Professional Documents
Culture Documents
[Studies in English language] Schneider, Edgar Werner_ Schreier, Daniel_ Trudgill, Peter_ Williams, Jeffrey Payne - Further studies in the lesser-known varieties of English (2015, Cambridge University Press).pdf
[Studies in English language] Schneider, Edgar Werner_ Schreier, Daniel_ Trudgill, Peter_ Williams, Jeffrey Payne - Further studies in the lesser-known varieties of English (2015, Cambridge University Press).pdf
VAR I E T I E S O F E N G L I S H
General editor
Merja Kytö (Uppsala University)
Editorial Board
Bas Aarts (University College London)
John Algeo (University of Georgia)
Susan Fitzmaurice (University of Sheffield)
Christian Mair (University of Freiburg)
Charles F. Meyer (University of Massachusetts)
The aim of this series is to provide a framework for original studies of English,
both present-day and past. All books are based securely on empirical research,
and represent theoretical and descriptive contributions to our knowledge of
national and international varieties of English, both written and spoken. The
series covers a broad range of topics and approaches, including syntax,
phonology, grammar, vocabulary, discourse, pragmatics, and sociolinguistics, and
is aimed at an international readership.
edited by
JEFFREY P. WILLIAMS
Texas Tech University
EDGAR W. SCHNEIDER
University of Regensburg
PE TER TRUDGILL
University of Agder
DANIEL SCHREIER
University of Zurich
University Printing House, Cambridge cb2 8bs, United Kingdom
www.cambridge.org
Information on this title: www.cambridge.org/9781107021204
C Cambridge University Press 2015
1 Introduction 1
Jeffrey P. Williams, Edgar W. Schneider, Peter Trudgill and
Daniel Schreier
part i europe
2 Maltese English 11
Manfred Krug
3 Gibraltar English 51
David Levey
4 Irish Traveller English 70
Maria Rieder
vii
viii Contents
9 The English of Gustavia, St. Barthélemy 198
Ken Decker
10 Anglo-Paraguayan English 219
Danae M. Perez-Inofuentes
11 Gullah West: Texas Afro-Seminole Creole 236
Ian Hancock
Index 344
Maps
ix
Tables
x
Contributors
Introduction
Jeffrey P. Williams, Edgar W. Schneider, Peter Trudgill
and Daniel Schreier
The last point deserves more discussion here than we were able to provide
in the first volume. The notion of endangered varieties of a seemingly
voraciously dominating language such as ‘English’ may seem insincere to
some. Wolfram (2008) has written passionately about the Ocracoke Brogue
as being an endangered language that challenges the established canon of
linguistic endangerment:
Europe
In this volume we extend our focus from the British Isles, as in the first
volume, to the broader geopolitical landscape of insular Europe. The three
varieties that are described by Krug, Rieder and Levey all developed and
are primarily spoken in island communities.
4 williams, schneider, trudgill and schreier
Malta, in spite of its proximity to Italy, is part of the British Common-
wealth and a member of the EU. Manfred Krug explains its multilin-
gual history, under the rule of Greeks, Romans, Phoenicians, Byzantines,
Arabs and Normans, which is reflected in the linguistic stratigraphy1 of the
English spoken there. In the early nineteenth century, Malta became part
of the British Empire and has remained an anglophone outpost ever since.
Maltese English, as opposed to most other lesser-known varieties, is not
endangered.
In his chapter, David Levey describes another insular European outpost
of English on Gibraltar: an overseas British territory located off the south-
ernmost tip of Spain. In spite of English’s status as the only official language
in Gibraltar, Yanito – the local variety that contains elements of Andulusian
Spanish and British English – is generally preferred in vernacular contexts.
However, as Levey points out through his description, the use of English
in such contexts is on the rise with younger speakers in particular.
Maria Rieder provides a foundational description of Irish Traveller
English. This formerly undocumented variety differs from other kinds of
English through its incorporation of Shelta (also known as ‘Gammon’ or
‘Cant’) into Irish English, which evidences archaisms and dialect mixture.
As Rieder points out, the combination of Shelta lexicon with Archaic Irish
English grammar creates an unintelligible code that promotes in-group
cohesion and solidarity.
1 The use of the term stratigraphy is intentional in this description. The intent is to draw a parallel
with the concept of stratigraphy in geosciences that dates back to the mid seventeenth century.
Biostratigraphy is likely to be the best analogue for linguistic stratigraphy since it accounts for the
formation and extinction of species. In this case, the layers of linguistic influence in the language
can provide clues to periods of influence and contact, and cultural realms of contact, as well as other
aspects of sociolinguistic history.
Introduction 5
Coggshall addresses the challenge of providing a unified description of a
set of disparate varieties that share some features that set them off from
other anglophone varieties.
The English language has a lengthy and complex history in its West
Indian setting, with its incipient transportation to the region in the early
seventeenth century. Williams (2012) estimates over sixty varieties of
English spoken throughout the region, but the vast majority of those are
lesser known. This volume enhances our knowledge of the lesser-known
West Indian varieties of English with chapters on Bequia in the Vincentian
Grenadines, Saba (one of the former Windward Netherlands Antilles),
St Barthélemy and St Eustatius (also one of the former Windward
Netherlands Antilles).
In spite of its relatively small size, the island of Bequia exhibits a great
deal of linguistic variation within its English-origin varieties. James A.
Walker and Miriam Meyerhoff discuss that variation within the context
of providing an overview of Bequia English. The variety of linguistic
inputs to the overall sociolinguistic landscape of the island – including
whalers from the northeastern US region, former indentured servants from
eastern Barbados, creole English speakers from other islands in the eastern
Caribbean – is not atypical for the insular speech communities of the
West Indies.
Saba, an island of less than thirteen square kilometers in area, also exhibits
significant village-level variation, like Bequia. Jeffrey P. Williams and
Caroline Myrick’s description of Saban English focuses on the varieties
spoken primarily in the villages of Windwardside and Hell’s Gate. Saban
English predates Bequia English by almost a century, with Saba being
colonized by anglophones in the mid seventeenth century. Internal isolation
has characterized Saban social interaction over the centuries of European
settlement and colonization, resulting in distinctive village dialects.
St Eustatius, or Statia, was one of the most important entrepôts for
African slaves during the middle to later seventeenth century. Its cos-
mopolitan character, based on a diversity of merchants, sets the island
apart from most others in the West Indies during the same period. Michael
Aceto’s contribution on St Eustatius is important because of the island’s
prominent place in the region’s history, as well as its distinctiveness vis-
à-vis other regional varieties. Aceto explains this divergence in terms of
socio-economic focus: St Eustatius was a commercial centre and not an
agricultural centre.2
References
Schneider, Edgar W. 2007. Postcolonial English: Varieties around the World.
Cambridge University Press.
Schreier, Daniel, Peter Trudgill, Edgar W. Schneider and Jeffrey P. Williams, eds.
2010. The Lesser-Known Varieties of English: An Introduction. Cambridge
University Press.
Trudgill, Peter. 2002. The history of the lesser-known varieties of English. In
Alan Watts and Peter Trudgill, eds., Alternative Histories of English. London:
Routledge, 27–44.
Williams, Jeffrey P. 2012. English varieties in the Caribbean. In Raymond Hickey,
ed., Areal Features of the Anglophone World. Berlin: Mouton de Gruyter,
133–60.
Wolfram, Walt. 2008. When islands lose dialects: the case of the Ocracoke Brogue.
Shima: The International Journal of Research into Island Cultures 2(1): 1–13.
part i
Europe
GOZO
COMINO
MALTA VALLETTA
5 km
Sicily Greece
Italy
Malta
Tunisia
Libya
Maltese English
Manfred Krug
1 Introduction
Due to migration, at least 100,000 speakers of English with a Maltese back-
ground (including approximately 50,000 speakers of Maltese, a language
historically derived from Arabic; see Fabri 2010; Stolz 2011) live outside
the Maltese islands, notably in Australia, Canada, the UK and the USA.
The designation Maltese English (or, for short, MaltE) in the linguistic
literature, however, applies only to the varieties spoken in the Republic of
Malta, i.e. on the archipelago located in the Mediterranean about 100 km
south of Sicily and about 300 km north of Libya and east of Tunisia. Its two
biggest inhabited islands are Malta (population of c. 380,000) and Gozo
(population of c. 30,000); the third inhabited Maltese island, Comino, has
a permanent population of less than ten (Census of Population and Housing
2005; preliminary report of Census of Population and Housing 2011). (See
Maps 2.1 and 2.2.)
It should be noted at the outset that, depending on the nature and
intensity of language contact, the English spoken by individual people
in Malta may be indistinguishable from a variety spoken and written in,
say, England or Australia. For obvious reasons, these are not the varieties
I will describe in the present chapter. Instead, I will focus on acrolectal
Maltese English,1 which is, of course, a cover term conflating a number
of language-internal and -external factors. The term is used here to refer
to an idealized language variety spoken by university-educated speakers,
who speak at least some English at home and at work, who have spent no
more than short periods of their lives in English-speaking countries outside
Malta and whose parents were both born in Malta.
1 In line with some earlier publications (e.g. Platt and Weber 1980: 46–7; Kachru and McArthur 1992:
506) and an increasing number of recent publications on World Englishes (e.g. Baskaran 2008: 611;
Bautista and Gonzalez 2006: 133, 137; Bolton 2006: 293; Mahboob 2008: 252; Mesthrie 2008b: 25–6;
2008c: 307; Tayao 2008: 294; Biewer 2012; Hickey 2012a: 392), I will use the term acrolect (as well as
mesolect and basilect) without presupposing a creole history or a postcreole continuum of the variety.
11
12 manfred krug
Acrolectal MaltE thus correlates with higher socioeconomic strata,
although this is more true of older than of younger speakers because Malta
has in recent times encouraged tertiary-level education by grants and has a
high proportion of young people from all social strata attending university.
When I use MaltE in this chapter without qualifications such as acrolectal
or basilectal, I will be referring to the English spoken by educated Maltese
speakers, who have typically received or are still receiving a tertiary-level
education, but with no assumptions regarding the languages spoken at
home or at the workplace. It is also this larger group of speakers whose
language forms the bulk of the material that is compiled for components
of the International Corpus of English (ICE). In addition to the pertinent
literature, this chapter is based on data from the Maltese component of ICE
under compilation at the University of Bamberg (hereafter ICE-Malta; see
Hilbert and Krug 2010 for details) and data from a questionnaire for lexical
and morphosyntactic variation in English (see Krug, Hilbert and Fabri, in
press, for detail).
As is often the case in places with a colonial history involving British
rule, the varieties of English that are spoken in Malta represent in actual
fact a continuum between an acrolectal variety (a near-RP pronunciation
with a grammar and lexicon that is very similar to standard BrE) on the
one hand, and basilectal varieties on the other. The latter are characterized
by typical EFL learner features and more structural parallels with Maltese,
i.e. contact features, plus extensive code-switching (cf. the continuum
described in Vella 1994, Bonnici 2010 and such notions as mixed Maltese
English).
As will be seen, MaltE exhibits not only interspeaker but also
intraspeaker, including stylistic, variation. I will concentrate here on the
more formal English spoken and written at the workplace rather than that
used at home or in private, informal emails (or other digitally transmit-
ted informal messages), but point to important stylistic differences where
necessary. In essence, acrolectal MaltE is thus taken to be a standardiz-
ing edulectal variety that is oriented towards the traditional exonormative
British standard, but – to trained linguists at least – noticeably different
from it.
The next section will offer more figures and explanations for the claims
laid out above.
2 See below on overt style markers of MaltE. It should be kept in mind, however, that the identity-
carrying function of language is probably characteristic of every speech community and indeed of
every form of human verbal interaction and not restricted to lesser-known varieties of English.
14 manfred krug
3 Italian and Sicilian together account for about 50 per cent of the Maltese lexicon; estimates for the
proportion of English loanwords range from about 5 to 20 per cent (see Brincat 2005).
16 manfred krug
Table 2.1 Question: How well do you speak . . . ? (Census of Population
and Housing 2005, data for population aged 10 years and over)
Other
Maltese English Italian French German Arabic language
Not only bilingualism, then, but also tri- and multilingualism are
widespread in Malta, an aspect that is also well documented in the literature
(e.g. Camilleri 1991; Sciriha 2001; Sciriha and Vassallo 2006). As Table 2.1
shows, 57 per cent of the Maltese population claim (at least some) compe-
tence in Italian, 21 per cent in French, 6 per cent in German and 4 per cent
in Arabic. Both in terms of speaker numbers and in terms of average com-
petence, these languages trail well behind Maltese and English. This mul-
tilingual situation is partly due to historical language contact (see above),
in particular as far as English and Italian are concerned. Other important
factors are trade and tourism (which includes a significant English language
teaching branch) in a country whose other official language has a small
number of native speakers by international standards. Hence, Maltese
has no privileged status in teaching syllabi outside Malta and the fact
that English is sufficient for communicative purposes on the archipelago
is a counterincentive for potential learners of Maltese as a foreign
language.
Obviously, Maltese language and education policies play a role as well:
Schooling in Malta is mandatory until the age of 16. There is no official
policy on the classroom use of languages, but the National Minimum
Curriculum from 1999 issued by the Ministry of Education emphasizes
the importance of English and Maltese as official languages and states
further that pupils in secondary schools are expected to learn a third or
fourth language. Both Maltese and English are used from school entry, but
Maltese is naturally more prominent in primary schools, while English is
more prominent in secondary schools and dominant at tertiary level (in
class). There is a tendency for state schools to use less English than Catholic
(church) and private (‘independent’) schools (see Fabri 2010 for detail). In
particular, English is the dominant language of reading and writing. The
language of instruction depends to a great extent on the language of the
Maltese English 17
textbooks, most of which are in (British) English. For the same reasons,
the subjects Maltese and History are largely taught in Maltese. In spoken
interaction, code-switching is widespread among both pupils and teachers,
although a change in teacher education has shifted the balance somewhat
towards Maltese: until the 1970s, teachers were trained by British religious
orders, but more recently teachers have been trained by bilingual Maltese
native-speaker scholars at the University of Malta.
4 Phonology
Compared to other linguistic levels, it is the MaltE phonology that is
probably most independent of exonormative standards. In this section I
use standard Southern British English pronunciation (commonly known
as Received Pronunciation or RP) as a reference point, which historically
was – and for many Maltese speakers still is – the exonormative standard.
This is in line with previous studies on MaltE phonetics and phonology,
from which much of what follows is adopted, though often adapted (cf. in
particular Vella 1994: 57–86; Calleja 1987; Mazzon 1992: 126–9; Camilleri
1991: 108–9; Beer 2011; Bonnici 2010; Borg and Azzopardi-Alexander 1997:
299–338).
4.1 Consonants
4.1.1 Clear and dark /l/
RP has complementary allophonic variation for the phoneme /l/. Word-
initially before a vowel (as in lead or like), RP has ‘clear’ (or ‘light’) /l/. In
prepausal and preconsonantal positions (as in bell or world), we find the
velarized allophone [ɫ], known as ‘dark’ /l/ in RP. Intervocalic /l/ (as in belly)
and /l/ before yod (a possible realization in lure) are somewhere in between,
but in RP (unlike General American, for short: GenAm) tending towards
clear /l/ (see Johnson and Britain 2007; Wrench and Scobbie 2003). The
vast majority of MaltE speakers have only clear /l/ (e.g. Vella 1994: 77–8)
and this applies also to the acrolectal end of the continuum. Vocalized
allophones of dark /l/, which are common in BrE in coda-position
(e.g. bottle, hill ), are rarely heard in MaltE.
(v) In ICE Malta, the high-frequency item going features less often the
cluster /ŋɡ/ in prevocalic contexts than running, for example.
(vi) An exception to (iv) above is the going to future, in which /ˈɡɔʊɪŋtʊ/
varies with /ˈɡɔʊɪntʊ/ and even /ˈɡɔnə/ (see below for the qual-
ity and variation of vowels). MaltE thus shows reductive frequency
effects, which are typical of grammaticalization generally and there-
fore common in varieties of English worldwide (see Krug 2000: ch. 5;
2011).
4.1.5 Rhoticity
MaltE is generally considered a rhotic variety (Mazzon 1992: 127; Vella
1994: 76, Beer 2011) although non-prevocalic /r/ is commonly not sounded,
especially at the acrolectal end of the continuum. Bonnici (2010: ch. 6), for
instance, finds that postvocalic /r/ in the speech of L1 English-dominant and
Maltese-English bilinguals favours the null realization in the vast majority
of cases (around 80 per cent of the time). Exceptions in her data are the
contractions you’re, they’re, we’re, which are more often than not /r/-ful,
probably due to functional reasons as the loss or vocalization of /r/ in
these would lead to the loss of an entire phoneme and potentially to
20 manfred krug
homophones (cf. your, their, there, which in Bonnici’s data are usually
non-rhotic).
Beer (2011) finds MaltE to be essentially rhotic if a different spectrum of
the society is analysed. With around 80 per cent of realized non-prevocalic
/r/, he obtains in fact almost exactly the inverse result of Bonnici (2010).
Nevertheless, the two studies are compatible (see Bonnici 2010: 205).
The main reason is that Beer’s informants are overwhelmingly Maltese-
dominant L1 speakers. Furthermore, in his subsample analysed for self-
reported home language, those speakers who report using mostly English
and only some Maltese at home show dramatically lower rhoticity rates
(of about 50 per cent) and a virtually categorical use of approximants, and
thus near-absence of the more consonantal taps and trills.
As regards allophonic variation, MaltE /r/ has four important allophones,
three of which are found in Maltese (on which see Borg and Azzopardi-
Alexander 1997; Stolz 2011): poly-vibrant trill [r], alveolar tap [ɾ] and a
retroflex approximant similar to AmE [ɻ]. The fourth MaltE allophone
is the postalveolar frictionless approximant [ɹ] known from RP. All /r/
allophones occur essentially in free variation, but the following tendencies
hold according to Beer (2011: ch. 4):
4.2 Vowels
The vowel inventory of Maltese provides the repository from which the
MaltE vowels are recruited for most speakers, except for those that, due to
extensive training or language contact with an inner-circle variety, approxi-
mate an exonormative standard (usually RP). Maltese has a large inventory
of diphthongs as well as long and short monophthongs covering almost the
entire vowel space; MaltE vocalic realizations are therefore as a rule fairly
close to their RP counterparts. A notable exception is schwa, which thus is
the first vowel in the detailed discussion that follows.
Triphthongs of RP
The RP triphthongs commonly found in tower, tire are typically [ɐʊɛr]
and [ɐɪɛr], respectively, in MaltE. Non-rhotic accents typically have [ɐʊɐ]
and [ɐɪɐ]. Diphthongization to [ɐːə] and monophthongization to [ɑː]
(or similar vowel qualities), known as ‘smoothing’ from BrE (Wells 1982:
238–9), are very rare in MaltE.
face
The face vowel /eɪ/ is [ɛi] in MaltE and thus similar to RP. As in many
other varieties of English, including RP (see Wells 1982: 240), monophthon-
gal realizations [ɛ(ː)] and [e(ː)] are also commonly heard for this phoneme
(see Stolz 2011: 243 on the variation between open and close allophones of
what is usually labelled the Maltese phoneme /ɛ/ in the literature).
goat
The RP pronunciation [əʊ] for goat is similar to the typical MaltE
pronunciation [ɔʊ]. As in many other varieties of English, including RP
(see Wells 1982: 240), monophthongs [o(ː)] and [ɔ(ː)] are heard for this
phoneme too (see Stolz 2011: 243 on the variation between open and close
realizations of the Maltese phoneme /ɔ/).
