex as in (19h) and (19i), but it is not clear how productive this pattern is.

(19) a. Then you dough it. ‘Then you knead it’ b. Half them with water. ‘Fill them halfway with water’ c. You broom now: next month you broom again. Broom now and come the ship another broom. ‘If you sweep now, next month you’ll sweep again. Sweep now, and when the ship comes, sweep again.’ e. We off the generator at two. ‘We turn the generator off at two’. f. On the light, boy! ‘Turn the light on, boy!’

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM Palmerston Island English 633

g. She bright her eyes. ‘She opened her eyes wide’ h. He knife-chopping. ‘He is chopping with a knife’ i. We going to go on the reef and do some rodfish. ‘We are going to go to the reef and fish with a rod.’

Two final features of Palmerston Island English that could be attributed to the influence of Cook Island M¯aori are both a matter of optional omission of elements that are obligatory in standard English, namely the copula (F176–F178) and pronominal arguments. Examples of clauses with omitted copula are shown in (20).

(20)a. After Mama Shirley Ø you.[=‘After Mama Shirley is you’, referring to turns in a game] b. Too small Ø the table [= ‘the table is too small’] c. They Ø in the tree

Both subject and object pro-drop occur (F42, F43), as shown in (21a) and (21b) respectively. It is also poss- ible to have an empty position where Standard English would have the dummy argument there, as shown in (21c). The example in (21d) shows the use of pro-drop in context. The omission of arguments in chained clauses occurs sometimes in conversational Standard English, but generally only when the subject is the same for each clause. In (21d), on the other hand, we have two examples of pro-dropped subjects which are not identical in reference to the subject of the previous clause: He saw the girl coming pass. Distracted him. and Boy tooken his hat off, distracted him, with the implied subject of distracted being the girl in each case. In the former example the subject of distracted is the most recent noun phrase, but in the latter example this is not the case.

(21) a. It’s really fun when Ø hear them speaking. b. That’s why you calls Ø a teacher. ‘That’s why you call me a teacher’ c. Cos engine room was burnt and the controls – Ø is none controls at the bridge. d. And on his way he saw the girl coming pass. Ø Distracted him. Boy tooken his hat off, Ø distracted him and he hit a big rock, Ø drop down, and he was on the ground suffering.

2.2 Features from the survey that are particularly interesting

The information in this survey is based on data collected in several field trips by two different researchers, Ehrhart in the 1990s and in 2007 (this second one with more sociolinguistic observations) and Hendery in 2009. There is around 15 year’s gap between the two main collection periods. For this reason there are com- plicating factors when it comes to assessing the presence or prevalence of a feature in the dialect. When a fea- ture or its frequency differs substantially between the two researchers’ assessments, this can be due to several possible reasons: the researchers spoke with different people and groups of people, and Palmerston Island English has a great deal of variation; Palmerston Islanders may have accommodated to the English of each re- searcher differently, as one was a New Zealander and the other a German; the researchers may be interpreting their data differently; or the language may have changed in the past 15 years. Probably a combination of these factors work together to produce the differences in the researchers’ findings. In this section we will briefly outline the main differences, paying particular attention to possible indi- cations of language change. We will begin with those features that were found in Ehrhart’s data from the early 1990s, but not in Hendery’s from 2009, as there are fewer of these than vice versa. The use of as/to as com- parative markers appears in the speech of the elderly speakers only in the 1990s. Hendery did not observe this feature at all, which may have been due to less contact with the older speakers (Ehrhart’s host family included people from the oldest generation, while Hendery only spoke with them in a few 1-hour interviews). Never- theless, if this feature is only in the speech of the older people, it is not likely to be present in the dialect for much longer.

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM 634 Rachel Hendery and Sabine Ehrhart

The other possible example of loss is perhaps rather an example of instability on the way to a new system. Ehrhart found regular marking of non-third person present forms with -s, and regular lack of this morpheme on the third person singular forms, i.e. the pattern shown in Table 2.

singular dual plural

1 incl I goes yami goes we/us goes 1 excl the two of us goes us lot goes 2 you goes you two goes you lot goes 3 he/she/it go them two goes they/them goes

Table 2: Present tense verb paradigm in Palmerston Island English (“traditional”?)