24 manfred krug
bath, palm
The Maltese phoneme system has only a central /a/, both long and short,
with the quality typically being [ɐ]. All varieties of MaltE use both long [ɐː]
and short [ɐ] for the bath and palm vowels more frequently than the more
back RP quality [ɑː]. The <l> in palm or calm is not consistently silent
across varieties of MaltE, with mesolectal and basilectal varieties favouring
/l/-ful pronunciations. If /l/ is pronounced in MaltE words like palm and
calm, then the vowel tends to be short.
happy, kit
Like in RP, the happY vowel is usually pronounced [i] or [iː] in MaltE.
Since /i/ is tenser in MaltE than in RP, the rather central allophone [i] found
in conservative RP is rarely used in Malta – except by English retirees.
dress
Like in RP, the dress vowel /e/ in MaltE is [ ̞e] or [ɛ̝ ] and thus auditorily
indistinguishable in the two varieties, even if the MaltE realization is
generally somewhat more centralized.
lot
The lot vowel in MaltE is typically [ɔ] and thus somewhat higher than the
RP realization [ɒ]. In addition, the MaltE realization tends to be marginally
more central than its RP counterpart.
strut
The quality of the RP strut vowel /ʌ/ is [ɐ] in MaltE and thus almost
identical to RP, if perhaps marginally more central. It can merge with the
vowels of trap, palm, start and, less commonly, bath (see above).
Maltese English 25
4.5.2 Rhythm
Maltese English has a number of characteristics that have an effect on the
perceived rhythm: fuller vowels relative to RP; /ɪ/ or /ə/ before /l/, /m/ or
/n/ instead of syllabic consonants (in words like bottle, bottom; see Section
4.2.1 above); greater phonological integrity of individual words; less syllable
compression and thus syllables of more equal length. In particular at the
basilectal end of the continuum, where these features are most in evidence,
MaltE sounds rather syllable-timed (a tendency noted already in Calleja
1987 and Mazzon 1992) and thus close to an Italian EFL variety in terms of
suprasegemental phonology. An ancillary role in this phenomenon is played
by the creation of open syllables through full release and concomitant schwa
epenthesis, which Vella (1994: 76–7) finds after closed syllables ending in
26 manfred krug
stops or affricates, as for instance in stop, red, but, huge giving [ˈstɔph ə],
[ˈrɛdə], [ˈbɐth ə], [ˈhju:ʤə] – a trait that figures similarly in Italian EFL
varieties, which commonly feature open syllables in such contexts, too, e.g.
[ˈstɔpe] or [ˈbatɛ].
4.5.3 Stress
Calleja (1987: 71–85) and Vella (1994: 79–85) find postponed stress in
MaltE for words that in RP have stress on the antepenultimate or an earlier
syllable, as in examples (1) to (3). On a related note, single-stress words
and compounds with an early primary stress in RP such as (4) and (5)
often have late stress in MaltE or receive two full stresses, in which case
a secondary stress is typically promoted to a primary one. Compare the
following examples (from Calleja, Vella, Bonnici and ICE Malta):
(1) RP ˈcri.ti.ci.sm vs MaltE cri.ti. ˈci.sm
(2) RP ˈe.xer.cise vs MaltE e.xer. ˈcise
(3) RP ˈcen.ti.me.tre vs MaltE cen.ti. ˈme.tre
(4) RP ˈdish.ˌwa.sher vs MaltE ˌdish. ˈwa.sher
(5) RP ˈtrai.ning ˌpart.ners vs MaltE ˈtrai.ning ˈpart.ners
It is also striking that MaltE words ending in -ism (like tourism, fascism,
plagiarism, socialism, communism) fairly consistently receive stress on the
penultimate syllable. With some exceptions, English words ending in -ism
have equivalents in Maltese ending in -iżmu, which in turn are loans from
Italian ending in -ismo. Both languages have their stress in these words on
the penultimate syllable. Furthermore, Maltese has regular stress on heavy
final syllables (i.e. syllables with a long vowel or diphthong, and syllables
with a short vowel followed by consonant clusters or geminates; Fabri 2010:
800). Hence, late stress in words like criticism in MaltE can be borrowed
directly from Italian or, even more likely, via Maltese.
More generally, transfer of the Maltese pattern ‘heavy late syllables receive
stress’ seems to be applied to MaltE commonly, e.g. in the above MaltE
examples (1) to (3): except in cases of vowel epenthesis, criticism features
a consonant cluster in the final syllable, exercise a diphthong; and -metre
a long vowel in the penultimate. Not all words are affected by the MaltE
tendency to postpone stress, however. Vella (1994: 80) notes exceptions
like messenger and characteristically, which are stressed as in RP. Apparently,
therefore, stress shifting in MaltE depends on more than just which syllable
Maltese English 27
is stressed in RP, Maltese or the Maltese equivalent of an English word, and
it seems certain that lexical effects and effects related to secondary stress
and degree of reduction in RP play important roles, too (see Vella 1994:
81).
4.5.4 Intonation
Calleja (1987: 112) finds that ‘changes in pitch patterns and the occurrence
of tonic stresses are much more frequent’ in MaltE than RP. This is con-
firmed in essence and refined by Vella (1994: ch. 5), who provides a detailed
investigation of intonation in MaltE interrogatives and demonstrates
Maltese influence on MaltE intonation. Notably, Vella finds sentence-final
high rise patterns and post-nuclear stressed syllables with a rather high
pitch across (almost) all investigated structures – polar questions, inter-
rogatives guised as statements and interrogatives with primary verbs and
auxiliaries (examples below from Vella 1994: 236–40). This is noteworthy,
because in RP questions often show a fall or rise–fall at the end.
(6) Is that /all?
(7) You under- /stand?
(8) Can we com- /pare?
(9) Is there an abandoned cot- /tage?
(10) Do you have the old /mill?
In MaltE, final rises also occur more frequently in declarative sentences
and imperatives than in RP. For instance in the Bamberg Questionnaire for
Lexical and Morphosyntactic Variation in English (see Krug et al. in press,
for the full questionnaire; Krug and Sell 2013 for methodological detail),
the sentences given below, read out by a female acrolectal bilingual speaker,
also have a final rise on but (here meaning ‘though’ or ‘however’; see next
section for a discussion of syntactic aspects) in (11); and a final rise on the
last syllable in each of the examples (12) to (16). Crucially, such post-nuclear
rising intonation patterns are no exceptions; they are paralleled by many
other examples in the questionnaire recordings and also occur in MaltE
spontaneous interaction.
(11) I like this painting, I prefer the other one, /but.
(12) My sister and me got along very well when we were youn- /ger.
(13) This car is more fast than the one I drove yester- /day.
28 manfred krug
(14) French I do not use a /lot.
(15) American English does not spell like British Eng- /lish.
(16) Don’t stay walking on the /grass!
5 Morphosyntax
Prominent morphosyntactic features of acrolectal MaltE include want
constructions with a subjective pronoun in the dependent clause, sentence-
final but, definite article omission and marked uses of the progressive.
5 In the spoken BNC (c. 10m words), uses of seasons with the definite article outnumber those without
the article by a margin of about 5:1 when they follow the preposition in. The ratios range from 4:1
(for winter) via 5:1 (for autumn) to 7:1 (for spring and summer).
34 manfred krug
the case of unambiguous posts, institutions, languages and ordinal num-
bers; or because the referent has been established in the discourse). Vice
versa, MaltE avoids the definite article commonly where the NP is not
definite but generic (as with summer/winter).
5.4 Progressives
In press language, MaltE usage of the progressive conforms closely to the
exonormative British standard, with the noteworthy difference that it com-
bines more frequently with modal verbs and modal constructions (Hilbert
and Krug 2012). As in BrE, progressives have a higher text frequency in
spoken MaltE than in journalistic prose. Spoken MaltE also features pro-
gressives with a limited number of stative verbs, notably have and be,
which then often indicate dynamic meaning (with have) and ‘temporary
behaviour’ or ‘transitory state’ (with be), e.g.:
(37) dance can be a wider, uh, can have a wider influence and use than
it’s currently having.
(38) Muscat may well score crucial political points by accusing Gonzi of
misleading the public. But the sad truth of the matter is both sides
are being deceptive here: . . .
Often, however, as in (39) and (40) for example, semantic changes to
temporariness or dynamic meaning – which would license progressive be
and have constructions in standard BrE or AmE (e.g. he is being funny; he
is having a bath) – are hard to detect in MaltE:
(39) so that’s why I think we’re having the the older population that we’re
having apart from uh science which nowadays getting more into act
which the older people are being uh more healthier
(40) but we have so much jargon here and there that the winding roads
that we are having it’s quite expensive yeah
As in most major varieties of English, the progressive in MaltE can have
future time reference, e.g.:
(41) We’re leaving Saturday.
The progressive in MaltE can also occur with habitual meaning, which
would normally trigger simple (present or past) tense in standard BrE:
(42) so every month we’re just gathering this all this this data
Maltese English 35
On a related note, MaltE has an additional aspectual marker, stay + V-ing
(Bonnici 2010, example taken from ibid.), which appears to overlap with
iterative meaning:
6 Discourse markers
Discourse markers have not received the attention they deserve in the lit-
erature on MaltE, even though their use, certainly in combination, seems
quite unique. What follows are findings based on the Bamberg question-
naire and corpus data.
6.5 Even I
Rather unobtrusive features of MaltE develop where an identical struc-
ture exists in norm-providing inner-circle varieties but the MaltE meaning
differs in one of the following ways: (i) pragmatic enrichment by implica-
tures; (ii) generalization by semantic bleaching; (iii) specification or subtle
semantic shift due to a lexical gap or under-/overdifferentiation in a con-
tact language. Even I, for instance, in MaltE often just means ‘me too’
or ‘I also’ (if a verb follows); it thus lacks the emphatic or counter-to-
expectation meaning aspects of, say, a British or American lexicon entry
for even.6 Maltese lacks a lexical item synonymous with English even (Ray
Fabri, p.c.), but it has two lexemes for ‘also’/‘too’: Semitic ukoll and the
Italian borrowing anki. And since such brief and common phraseological
units like anchˈio may have been borrowed into Maltese in toto during the
long era of Maltese–Italian contact, the MaltE usage of even I might be
a calque from Italian that made its way into MaltE via Maltese anki jien
(‘also I’). Such subtle semantic differences would be interesting topics for
further research. But even in the absence of more detailed studies, the fact
6 I am grateful to Sarah Grech for bringing this example and the typical Maltese interpretation to
my attention. She also pointed out to me that even I is identical structurally and semantically with
Italian anch’io ‘me too’.
40 manfred krug
that the discourse frequency of even in the spoken ICE Malta material is
three times higher than in the spoken BNC strongly suggests that MaltE
indeed employs even in the more general meaning ‘too’.
7 The apparently incompatible tendencies towards backwards and forward in MaltE may be due to
the fact that nominal uses of forward known from the domain of sports (e.g. centre forward) affect
people’s ratings.
42 manfred krug
[center], holiday [vacation], jacket potato [baked potato], globalisation, liber-
alisation, organisation [-ization], lift [elevator]. It is apparently the effect of
formal teaching that encourages the use of traditional British variants for
such salient and easily learnable items.
9 Conclusion
9.1 Convergence, divergence and globalization in
different linguistic domains
Maltese English is not overtly codified and, from a bird’s-eye perspective, it
still displays a clear orientation towards British English, most obviously so
in its lexicon and morphosyntax. As early as 1998, Trudgill discusses con-
vergence and divergence scenarios in the domains of phonology, grammar
and lexicon. He points out that while the situation for grammar is difficult
to determine, the ‘homogenisation in the direction of North American
usage’ at the lexical level contrasts with divergence between British and
American English – and between varieties of English around the world in
general – for the level of phonology (1998: 30–2). The data reported here
essentially confirm this view. It was seen that differences between acrolectal
MaltE and standard Southern British English are more obvious for phono-
logical features and lexical items than for morphosyntax, and that some
lexical items show signs of Americanization or globalization (e.g. towards
truck, sick, sports, package, for rent). In acrolectal MaltE grammar, especially
in formal written genres, the exonormative BrE model is still very much
adhered to, although differences in usage can be observed, for instance in
the domains of progressives, verb complementation and determiners.
References
Baskaran, Loga. 2008. Malaysian English: morphology and syntax. In Mesthrie,
ed., 610–23.
Bautista, Maria Lourdes S. and Andrew B. Gonzalez. 2006. Southeast Asian
Englishes. In Kachru et al., eds., 130–44.
Beer, Max. 2011. Rhoticity in Maltese English. Unpublished Master’s thesis, Uni-
versity of Bamberg.
Biewer, Carolin. 2012. Acrolectal English in the South Pacific: emerging standards?
In Hickey, ed., 336–53.
Bolton, Kingsley. 2006. Varieties of World Englishes. In Kachru et al., eds., 289–
312.
Bonnici, Lisa. 2010. Variation in Maltese English: the interplay of the local and the
global in an emerging postcolonial variety. Unpublished PhD dissertation,
University of California, Davis. Available at http://linguistics.ucdavis.edu/
pics-and-pdfs/DissertationBonnici.pdf; accessed 12 Dec. 2012.
Bonnici, Lisa, Michaela Hilbert and Manfred Krug. 2012. Maltese English. In
Bernd Kortmann and Kerstin Lunkenheimer, eds., The Mouton World Atlas
of Variation in English. Berlin and New York: Mouton de Gruyter, 653–68.
Borg, Albert and Marie Azzopardi-Alexander. 1997. Maltese. London and New
York: Routledge.
Brincat, Joseph M. 2005. Maltese: an unusual formula. MED Magazine (The
monthly webzine of the Macmillan English Dictionaries) 27. Available
at http://macmillandictionaries.com/MED-Magazine/February2005/27-LI-
Maltese-print.htm; accessed 12 Dec. 2012.
Calleja, Marisa. 1987. A study of stress and rhythm as used by Maltese speakers of
English. Unpublished B.Ed. thesis, University of Malta.
48 manfred krug
Camilleri, Antoinette. 1991. Crosslinguistic influence in a bilingual classroom: the
example of Maltese and English. Edinburgh Working Papers in Linguistics 2:
101–11.
Census of Population and Housing 2005. Valletta: National Statistics Office, Malta (2
vols., published 2007, available at www.nso.gov. mt; accessed 12 Dec. 2012).
Census of Population and Housing 2011: Preliminary Report. Valletta: National Statis-
tics Office, Malta (published 2012, available at www.nso.gov. mt; accessed
12 Dec. 2012).
Fabri, Raymond. 1995. The tense and aspect system of Maltese. In Rolf Thieroff,
ed., Tense Systems in European Languages II (Linguistische Arbeiten, 338).
Tübingen: Niemeyer, 327–43.
2008. Maltese-English language contact. Paper given at Bamberg, July.
2010. Maltese. In Christian Delcourt and Piet van Sterkenburg, eds., The Lan-
guages of the New EU Member States. Revue Belge de Philologie et d’Histoire
88(3): 791–816.
Hickey, Raymond. 2012a. Glossary. In Raymond Hickey, ed., Standards of English:
Codified Varieties around the World. Cambridge University Press, 392–406.
ed. 2012b. Standards of English: Codified Varieties around the World. Cambridge
University Press
Hilbert, Michaela and Manfred Krug. 2010. The compilation of ICE Malta: state
of the art and challenges along the way. ICAME Journal 34: 54–63.
2012. Progressives in Maltese English: a comparison with spoken and written
text types of British and American English. In Ulrike Gut and Marianne
Hundt, eds., Mapping Unity in Diversity (Varieties of English around the
World). Amsterdam: Benjamins, 103–36.
Johnson, Wyn and David Britain. 2007. L-vocalisation as a natural phenomenon:
explorations in sociophonology. Language Sciences 29 (special issue on Issues
in English Phonology, ed. Philip Carr and Patrick Honeybone): 294–315.
Kachru, Braj B. and Tom McArthur. 1992. Indian English. In McArthur, ed.,
504–8.
Kachru, Braj B., Yamuna Kachru and Cecil L. Nelson, eds. 2006. The Handbook
of World Englishes. Oxford: Blackwell.
Kmetova, Tanya and Lydia Sciriha. 1992. Why bother about those tenses?
Sǎpostavitelno ezikoznanie 17(3): 45–57.
Kortmann, Bernd and Kerstin Lunkenheimer, eds. 2011. The Electronic World
Atlas of Varieties of English [eWAVE]. Leipzig: Max Planck Institute for
Evolutionary Anthropology. www.ewave-atlas.org.
Kortmann, Bernd and Kerstin Lunkenheimer, eds. 2012. The Mouton World Atlas
of Variation in English. Berlin and New York: De Gruyter Mouton.
Krug, Manfred. 1998. String frequency: a cognitive motivating factor in coales-
cence, language processing and linguistic change. Journal of English Linguistics
26(4): 286–320.
2000. Emerging English Modals: A Corpus-Based Study of Grammaticalization.
Berlin and New York: Mouton de Gruyter.
Maltese English 49
2007. Modern methodologies and changing standards in English linguistics. In
Marı́a Losada Friend, Pilar Ron Vaz, Sonia Hernández Santano and Jorge
Casanova, eds., Proceedings of the 30th International AEDEAN Conference.
Huelva: Servicio de Publicaciones de la Universidad de Huelva. [CD-Rom]
2011. Auxiliaries and grammaticalization. In Bernd Heine and Heiko Narrog,
eds., Handbook of Grammaticalization. Oxford University Press, 547–58.
Krug, Manfred and Anna Rosen. 2012. Standards of English in Malta and the
Channel Islands. In Raymond Hickey, ed., Standards of English: Codified
Varieties around the World. Cambridge University Press, 117–38.
Krug, Manfred and Katrin Sell. 2013. Designing and conducting interviews and
questionnaires. In Manfred Krug and Julia Schlüter, eds., Research Methods
in Language Variation and Change. Cambridge University Press.
Krug, Manfred, Michaela Hilbert and Ray Fabri. In press. Maltese English mor-
phosyntax: corpus-based and questionnaire-based studies. Il-Lingwa Taghna
(special issue: Towards a Description of Maltese English, ed. Alexandra Vella
and Ray Fabri).
Labov, William. 1990. The intersection of sex and social class in the course of
linguistic change. Language Variation and Change 2: 205–54.
McArthur, Tom, ed. 1992. The Oxford Companion to the English Language. Oxford
University Press.
Mahboob, Ahmar. 2008. Pakistani English: morphology and syntax. In Mesthrie,
ed., 244–58.
Mazzon, Gabriella. 1992. L’inglese di Malta. Naples: Liguori Editore.
Mesthrie, Rajend, ed. 2008a. Varieties of English, vol. 4: Africa, South and South
East Asia. Berlin and New York: Mouton de Gruyter.
2008b. Introduction: varieties of English in Africa, South and South East Asia.
In Mesthrie, ed., 23–31.
2008c. Synopsis: the phonology of English in Africa and South and Southeast
Asia. In Mesthrie, ed., 307–19.