Hendery’s data shows irregular use of the -s morpheme, with some speakers using it for all verb forms (F171), some for none (F170), and some varying seemingly at random. Even controlling for genre and speech partner, Hendery found some Palmerston Island English speakers who use -s on third person singular verb forms 100% of the time (i.e. as in the Standard English pattern), including people who have rarely left the island. Some other speakers in Hendery’s data use it 0%of the time (i.e. the “traditional Palmerston Island” pattern he go, etc, found by Ehrhart), and these include the two oldest speakers, but also one young woman in her twenties. The use of -s with non-third-person present tense verbs (we goes, etc) is comparatively infrequent in Hendery’s data: even the speakers who use this pattern the most only use it 29% of the time. This suggests that if the paradigm observed by Ehrhart is indeed the traditional one, and if it is changing, the change is tak- ing place first by loss of the -s on non-third-person verbs, then (for some speakers) by the adoption of the Standard English third-person singular -s. Some speakers have perhaps stabilised with their new pattern of no present tense verb inflection at all. The opposite sort of change – development of new features, or increase in their use – is perhaps more obvious in the data. There are at least six features from the present survey that Ehrhart does not have repre- sented in her data but Hendery finds very frequently. At least three of these can be attributed to the increasing influence of New Zealand English on the Palmerston Island speakers, due to increased ease of travel, visits to and from family, and two-way migration, as well as the presence of a speaker of New Zealand English as a teacher in the school over the past five years. These three features are the use of go-based future markers (F114; He’s gonna sack me; You going to play?); the use of an eh tag (F165; Very good fella, eh?); and the use of like as a focussing device (F234; Does that recorder record like shit talk?). The latter feature was only present in the speech of young people, but the former two were found frequently throughout the community. Then there are three features that seem to have arisen in the past 15 years without a possible source in New Zealand English. The first of these is the use of object pronoun forms in subject function (F31), as in (22).

(22) a. Us is talking. b. Me got a interview now. c. Him got it. d. Ten o’clock, them be back.

The oldest generation speakers in Hendery’s data are not observed to use this feature, but it is present in the speech of the second-oldest generation, and very frequent in the speech of teenagers and children. Another such feature that is frequent in Hendery’s data and absent in Ehrhart’s is the use of finish as a completive marker (F110). This is illustrated in (23).

(23) a. Scale the fish, cleaned it up finish. b. I went feed my pig finish, come sit here.

This feature seems to have a similar distribution to the previous one: i.e. it is not present in the speech of the oldest Palmerston Islanders. Both could conceivably have been brought to the island by speakers of other Polynesian varieties of English, who do occasionally spend time on Palmerston Island. During Hendery’s fieldwork, a dentist from Pukapuka, a missionary couple originally from the Cook Islands, but who learned

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM Palmerston Island English 635

English while living in the Marshall Islands, and a young man who grew up in Samoa were all living tempor- arily on Palmerston Island, for example. A final feature that is frequent in Hendery’s data and absent in the earlier records is that the degree modi- fier adverbs and adjectives have the same morphological form (F220). Not only do bare forms like real per- fect or real sad occur, but some speakers also overgeneralise the form really to adjective position instead or as well: the really language or my really father. A similar hypercorrection is frequently found with adverbs that do not have the -ly form in Standard English, so that we find examples such as (24). All of these could be in- dependent innovations on Palmerston Island, although they are features that are found in other English var- ieties as well.

(24) I’ll be myself morely with the childrens.

Finally we would like to mention three features which were present in both Ehrhart and Hendery’s data, but seem to have increased in use since the early 1990s. The first of these is the double marking of past tense by the addition of the -ed morpheme to verbs that already contain it, or to irregular verbs whose stem is already in the past tense form (F133). This seems to occur in both the preterite and in participle forms. Some examples of this double marking are shown in (25).