National Minimum Curriculum. 1999. Issued by the Maltese Ministry of Educa-
tion. Available at www.curriculum.gov.mt/docs/nmc english.pdf ; accessed
12 Dec. 2012.
Philipp, Isabell. 2008. Maltese English: the variable use of prepositions. Unpub-
lished Master’s thesis, University of Bamberg.
Platt, John and Heidi Weber. 1980. English in Singapore and Malaysia: Status,
Features. Functions. Kuala Lumpur: Oxford University Press.
Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech, Jan Svartvik. 1985. A
Comprehensive Grammar of the English Language. London: Longman.
Rohdenburg, Günter and Julia Schlüter. 2009. New departures. In Günter
Rohdenburg and Julia Schlüter, eds., One Language, Two Grammars? Dif-
ferences between British and American English. Cambridge University Press,
364–423.
Schneider, Edgar W. 2007. Postcolonial English: Varieties around the World.
Cambridge University Press.
50 manfred krug
Schreier, Daniel, Peter Trudgill, Edgar W. Schneider and Jeffrey P. Williams. 2010.
Introduction. In Daniel Schreier, Peter Trudgill, Edgar W. Schneider and
Jeffrey P. Williams, eds., The Lesser-Known Varieties of English. Cambridge
University Press, 1–14.
Sciriha, Lydia. 2001. Trilingualism in Malta: social and educational perspectives.
International Journal of Bilingual Education and Bilingualism 4(1): 23–37.
Sciriha, Lydia and Mario Vassallo. 2006. Living Languages in Malta. Malta: Print
IT Printing Services.
Stolz, Thomas. 2011. Maltese. In Bernd Kortmann and Johan van der Auwera, eds.,
The Languages and Linguistics of Europe. Berlin and New York: De Gruyter
Mouton, 241–55.
Tayao, Ma. Lourdes G. 2008. Philippine English: phonology. In Mesthrie, ed.,
292–306.
Thusat, Joshua, Emily Anderson, Shante Davis, Mike Ferris, Amber Javed, Angela
Laughlin, Christopher McFarland, Raknakwan Sangsiri, Judith Sinclair, Vic-
toria Vastalo, Win Whelan and Jessica Wrubel. 2009. Maltese English and
the nativization phase of the dynamic model. English Today 25: 25–32.
Trudgill, Peter. 1998. World Englishes: convergence or divergence? In Hans
Lindquist, Staffan Klintborg, Magnus Levin and Maria Estling, eds., The
Major Varieties of English: Papers from MAVEN 97. Växjö: Acta Wexionensia,
29–34.
2004. Dialects. 2nd edition. New York: Routledge.
Vella, Alexandra. 1994. Prosodic structure and intonation in Maltese and its influ-
ence on Maltese English. Unpublished PhD thesis, University of Edinburgh.
Walshe, Shane. 2009. Irish English as Represented in Film (University of Bamberg
Studies in English Linguistics, 53). Frankfurt am Main: Lang.
Wells, J. C. 1982. Accents of English, vol 1. Cambridge University Press.
Wrench, Alan A. and James M. Scobbie. 2003. Categorising vocalisation of English
/l/ using EPG, EMA and ultrasound. Proceedings of the 6th International
Seminar on Speech Production, 314–19.
ch a p ter 3
Gibraltar English
David Levey
1 Introduction
Gibraltar is a British overseas territory situated on the southern tip of the
Iberian Peninsula at the western entrance to the Mediterranean with the
northern coast of Africa lying just 30km across the Strait of Gibraltar. It
measures just 6.8 km2 and is one of the most densely populated areas in
Europe (4.9 people per km2 ). It has very few natural resources and its
economy is based mainly on tourism, shipping and financial services. Most
of the population lives in the small town area in the shadow of the famous
Rock, or el Peñón as it is known in Spanish, which soars magnificently
to a height of some 426 metres. This Rock together with Jebel Musa, the
mountain on the Moroccan coast opposite, are said to have formed the
mythological Pillars of Hercules which, legend has it, were forced apart to
form the African and European continents.
The British have been in Gibraltar for more than three centuries although
the question of sovereignty remains contentious, and Spain has sought
its return ever since it was ceded to Britain by the terms of the Treaty
of Utrecht in 1713. Gibraltar and its people, for their part, have always
rejected any suggestion that they are Spanish and have resisted all attempts
by Spain to reclaim the Rock, vehemently defending their right to self-
determination. This has been shown in the two referendums which have
been held on the matter. In 1967, 99.6 per cent voted against the proposal
of Spanish sovereignty and thirty-five years later, in 2002, 98.5 per cent of
Gibraltarians voted against the option of joint sovereignty. Gibraltarians
have traditionally felt strongly British and manifestations of this sentiment
are clearly visible on the Rock in the form of Union Jacks, pictures of
the Queen and pro-British slogans. In recent years, however, particularly
when the local population has felt let down or abandoned by the UK
government, a stronger sense of Gibraltarian nationalism has emerged.
Although Gibraltar remains under the jurisdiction of the United Kingdom,
51
52 david levey
which is still responsible for matters such as defence and foreign affairs, the
new 2006 Constitution gave the local government increased autonomy in
running its own affairs.
Of the 29,752 people who live in Gibraltar, 24,288 of these are ‘Gibral-
tarians’, 3,042 are ‘UK British’ and 2,422 are classified as ‘Others’.1 The
Gibraltarian population is predominantly Roman Catholic (88 per cent)
with Church of England and other Christians accounting for 6.6 per cent
and a further 2 per cent of the population are Jewish. Hindus, Muslims,
Atheists/Agnostics and ‘Other’ make up the remaining 3.5 per cent.2
English is the only official language in Gibraltar, yet largely due to
geographical proximity and historical as well as family ties, Spanish and/or
the local variant Yanito (also spelled Llanito) still arguably remain the
most common forms of expression in the home domain and in informal
contexts. The situation, however, appears to be changing and English use
is increasing, particularly amongst younger speakers (see Levey 2008a: 95–
8). While the Spanish spoken in Gibraltar has certain distinctive features
unique to the Rock, its dialectal form and accent, in general terms, are not
dissimilar to, and are sometimes indistinguishable from, those used in the
neighbouring Andalusian towns.3
Yanito is tricky to define and classify since it often implies different
things to different people. For some, it is simply Gibraltarian Spanish, a
variant of Andalusian Spanish but with some locally specific lexical items
incorporated. These are often words borrowed or adapted, not only from
English, but also from the languages of the immigrant communities who
have settled on the Rock. For others, however, Yanito refers fundamentally
to the local tendency to code-switch between Spanish and English.4 Some
would therefore argue that there are three distinct languages spoken in
Gibraltar: English, Spanish and Yanito (Ballantine 2000: 118–19), albeit
with considerable overlap.
Most Gibraltarians can converse, to varying degrees, in two or more lan-
guages, sometimes independently and sometimes simultaneously. Although
many would consider themselves multilingual, this does not mean that
1 Figures were kindly supplied by the Statistics Office, Government of Gibraltar and are correct as of
31 December 2013. The last full census was carried out in 2001.
2 Figures are taken from the Census of Gibraltar (2001).
3 According to Lipski (1986: 417–19) and Ballantine (2000: 119), Gibraltarian Spanish shares certain
phonetic features with the Canary Islands and South America.
4 As well as being the name given to the local vernacular, Yanito is also the demonym, used both
locally and in neighbouring Spain, to describe someone from Gibraltar. It probably derives from
‘Gianni’ (the diminutive of the name ‘Giovanni’), and harks back to the eighteenth and nineteenth
centuries when Italians, particularly Genoese, abounded in Gibraltar.
Gibraltar English 53
they speak each language with equal ease and proficiency. Language choice
and preference often depend on the situation and domain and are con-
ditioned by factors such as age, social class, education and ethnic back-
ground (Levey 2008a: 95–8). Older speakers, for example, who have not
gone on to further education, may prefer to speak in Spanish or Yanito,
and may find difficulties maintaining an extended complex conversation
in English. Younger generations, on the other hand, having been educated
in English, tend to consider their vocabulary is more extensive in English,
and may feel less comfortable speaking formal Spanish in non-colloquial
environments.
It is not unusual to find different languages spoken within the same
nuclear family. While parents may choose to speak to their children in
English for the benefit of their education, they may speak to each other in
Yanito and to their parents in Spanish. Although English is seen as the pres-
tige language and encouraged, this does not mean that other languages are
rejected. Gibraltar has always been a multicultural speech community and
language is seen fundamentally as a means of communication. Although
some may view language choice as an act of identity or a declaration of
allegiance, for the most part, Gibraltarians do not consider it paradox-
ical or a contradiction to feel staunchly British yet choose to speak in
Spanish.
5 Article X of the Treaty of Utrecht (1713) states: ‘The Catholic King does hereby, for himself, his heirs
and successors, yield to the Crown of Great Britain the full and entire propriety of the town and
castle of Gibraltar, together with the port, fortifications, and forts thereunto belonging; and he gives
up the said propriety to be held and enjoyed absolutely with all manner of right forever, without any
exception or impediment whatsoever’ (translation from the original Latin from Valle Gálvez and
González Garcı́a 2004: 461).
6 The fifth paragraph of Article X states: ‘Her Britannic Majesty, at the request of the Catholic King,
consents and agrees that no permission shall be given, under any condition, to neither Jew nor Moor
to reside or dwell in the said town of Gibraltar.’
Gibraltar English 55
in Spanish and Italian as well as in English. Several languages and dialects,
mostly of Romance origin, would have co-existed on the streets and docks
of Gibraltar. Many of the Jews would have spoken Ladino (Judeo-Español),
a language derived from Old Castilian, which was widely spoken by the
exiled Sephardic Jews since their expulsion from Spain at the end of the
fifteenth century. It has been suggested that a type of Mediterranean lingua
franca or ‘Romance-based pidgin’ (Kramer 1986: 53) was also spoken during
the eighteenth century. However, given the limited evidence, it is difficult
to know for sure whether this did indeed exist as such and, if it did, how
uniform or widespread it was. Speculation seems to partially stem from
a brief and tantalizing allusion made by an eighteenth-century Spanish
historian. In his Historı́a de Gibraltar published in 1782, Ignacio López
de Ayala mentions the existence of an international vernacular spoken in
Gibraltar which was apparently understood by all:
i tanto éstos (los Genoveses) como los Judios hablan bien ó mal el Castellano
é Inglés, i un dialecto ó jerga común a todas las naciones, sin excluir las
Africanas. (López de Ayala 1782: 374)
[both the Genoese as well as the Jews speak Castilian and English well or
badly as well as a dialect or jargon common to all nations including Africans]
In the first half of the eighteenth century the population was to fluctuate
considerably as deadly epidemics ravaged Europe. The demographics were
also to change as people from different nations sought refuge from the wars
and upheavals that marked the period. A new wave of Genoese immigrants,
both Christian and Jewish, headed for the Iberian peninsula after the
‘Ligurian Republic’, as Napoleon Bonaparte renamed it, was annexed by
France in 1805. There was also immigration from France, Portugal, Spain
and also Minorca, which, like Gibraltar, had been ceded to Britain by the
terms of the Treaty of Utrecht. Italian-speaking Maltese immigrants, who
were to play an important role in Gibraltar’s sociolinguistic and cultural
development, began to arrive after the Mediterranean archipelago became
part of the British Empire in 1814.
While other European ports and cities saw their commercial activities
badly affected by events in Europe, Gibraltar was able to prosper and hang
up a ‘business as usual’ sign. As the population increased, overcrowding
became a major problem and it was not uncommon for six or seven family
members to live in one room. In these cramped and usually unsanitary
conditions, the contagious diseases such as influenza, cholera and typhoid
which appeared in waves in Europe in the first half of the nineteenth
century spread particularly rapidly, devastating the local population. Yet
56 david levey
there was no shortage of new immigrants prepared to take their place and
by 1871 official census figures reveal the population had grown to 18,695.
As a result, the local authorities felt it necessary to curb immigration by
introducing the ‘Aliens Order in Council’ in 1873 which permitted free
access to the Rock only to the British and required ‘Aliens’ to obtain special
permits. During the twentieth century, immigration continued, mostly
from other British colonies such as Malta and later from India. Economic
migrants from Morocco also began arriving in the 1970s and 1980s to fill
the labour vacuum left after the closing of the border with Spain, adding
to the racial and cultural fusion.7
Tentative attempts by the authorities to encourage the use of English had
limited success. English was the official language, the language of education
and administration, yet Spanish and its variants were the everyday means
of communication. This of course is not to say that nobody spoke English.
The UK serviceman and their families naturally spoke English along with
other UK residents in Gibraltar. The more affluent classes, who could
afford to, sent their children to private English schools. Many Jews and
Indians, whose home language was often not Spanish, placed considerable
importance on multilingualism. However, taking the population as a whole,
it seems fair to say that levels of English proficiency remained low until
well into the second half of the twentieth century.
The Second World War was to be an important turning point in Gibral-
tar’s social and linguistic development. The Rock was of vital strategic
importance to Britain and largely for its own safety the whole civil popula-
tion of Gibraltar was evacuated for the duration of the war. The majority
eventually went to the UK. For many it was the first time they had left
their homes and were put in a situation where they were obliged to speak
English. When they were eventually repatriated, they returned to Gibraltar
with a stronger sense of national pride and a better level of English. The
first seeds had been sown but there was no revolutionary change in the
language habits. Many, or perhaps most, had started to acquire English
as a second language. On returning to their homes, despite the best of
7 Whereas in 1981, during the Spanish blockade, there were 2,140 Moroccans listed as ‘usually resident’,
the 2001 census figures showed that numbers had dropped by 55 per cent to just 961. Although this
considerable fall in numbers may partially be explained by ‘the inherent difficulties of enumerating
the Moroccan community’, this does not detract from the fact that there is ‘a significant decrease
in the numbers of Moroccans living and working in Gibraltar’ as employment figures clearly show
(Census of Gibraltar 2001: xvi). After the reopening of the border and increased freedom of movement
for EC members, it has become increasingly difficult for Moroccan workers to obtain work permits
and work.
Gibraltar English 57
intentions in some cases, it inevitably proved more comfortable and com-
forting to return to familiar language forms.
Although English language competence increased gradually after the
war, the policies adopted by Spain under General Franco’s dictatorship
arguably did more, directly and indirectly, to improve the levels of English
in Gibraltar than any local or UK initiatives could. In 1969, in a misguided
attempt to starve the local population out, the frontier or verja, as it is
known, was closed overnight, thereby cutting Gibraltar off from Spain
and the rest of Europe. The decision had drastic and sometimes tragic
consequences for those living on both sides of the border. The blockade
was to last thirteen years and Gibraltarians, particularly older ones, find it
hard to forgive and forget this singular action which divided families and
changed lives. Besides the political consequences, the closing of the frontier
had considerable effects on Gibraltar’s linguistic development.
Resentment and hostility towards Spain intensified and for some lan-
guage choice became a declaration of allegiance and provided a motivation
for learning and speaking English. At the same time, the need or excuse
to speak Spanish diminished. With the frontier closed, regular social con-
tact with Spanish friends and family ceased as did commercial relations
with Spain. Spanish labourers, domestic workers and childminders, who
had always provided an important language input in the home and the
workplace, and have been cited as one of the chief reasons for language
maintenance, could no longer cross the border. As contact with Spain and
its language decreased, so language contact with the UK increased. Coin-
ciding with the 1970s tourist boom, there were now more regular flight
connections between Gibraltar and the UK. With trips and holidays to
Spain no longer possible, those who could afford to, flew to Britain. Dur-
ing this period, young Gibraltarians, helped by government grants, began
to further their studies at UK universities in greater numbers. Therefore,
when the border finally reopened on 15 December 1982 after thirteen
years of isolation from its Spanish neighbours, a new speech community
emerged with a stronger national identity and with greater English language
confidence.
Many years have passed since then and, on the surface at least, it would
appear that things have returned to relative normality. Cross-border rela-
tions have resumed, but some would argue that the close relationships that
once existed have been lost and time is still needed to rebuild bridges.
English language competence has notably increased and Gibraltar English
is more widely used than ever before, but this does not mean that Spanish
has been displaced.
58 david levey
3.1 Vowels
kit/fleece
Previous studies (West 1956; Ballantine 1983; Enriles 1992) note that Gibral-
tarians often fail to distinguish between kit and fleece vowels, thus chip
and cheap may be perceived and produced identically.8 fleece tends to
be shortened and realized as [i] or [ ̞i] and kit may be produced with a
certain degree of tensing, especially before /l/. While kit–fleece merger
still features in the speech of some Gibraltarians, particularly older ones,
younger speakers today generally distinguish the two vowels by quality (see
Cal Varela 2001; Levey 2008a).
foot/goose
Enriles (1992: 30); Errico (1997: 141) and Kellermann (2001: 362) noted the
absence of a long close back vowel. More than twenty years ago, Enriles
(1992: 26) wrote that foot and goose tend to merge and were both
pronounced [u̞ + ]. Although short realizations are still evident, a longer
goose vowel is becoming more common and complete foot–goose
merger was found in less than 10 per cent of pre-adolescents and adolescents
Levey (2008a: 104–5).
lot/thought
The thought vowel in Gibraltar tends to be noticeably short and may
merge with lot, with both being realized as [ɔ] ̞ and therefore cot/caught
or wok/walk may not be differentiated. While merger is the norm amongst
speakers over the age of 45, the two vowels are now distinguished by younger
Gibraltarians with lot being lower than thought.
8 This is also suggested by spellings in local Yanito dictionaries (e.g. scrin ‘screen’, tipá ‘teapot’, pisup
‘pea soup’).
Gibraltar English 59
cat/cut or match/much are often not perceived. Recent studies, however,
reveal that more than 80 per cent of children and adolescents differentiated
between the two vowels (Levey 2008: 110) with trap increasingly being
produced as a more open and front vowel, tending towards [æ].
Although some speakers pronounce bath or start with a back vowel,
a short centralized short vowel is favoured, often overlapping with strut.
Amongst the less anglophone speakers, triple trap/strut and start
merger may occur with all three vowels being realized as [ä].
nurse
In traditional Gibraltar English, the pronunciation of this vowel may be
conditioned by orthography and accompanied by postvocalic /r/. Thus
shirt and bird would be pronounced [ir], nurse and turn [ur], earth and
earn [er] and word or work [or]. In modern Gibraltar English, which is
non-rhotic, nurse is a realized as a notably short front vowel [ɛ] although
[ɜː] is making inroads.
letter
The use of weak vowels is not common in Gibraltar. letter tends to
be produced as [a] by older speakers and as [ɐ] by younger speakers. The
schwa is also beginning to appear.
3.2 Consonants
/p, t, k/
Initial plosives are generally released with aspiration with voice-onset time
(VOT) values similar to those found in standard British English (Cal Varela
2001: 37). Final /p, t, k/ tend to be audibly released and, in the case of /t/,
may be accompanied with some affrication.
/b/ /v/
Labiodental /v/ has no phonemic status in Spanish where the letters <b>
and <v> are both produced bilabially. In Gibraltar English /b/ /v/ merger
occurred very occasionally among older speakers. In a small number of
60 david levey
cases, the use of [ʋ] substituted /v/, particularly after nasals (e.g. involved
[iɱˈʋɔlvd]).
/ʃ/ /tʃ/
/ʃ/ /tʃ/ merger was once a typical feature of Gibraltar English and older
Gibraltarians may not distinguish between minimal pairs (e.g. shoes/choose;
she’s/cheese; wash/watch), realizing both as fricatives. This, however, is con-
siderably less common amongst the new generation.9
Initial s + consonant
Hispanophones speaking English often insert an epenthetic vowel before
words beginning with /s/ + consonant. This is also a feature of traditional
Gibraltar English (e.g. start [eˈstat]; strong [eˈstron]), although it is not
widespread in modern Gibraltar English.
/r/
Gibraltar English is non-rhotic. Approximant [ɹ] and flap [ɾ] cohabit in
Gibraltar. The use of Spanish coloured trill [r], which Kellermann (2001:
398) notes in a few informants born in the 1940s and 1950s, has now
practically disappeared.
[l]
Gibraltar English /l/ is notably clear in all positions although there are signs
of darkening amongst younger informants (see Levey 2008a: 158–9).
H-dropping
Although an aspirated [h] realization is the norm in Gibraltar English, H-
dropping is occasionally present in intervocalic environments (e.g. behind
[bɪˈaind], behave [bɪˈeiv]) and in initial <hu> lexical sets (e.g. huge [judz];
human [ˈjuman].
T-glottalling
T-glottalling, was found to be present sporadically in the speech of young
Gibraltarians, but rarely occurred in intervocalic positions. No cases were
recorded for speakers older than 45. As was the case with TH-fronting, it
is significant that the glottal stop is not a feature of Spanish and indeed
is quite difficult for hispanophones to produce. While T-glottalling in
the UK has traditionally been associated with less prestigious lower-class
accents, although this perception is undoubtedly changing, in Gibraltar, it
is not a social marker and is essentially a fast speech phenomenon reflecting
English language fluency.
3.3 Prosody
Little has been written on prosody in Gibraltar and it is an area, which
is worthy of further study. Here I can merely offer a few brief impres-
sions. Gibraltar English is fundamentally different from most UK regional
varieties in that it has a syllable-timed rhythm rather than a stress-
timed one and weak forms are rarely used. This, to some ears, makes
it sound ‘non-native’. It has distinctive stress and intonation patterns.
Compound nouns such as seashell, car ferry or dockyard, for example, are
stressed on the second syllable and accompanied by a characteristic rising
intonation.
It is often claimed that Gibraltar English is simply English coloured by
Spanish. Findings suggest that this is a fallacy. When a Gibraltarian speaks
English it is usually evident that he or she is not from the UK. But, to
the trained ear, it is also clear that his or her accent is markedly different
from that of an Andalusian speaking English. The Gibraltarian’s tone and
rhythm reveal different primary and secondary influences, perhaps as the
direct or indirect result of a Genoese, Sephardic or Maltese ancestry.
4 Lexicon
Gibraltarians are renowned for their code switching. On entering Gibraltar
one is soon struck by they way that locals maintain conversations in more
than one language. As an example of the way this works I cite a short
exchange I overheard in a shop on my last visit to Gibraltar:
62 david levey
10 In addition to my own research and observation, four main sources were consulted when compiling
this section: Cavilla’s (1990) Diccionario yanito, Vallejo’s (2001) The Yanito Dictionary and Montero
Sánchez’s (2010) El habla del campo de Gibraltar and Kramer’s (1986) English and Spanish in
Gibraltar.
11 The presence of Anglicisms and English transfer in the Spanish of the towns and villages in the
surrounding area known as el Campo de Gibraltár has been well documented (Garcı́a Martı́n 1996,
1997; Montero Sánchez 2010). These are particularly notable in La Linea de la Concepción where,
for example, there is a well-known street and car park near the border called and spelt Focona which
derives from ‘Four Corners’.
Gibraltar English 63
often attributed to the speech of Gibraltarians are arguably not nearly as
widespread as claimed, or at least not now.
Many of the entries in Yanito word lists are English borrowings which
originated in past times when little English was spoken. Words such as
beki sangui ‘bacon sandwich’, trafilai ‘traffic lights’ or siticonsi ‘City Coun-
cil’ are really phonetic representations of Andalusian Spanish transfer.12 If
these realizations were heard over the border in La Linea (Spain) in an ESL
class they would probably be treated as English pronunciation errors and
corrected. This raises the hundred-dollar question whether these should be
considered bona fide lexical items or not. To take the debate one stage fur-
ther, in terms of their pronunciation, given that English is the only official
language in Gibraltar and Gibraltar is British, should they be accepted as
legitimate British regional variants?
The idea of the Gibraltarian speaking Spanish with a thick Andalusian
accent throwing in a word of badly pronounced English has a quaint
and sometimes comical appeal. While this stereotype might have been the
case in previous generations and may still exist in some older speakers,
times have changed and continue to change. As levels of education and
contact with the English language increase, there is a new generation of
Gibraltarians who are competent in English and speak with a less marked
pronunciation. Spanish transfer is less evident and if, for example, they do
say rolipó, it is not because they can’t pronounce ‘lollypop’. If they choose
to pronounce it in the ‘Yanito’ way it may be for comic effect (Gibraltarians
have a keen sense of humour and are not adverse to self-parody!), or, in
some cases, it may be a case of accommodating to the speech of their
interlocutor.
Much of the food imported to cater for British tastes from the late
nineteenth century to the second half of the twentieth century was new to
Gibraltar. The autochthonous population pronounced these new products
as heard or read. This gave rise to: arishu ‘Irish Stew’; combi or carne con
bı́ ‘corned beef’; greivi ‘gravy’; Oso ‘Oxo stock cube’; Quekaro ‘porridge’
(from Quaker Oats); chinchibı́a ‘ginger beer’; saltipina ‘salted peanuts’;
bequipagua ‘baking powder’; capotı́n ‘cup of tea’; liqueribar ‘liquorice bar’;
chingá ‘chewing gum’.
As is to be expected, there are numerous words related to work, particu-
larly in the docks and construction. The fact that Gibraltarians have always
worked side by side with Spanish workers may also partially explain why
12 Andalusian Spanish tends to elide certain intervocalic and final consonants, thus the city of Cádiz
is commonly pronounced (and popularly written) Cai.
64 david levey
the pronunciation of English words was not anglicised. Some of the most
common words included: doquia ‘dockyard’; forme ‘foreman’; cren ‘crane’;
cimen ‘cement’; esprin ‘spring’; guasha ‘(tap) washer’; gerda ‘girder’; winchi
‘winch’; manpagua ‘manpower’; iunio or working iunio ‘trade union’; penshi
‘pension’. At home, there was juva ‘vaccum cleaner’ (from Hoover), hacer
londri ‘do the laundry’.
While Spanish remained the principal home language, English was the
enforced language of education. Not only was this reflected in the language
of the classroom (e.g. tishe/tisha ‘teacher’; chó ‘chalk’; sepli ‘say please’)
but also extended into the playground. The game of meblis ‘marbles’, for
example, was popular amongst past generations of Gibraltarian children
and had its own language to convey the dos and don’ts of flicking or
shooting (e.g. follinacle ‘fold in knuckle’; fondinga ‘fold in finger’). It is
interesting to note that the game, along with its corresponding English
lexicon, spread well outside the confines of Gibraltar, as attested to by
numerous Spanish colleagues who grew up in the 1960s and 1970s in the
provinces of Cádiz and Málaga. Needless to say, in the modern computer
age, the game and its vocabulary mean little to the younger generations.
English gerunds are commonly used after Spanish verb forms, giving rise
to: tomamos un trinqui(ng) ‘Let’s have a drink’, tomar a guashi(ng) ‘have a
wash’, hacer nitin a champa ‘to knit a jumper’. Amongst younger speakers
today, on both sides of the border, this tendency continues: voy shopping
‘I’m going shopping’, tengo training hoy ‘I’ve got training today’. It is
also common to create Spanish-sounding words from English words: afor-
darse ‘to afford’; chiterı́a ‘cheating’; dampista ‘dumptruck driver’; pipando
‘piping hot’; plomero ‘plumber’. Perhaps the most well-known example of
what some would term Spanglish is the verb afolinarse which means to
‘line up’ and comes from the British military term ‘to fall in’. There is
another slightly different variation of the hispanification process whereby,
due to influence from English, existing Spanish words are given a dif-
ferent or second meaning, which they do not usually have. Examples of
these ‘false friends’ include: aplicación ‘(job) application’ (Sp = solicitud );
apologı́a ‘apology’ (Sp = disculpa); soportar ‘support’ (Sp = apoyar); vacancia
‘vacancy’ (Sp = vacante); documentario ‘documentary’ (Sp = documental );
orden ‘order’ (Sp = pedido). There are other Yanito words such as estación
de policı́a ‘police station’ or teatro de operaciones, ‘operating theatre’ which
are literal translations from English but make no sense in standard Spanish.
While most of Gibraltar’s lexicon is English or Spanish in origin, there
are several items, which have come about through contact with those
immigrant communities who settled on the Rock. While the Maltese and
Gibraltar English 65
Italian influence is notable in several local names and some of the typical
Gibraltarian dishes such as rosto, panissa and calentita, the lexical legacy,
which was undoubtedly once more evident, is now limited to a handful
of words which are increasingly disappearing from use. In his Diccionario
yanito, Cavilla includes the following examples: bucherı́o ‘a din or racket’
(It. = buscherio); tana ‘hiding place’ (It. = tana); chufo ‘tuft of hair’ (It. =
ciuffo). The Jewish community has also contributed various words. Some
unsubstantiated sources have put the figure as high as 500, but my own
research seems to suggest that this figure is greatly exaggerated.13 Amongst
those words of Hebrew or Haketia origins which are recognized by the wider
Gibraltarian community are: bizim ‘balls, guts’; ha ham ’important person’;
haiznear ‘observe carefully’; las nogas ‘synagogue’; haremos who ‘what can
we do?’ (expression of resignation); echar el who ‘to curse someone’. Arabic
has provided words such as jará ‘pigsty’; flus ‘money’; zup ‘penis’; bicef
‘enough’; and chuni ‘nice’ might come from German (schön). There are
other items whose origins are less clear such as aliquindoi, meaning ‘keep
a lookout’, which Vallejo (2001: 25) suggests might come from Caló, a
Gypsy language spoken in Spain, or possibly from the French un clin d’oeil
‘a twinkling of the eye’.14
Although it might have once been considered a sign of lexical deficiency
or semi-lingualism, Yanito is seen by many as a badge of identity which is
worn with evident pride. Although not prestigious, it is not overtly stigma-
tized either and Gibraltarians generally look on it with a certain fondness
and see it is a linguistic expression of their unique cultural heritage (Keller-
mann 2001: 134–5; Fernández Martı́n 2003: 190–1; Levey 2006: 725). The
new generation of Gibraltarians is increasingly competent in two languages
and has a degree of language confidence that many of their parents and
grandparents lacked. If they use ‘Spanglish’ it does not necessarily imply
language weakness and ‘Yanitadas’, as they are known locally, are often
expressions of multilingual word play. Thus, when Gibraltarians say es un
cachofinger ‘it’s a joke’, they are consciously playing with different languages
and dialects.15 There is also a recent trend to add the English suffix -tion
to Spanish words for deliberately absurd effect. This may take place when
13 I would like to thank all those who took part in my survey. I am particularly grateful to Levi Attius
and Albert Borrell for their very helpful comments and clarifications.
14 The use of aliquindoi is not confined to Gibraltar. Estar al liquindoi or al liqui are recorded in Pedro
Payán’s (1991) popular lexicon of Cádiz. The expression is also used in Málaga and can be found in
Antonio del Pozo’s (2005) and Juan Cepas’ (2010) respective lexicons. Pozo suggests that it comes
from the English ‘look and do it’.
15 The correct Spanish word is cachondeo, but as Andalusians tend to omit intervocalic <d>, the final
element is hypercorrected to dedo and then translated as ‘finger’ for comic effect.
66 david levey
the speaker cannot call to mind the English word or when there is no easy
equivalent (e.g. armondigations ‘meat balls’ [Sp = albondigas]; asergation de
Spanish torti ‘chard pie’ [Sp = torta de acelgas]).
5 Conclusion
At present, it is difficult to talk about a homogeneous and uniform Gibral-
tarian speech community. Education, social class, ethnic background and
particularly the age of the speaker condition language choice and pro-
ficiency (cf. Levey 2008a). That children speak differently from parents
and grandparents in any speech community is of course usual, but social
and political events as well as educational policies have combined and
contributed to widening the generation gap in the case of Gibraltar. The
closing of the frontier from 1969 to 1982 was to leave its mark on the speech
community and served as a catalyst for language change.
Although most Gibraltarians can communicate, to varying degrees, in
two or more languages, it would be wrong to assume that everyone in
Gibraltar is multilingual. There have always been an important number
of monolingual UK residents in Gibraltar and monolingual Spanish day
workers have been crossing over the border for centuries. English and
Spanish have always cohabited in Gibraltar although the community’s true
bilingual potential has never been exploited. English is the only official
language on the Rock and given the contentious question of sovereignty,
it seems inconceivable that Spanish could ever be given any official status.
English has gained ground and today it is difficult to find young Gibral-
tarians who are not reasonably fluent. However the question is whether
or not this has taken place at the expense of Spanish. Although colloquial
Spanish variants are widely spoken in homes and on the streets, concerns
have been voiced in certain sectors that the levels of formal and written
Spanish have declined in recent years. Further research is necessary to gauge
whether this is indeed the case. Spanish is not obligatory in schools but stu-
dents can choose it as a second language or alternatively French, Russian,
Portuguese or Italian. Perhaps not surprisingly, of all the GCSE and A level
subjects available, Spanish is one of the most popular.16 English as a second
language is of course not an option in Gibraltar, which follows the UK
16 Official government statistics reveal that in 2013, 405 Gibraltarian schoolchildren took Spanish
GCSE with a pass rate of 88%. Of these passes 68% obtained A or A∗ grades. All but one of the
167 students who took Spanish A level passed, with 18% obtaining A or A∗ (Abstract of Statistics
2013: 41–4).
Gibraltar English 67
National Curriculum, but it is interesting to note that English Literature
is the third most popular A level subject after Spanish and Psychology.
It is important to point out that at present, Gibraltar does not have its
own university and so practically all nationals who want to further their
studies and obtain a university degree have to do so in the UK.17 Despite
proximity, there are no Gibraltarians currently studying in Spain, except
those sent there on Erasmus exchange programmes from UK universities.
The reason for this partially lies in the incompatibility of the respective
education systems. Despite the Bologna Agreement, recognizing UK qual-
ifications is not automatic in Spain and can be a lengthy and complicated
process.18 This inevitably means that it is difficult for Gibraltarians to attain
a high academic knowledge of Spanish. While their Spanish language and
vocabulary are sufficient for dealing with everyday situations, they may be
lacking in more formal or complex contexts.
In an interesting recent development, the Spanish government has now
established an Instituto Cervantes in Gibraltar.19 This initiative was met with
mistrust by certain sectors of the local population who see it as a Spanish
Trojan Horse and have questioned why the Instituto Cervantes, ‘a public
institution which promotes Spanish language and culture throughout the
world’, would want to set up shop in a place where Spanish is apparently
already widely spoken. The Instituto finally opened its doors in spring 2011
and according to the Spanish news agency Europa Press (5 May 2011) it had
150 enrolments divided into 20 groups before courses had even started.
Gibraltar enjoys a love–hate relationship with its Spanish neighbours.
Periodically there is cross-border tension for one of a number of reasons:
disputes over fishing rights, accusations of encroachment into territorial
waters, protests over planned royal visits, fears over bilateral talks between
UK and Spanish governments in which the future of Gibraltar may be
discussed. Gibraltarians will complain about the exasperatingly long queues
at the Spanish border posts, which they see as deliberate provocation, while
the Spanish authorities will justify the need to control contraband and
illegal immigration. The press on both sides of the border add fuel to the
fire and the question of sovereignty will inevitably be raised once again.
But then things die down and normality returns. Both neighbours have
learnt to live with the situation and with each other. As long as the sensitive
issues are avoided they get on fine and have much in common. National
17 In 2013, there were 247 student enrolments at UK universities (Abstract of Statistics 2013: 46).
18 At present the Spanish Ministry of Education has a huge backlog of unresolved applications from
European nationals, many of whom have been waiting for up to two years for a resolution.
19 This was one of the five resolutions of the trilateral Córdoba Agreement of 18 September 2006.
68 david levey
sentiment does not preclude Gibraltarians from chatting in Spanish and
enjoying Andalusian food. Although some will follow the BBC on cable
television, others will watch popular Spanish programmes and read Spanish
magazines and newspapers. There is no contradiction or incongruence, as
far as they are concerned, in feeling Gibraltarian and British while enjoying
their neighbours’ language, customs and culture. This has always happened
and there is no reason to suspect that it will change substantially in the
near future. The Gibraltarian view is clearly summed up by Luis Montiel,
the former Minister for Employment.
The privilege of the Gibraltarian is to live two cultures, two worlds: the
Anglo-Saxon culture and the Spanish culture. We like the good things of
both countries. So we live two cultures and enjoy the best of each. We reject
the worst of one and the worst of the other. But we choose what we want.
That’s the privilege of the Gibraltarian.20
References
Abstract of Statistics 2013. Government of Gibraltar.
Ballantine, S. 1983. A study of the effects of English-medium education on ini-
tially monoglot Spanish-speaking Gibraltarian children. MA dissertation,
University of Valencia.
2000. English and Spanish in Gibraltar: development and characteristics of two
languages in a bilingual community. Gibraltar Heritage Journal 7: 115–24.
Cal Varela, M. 2001. Algunos aspectos sociolingüı́sticos del inglés gibraltareño: análisis
cuantitativo de tres variables en el nivel fónico. Santiago de Compostela: Uni-
versidad de Santiago de Compostela servicio de publicaciones.
Cavilla, M. 1990. Diccionario yanito, 2nd edition. Gibraltar: MedSUN.
Census of Gibraltar 2001. Government of Gibraltar.
Cepas González, Juan. 2010. Vocabulario popular malagueño. Malaga: Arguval.
Enriles, J. M. 1992. The vowel system of Gibraltarian English. MA dissertation,
University College London.
Errico, E. 1997. Gibraltar: a hybrid of language and culture. PhD dissertation,
University of Bologna.
Fernández Martı́n, C. 2003. An Approach to Language Attitudes in Gibraltar.
Madrid: Umi-ProQuest information on Learning.
Garcı́a Martı́n, J. M. 1996. Materiales para el estudio del español de Gibraltar:
aproximación sociolingüı́stica al léxico español de los estudiantes de enseñanza
secundaria. Cádiz: Servicio de Publicaciones de la Universidad de Cádiz.
1997. El español en Gibraltar: panorama general. Demófilo. Revista de cultura
tradicional de Andalucı́a 22: 141–54.
20 The quote is from an interview with Luis Montiel in Spanish, which is published in Oda-Ángel
(1998: 45). My English translation originally appeared in Levey (2011: 76).
Gibraltar English 69
Howes, H. W. 1991. The Gibraltarian: The Origin and Development of the People
of Gibraltar, 3rd edition. Gibraltar: MedSUN.