(25) a. A man passeded him with his goat. b. This island is blesseded by God. c. So I end off got stucked here. ‘So I ended up getting stuck here.’

Ehrhart observed this double marking only in the speech of the older Palmerston Islanders, and not fre- quently even there. Hendery found it at all age-levels (25b was uttered by an eight year old), and it seemed to be very common: for some speakers it is ubiquitous. It resembles a little the way the present participle is con- structed in Palmerston Island English: with an -ening form, so that instead of fishing or singing, we find fis- hening and singening. This too looks a little like double marking: as though the -en of the past participle has been added before the -ing of the present. However, the verbs it most frequently occurs with do not take a past participle in -en. So if this was the origin, it must have started on other verbs and then spread. Another feature that is rare in the Ehrhart corpus and common in Hendery’s later data is the lack of for- mal plural marking on semantically plural nouns (F57,F58; When the tourist come, I tell them … ). This was il- lustrated and discussed in section 2.1, so we will not go into it again here, except to state that it is extremely frequent in Hendery’s data, and seemingly independent of social factors. Of the 18 adult native speakers of Palmerston Island English in Hendery’s corpus, there are four who leave semantically plural nouns unin- flected more than 20% of the time. One of these is from the oldest generation, but the others are not; one has spent a long time in New Zealand, but the others have not; they include representatives of all three branches of the family; two women and two men. Nor is there any factor in common among the three speakers who do not ever use bare plurals in Hendery’s corpus. The final feature that seems to have increased since Ehrhart’s recordings in the 1990s is the omission of Standard English prepositions (F216). In particular, the verb look is never found with the preposition at or for or any other in Hendery’s 2009 data. Rather it is used as illustrated in (26).

(26) a. I went to look my uncle. b. Go and look some food. c. The dog is up on the bottom, jump out of the window, and the boy looking dog. d. The child and the dog is looking the frog

Interestingly, this idiom is also found in the speech of the non-Palmerston-Islanders who were present during Hendery’s stay: speakers of L2 varieties of Polynesian English, one of whom produced The dog is looking the bees when telling the same frog story as the two speakers in (26c) and (26d) (whom he did not hear). It is poss- ible therefore that it has spread to Palmerston Island from elsewhere in the Cook Islands or beyond. Alter- natively, these speakers from outside may have picked the idiom up during their stay on Palmerston Island. It is impossible to be sure of the direction of spread without more research on Cook Island L2 English.

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM 636 Rachel Hendery and Sabine Ehrhart

2.3 Overlap with regard to presence/absence and pervasiveness with other varieties (“angloversals”)

It is difficult to compare Palmerston Island English to other English varieties because it does not fall neatly into any of the common classifications used. It is no longer an L2 variety, as all speakers acquired it as their first language, and this has been the case now for many generations. It differs from the “core” or L1 group, on the other hand, exactly because it arose as an L2 variety – at least we must assume that the original female settlers on the island spoke something akin to modern Palmerston Island English and that this is the source of many of today’s features. The cross-pollination between William Marsters’ English (probably Leicestershire) dialect and his wives’ Cook Island M¯aori fed into the next generation, and therefore the language also has much in common with pidgins and creoles. “Angloversals” that have been proposed in the literature are usually suggested to apply to a specific “type” of English. For example, Kortmann and Szmrecsanyi (2009: 272–273) note seven “ornamentally com- plex” features that are commonly found in L1 or “core” English varieties, of which Palmerston Island English has at least three (depending on how frequently a feature has to occur for it to “count”: non-coordinated ob- ject pronoun forms in subject function (corresponds to WAVE feature F31); be as perfect auxiliary (F102), and a-prefixing on ing-forms (F135).The “traditional” Palmerston Island agreement system also looks as though it may have derived from the Northern Subject Rule (F183), which is a further one of these “ornamentally com- plex features”. After this, three “simplifying” features are listed, of which Palmerston Island English has one: a loosen- ing of the sequence of tenses rule (F113). Then they list 10 features that are often found in L2 varieties. Palmer- ston Island English has five or six of these:

• would in if-clauses (F120) • regularization of irregular verb paradigms (F128) • zero past tense forms of regular verbs (F132) • invariant don’t in the present tense (F158) • invariant present tense forms: no marking for the 3rd person singular (F170)1 • deletion of be (loss of copula) (F176–F178)

Just on the basis of these selected examples (Kortmann and Szmrecsanyi’s database of features comprises many more), we can see that Palmerston Island English does not pattern with any of the grouped English types: “traditional L1” have, on average, 2.4 of the “ornamentally complex” features, about a third of the “simplifying” features and only a quarter of the L2 simple features. So Palmerston Island English has more L2 features than most core varieties. But it does not pattern with the L2 languages either, as these tend to have very few “ornamentally complex” features. Pidgins and creoles have even fewer of the complex features, but about half of the L2-simple features, so in this latter respect Palmerston Island English looks more like a creole. Because of this difficulty of classifying Palmerston Island English, in the rest of this section we will in- clude in our discussion Angloversals that have been proposed for any of the three types of variety. We will not, however, discuss those which seem to exist in Palmerston Island English only marginally, or for which we do not have sufficient data to be certain of their existence. Rather we will focus on those features that are either clearly present and common in Palmerston Island English, or clearly non-existent. Lack of inversion in wh-questions (F228) along with the presence of inversion in embedded wh-questions (F227), as shown in (10) and (12), have been proposed as “angloversals” at various times (Platt, Weber and Ho 1984: 127–128 and Kortmann and Szmrecsanyi 2004: 1200). It is usually suggested that this in particular is a marker of L2 English varieties. In Palmerston Island English, as discussed above, it is probably not the case

1 Whether Palmerston Island English counts as having this paradigm, which is not really true of Palmerston English, ex- feature depends on whether the key element is the lack of 3rd cept for a few speakers. Most other speakers have at least the person marking on present tense verbs, which certainly is true option of inflecting non-3rd person verbs with final -s, so the for Palmerston English; or the invariance of the present tense paradigm as a whole is not invariant

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM Palmerston Island English 637 that inversion in embedded questions is the norm, but it does occur. Other angloversals that clearly exist in Palmerston Island English include reduction in inflection, including some zero past tense forms (F132), lack of plural marking on nouns, and some tendency to invariant marking of the present tense verb forms (F170, F171). This latter is something that may be undergoing change at the moment, as was described in more detail above. The lack of copula (F176–F178), as shown above in (20), has also frequently been proposed as an anglo- versal for “vernacular” or L2 varieties (Chambers 2004: 129, Crystal 2003: 153, Szmrecsanyi and Kortmann 2009). Chambers also has multiple negation (F154) as an angloversal and Kortmann mentions the use of don’t as an invariant negator (F158), both of which are also found in Palmerston Island English. Palmerston Island English has a few different methods of negation. The most frequent is probably never (F159), which does not seem to have the same semantics as in Standard English. The form ain’t (e.g. F156) is sometimes used as well. Examples of the various negation constructions are shown in (27) and (28).

(27) SM After this one [carving] I got another project RH What is it? SM Fixing Kori’s door. RH What’s wrong with Kori’s door? SM He haven’t got no door. (28)a. He ain’t bought the islet. b. They never give me a invitation.

Crystal’s (2003: 153–156) list of “potentially distinctive features” of new English varieties is longer than most. It further includes the following features, all of which are found to some extent in Palmerston Island English:

• Non-standard complementation • SV order in questions • Object deletion • Adverbial position more flexible • Topicalization • New aspect/tense constructions • New phrasal verbs • Preposed elements: • Non-standard article use • Pronoun deletion • Non-standard prepositions • Non-standard comparatives • Non-standard lexical morphology

There are further features in Crystal’s list that are not found in Palmerston Island English: non-standard rhe- torical questions, and sentence-final particles, to name just two examples. Szmrecsanyi and Kortmann (2009) propose some implicational “varioversals” that they have found to be exceptionless in their sample. Interestingly, Palmerston Island English is a counterexample to one of these. Their “two-way” universal, that if a language has a past tense or anterior marker been, then it has no as a pre- verbal negator and vice versa (2009: 41) does not hold. Palmerston Island English does have a past tense con- struction using been, as illustrated in (29), but it does not have the preverbal no negator (160).