Kellermann, A. 2001. A New New English: Language, Politics and Identity in Gibral-
tar. Norderstedt: Books on Demand.
Kramer, J. 1986. English and Spanish in Gibraltar. Hamburg: Helmut Buske Verlag.
Levey, D. 2006. Yanito. In K. Brown, ed., Encyclopedia of Language and Linguistics,
2nd edition, vol. 13. Oxford: Elsevier, 724–5.
2008a. Language Change and Variation in Gibraltar. Amsterdam and Philadel-
phia: John Benjamins.
2008b. The changing face of Gibraltarian English: TH-fronting on the Rock. In
Ewa Waniek-Klimczak, ed., Issues in Accents of English Language Change and
Variation in Gibraltar. Newcastle: Cambridge Scholars Publishing, 91–101.
2011. National identity and allegiance in Gibraltar. In C. McGlynn, A. Mycock
and J. W. McAuley, eds., Britishness, Identity and Citizenship: The View from
Abroad. Bern: Peter Lang, 73–93.
Lipski, J. M. 1986. Sobre el biliongüismo anglo-hispánico en Gibraltar. Neuphilol-
ogische Mitteilungen 87(3): 414–27.
López de Ayala, I. 1782. Historia de Gibraltar. Madrid: Antonio de Sancha.
Montero Sánchez, Sebastián. 2010. El habla del Campo de Gibraltar. Cádiz:
Quorum.
Moyer, M. 1993. Analysis of code-switching in Gibraltar. PhD dissertation, Uni-
versity of Barcelona.
Oda-Ángel, F. 1998. Gibraltar: la herencia oblicua: aproximación sociológica al
contencioso. Cádiz: Servicio de Publicaciones de la Diputación de Cádiz.
Payán Sotomayor, Pedro. 1991. El habla de Cádiz. Cádiz: Quorum.
Pozo, Antonio del. 2005. El habla de Málaga. Málaga: Miramar.
Valle Gálvez, A. del and González Garcı́a, I., eds. 2004. Gibraltar, 300 años. Cádiz:
Servicio de Publicaciones Universidad de Cádiz.
Vallejo, T. 2001. The Yanito Dictionary. Gibraltar: Panorama Publishing.
Wells, J. C. 1982. Accents of English, 3 volumes. Cambridge University Press.
West, M. 1956. Bilingualism in Gibraltar. Overseas Education 27: 148–53.
ch a p ter 4
1 Introduction
Ireland is often considered to be a homogeneous bilingual country, and
before the influx of foreigners in the boom years of the Celtic Tiger the
Irish themselves may have regarded their country as a monocultural society.
However, even though still widely disregarded in their cultural distinctive-
ness, the Irish Travellers have stood out as a separate community in Irish
society for several centuries. The Irish Travellers are a native Irish com-
munity with a nomadic background, who naturally share a lot of history
with settled Irish people, but culturally, religiously and linguistically they
have preserved their own identity. According to researchers and the Irish
Travellers themselves, there is no or very little genetic connection with
other European nomadic or Gypsy groups, even though they may share
with them many traditions and values, such as the preference of self-
employment, birth, marriage and burial customs, and values concerning
morality, taboos and purity (Freese 1980: 53–63).
The Irish Travellers are recognized as an ethnic group in Northern
Ireland and in the UK. In the Republic of Ireland their legal status has
been widely discussed but remains insecure.1 According to the Census of
2011, there are 29,573 Irish Travellers living and travelling in the Republic
of Ireland, which accounts for 0.6 per cent of the overall population in
Ireland.2 There are also Irish Travellers living in Britain, Australia and the
USA.
Linguistically, Irish Travellers differ from the settled community in a
twofold way. Firstly, their in-group code ‘Shelta’ (also known as ‘Gammon’
or ‘Cant’), a distinctive communicative tool used in specific, Traveller-
related contexts, provides the possibility to have private conversations in
situations where settled people are present, such as trade and business
70
Irish Traveller English 71
situations, contexts where warnings need to be exchanged or when talk-
ing about taboo topics. Morphosyntactically, Shelta is a mixture of Irish
English grammar and the Travellers’ own lexicon, a majority of which is
derived from Irish Gaelic and disguised in various ways by means of trans-
position (deliberate switching around of consonants, insertion and deletion
of syllables, etc.; see Hickey 2007b: 382 for an overview, Ó hAodha 2002
for a detailed description of the Shelta lexicon), while a smaller amount is
of unknown, though possibly very old origin. The combination of Shelta
lexicon with Irish English grammar allows Travellers to speak privately
without raising suspicion (for more information on Shelta: Binchy 1994,
1995, 2002; Browne 2002; Cauley 2006; Grant 1994; Hancock 1973, 1984,
1986; Macalister 1937; Nı́ Shuinéar 2002; Ó hAodha 2002).
Besides Shelta, also the Irish Travellers’ variety of English distin-
guishes Travellers from settled speakers and general Irish English. However,
Traveller English has not yet been researched as a variety of its own and
therefore the term ‘Traveller English’ is not yet commonly established. The
linguistic analysis of the variety in this contribution will be based on a
modest corpus of 40,000 words (Rieder, unpublished data). The corpus
stems from a two-year ethnographic project carried out among the Irish
Traveller community in the West of Ireland and consists of seven audio-
recorded, semi-structured group interviews of about 40 minutes each. Both
men and women were interviewed; the age group ranged from 18 to 65 and
the participants came from varied socioeconomic backgrounds.
In what follows, the phonological, morphosyntactic and lexical charac-
teristics of Traveller English will be analysed in depth, after a brief account
of the Travellers’ sociolinguistic history.
4 CSO 2006.
74 maria rieder
christenings and other celebrations kept the contact with the extended
family alive, and were also the setting where partners were found and new
families planned (McDonagh 1994: 89). Travelling therefore contributed
to the reinforcement of family ties. Families are traditionally quite large
and the age profile is very young with 41 per cent of the community under
14 years of age. Life expectancy, in turn, is considerably lower than in the
settled community.5
Connected to social aspects, nomadism also had a cultural function.
By regularly meeting friends or family, traditions, folk wisdoms, lan-
guage, values and beliefs could be shared and kept alive (McDonagh
1994: 89). Most values and the moral code centre around the Roman
Catholic faith, in which a number of the Travellers’ own religious prac-
tices have been integrated. Some of these are older Catholic practices,
for example novenas, praying for special intentions such as illnesses and
relatives, faith in and visiting spiritual healers, old Irish superstitions and
omens of good and bad luck. Traveller women’s faith is very strong and
openly expressed by wearing religious jewellery and carrying religious items,
such as prayer books, prayer cards, saints’ images, holy water, oils and
ointments. Most Traveller men usually display their faith less strongly
than women. They participate in the sequence of sacraments but usually
attend mass only on special occasions. Family meetings at events such
as funerals, christenings and weddings usually draw a large number of
Travellers from all over the country and keep these values and beliefs
alive.
In conclusion, impeding the freedom of movement of Travellers is slowly
wiping out traditional core values. Nevertheless, the Traveller community is
trying to keep their cultural identity alive while at the same time expanding
their networks towards the settled community.
5 CSO 2006.
Irish Traveller English 75
a certain region, but are instead reproducing a peculiar dialect that exhibits
mixed dialectal characteristics.
Phonological, morphosyntactic, lexical and pragmatic features of Trav-
eller English differ slightly from ‘mainstream’ general Irish English of today
in two respects. Firstly, Traveller English has retained Irish English features
that would have been widespread decades ago among settled Irish people
and can be called archaic Irish English features, which are becoming reces-
sive in general Irish English. These features will be referred to as ‘Archaic
Irish English’ in the following sections (for a detailed list of Irish English
feature see Bliss 1979; Kallen 1994; Wells 1982), in contrast to the ‘General
Irish English’ spoken by settled people today, as collected in the ICE-
Ireland Project (Kallen and Kirk, 2008), which takes into consideration
the many different regional dialects of settled Irish English in the Republic
of Ireland. Irish English features have been described as resulting from
the influence of an Irish substrate, as well as from historical and dialectal
features of English settlers, that were retained due to a long period of lan-
guage shift from Irish to English, and lastly from ‘other, primarily internal,
principles of historical change and variation’ (Kallen 1997: 3ff.). Looking at
the historical and modern situation of Irish Travellers, a secluded life, lim-
ited contact with the wider Irish society and learning English mainly from
speakers based in rural areas may have preserved and reinforced Archaic
Irish English features.
Secondly, Traveller English is experienced as being different beyond
these Archaic Irish English features. To out-group speakers, Travellers are
instantly recognizable by their language and often difficult to understand
even for native Irish people for reasons that will be pointed out below.
This led to suggestions of another substratum that could underlie Traveller
English. Nı́ Shuinéar, for example, suggested a ‘Gammon underlay’ (1994:
58), i.e. a possible but lost Cant grammar, which still influences Traveller
English besides Archaic Irish English features (Ó Baoill 1994: 157). The
distinctive elements lie primarily in the phonology and prosody, particularly
the intonation of Traveller English, but are also found on all the other levels
of language. The theory of a Gammon substratum, however, has not yet
been proved historically.
Travellers themselves are well aware of their distinctive variety of English
and have described it as a ‘flat accent’ in contrast to settled Irish English
(Rieder, unpublished data), which again refers to phonology primarily. In
terms of attitudes, Travellers usually defend their variety strongly against
outsiders who might feel General Irish English to be superior to Traveller
English, and claim that they would always refuse to adapt their speech
76 maria rieder
in situations of contact with settled Irish people. In real-life situations,
however, one can perceive a degree of convergence towards General Irish
English, with a gradual appropriation of standard features also in in-group
situations (O’Sullivan 2008: 55).
3.1 Phonology
The influence of strong Archaic Irish English features, as well as other,
unknown developments, is most strongly felt in Traveller English phonol-
ogy. Particularly the greater tendency to mid-centralize kit, trap, strut
and unstressed vowels towards /ə/, the close mid or mid-central onsets of
many diphthongs, and rhoticity, pre-R breaking and pre-Schwa laxing pro-
cesses surrounding near, square, cure centring diphthongs distinguish
Traveller English from settled Irish English. Apart from these observa-
tions, however, single vowel realizations do not differ much from the Irish
English still heard in rural areas, and it may rather be for prosodic reasons,
i.e. intonation, rhythm etc. that Traveller English is perceived to be dif-
ferent. In what follows, peculiarities of Traveller English will be described
in detail, especially in comparison to features of the General Irish English
of the settled community (as described in Kallen 1994; Wells 1982; Hickey
2007a).
dress /e/̝
While kit words often become centralized in a [ə] or at least a [ʚ] sound,
words that would be in the dress category in RP, especially vowels followed
by a nasal such as den, Ben, then, but also in get, settled etc., often have a
raised [ ̝e] or kit [ɪ]. This is also a very common feature of General Irish
English.
Irish Traveller English 77
trap /æ/
The trap group displays a great deal of variation in Traveller English.
trap vowels are often more raised to [æ̝ ] and can approximate the open-
mid vowel [ɛ] in syllable-final position. Also the words many, any, which
most settled Irish realize as [æ] instead of the RP [ɛ] and which is seen as
a ‘striking Irishism’ (Wells 1982: 423), are mostly pronounced with a more
raised [æ̝ ] by Irish Travellers.
lot /ɒ/
As in General Irish English, LOT words frequently have the unrounded
variant [ɑ] in Traveller English, especially before nasal consonants (e.g.
long). Some words in this group are raised as far as to [ʌ], e.g. clock. In all
other instances it is usually [ɒ]: got.
strut /ɔ̈ /
Realizations of the strut vowels can be similar to the General Irish English
‘mid centralized back somewhat rounded vowel’ [ɔ̈ ] (Wells 1982: 422), an
intermediate between [ɔ] and [ʊ], as in bus [bɔ̈ s), summer [ˈsɔ̈ mə˞]. The
realization of [ɔ̈ ] can be found in some lexemes in Traveller English, but is
less pronounced than in General Irish English. Most realizations of strut
would either have [ɔ], which for some words may be influenced by the
spelling, e.g. come, done, other, but also pub, but etc. are pronounced with
an [ɔ]. Some words have [ɔ̈ ] or [ɑ], e.g. in husband, run, and many others
are pronounced with a foot [ʊ], e.g. spuds.
As in General Irish English, some words that would have an onset strut
vowel in General English can be realized with a kit [i] in Traveller English,
e.g. onion [ˈɪɲən].
foot /ʊ/
Some words in the strut group have not even been lowered to /ɔ/ or /ɔ̈ /,
but have a foot /ʊ/ vowel. This indicates that the foot–strut Split has
not entirely taken place and words such as spuds, cut, bucket have an /ʊ/,
resulting in some homophones with foot words, e.g. look and luck, which
can still be found in vernacular forms of General Irish English.
Many foot words have retained the historical /uː/, which is also still
present in General Irish English, a retention of the Middle English /oː/
which underwent the raising but not the shortening. Therefore, many
words of the foot group can be included in the mood group, such as
78 maria rieder
book, cook, crook (see also Wells 1982: 423). This feature is also present in
General Irish English, though it is becoming recessive.
A peculiarity of Traveller English absent in General Irish English is
that words with an onset <u> are often aspirated, e.g. us [hʊs], under
[hɔ̈ ndə˞].
Weak vowels
Similar to General Irish English, Traveller English uses schwa extensively,
especially in unstressed word-final syllables, where /i/ and /ə/ are often
merged, e.g. happy [ˈhæpə]. Words ending in -er either have an r-coloured
mid-central vowel: e.g. letter [ˈlɛtɚ] or, more commonly, the schwa
absorbed, e.g. better [ˈbɛtr]. These features are also found in General Irish
English, though especially the absorbed schwa would be more common
in Traveller English.
Words ending in -ow also commonly have a [ə], and can even be raised
to an [ɪ]: follow [ˈfɒlɪ], which can be lengthened: window [wɪndiː]. The
raising to [ɪ] or [iː] is peculiar to Traveller English and may not be found
in General Irish English.
The -ing suffix is mostly reduced to /-ɪn/ or /-ən/ in Traveller English,
while schwa absorption (Wells 1982: 434) is common for words ending in
dentals: putting [ˈpʊtn], sitting [ˈsɪtn]. Likewise the endings of morning
and evening are usually reduced to /ən/. O’Sullivan (2008: 34ff.) studied
the reduction of the -ing suffix by the example of doing and going in a
comparison of the Limerick Corpus of Irish English and her own corpora of
Traveller English and found that this feature occurred in almost 100 per cent
of all -ing forms used by Irish Travellers, in contrast to about 7 per cent by
General Irish English speakers.
To in all meanings is usually weak in Traveller English and has either
a schwa, [tə], or an unstressed front open vowel [ta]. Other, normally
stressed words such as I, what, when, and occasionally verbs like went
followed by a stressed preposition, are often used in their weak forms
with a schwa. Also the weak form of my [mi] is very commonly used.
Traveller English is very similar here to General Irish English, but again
there may be quantitative differences in regard to the articulation of weak
vowels and further quantitative research would be required for more precise
distinctions.
Unstressed prefixes of multisyllabic verbs are often not audible: remember
[ˈmembə], I decided [ai ˈsaidəd]. This may occasionally occur in General
Irish English in connected speech, but is used very noticeably and consis-
tently in Traveller English.
Irish Traveller English 79
3.1.2 Diphthongs
choice /aɪ/
The realization of choice diphthongs is typically shifted to price /aı/ as
in boy [baı], noise [naɪs], annoyed [aˈnaid] and is more advanced/fronted
than the General Irish English /bɑɪ/.
price /aɪ/
This diphthong is unremarkable in Traveller English. While General Irish
English tends to neutralize the opposition /ai/ and /ɔi/ by a low central
onset: Irish [ˈəɪrɪʃ] [ˈɔɪrɪʃ] [ˈɔ̈ ɪrɪʃ] [ˈʊɪrɪʃ], Traveller English usu-
ally shifts both diphthongs into the price direction with a slightly more
advanced/fronted onset: Traveller English Irish [ˈaɪriʃ], boy [baɪ].
mouth /oʊ/
mouth diphthongs have close/mid-back onsets: /oʊ/ /ɔʊ/ as in Traveller
English bouncer [ˈboʊnsɚ] [ˈbɔʊnsɚ], in contrast to a low central onset
in General Irish English.
start /a˞/
start words tend to be realized as /æ/ plus pre-R schwa: mark [ma˞k]
[mæə˞k] and may have a somewhat shorter vowel than General Irish English
for some members of the community. Travellers therefore seem not to have
appropriated the pre-fricative lengthening that was completed around the
end of the seventeenth century for RP (Wells 1982: 203ff.) and which
General Irish English seems to perform to a greater degree than Traveller
English.
80 maria rieder
nurse /ɜ˞/
The nurse merger is not completely carried through for Traveller English
and therefore displays a great variation. nurse words that were pronounced
as /ɪ/ or /ɛ/ before the merger was completed by the seventeenth century
(Wells 1982: 196), usually have an /ɜ˞/ realization, e.g. heard [hɜ˞d], bird
[bɜ˞d], in Traveller English, and may have a lengthened vowel as in Germany
[dʒɜːʳmənɪ]. Words that have an <i> are pronounced as [ɛ]: girl [gɛ˞l].
Words spelled with <o> tend towards an /ɔ/ sound: word [wɔ˞d], and
those spelled with a <u> usually have a centralised /ʌ/ or /ɔ/ vowel, e.g.
curb [kʌ̈ ˞b], turnip [tɔ̈ ˞nəp].
north /ɔ˞/
Like many other dialects north is merged with force in Traveller
English. Both vowel groups have a shorter vowel than General Irish English:
north [nɔ˞θ].
face /eː/
General Irish English face has not or only variably undergone the long-
mid diphthonging completed around 1800 (Wells 1982: 211), hence also
Travellers use mostly an /eː/ vowel for the face group, though more
consistently than settled Irish people would, e.g. today [tədeː], name [neːm].
Words ending in the /eɪ/ diphthong typically approximate fleece /iː/, as
in they [d̪ i:], or /ɛ/ in say [sɛ].
bath/palm /æː/
The vowels /aː/, /æː/, /æ/ may not be distinct in Traveller English, and gen-
erally the bath/palm vowel can be slightly more raised in Traveller English
Irish Traveller English 81
than in General Irish English and RP depending on the environment.
Therefore, words such as calm, balm would have a vowel approximating
the RP vowel /aː/, whereas man, Ann, tend towards /æː/. Father, which in
General Irish English is often pronounced as [fɔːðə˞], usually has a slightly
raised short [æ]. Likewise, the vowels in dance, advance and similar words
are shortened and would therefore fall into the trap category.
thought /ɔː/
The thought vowel is unremarkable in Traveller English, though it may
be a slightly shorter [ɔ] than in RP.
goat /oː/
Travellers have preserved the traditional use of the monophthong /oː/ for
/əʊ/, which is, similar to the face group, a sign of the absence of long-mid
diphthonging. This is also a feature of General Irish English, though it
would be more consistent in Traveller English. A similarly recessive feature
in General Irish English, but widespread in Traveller English, is that some
of the goat words have a second variant with a mouth [aʊ]: old [oːld] /
[aʊld], bold [boːld] / [baʊld], which has a jocular and non-literal meaning
(Wells 1982: 427). [aʊld] has a sentimental connotation when talking about
times long gone by or affectionately about other people. Other words have
as their only realization an approximation towards an /aʊ/ diphthong, e.g.
cold [kaʊld], told [taʊld], shoulder [ˈʃaʊldɚ]. This feature is recessive in
General Irish English of today (O’Sullivan 2008: 48) but very present in
Traveller English.
mood /uː/
The mood vowel is a very close, back long vowel in Traveller English.