(29) a. She been record you? b. They never been make it. c. Yep, I been eat one.

This varioversal is only supposed to apply to L2 varieties, so it is possible that Palmerston Island English’s ex- ception is due in part at least to its non-prototypical type. An English variety that it is worth comparing Palmerston Island English to is New Zealand M¯aori English. In both cases we are talking about an originally L2 English with aM¯a ori “substrate”, and the time depth since the first non-native-English-speaker generation is very similar. Jacob (1990) carried out a study of the English

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM 638 Rachel Hendery and Sabine Ehrhart of M¯aori and P¯akeh¯a (non-M¯a ori) women in New Zealand of the same age and social class. She found the fol- lowing features were distinctive of the M¯aori women’s English:

• the use of come, seen, run, been as finite past tense verbs • the use of -s forms of present tense verbs in non-3rd persons • lack of auxiliaries in constructions with going to (I/he going to etc) • double negatives • lack of plural marking on semantically plural nouns

All of these features are common in Palmerston Island English as well. Benton (1966) mentions several other features common in NZ M¯aori English: a construction went to or went and followed by a bare verb, used to mark past tense (He went and knock), a lack of morphological distinction between count and non-count nouns (the firewoods), and the use of the preposition on with vehicles (He went on the car). The latter two are features of Palmerston Island English, while the went construction is rare or absent. Some of the similarity between the two varieties might be due to recent contact: many Palmerston Island- ers have spent time with relatives in New Zealand, many of whom live in South Auckland, a region where M¯aori English is widely spoken. Other features, as suggested in 2.1, might be due to “substrate” influence, and Cook Island M¯aori and New Zealand M¯aori are similar enough that we would expect them to have similar in- fluences on English in each case.

3Problematic features

In this section we would like to mention several features in the survey that are problematic when we try to find them in Palmerston Island English. Palmerston Island English has a general phonetic rule of reduction of consonant clusters, such that [ts] would generally become [s]. This means that its and is are homophones for most speakers. It is therefore difficult to tell how common the deletion of it is in referential and non-referen- tial it is constructions. The examples given in (30a) and (30c) could well be understood as (b) and (d) with [ts] > [s] reduction instead.

(30)a. My grandfather and grandmother speaks it […] and Ø is still in me b. My grandfather and grandmother speaks it […] and i[t’]s still in me c. Ø Is a real yacht on a way. d. I[t’]s a real yacht on a way.

The use of non-referential it in a construction such as (30d) would be acceptable in Palmerston Island Eng- lish, even though in Standard English there would be required (F173). Examples of it instead of dummy there are given in some rare non-reduced examples in (31) as a comparison.

(31) a. It’s better chances for me being Prime Minister there. b. It’s ants! [upon opening a sandwich, to find it full of ants] c. It’s no dog on this side here. d. When you finds one [a coconut], you bring it home because we don’t really have one on the tree and it’s only a few trees.

Although it is impossible to tell whether the deletions in (31) are phonetic or lexical, we can, however, assume that lexical deletion would be acceptable here, since subject pro-drop is generally permissible in Palmerston Island English (F43, F44). On a speculative note, this it’s/is ambiguity may have provided a pathway for the development of pro-drop in Palmerston Island English in the first place. To support this theory, however, we would need to find evidence that there was a stage during which subject pro-drop was only permissible with the verb be. Another feature in the survey that is not strictly problematic, but needs clarification, is F221 “other ad- verbs have the same form as adjectives”. This is complicated for Palmerston Island English because the regu-

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM Palmerston Island English 639 larization of forms seems to be working in two directions at once, leading to a continuation of irregularity at the level of the whole system. Thus we find “bare” forms being used as adverbs (32a). But we also find -ly forms used as adjectives (32b, c). In (32d), we also see how adverbs that do not end in -ly in standard English are sometimes regularised to -ly forms in Palmerston Island English. This is a further complicating factor.