As mentioned above, some RP foot words have a long /u:/ in Traveller
English: cook [kuːk], book [buːk]. This feature is becoming recessive in
General Irish English.
3.1.5 Consonants
As for vowel realizations, many consonantal features of General Irish
English vernacular that are already or are becoming recessive can still
be found extensively in Traveller English. A detailed comparative research
and analysis will be required for a clear picture as to the quantitative dif-
ference of usage between General Irish English and Traveller English of
the features summarized below. Those characteristics that may distinguish
Traveller English from General Irish English and may have different origins
82 maria rieder
are clearly marked and listed at the end of each subsection (examples from
Rieder, unpublished data):
3.2 Morphosyntax
Also in terms of morphosyntax, Traveller English displays strong vernacu-
lar and Archaic Irish English features, which, according to Forde’s (2005)
corpus-linguistic, lexico-grammatical analysis of modern Irish English
speakers, are slowly being abandoned by the settled Irish population. Irish
Travellers in turn, have held on to most features outlined in Forde and this
section will therefore align itself to his taxonomy, while also pointing out
some distinguishing characteristics of Traveller English.
3.2.1.3 Pronouns
The pronoun systems are fairly standard. Subject pronouns follow general
patterns of most standard varieties: I, you, he/she/it for singular, we for
the first-person plural, the second-person plural pronoun is distinguished
from the singular by ye, and they is the third-person plural pronoun. Object
pronouns are unremarkable.
Reflexive pronouns can differ slightly from standard varieties as even
the plural pronouns are usually composed of a possessive adjective plus the
singular of self: meself, yourself, hisself, herself, ourself, yerself, theirself.
4 Conclusion
Phonological, morphosyntactic, lexical and pragmatic features of Traveller
English have been shown to combine to create a variety of English that is
rich in unique cultural characteristics, while at the same time it displays a
great many Archaic Irish English features that are slowly being left behind
by the settled Irish population. In its distinctiveness from and similari-
ties to General Irish English, Traveller English reflects the positioning of
Irish Travellers with regard to the settled Irish population. Despite their
indigenously Irish origin, a secluded way of life separate from mainstream
society as well as strong family ties and distinctive cultural aspects have
characterized Irish Travellers for centuries, and perpetuated their variety of
English.
It needs to be mentioned though that Traveller English is not a homoge-
neous variety, and the degree to which the vernacular is spoken depends very
94 maria rieder
much on the level of education, accommodation and nature of networks of
the individual speaker. Until recently, the women in the community used
to be confined to their homes looking after a big family. This is now slowly
changing with women starting to look beyond the community boundaries
for work or free-time activities, which therefore opens up and loosens their
network ties. This development may have an impact on the use of vernac-
ular norms and may bring about language change. O’Sullivan’s (2008: 55)
study of communicative shifts in Travellers’ casual speech revealed a cer-
tain degree of accommodation towards General Irish English with regard to
several pragmatic and morphosyntactic features, such as subject–verb con-
cord, nonstandard negation, use of learn and teach, etc. However, Traveller
English phonological characteristics seem to be among the most resistant
to change. Pronunciation and intonation may also serve as a way to ‘dif-
ferentiate themselves favorably from the out-group in order to maintain a
positive social identity’ (O’Sullivan 2008: 14). After all, the Irish Travellers
are and perceive themselves as a separate cultural group. Years of denigra-
tion have led to a lot of opposition as well as the acquisition of a certain
pride, which may be symbolically expressed by linguistic separation and
the strong identification with their own variety of English.
References
Bakker, Peter and Maarten Mous, eds. 1994. Mixed Languages: 15 Case Studies in
Language Intertwining. Amsterdam: IFOTT/Foris.
Bhreathnach, Aoife. 2007. Confusing origins and histories: the case of Irish Trav-
ellers. Irish Journal of Anthropology 10(1): 30–5.
Binchy, Alice. 1994. Travellers’ language: a sociolinguistic perspective. In McCann
et al., eds., 134–54.
1995. Travellers language. In A Heritage Ahead: Cultural Action and Travellers.
Dublin: Pavee Point Publications, 87–93.
2002. ‘Travellers’ use of Shelta. In Kirk and Ó Baoill, eds., 11–16.
Bliss, Alan. 1979. Spoken English in Ireland, 1600–1740. Dublin: Dolmen Press.
Britain, David, ed. 2007. Language in the British Isles. Cambridge University Press.
Browne, Marian. 2002. The syntactic structure of present-day Cant. In Kirk and
Ó Baoill, eds., 65–78.
Cauley, William. 2006. Canting with Cauley: A Glossary of Travellers’ Cant/
Gammon. Dublin: A. & A. Farmar.
Clancy, Brian. 2011. Complementary perspectives on hedging behaviour in family
discourse. International Journal of Corpus Linguistics 16(3): 371–90.
Forde, Kieran. 2005. A corpus-based lexico-grammatical analysis of the Limerick
Corpus of Irish English (L-CIrish English): some nominal and verbal features.
Unpublished MA thesis, University of Limerick.
Irish Traveller English 95
Freese, Christoph. 1980. Zur Geschichte und Gegenwart der Zigeuner und Land-
fahrer in Deutschland: Versuch einer subkulturellen Erklärung. University of
Nürnberg-Erlangen.
Gmelch, Sharon and George Gmelch. 1976. The emergence of an ethnic group:
the Irish Tinkers. Anthropological Quarterly 49(4): 225–38.
Gmelch, Sharon, Pat Langan and George Gmelch. 1975. Tinkers and Travellers.
Dublin: O’Brien Press.
Grant, Anthony P. 1994. Shelta: the secret language of Irish Travellers viewed as a
mixed language. In Bakker and Mous, eds., 123–50.
Hancock, Ian F. 1973. Shelta, a problem of classification. Journal of the Gypsy Lore
Society 52: 80–6.
1984. Shelta and Polari. In Trudgill, ed., 384–403.
1986. The cryptolectal speech of the American roads: Traveler Cant and Amer-
ican Angloromani. American Speech 61(3): 206–20.
Harper, Jared. 1973. Irish Traveler Cant in its social setting. Southern Folklore
Quarterly 37: 101–14.
Helleiner, Jane Leslie. 2000. Irish Travellers: Racism and the Politics of Culture.
University of Toronto Press.
Hickey, Raymond. 2007a. Southern Irish English. In Britain, ed., 135–51.
2007b. Irish English: History and Present-Day Forms. Cambridge University
Press.
Kallen, Jeffrey L. 1994. English in Ireland. In R. W. Burchfield, ed., The Cambridge
History of the English Language, vol. 5. Cambridge University Press, 148–96.
1997. Focus on Ireland. Amsterdam and Philadelphia: John Benjamins.
2005. Internal and external factors in phonological convergence: the case of
English /t/ lenition. In Peter Auer, Frans Hinskens and Paul Kerswill,
eds., Dialect Change: Convergence and Divergence in European Languages.
Cambridge University Press, 51–80.
Kallen, J. L. and J. M. Kirk. 2008. IC-Ireland: A User’s Guide. Belfast: Cló Ollscoil
na Banrı́ona.
Kirk, John M. and Dónall Ó Baoill, eds. 2002. Travellers and Their Language.
Belfast: Cló Ollscoil na Banrı́ona.
Macalister, Stewart R. A. 1937. The Secret Languages of Ireland: With Special Refer-
ence to the Origin and Nature of the Shelta Language. Cambridge University
Press.
MacNeill, Eoin. 1937. Phases of Irish History. Dublin: M. H. Gill & Sons.
McCann, May, Séamas Ó Sı́ocháin and Joseph Ruane, eds. 1994. Irish Travellers:
Culture and Ethnicity. Queen’s University of Belfast.
McDonagh, Michael. 1994. Nomadism in Irish Travellers’ identity. In McCann
et al., eds., 95–109.
Nı́ Shuinéar, Sinéad. 1994. Irish Travellers, ethnicity and the origin question. In
McCann et al., eds., 54–77.
2002. The curious case of Shelta. In Kirk and Ó Baoill, eds., 21–41.
96 maria rieder
Ó hAodha, Mı́cheál. 2002. Travellers’ language: some Irish language perspectives.
In Kirk and Ó Baoill, eds., 47–63.
Ó Baoill, Dónall P. 1994. Travellers’ Cant – language or register? In McCann
et al., eds., 155–67.
O’Sullivan, Joan. 2008. ‘Talkin’ Different’: Linguistic Diversity and the Irish Trav-
eller Minority. Newcastle upon Tyne: Cambridge Scholars.
Trudgill, Peter, ed. 1984. Language in the British Isles. Cambridge University Press.
Trudgill, Peter and Jean Hannah. 2002. International English: A Guide to the
Varieties of Standard English. London: Arnold.
Wells, J. C. 1982. Accents of English, vol. 2. Cambridge University Press.
part ii
The Americas
c h a p ter 5
1 Introduction
American Indian English (hereafter AIE) is best described as a constel-
lation of several varieties of English rather than a single variety. Because
the indigenous peoples of North America are not an undifferentiated,
monolithic group, it only follows that the English they speak is also not a
singular variety. But there is a relationship among various varieties of AIE,
both linguistically and socially, and it is this relationship that I focus on
in this chapter. AIE is spoken throughout the United States and Canada
primarily by people of indigenous heritage, and the speakers may number
as many as four million.1 However, the exact delineation of who is and
who isn’t American Indian2 is not merely a social matter or based on a box
checked on the census, but rather, it is a long-fought political issue. As
Louise Erdrich (Chippewa) writes in her novel The Round House:
You can’t tell if a person is an Indian from a set of fingerprints. You can’t
tell from a name . . . You can’t tell from a picture . . . From the government’s
point of view, the only way you can tell an Indian is an Indian is to look at
that person’s history. There must be ancestors from way back who signed
some document or were recorded as Indians by the US government, someone
identified as a member of a tribe. And then after that you have to look at that
person’s blood quantum, how much Indian blood they’ve got that belongs
to one tribe . . . On the other hand, Indians know other Indians without the
1 Based on the 2010 US Census (Humes, Jones, and Ramirez 2011) and the 2006 Canadian Census
(www.statcan.gc.ca/pub/89–645-x/2010001/count-pop-denombrement-eng.htm); of the 4 million,
about 685,000 also identified as Hispanic or Latino on the US Census, meaning that they are
probably not in the population in question because they most likely originate from Central or South
America or the Caribbean.
2 A note on terms: The “correct” term to use when referring to the indigenous peoples above the Rio
Grande is a source of contention, usually between “Native American” and “American Indian.” For
this chapter, I use the two terms interchangeably. I use “First Nations” or “aboriginal peoples” when
discussing the indigenous people of Canada (the term “First Nations” does not include the Inuit or
Métis), and the name of the particular tribe when appropriate.
99
100 elizabeth l. coggshall
need for a federal pedigree, and this knowledge . . . has nothing to do with
the government. (2012: 29–30)
3 Leechman and Hall (1955) suggest, based on historical documents, that features such as these were
actually once common and are indications of an earlier “American Indian Pidgin English.” See also
Miller (1967).
American Indian English 101
pronouns (he/she, him/her, his/hers) are rarely found in other varieties of
English.
In what follows, I first demonstrate the status of AIE as a lesser-known
variety of English, as well as how AIE came about both as a variety and as
a code of indigenous identity. The challenge of my chapter is to provide
a unified description of these disparate varieties. To this end, I bring
together the various strands of research on AIE, showing commonalities
among different varieties of AIE as well as differences. Finally, I show the
need for greater research on AIE and general avenues of possible future
research.
4 Besides the English, the Spanish, French, Russians, and Dutch also had a presence in the areas under
question.
American Indian English 103
After 1776, Indian affairs became the burden of the newly formed US
government, which assigned this task to the War Department, though it
moved to the Departments of the Interior in 1849. Today, the department in
charge of Indian policy is the Bureau of Indian Affairs (BIA). The wars and
epidemics continued, and the US government signed several treaties with
different tribes. American Indians were not US citizens but instead existed
in a no-man’s land between sovereign nations and regular citizens; they
finally became full citizens in 1924. These treaties became important later
to determine who counts as American Indian for government purposes, as
explained above.
In 1830, Congress passed the Indian Removal Act, allowing for the forced
relocation of the eastern tribes to “Indian Territory” in the west. Most
notable of these removals was the Trail of Tears in 1838, where approximately
4,000 Cherokees died on a forced march to present-day Oklahoma from the
Carolinas and Georgia. The reservation system that is still in existence in the
United States started in the 1850s. The reservation system was always seen
as suboptimal by the federal government, for whom the ultimate goal was
the “civilization” (i.e. assimilation) of Native Americans. In keeping with
this goal, in 1877, Congress appropriated the first funds to create schools
for Native American children (Calloway 2004: 345). “Like the children of
European immigrants, Indian children were expected to jettison their old
ways and language and become English-speaking ‘Americans’” (Calloway
2004: 344). Boarding schools were typically placed far from the home
reservations of the children, who were (often forcibly) removed from their
families. In 1887, an English-only policy was instituted (Fear 1980: 14), and
speaking a heritage language was grounds for (often corporal) punishment.
Most boarding schools separated children from the same tribe as a strategy
to force children’s use of English. This language education was part of
a larger plan of transformation to “remake them as individual citizens,
not tribal members” (Calloway 2004: 347). The boarding schools had an
impact on Native American language far outside their walls: graduates of
such institutions feared teaching their children anything other than English
lest these children would have to endure the same humiliation and torture
that their parents had endured.
The federal government did much more in the interest of “civilizing”
the “savages,” though many of the government officials in charge of Indian
policy were convinced they were doing what was best for a beleaguered
people. In 1887, in order to show Native Americans in the US the impor-
tance of private ownership of land, the Allotment Act was passed, which
divided many reservations into parcels of land that the American Indian
104 elizabeth l. coggshall
“owners” could retain or, in many cases, sell to non-Indians. In 1934, the
Indian Reorganization Act was passed that set up self-governing bodies on
many reservations. Soon after, however, Indian policy in the US turned
toward the termination of reservations and the special status of American
Indians. To this end, the Indian Claims Commission was set up in 1946 to
buy back land from tribes, and in 1956, a system of relocation was set up
to move Native Americans from rural reservations to urban centers. There
is an ironic consequence of the policies discussed thus far: the goal of the
termination and relocation policies, along with the English-only boarding
schools, had been to end the “Indian problem” through assimilation. But
instead of individuals’ ties to Indian identity being weakened, a panethnic
identity was created by these pressures, and the population of people claim-
ing American Indian heritage grew rather than shrank (Lopez and Espiritu
1990; Nagel 1996; Nagel and Snipp 1993). In 1968, the Indian Civil Rights
Act was passed, and in 1975, Richard Nixon declared that the federal gov-
ernment would now have a policy of self-determination, allowing tribes to
do what they think is best for themselves, with certain restrictions based on
the fact that they are not sovereign nations but part of the United States.
The Canadian government’s history with aboriginal peoples followed
a similar trajectory (see Dickason 2006), with most likely similar results
as far as their use of English is concerned. Waves of epidemics and wars
swept through the First Nations upon contact (and for hundreds of years),
decimating much of the indigenous population. Much of the initial contact
was made by French fur traders in search of beaver pelts, rather than by
English-speaking settlers moving west in search of land as was the case
in the US. A policy of assimilation gained momentum in the nineteenth
century, mostly based around what is called the Indian Act, passed in 1876.
Reserves were created for First Nations to live on, and then they were broken
apart for individual allotments. The Indian Act created Indian residential
schools to speed the acquisition of civilization, and used force to push
an English-only environment. Policies of denying religious and linguistic
freedom, as well as rights to use of land for hunting were maintained
into the twentieth century; aboriginal peoples were not given the right to
vote until 1956. In the 1969, the Minister of Indian Affairs suggested the
abolition of the Indian Act and thus the special status held by First Nations,
but the relevant legislation has never been passed. Since then, First Nations
have been continuing the struggle for civil rights and representation in the
government.
The history of contact in the United States and Canada, as well as the
sorts of policies the governments of the two countries enacted, had a large
American Indian English 105
impact on language use in Native American communities. Most obvious is
the extinction, or near extinction, of the diverse languages spoken before
1492. The adoption of English may seem a straightforward reaction to
the events discussed above, but not all American Indians have become
monolingual English speakers, and those who are English speakers do not
necessarily speak a standard version of American or Canadian English.
The assimilation that these policies were meant to accomplish has not
been successful, and, in the last few decades, has been replaced by pro-
multicultural policies and attitudes in the general public. This lack of
assimilation allowed for the creation of separate varieties of English, ones
that often replaced heritage languages as a locus for speakers to express
American Indian identity.
AIE fits most of the criteria of a lesser-known variety (LKVE) of English
as put forth by Schreier et al. (2010: 4) in their introduction. AIE is most
definitely “a variety that is lesser known, at least to the outside world”
(Schreier et al. 2010: 4) since much of present-day American Indian culture
is not understood or known to non-Indians; further, AIE “has not received
much attention in the literature” (Schreier et al. 2010: 4), especially in the
last two decades, happening to coincide with the publication of American
Indian English by William Leap in 1993. Before that date, in the 1970s and
early 1980s, the vast majority of the work was done by a small number of
scholars: Leap, H. Guillermo Bartelt, and Susan Penfield Jasper. Of these
three scholars, only Bartelt continues to work on AIE.
AIE fits many of the more specific criteria of an LKVE as well. AIE is
the first language of most of its speakers, and sometimes the only code
they know. AIE is considered its own variety, separate from those spoken
by non-Indians, the variety spoken in school, and other varieties spoken
by AIE speakers (Leap 1993). AIE is spoken by people in “stable communi-
ties,” many of these communities being several hundred if not thousands
of years old. AIE is spoken by minorities, American Indians and First
Nations being some of the most marginalized people, both socially and
geographically, in the US and Canada. They are among the poorest peo-
ple, they are ethnically separate, and for generations they have mostly lived
in reservations or reserves far outside population centers. They have also
resisted outsiders’ attempts to assimilate them into the mainstream, such
as allotment, termination, and relocation. AIE was not really “transmit-
ted by settler communities or adopted by newly-formed social groups”
(Schreier et al. 2010: 4) in that many varieties are not directly derived from
British English during colonial times but have more of a connection with
American and Canadian English. Though there is some controversy over
106 elizabeth l. coggshall
the exact genesis of AIE, no one would deny that language contact had a
hand in its formation. AIE is also an important identity marker for many
of its speakers. It is the only Native American-related code available to
them because of the lack of an indigenous language in the community
either through the eradication or slow depletion of the language. AIE is
also a way for American Indians to identify each other, and some even
claim they can tell what tribe someone is from based on their English.
AIE is not, however, endangered. I argue that as long as American Indi-
ans exist, there will be AIE. I do so for two reasons. First, the isolation
described above, especially the geographic isolation, promotes the existence
of a separate code for American Indians. Second, since the rise of the Red
Power movement in the 1960s/70s, Native American identity has earned
cachet, and so any identity markers associated with it will most likely be
maintained.