(32) a. They might just talk normal only b. And now his really father is me c. Have something to eat and do a bit of restfully time. d. But I think we been morely taught how to speak a bit more proper than how we usually talk.

In general, therefore, any given adverb in Palmerston Island English may or may not resemble “the” proto- typical Palmerston or standard English adjective form, but there is no longer a strict division between forms used for adjective functions and forms used for adverb functions.

4Conclusions

In summary, Palmerston Island English is a dialect of English that appears to have been heavily influenced by L2 English and Cook Island M¯aori structure and lexicon in the earliest generations. It does not behave exactly like a creole, but nor does it fit neatly in a category together with traditional L1 Englishes or the L2 Englishes such as or Indian English. Its unique history and current social situation have brought about large amounts of inter- and intra-speaker variation, some of which has been stable over decades. Recent influences from New Zealand English and various Pacific L2 Englishes may be contributing to the innovations and changes that can be observed in a comparison of Ehrhart’s material from the early 1990s with Hendery’s material from 20 years later. We expect to be able to identify further changes by comparison of both these sets of material with early archival sources such as letters from the island’s Resident Agents, coun- cil meeting records, and John Burland’s interview with Ned Masters in 1959. While this volume is focussed on the morphology and morphosyntax of English varieties, the phonetic and phonological features of Palmerston Island English, as well as the purely syntactic features, show a simi- lar combination of Cook Island M¯aori influence and Northern English dialect features, and similar degree of variation among speakers and within the speech of individuals. Further information about these elements of the dialect can be found in Ehrhart-Kneher (1996).

Appendix: Overview of WAVE features attested in Palmerston Island English

# feature PalmE example Rating I. Pronouns, pronoun exchange, nominal gender 2 he/him used for inanimate referents And then he grows bigger [referring to a coconut] B 3 alternative forms/phrases for referential Well last month we saw a seal here. And then the thing just went B (non-dummy) it out and just disappeared. 5 generalized 3rd person singular pronoun: That girl, it is living on Aitutaki now. C subject pronouns 6 generalized 3rd person singular pronoun: Can’t you hit that girl? Why don’t you hit it then? C object pronouns 14 no number distinction in reflexives (i.e. We have to set ourself. C plural forms ending in -self) 22 you as possessive pronoun and then you got to sit on you backside … A 23 2nd person pronoun forms other than you you do you two job A as possessive pronoun

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM 640 Rachel Hendery and Sabine Ehrhart