While AIE is a lesser-known variety of English, some studies of AIE
in the US (but none, to my knowledge, in Canada) have been done, but
much of the work that was done was based on data gathered in the 1960s
and 1970s. The goal of much of this work was to ascertain the provenance
of these features, be it interference from the indigenous language (e.g.
Wolfram 1980), influence from other nonstandard varieties (e.g. Dillard
1972), fossilized features of second language learning (e.g. Leap 1974), or
remnants from a past English-lexifier creole or pidgin (e.g. Craig 1991).
Further, most of the research was funded by institutions, like the National
Institute of Education in the case of Wolfram, Christian, Leap, and Potter
(1979), that were primarily concerned with the educational application of
linguistic research.
3.1 Phonology
The phonology of AIE has not been studied as extensively as the syn-
tax. Many of the unique/interesting features of the phonology are related
to ancestral language traditions. However, recent research suggests that
some features, namely the extensive use of glottal stops (Rowicka 2005)
and a syllable-timed prosodic rhythm (Coggshall 2008), appear to have
spread from some varieties of AIE to other varieties. These findings lead to
108 elizabeth l. coggshall
some interesting questions that should be addressed in future research, as
described in the conclusion below.
3.1.1 Consonants
A noteworthy aspect of consonants in AIE involves differences between
these varieties and non-Indian varieties of English in consonant inventory.
Below, two such examples are discussed that occur in many varieties of AIE:
TH-stopping, where the interdental fricatives are lost from the inventory,
and glottal stops, an addition of a sound. There are more such examples
of differences in consonant inventory. For instance, a loss of consonant
distinctions can be found in Navajo English, Pima English, and Tsimshian
English, with the interdental fricatives /f, v/ replaced by the voiced bil-
abial stop [b] (Cook and Sharp 1966: 24; Nelson- Barber 1982: 125; Mulder
1982: 100). Further, the set of alveolar and palato-alveolar fricatives has
undergone changes in several varieties. In Tsimshian English and Pima
English, /z, ʃ, ʒ/ are realized as [s] (Cook and Sharp 1966: 24; Nelson-
Barber 1982: 124), and Kotzebue English lacks a distinction between these
four phones, using them interchangeably (Vandergriff 1982: 138–9). Some
varieties have consonants that are not found in non-Indian varieties of
English. For instance, Quinault English has a labialized voiceless velar frica-
tive [xw ] in words beginning with wh- (Rowicka 2005: 307), and Kotzebue
English has a voiced velar fricative [ɣ] that replaces /g/ (Vandergriff 1982:
142).
The other process involving consonants that is common to many vari-
eties of AIE is consonant cluster reduction (CCR). CCR occurs usually,
but not always, at the end of words. The reduction refers to the deletion of
whole segments, like han’ for hand or des’ for desk. CCR is noted in almost
every description of varieties of AIE. CCR has been documented in the
Quinault (Rowicka 2005), Isletan (Leap 1977a), Mohave (Penfield 1977),
Hopi (Penfield 1977), Navajo (Penfield 1977), Cheyenne (Alford 1974), San
Juan (Wolfram et al. 1979; Wolfram 1980), Laguna (Wolfram et al. 1979;
Wolfram 1980), Lakota (Flanigan 1984, 1985), Lumbee (Torbert 2001), and
Brandywine (Gilbert 1986) varieties of AIE, as well as many non-Indian
varieties of English.
3.1.1.1 TH-stopping
TH-stopping, that is, a stop where a standard variety would have one of
the interdental fricatives /ð, θ/, is one of the most commonly cited features
of varieties of AIE. The stops in question are dental or, less commonly,
alveolar. For instance, in Tsmshian English the is pronounced [də] with
American Indian English 109
the voiced alveolar stop, and northland is pronounced [nortlənd] with the
voiceless alveolar stop (Mulder 1982: 100). The fricatives are not always
replaced by stops; in Western Apache English and Navajo English, while
/ð/ is always the dental stop [d̪ ] and word-initial /θ/ becomes [d̪ ], word-final
and word-medial /θ/ become [f] (Bartelt 1986: 692). TH-stopping has also
been documented in the Brandywine (Gilbert 1986), Kotzebue (Vandergriff
1982), Alabama-Coushatta (Hoffer 1982), Hopi (Penfield 1977), and Pima
(Miller 1977) varieties of AIE.
3.1.2 Vowels
Unlike the patterns noted with consonants, there are few commonali-
ties between different varieties of AIE in regard to vowels. While AIE
vowels are definitely different from those in other varieties of English,
they also vary from variety to variety. In some varieties, such as Eastern
Cherokee English (Anderson 1999, Coggshall 2006) and Lumbee English
(Schilling-Estes 2004; Coggshall 2006), the vowels used by AIE speakers
and speakers of non-Indian English are very similar and show little influ-
ence from the substrate language. For instance, Eastern Cherokee English
has an extremely fronted goose vowel in line with the regional standard,
Appalachian English, even though the Eastern Cherokee language has a
fully backed goose vowel (Coggshall 2006: 59). Other varieties of AIE
show more influence from the substrate. For instance, Quinault English
110 elizabeth l. coggshall
(Rowicka 2005: 308–9) and Navajo English (Cook and Sharp 1966: 22–3;
Bartelt 1986: 692) both lack glides on face and goat vowels due to a sim-
ilar lack in Quinault and Navajo. Other varieties such as Navajo English
(Cook and Sharp 1966: 22–3) and Isletan English (Leap 1993: 46) lose
distinctions between tense and lax vowels.
3.1.3 Prosody
The suprasegmental phonology of AIE is one of the most marked aspects
of the varieties, as well as one of the least studied. Leap hypothesizes
that suprasegmental “features contribute substantially to contrasts with
standard English – and to contrasts that distinguish Indian English codes
from different tribal communities” (Leap 1993: 50). These differences lead
to an impression that AIE speakers “talk in more subdued tones, show little
expression or emotion in their voices, speak in a monotone, or speak in
sing song voice” (Leap 1993: 52). Such impressions can be linked at least
in part to the prosodic features explored below: a smaller pitch range, high
rising terminal, and syllable timing.
Penfield (1977), in her work with speakers of Mohave English, Navajo
English, and Hopi English, found that they spoke with few pitch changes
and thus display a smaller pitch range than speakers of other varieties
of English. They remained level even when asking questions; in non-
Indian varieties, questions usually have a rise in pitch at the end. In con-
trast to the varieties described by Penfield, other varieties have a high
rising terminal on declarative sentences. That is, there is a higher pitch
at the end of the sentence than in the rest of the sentence. This feature
has been attested in the Alabama-Coushatta (Hoffer 1982), Brandywine
(Gilbert 1986), and Tsimshian (Mulder 1982) varieties of AIE. Another
salient aspect of AIE prosody is rhythm, i.e. the relative length of adja-
cent syllables. Most non-Indian varieties of English have a stress-timed
rhythm such that stressed syllables are lengthened and unstressed ones
reduced, resulting in a difference in syllable length between the two types.
The resulting pattern has been likened to Morse code; it is found in
Germanic languages generally. In contrast, many varieties of AIE have
a syllable-timed rhythm, similar to that which occurs in most Romance
languages. That is, all syllables are of similar length, creating a rhythm
that is compared to a metronome. Syllable timing has been attested
to in Kotzebue (Vandergriff 1982), Brandywine (Gilbert 1986), Lumbee
(Coggshall 2008), and Eastern Cherokee (Coggshall 2008) varieties of
AIE.
American Indian English 111
3.2 Morphosyntax
The sentence and word structure of several varieties of AIE (namely Isle-
tan, Lakota, Lumbee, Mohave, Laguna, and San Juan) have been studied
extensively, and several of these structures are discussed below. I start with
variation in copula usage, and then move to other verbs, specifically to
features dealing with tense and aspect, as well as agreement. A short dis-
cussion of variation in nouns and pronouns follows, along with negative
concord. Variation among pronoun usage as well as deletion of pronouns
is explored after that, and I finish with a look at nonstandard word order
in AIE.
(6) Navajo: I live by the beliefs that coming from both the Navajo culture
and Christianity. (Bartelt 1985: 50–1)
In other cases that are progressive, the -ing suffix is dropped, as in (7)
and (8).
(7) Lakota English: Our childrens are start, you know, really mixing it up.
(Flanigan 1985: 222)
(8) Yakima English: You’re going to be the one bring the money home.
(Chessin and Aurbach 1982: 180)
American Indian English 113
The most studied feature of tense in AIE is unmarked tense, where verbs
that are meant to describe something that happened in the past lack any
indication of past tense. Some of the instances of absence of overt tense
marking may be due to CCR, where the past-tense suffix of weak verbs is
deleted due to this phonological constraint, as shown in (9) and (10):
(9) Mohave English: One guy got mash_ bad. (Penfield Jasper 1980: 82)
(10) Tohono O’odham English: His shirt got unzipper. (Bayles and Harris
1982: 6)
However, there are other examples where CCR cannot explain the lack
of tense, as in (11)–(13).
(11) San Juan English: Remember the time they fight for Unge. (Wolfram
et al. 1979: 49)
(12) Eastern Cherokee English: You’d go to the home they all speak in
Cherokee. (Coggshall 2005)
(13) Lakota English: He begin to look for her. (Flanigan 1985: 225)
Unmarked tense has also been documented in the Brandywine (Gilbert
1986), Laguna (Wolfram 1980, 1984, 1986; Wolfram et al. 1979), Western
Apache (Bartelt 1985, 1986), Isletan (Leap 1993), Quinault (Rowicka 2005),
Nisqually (Chessin and Aurbach 1982), Colville (Chessin and Aurbach
1982), and Navajo (Cook and Sharp 1966) varieties of AIE.
3.2.3 Nouns
Inflectional endings on nouns, namely plural (21) and (22) and possessive
-s (23), are optional in varieties of AIE. The information contained in
these morphemes is often expressed instead through other means, such as
through overt expressions of number elsewhere in the sentence or by word
order.
(21) Quinault English: My nephew – they are dead people. (Rowicka
2005: 309)
(22) San Juan English: Three other place we went. (Wolfram et al. 1979:
144)
(23) Navajo English: my sister husband, Jack father, my grandma house
(Cook and Sharp 1966: 25)
These features are also documented in the Mohave (Penfield Jasper 1980),
Laguna (Wolfram et al. 1979), Lakota (Flanigan 1984), and Brandywine
(Gilbert 1986) varieties of AIE.
American Indian English 115
The grammatical distinctions between count and mass nouns can be
lost, as shown in (24), where the count noun horse receives the modifier
much, which in most non-Indian varieties of English can only be attached
to mass nouns. (25) shows the opposite, where the mass noun pottery has
the plural morpheme typically reserved for count nouns.
(24) Lakota English: We ride much horses. (Flanigan 1985: 223)
(25) Mohave English: There’s a lot of potteries around there. (Penfield
Jasper 1980: 89)
This feature has also been attested in Koyukon English (Kwachka 1988).
3.2.4 Pronouns
Some varieties of AIE behave as pro-drop languages, where pronouns are
optional, as shown in (26) and (27).
(26) Hoopa English: Now when __ hear that some of my friends are
getting married, it a sad occasion. (Chessin and Aurbach 1982: 117)
(27) Lummi English: Something should be done to make __ possible.
(Chessin and Aurbach 1982: 117)
Pro-drop is also found in the Tlingit-Haida (Chessin and Aurbach 1982),
Swinomish (Chessin and Aurbach 1982), Tsimshian (Mulder 1982), Ute
(Leap 1993), and Mohave (Penfield Jasper 1980) varieties of AIE.
A particularly marked feature of many varieties of AIE is the variable
loss of gender distinction in the third person singular pronouns, he, she,
him, her, his, and hers. This leads to utterances such as (28) and (29) where
the pronoun does not have the same gender as its antecedent.
(28) Tohono O’odham English: The boy’s zipper got caught in her jacket.
(Bayles and Harris 1982: 6)
(29) Mohave English: My aunt plants corn in his own garden. (Penfield
Jasper 1980: 75)
This feature has also been documented in the Navajo (Cook and
Sharp 1966), Tsimshian (Mulder 1982), Lakota (Flanigan 1984, 1985), and
Cheyenne (Alford 1974) varieties of AIE.
3.2.5 Articles
In some varieties of AIE the articles a(n) and the are optional, as shown
in (30)–(32). This feature may be a system of marking specific versus non-
specific or what is known to the speaker versus what is known to the
116 elizabeth l. coggshall
listener, as is found in creoles (John Singler p.c.), so an avenue of further
research is to determine if there is a pattern to article deletion or not.
(30) Nisqually English: You’re __ nice person. (Chessin and Aurbach
1982: 178)
(31) Navajo English: They found __ bone in __ dumpyard. (Cook and
Sharp 1966: 25)
(32) Brandywine English: And __ fellow looked around. (Gilbert 1986:
107)
This feature is also documented in the Quinault (Rowicka 2005),
Mohave (Penfield Jasper 1980), Swinomish (Chessin and Aurbach 1982),
and Yakima (Chessin and Aurbach 1982) varieties of AIE.
3.2.6 Negation
A common feature of nonstandard varieties of English in general is the use
of multiple negative items in a single clause, what is referred to as negative
concord. Two examples of negative concord in AIE are shown in (33) and
(34).
(33) Eastern Cherokee English: My mother didn’t make no pottery.
(Coggshall 2005)
(34) Laguna English: Then no police didn’t catch us. (Stout 1979: 67)
Negative concord has also been found in the Mohave (Penfield Jasper
1980), San Juan (Wolfram et al. 1979), Lakota (Flanigan 1984, 1985), and
Isletan (Leap 1977b, 1974) varieties of AIE.
3.2.7 Prepositions
Prepositions are another source of difference between AIE and other vari-
eties of English. Many varieties show nonstandard use of prepositions, as
in (35)–(37).
(35) Tohono O’odham English: They were at fishing. (Bayles and Harris
1982: 17)
(36) Mohave English: He got fired of the church. (Penfield Jasper 1980:
145)
(37) Cheyenne English: Let’s ride on your car to Pizza Hut. (Alford 1974:
8)
American Indian English 117
This kind of prepositional usage has also been documented in the Navajo
(Cook and Sharp 1966), Yakima (Chessin and Aurbach 1982), Brandywine
(Gilbert 1986), Tlingit-Haida (Chessin and Aurbach 1982), and Tsimshian
(Mulder 1982) varieties of AIE. Prepositions can also be optional in some
varieties of AIE, as shown in (38) and (39).
(38) Mohave English: He lives __ that second house. (Penfield Jasper
1980: 144)
(39) Lakota English: They live __ New York. (Flanigan 1984: 92)
Preposition deletion has also been documented in Nisqually (Chessin
and Aurbach 1982), and Brandywine (Gilbert 1986) English.
3.3 Lexicon
The lexicon of AIE is understudied, with almost nothing written about it.
However, there are mentions of various sources of lexical items. Borrowings
from ancestral languages are obviously a major source for lexical innovation.
Kotzebue English has lexical items that come from the ancestral language,
loan translations, and the local dialect of English (Vandergriff 1982: 122).
Mulder (1982: 106–7) details the kinds of words from Tsimshian that
are found in Tsimshian English; such borrowings are usually for terms
that English lacks. Innovative lexical items in Lumbee English have been
catalogued, including ellick for coffee, toten for a sign of impending death
or evil given by a spirit, juvember for slingshot, and brickhouse Indian for a
rich Lumbee (Brewer and Reising 1982, Wolfram et al. 2002: 63). Finally,
innit is a lexical item that spans many varieties of AIE. Innit, sometimes
spelled enit or ennit, is a tag question akin to y’know in other varieties of
English, but available for broader usage (Johansen 2007: 336). I have also
heard the variation is it spoken by Navajo youth to express surprise or
disbelief. Examples of enit can be found in the works of the novelist and
short story writer Sherman Alexie (Spokane/Coeur d’Alene), as shown in
(44)–(47).
(44) Why don’t you get in your BMW, that’s what you drive, enit? (Alexie
2000: 50)
(45) “You’re a fighter, enit?”
I threw in the “enit,” a reservation colloquialism, because I wanted the
fighter to know that I had grown up on the rez, in the woods, with every
Indian in the world (Alexie 2012: 33, emphasis in original)
(46) So you must have eight or nine spirits going on inside you of you,
enit? (Alexie 2004: 183)
(47) “Don’t worry about the money,” Thomas said. “It don’t make any
difference anyhow.”
“Probably not, enit?” (Alexie 1994: 74)
3.4 Pragmatics
Pragmatics in AIE has been the subject of intense study, especially by
those interested in improving American Indian children’s performance in
schools. This focus results from extensive intercultural miscommunication
American Indian English 119
between American Indians and non-Indians. Philips stresses this point,
stating “Educators cannot assume that because Indian children . . . speak
English . . . that they have also assimilated all of the sociolinguistic rules
underlying interaction in classrooms and other non-Indian social situations
where English is spoken” (1972: 392). Below are a few of the more salient
pragmatic features.
3.4.1 Silence
Compared to groups in contact with American Indians, speakers of AIE
are extremely quiet, to the point that may seem baffling or even rude.
Dumont, working in classrooms on the Cherokee and Sioux reservations,
described the “mask of silence” the students used while in the classroom
(1972: 346). Basso (1970) used the indigenous term for the extensive silence
his Western Apache informants used: “To give up on words.” He found
six particular instances where not talking was considered the correct thing
to do: when meeting strangers, when courting, when children came home
from boarding school, when getting “cussed out,” when being with some-
one in mourning, and being with someone undergoing a healing ceremony
(Basso 1970: 217–24). He summarized this pattern thus: “keeping silent in
Western Apache culture is associated with social situations in which par-
ticipants perceive their relationships vis a vis one another to be ambiguous
and/or unpredictable” (Basso 1970: 226).
3.4.3 Humor
While not a locus for intercultural miscommunication, humor is a large
part of language use in American Indian communities. Basso, in his classic
study, Portraits of the “Whiteman”, shows humor in action in the Western
Apache community he studied (1979). However, that study was on Apache
120 elizabeth l. coggshall
speakers, and thus outside the purview of this chapter, but it does allow us
to see that humor pervades American Indian culture.
Vine Deloria, Jr. (Standing Rock Sioux), a prominent writer and activist,
has bemoaned the fact that this aspect of American Indians is not more
well known: “It has always been a great disappointment to Indian people
that the humorous side of Indian life has not been mentioned by professed
experts on Indian Affairs” (Deloria 1988: 146). In his book Custer Died for
Your Sins (its title itself being an example of this humor), he details many
examples of humor in AIE, usually satirical in nature, often lampooning
some of the worst things to ever happen to American Indians: Columbus,
Custer, the Bureau of Indian Affairs, missionaries, and white people in
general, as well as “razzing” members of other tribes (1988: 146–167). In
Custer Died for Your Sins, Deloria gives a few examples of his favorite jokes,
in (48)–(50).