25 object pronoun forms as possessive dem lot B pronouns: 3rd person plural 26 object pronoun forms as possessive Check here on me birthday, cos I’ll go and visit Mama. C pronouns: 1st person singular 27 object pronoun forms as possessive A pronouns: 1st person plural 28 use of us + NP in subject function A 31 non-coordinated object pronoun forms in Us is talking. B subject function 34 forms or phrases for the 2nd person plural “What yous playing?” “You lot playing ping pong?” “You all’ll get A pronoun other than you one.” 36 distinct forms for inclusive/exclusive 1st That one for mummy, this one for yami. [yami = 1pl incl] A person non-singular 37 more number distinctions in personal We two is going … A pronouns than simply singular vs. plural 38 specialized plural markers for pronouns What time you lot going fishening? A 41 singular it for plural they in anaphoric use Well the trees broke down, some of it. B (with non-human referents) 42 object pronoun drop That’s why you calls Ø a teacher.(Referent = “me”) B 43 subject pronoun drop: referential It’s really fun when Ø hear them speaking. B pronouns 44 subject pronoun drop: dummy pronouns Cos engine room was burnt and the controls – Ø is none controls B at the bridge. 46 deletion of it in referential it My grandfather and grandmother speaks it and that’s how I pick B is-constructions it up and Ø is still in me. 47 deletion of it in non-referential it Is a real yacht on a way. B is-constructions II. Noun phrase 48 regularization of plural formation: And then the mens sung ABC A extension of -s to StE irregular plurals 52 associative plural marked by postposed Pastor Tere and them there C and them/them all/dem 55 different count/mass noun distinctions most of the crews (= crew members) B resulting in use of plural for StE singular 57 plural marking generally optional: for When the tourist come, I tell them … B nouns with human referents 58 plural marking generally optional: for When he was counting his basket … B nouns with non-human referents 60 use of definite article where StE has You’re the big liar. A indefinite article 61 use of indefinite article where StE has It’s a wrong way. C definite article 64 use of definite article where StE favours The crocodile we have here is the half a crocodile. B zero 65 use of indefinite article where StE favours „Alex, you want a cabbage? (referring to cut up and cooked B zero cabbage stew) 66 indefinite article one/wan We been go in a group and get to to one place with show items C and things. 67 demonstratives for definite articles I wouldn’t know how to call them grandparents [= I wouldn’t A know what to call [the/my] grandparents (because they were dead before I was born)] 68 them instead of demonstrative those Them two houses A 69 yon/yonder indicating remoteness Look yonder! A 70 proximal and distal demonstratives with That thing there. A ‘here’ and ‘there’ 71 no number distinction in demonstratives That trees over there. C 77 omission of genitive suffix; possession And then you got to sit on you backside and then you now to do C expressed through bare juxtaposition you paperwork. 78 double comparatives and superlatives It’s more easier to have time for family these days. C

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM Palmerston Island English 641

79 regularized comparison strategies: Some of them, they catch ten, and some get lesser than ten. B extension of synthetic marking 80 regularized comparison strategies: C extension of analytic marking 82 as/to as comparative markers C 87 attributive adjectival modifiers follow C head noun III. Verb phrase: tense and aspect 96 there with past participle in resultative There was someone already been there before. C contexts 98 after-perfect C 102 be as perfect auxiliary I been done that. C 110 finish-derived completive markers Scale the fish, cleaned it up finish. C 111 past tense/anterior marker been I been eat my lunch already. A 113 loosening of sequence of tenses rule The farmer said that that will be enough. A 117 present tense forms for neutral future I helped Thomas so that one day she help Tuakanamoe A reference 120 would in if-clauses If I would say to them ‘the shubble’, they would get angry at me. A IV. Verb phrase: modal verbs 123 present tense forms of modals used where he took the anchor and dump it in the sea so that he can claim the C StE has past tense forms land 125 new quasi-modals: core modal meanings A V. Verb phrase: verb morphology 128 levelling of past tense/past participle verb I drinked the water from the well when I was a gal A forms: regularization 129 levelling of past tense/past participle verb They never give me a invitation. A forms: unmarked forms 131 levelling of past tense/past participle verb Last night he done his own thing. A forms: past participle for past tense 132 zero past tense forms of regular verbs He hit a big rock, drop down and was on the ground suffering. A 133 double marking of past tense A man passeded him with his goat. C 135 a-prefixing on elements other than I’s a told him C ing-forms 146 use of verbal suffix -ing with forms other the boy has falling down B than present participle/gerund 147 was for conditional were A VII. Negation 154 multiple negation / negative concord He haven’t got no door. A 156 ain’t as the negated form of have he ain’t bought the islet C 158 invariant don’t for all persons in the it don’t work C present tense 159 never as preverbal past tense negator I never give the taro. A 161 not as a preverbal negator A 169 non-standard system underlying B responses to negative yes/no questions VIII. Agreement 172 existential / presentational there’s/there there’s three or four different kinds knocks B is/there was with plural subjects 173 variant forms of dummy subject there in Was a cyclone that day. B existential clauses, e.g. they, it or zero 174 deletion of auxiliary be: before You going to play with the childrens? C progressive 176 deletion of copula be: before NPs After Mama Shirley Ø you. [referring to turns in a game] B 177 deletion of copula be: before AdjPs too small Ø the table [= ‘the table is too small’] B 178 deletion of copula be: before locatives they Ø in the tree C 179 deletion of auxiliary have Twice we Ø been to Fiji B IX. Relativization 193 gapping/zero-relativization in subject I think you seen coconuts Ø is all mashed up on the inside. A position

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM 642 Rachel Hendery and Sabine Ehrhart

X. Complementation 201 for-based complementizers … for I was went on. [= because] C 205 existentials with forms of get they got the torou and the kupa [listing types of fish found on an C island] XI. Adverbial subordination 213 no subordination; chaining construction He hit a big rock, drop down. B linking two main verbs XII. Adverbs and prepositions 216 omission of StE prepositions I went to look my uncle Edward C 221 other adverbs have the same form as instead of saying ‘sure’ slow, we say it kind of fast C adjectives 222 too; too much; very much ‘very’ as I’m not too sure C qualifier XIII. Discourse organization and word order 223 other options for clefting than StE The opposite of sack, that’s what he’s going to do. C 224 other possibilities for fronting than StE Too small the table B 225 sentence-initial focus marker for, e.g. for me, I use ‘um’ C 227 inverted word order in indirect questions Can you remember how old was you? B 228 no inversion/no auxiliaries in Where you was? A wh-questions 229 no inversion/no auxiliaries in main clause you find any? A yes/no questions 232 either order of objects in double object Can you pass it me? Give me that! C constructions 233 presence of subject in imperatives You tell your own story. B 234 like as a focussing device Does that recorder record like shit talk? B

References

Benton, Richard A. 1966. Research into the (eds), Atlas of Languages of Intercultural Communication Difficulties of Maori Schoolchildren: 1963–64. Wellington: in the Pacific, Asia and the Americas, 523–536. Berlin: Maori Education Foundation. Mouton de Gruyter. Buse, Jasper E. 1960. Rarotongan personal pronouns: Jacob, Jenny. 1990. A grammatical comparison of the spoken Form and distribution. Bulletin of the School of Oriental English of Maori and Pakeha women in Levin. M.A. thesis. and African Studies, University of London 23 (1): Wellington, Victoria University. 123–127. Kortmann, Bernd, and Benedikt Szmrecsanyi. 2004. Global Buse, Jasper E. 1963. The structure of Rarotongan nominal, synopsis: morphological and syntactic variation in Eng- negative and conjunctival pieces. Bulletin of the School of lish. In: Bernd Kortmann, and Edgar W. Schneider (eds.) A Oriental and African Studies, University of London 26: Handbook of Varieties of English: A Multimedia Reference 393–419. Tool. Vol. 4: Morphology and Syntax, 1142–1202. Chambers, J. K. 2004. Dynamic typology and vernacular uni- Kortmann, Bernd, and Benedikt Szmrecsanyi. 2009. World versals. In: Bernd Kortmann (ed.), Dialectology Meets Englishes between simplification and complexification. Typology: Dialect Grammar from a Cross-linguistic Per- In: Lucia Siebers, and Thomas Hoffmann (eds), World Eng- spective, 127–146. Berlin: Walter de Gruyter. lishes – Problems, Properties and Prospects: Selected Crystal, David. 2003. English as a Global Language. Cam- Papers from the 13th IAWE Conference, 265–285. Amster- bridge: Cambridge University Press. dam: John Benjamins. Dorian, Nancy C. 1994. Varieties of variation in a very small Platt, John, Heidi Weber, and Ho Mian Lian. 1984. The New Eng- place: Social homogeneity, prestige norms, and linguistic lishes. London/Melbourne: Routledge and Kegan Paul. variation. Language 70 (4): 631–696. Szmrecsanyi, Benedikt, and Bernd Kortmann. 2009. The mor- Ehrhart-Kneher, Sabine. 1996. Palmerston English. In: phosyntax of varieties of English worldwide: a quanti- Stephen A. Wurm, Peter Mühlhäusler, and Darrell T. Tryon tative perspective. Lingua 119: 1643–1663.

Brought to you by | Australian National University Authenticated | [email protected] Download Date | 1/23/13 1:14 AM