(48) We also had a saying that in case of fire call the BIA and they would
handle it because they put a wet blanket on everything. (Deloria
1988: 147–8)
(49) It is said that when Columbus landed, one Indian turned to the
another and said, ‘Well, there goes the neighborhood.’ Another ver-
sion has two Indians watching Columbus land and one saying to the
other, ‘Maybe if we leave them alone they will go away.’ (Deloria
1988: 148)
(50) Custer’s Last Words occupy a revered place in Indian Humor. One
source states that as he was falling mortally wounded he cried, ‘Take
no prisoners!’ Other versions, most of them off color, concentrate
on where those ∗∗∗∗ Indians are coming from. (Deloria 1988: 149)
Deloria hypothesizes that this emphasis on humor comes from the pre-
contact method of social control where individuals who did not follow
cultural conventions were teased by other members of the tribe. This
kind of teasing was done in order to preserve the face of the individual:
since teasing is an indirect form of social control; such teasing was then
anticipated and making fun of oneself was an act of humility (Deloria 1988:
147).
4 Conclusion
As long as American Indians exist as a separate social entity, AIE will
exist in some form. The reservation system in the US and the reserves in
American Indian English 121
Canada will also lead to the survival of AIE, since it is well attested that
isolation, either social or geographical, is a major factor in the creation
and maintenance of dialects (Labov and Harris 1986). And because it
will continue to exist, more research is needed on AIE. I suggest that
further work on AIE should take a particular path: new data on more
communities, especially in the growing urban communities, with a focus
beyond documentation to issues of identity work and changes in ethnicity.
Past research was mostly conducted in the 1970s, in the western part of the
United States, with an eye towards educational applications and questions
of genesis and documentation.
Because most of the research that has been presented in this study dates
back several decades, new research is needed to see how varieties of AIE have
changed over time, during decades of changing demographics and attitudes
towards Native Americans. Linguistic science has also progressed signifi-
cantly since the 1970s, especially in field recording and acoustic phonetics.
Present-day work on AIE can use tools such as these to better understand
the workings of AIE, getting more precise data on the phonological features
and a larger corpus in which to look at morphosyntactic features.
Further, the majority of the work on AIE has focused on the western half
of the United States: e.g. Lakota, Navajo, Isletan, Mohave. This concen-
tration is problematic because many Native American contact situations in
the eastern United States are significantly different from those in the west.
Eastern tribes were in contact with English speakers long before those in
the west. Further, eastern tribes were decimated early by disease and war;
they lack treaties signed between them and the United States government
and thus often cannot claim the special status of many western tribes;
they adopted English early on and lost their indigenous languages early,
too. These differences can perhaps lead to the reliance on and adoption of
cultural markers, such as linguistic features, to lay claim to an authentic
American Indian identity (Coggshall 2008). Further, little work has been
done on Canadian aboriginal peoples (Ball et al. 2006; Ball and Bernhardt
2008), who have yet a different history with regard to colonization and the
English language. Not only do these communities of AIE speakers have
different histories, the English speakers around them today can have an
effect on AIE. Language contact is a major force in language change and
new dialect formation, and studying contact situations that have been in
effect for various lengths of time enables a better understanding of the
linguistic and social facts of contact.
Another important area of concern is the large and growing population
of American Indians living off reservations and other rural enclaves, mainly
122 elizabeth l. coggshall
in urban centers throughout North America. In fact, in the US, more than
half of all people of indigenous descent live in cities. Urban Indian culture
differs greatly from that found on reservations and elsewhere (e.g. Lobo
and Peters 2001). So far, only one small study on urban Indians has been
published; Bartelt (1993) looked at a speech at a powwow in Los Angeles.
Perhaps the most important factor is that urban communities not only
include people from many tribes but also people of many other ethnicities.
This change in environment may lead to ethnic change. American Indian
ethnicity has undergone extensive reorganization since the 1970s (Nagel
1996), most strikingly in the advent of a pan-Indian identity for many
Native Americans, particularly in urban centers (Lopez and Espiritu 1990).
These changes in demographics and ethnicity suggest that there may
be repercussions in AIE. Recent studies (Coggshall 2008; Rowicka 2005)
suggest that the strict compartmentalization by tribe (as described above
in Section 3) may be deteriorating, and that certain features appear to be
spreading from one variety of AIE to another (see Leap 1982, 1993). This
change may be a result of panethnicity (Lopez and Espiritu 1990), where
the scope of identity expands to include a larger “Indian” identity on top
of a tribal-level identity. As the scope of identity changes, the language
used to express this identity may change as well, leading to a convergence
of different varieties of AIE. Leap (1993), on the other hand, has stated
categorically that there is no general AIE variety and that there will never
be one. Only further research can answer this question.
References
Alexie, Sherman. 1994. The Lone Ranger and Tonto Fist Fight in Heaven. New
York: Harper Perennial.
2000. The Toughest Indian in the World. New York: Grove Press.
2004. Ten Little Indians. New York: Grove Press.
2012. Blasphemy. New York: Grove Press.
Alford, D. 1974. The Cheyenne dialect of English and its educational implications.
Unpublished manuscript. Lame Deer, MT: Northern Cheyenne Title VII
Project.
Anderson, Bridget L. 1999. Source-language transfer and vowel accommodation in
the patterning of Cherokee English /ai/ and /oi/. American Speech 74: 339–68.
Bakker, Peter and Robert Papen. 1997. Michif: a mixed language based on Cree and
French. In Sarah G. Thomason, ed., Contact Languages: A Wider Perspective.
Amsterdam and Philadelphia: John Benjamins, 295–363.
Ball, Jessica and Barbara Bernhardt. 2008. First Nations English dialects in Canada:
implications for speech-language pathology. Clinical Linguistics & Phonetics
22(8): 570–88.
American Indian English 123
Ball, Jessica, Barbara Bernhardt, and Jeff Deby. 2006. First Nations English
dialects. Unpublished monograph. University of Victoria and University
of British Columbia.
Bartelt, H. Guillermo. 1986. Language contact in Arizona: the case of Apachean
English. Anthropos 81: 4–6.
1993. Urban American Indian intertribal discourse. English World-Wide 14(1):
57–70.
Basso, Keith. 1970. ‘To give up on words’: silence in Western Apache culture.
Southwest Journal of Anthropology 26(3): 213–38.
1979. Portrait of the Whiteman. Cambridge University Press.
Bayles, Kathryn A. and Gail A. Harris. 1982. Evaluating speech-language skills in
Papago Indian children. Journal of American Indian Education 21(2): 11–20.
Boas, Franz. 1911. Handbook of American Indian languages, vol. 1. Washington,
DC: Government Printing Office.
Brewer, Jeutonne P. and R. W. Reising. 1982. Tokens in the Pocosin: Lumbee
English in North Carolina. American Speech 57(2): 108–20.
Calloway, Colin G. 2004. First Peoples: A Documentary Survey of American Indian
History. Boston: Bedford/St. Martin’s.
Chambers, J. K. 2004. Dynamic typology and vernacular universals. In Bernd
Kortman, ed., Dialectology Meets Typology: Dialect Grammar from a Cross-
Linguistic Perspective. Berlin: Walter de Gruyter, 127–45.
Chessin, Linda and Emily Aurbach. 1982. Teaching composition to Northwest
American Indians. In H. Guillermo Bartelt, Susan Penfield Jasper, and Bates
Hoffer, eds., Essays in Native American English. San Antonio, TX: Trinity
University, 173–86.
Coggshall, Elizabeth L. 2005. Cherokee English in the Appalachian context. Paper
presented at the Southeastern Conference on Linguistics 72, North Carolina
State University, April.
2006. Differential vowel accommodation among two Native American groups.
Unpublished Master’s thesis, North Carolina State University.
2008. The prosodic rhythm of two varieties of native American English. Penn
Working Papers in Linguistics 14(2): 1–9.
Cook, Mary Jane and Margaret Amy Sharp. 1966. Problems of Navajo speakers in
learning English. Language Learning 16: 21–9.
Craig, Beth. 1991. American Indian English. English World-Wide 12(1): 25–61.
Dannenberg, Clare. 1999. Grammatical and phonological manifestations of null
copula in a tri-ethnic contact situation. Journal of English Linguistics 27(4):
356–70.
2002. Sociolinguistic Constructs of Ethnic Identity: The Syntactic Delineation of
an American Indian English. Raleigh, NC: Duke University Press for the
American Dialect Society.
Dannenberg, Clare and Walt Wolfram. 1998. Ethnic identity and grammatical
restructuring: be (s) in Lumbee English. American Speech 73(2): 139–59.
Deloria, Vine. 1988. Custer Died for your Sins: An Indian Manifesto. Norman, OK:
University of Oklahoma Press.
124 elizabeth l. coggshall
Dickason, Olive Patricia and William Newbigging. 2006.A Concise History of
Canada’s First Nations. New York: Oxford University Press.
Dillard, Joey Lee. 1972. Black English: Its History and Usage in the United States.
New York: Vintage Books.
Dozier, Edward. 1956. Two examples of linguistic acculturation: the Yaqui
of Sonora and Arizona and the Tewa of New Mexico. Language 32(1):
146–7.
Drechsel, Emanuel J. 1983. Towards an ethnohistory of speaking: the case of
Mobilian Jargon, an American Indian pidgin of the Lower Mississippi Valley.
Ethnohistory 30(3): 165–76.
Dumont, Robert V., Jr. 1972. Learning English and how to be silent: studies in
Sioux and Cherokee classrooms. In Courtney B. Cazden, Vera P. John, and
Dell Hymes, eds., Functions of Language in the Classroom. New York: Teachers
College Press, 344–69.
Erdrich, Louise. 2012. The Round House. New York: Harper Collins.
Fear, Jacqueline. 1980. English versus the vernacular: the suppression of Indian
languages in reservation schools at the end of the nineteenth century. Revue
française d’études américaines 9: 13–24.
Feinstein, Mark. 1980. Ethnicity and topicalization in New York City English.
International Journal of the Sociology of Language 26: 15–24.
Flanigan, Beverly Olson. 1984. Bilingual education for Native Americans: the
argument from studies of variational English. In Jean Hanscome, R. Orem,
and P. B. Taylor, eds., On TESOL ’83: The Question of Control: Selected
Papers from the Annual Convention of Teachers of English to Speakers of Other
Languages. Washington, DC: TESOL, 81–93.
1985. American Indian English and error analysis: the case of Lakota English.
English World-Wide 6(2): 217–36.
Gilbert, Glenn. 1986. The English of the Brandywine population: a triracial iso-
late in Southern Maryland. In Michael Montgomery and Guy Bailey, eds.,
Language Variety in the South: Perspectives in Black and White. Tuscaloosa,
AL: University of Alabama Press, 102–10.
Goddard, Ives. 1997. Pidgin Delaware. In Sarah G. Thomason, ed., Contact Lan-
guages: A Wider Perspective. Amsterdam and Philadelphia: John Benjamins,
43–98.
Harvey, Gina Cantoni. 1974. Dormitory English: implications for the classroom
teacher. In G. Bills, ed., Southwest Areal Linguistics. San Diego: Institute for
Cultural Pluralism, 283–92.
Hazen, Kirk. 2002. Identity and language variation in a rural community. Language
78(2): 240–57.
Hinton, Leanne. 1999. Trading tongues: loss of heritage languages in the United
States. English Today 15(4): 21–30.
Hoffer, Bates L. 1982. The Alabama-Coushatta of East Texas: a case for multiple-
source dialects. In H. Guillermo Bartelt, Susan Penfield Jasper, and Bates
Hoffer, eds., Essays in Native American English. San Antonio, TX: Trinity
University Press, 55–66.
American Indian English 125
Humes, Karen R., Nicholas A. Jones, and Roberto R. Ramirez. 2011. Overview
of Race and Hispanic Origin: 2010. Washington, DC: United States Census
Bureau.
Johansen, Bruce. 2007. The Praeger Handbook on Contemporary Issues in Native
America: Linguistic, Ethnic, and Economic Revival, vol. 1. Westport, CT:
Greenwood Publishing Group.
Kwachka, Patricia. 1988. Oral and Written English of the Koyukon Athabaskan
Area: A Preliminary Analysis. APEL Research Report vol. 4. Nenana, AK:
Yukon-Koyukon School District.
Labov, William. 1969. Contraction, deletion, and inherent variability of the
English copula. Language 45(4): 715–62.
Labov, William and Wendell Harris. 1986. De facto segregation of black and white
vernaculars. In David Sankoff, ed., Diversity and Diachrony. Amsterdam and
Philadelphia: John Benjamins, 1–24.
Leap, William L. 1974. On grammaticality in Native American English: the evi-
dence from Isleta. International Journal of the Sociology of Language 2: 79–90.
1977a. On consonant cluster simplification in Isletan English and elsewhere. In
William L. Leap, ed., Studies in Southwestern Indian English. San Antonio,
TX: Trinity University Press, 45–53.
1977b. Two examples of Isletan English syntax. In William L. Leap, ed., Studies
in Southwestern Indian English. San Antonio, TX: Trinity University, 65–78.
1977c. Uninflected be in Isletan English: a problem in accountability. In William
L. Leap, ed., Studies in Southwestern Indian English. San Antonio, TX: Trinity
University Press, 79–100.
1977d. A note on subject–verb agreement in Isletan English. In William L.
Leap, ed., Studies in Southwestern Indian English. San Antonio, TX: Trinity
University Press, 121–30.
1982. The study of American Indian English: status and direction of inquiry. In
H. Guillermo Bartelt, Susan Penfield Jasper, and Bates Hoffer, eds., Essays in
Native American English. San Antonio, TX: Trinity University Press, 1–22.
1993. American Indian English. Salt Lake City: University of Utah Press.
Leechman, Douglas and Robert A. Hall, Jr. 1955. American Indian Pidgin English:
attestations and grammatical peculiarities. American Speech 30(3): 163–71.
Liebe-Harkort, Marie-Louise. 1983. A note on the English spoken by Apaches.
International Journal of American Linguistics 2: 207–8.
Lobo, Susan and Kurt Peters. 2001. American Indians and the Urban Experience,
vol. 5. Lanham, MA: Altamira Press.
Lopez, David and Yen Espiritu. 1990. Panethnicity in the United States: a theo-
retical framework. Ethnic and Racial Studies 13(2): 198–224.
Malancon, Richard and Mary Jo Malancon. 1977. Indian English at Haskell
Institute, 1915. In William L. Leap, ed., Studies in Southwestern Indian English.
San Antonio, TX: Trinity University Press, 141–54.
Maynor Lowry, Malinda. 2010. Lumbee Indians in the Jim Crow South: Race,
Identity, and the Making of a Nation. Chapel Hill, NC: University of North
Carolina Press.
126 elizabeth l. coggshall
Meek, Barbara. 2006. “And the Injun goes ‘How!’”: representations of Amer-
ican Indian English in white public space. Language in Society 35(1): 93–
128.
Miller, Mary Rita. 1967. Attestations of American Indian Pidgin English in fiction
and nonfiction. American Speech 42(2): 142–7.
1977. Children of the Salt River: First and Second Language Acquisition among
Pima Children. Bloomington: Indiana University Press.
Mithun, Marianne. 1999.The Languages of Native North America. Cambridge Uni-
versity Press.
Mulder, Jean. 1982. The Tsimshian English dialect: the result of language inter-
ference. In H. Guillermo Bartelt, Susan Penfield Jasper, and Bates Hoffer,
eds., Essays in Native American English. San Antonio, TX: Trinity University,
95–112.
Nagel, Joane. 1996. American Indian Ethnic Renewal: Red Power and the Resurgence
of Identity and Culture. Oxford University Press.
Nagel, Joane and C. Matthew Snipp. 1993. Ethnic reorganization: American Indian
social, economic, political, and cultural strategies for survival. Ethnic and
Racial Studies 16(2): 203–35.
Nelson-Barber, S. S. 1982. Phonologic variations of Pima English. In Robert
St. Clair and William Leap, eds., Language Renewal among American Indian
Tribes: Issues, Problems, and Prospects. Arlington, VA: National Clearing house
for Bilingual Education, 122–3.
Penfield, Susan. 1977. Some examples of Southwestern Indian English compared.
In William L. Leap, ed., Studies in Southwestern Indian English. San Antonio,
TX: Trinity University, 23–43.
1980. Selected grammatical characteristics of Mohave English. Unpublished
PhD dissertation, University of Arizona.
Philips, Susan U. 1972. Participant structures and communicative competence:
Warm Springs children in community and classroom. In Courtney B. Caz-
den, Vera P. John, and Dell Hymes, eds., Functions of Language in the Class-
room. New York: Teachers College Press, 370–94.
Rowicka, Grażyna J. 2005. American Indian English: the Quinault case. English
World-Wide 26(3): 301–24.
Sapir, Edward. 1933. La réalité psychologique des phonemes. Journal de Psychologie
Normale et Pathologique 30: 247–65.
Schilling-Estes, Natalie. 2004. Constructing ethnicity in interaction. Journal of
Sociolinguistics 8(2): 163–95.
Schreier, Daniel, Peter Trudgill, Edgar W. Schneider, and Jeffrey P. Williams,
eds. 2010. The Lesser-Known Varieties of English: An Introduction. Cambridge
University Press.
Shuy, Roger. 1964. The problem of American Indian English. In Roger W. Shuy,
ed., Social Dialects and Language Learning. Urbana, IL: National Council of
Teachers of English, 52–4.
Silverstein, Michael. 1972. Chinook jargon: language contact and the problem of
multi-level generative systems. Language 48(2): 378–406.
American Indian English 127
Spicer, Edward Holland. 1967. Cycles of Conquest: The Impact of Spain, Mexico,
and the United States on Indians of the Southwest, 1533–1960. Tucson, AZ:
University of Arizona Press.
Stout, Steven. 1977. A comment on selection control in English expression at
Santa Ana. In William L. Leap, ed., Studies in Southwestern Indian English.
San Antonio, TX: Trinity University, 55–63.
1979. Sociolinguistic aspects of English diversity among elementary-aged
students from Laguna Pueblo. Unpublished PhD dissertation, American
University.
Torbert, Benjamin. 2001. Language history traced through consonant cluster
reduction: the case of Native American Lumbee English. American Speech
76(4): 361–87.
Trujillo, Octaviana. 1997. A tribal approach to language and literacy development
in a trilingual setting. In J. Reyhner, ed., Teaching Indigenous Languages.
Flagstaff, AZ: Northern Arizona University, 10–21.
Vandergriff, Jim. 1982. Kotzebue English: some notes on Inupiaq English. In
H. Guillermo Bartelt, Susan Penfield Jasper, and Bates Hoffer, eds., Essays in
Native American English. San Antonio, TX: Trinity University, 121–56.
Whalen, Douglas H. and Gary Simons. 2012. Endangered language families.
Language 88(1): 155–73.
Whorf, Benjamin Lee. 1941. The relation of habitual thought and behavior to
language. In Leslie Spier, A. Irving Hallowell, and Stanley S. Newman, eds.,
Language, Culture and Reality: Essays in Memory of Edward Sapir. Westport,
CT: Greenwood Press, 75–93.
Wolfram, Walt. 1980. Dynamic dimensions of language influence: the case of
American Indian English. In Howard Giles, W. Peter Robinson, and Philip
M. Smith, eds., Language: Social Psychological Perspectives: Selected Papers from
the First International Conference on Social Psychology and Language. Oxford:
Pergamon Press, 377–88.
1984. Unmarked tense in American Indian English. American Speech 59(1): 31–
50.
Wolfram, Walt, Donna Christian, Leap William L., and Lance Potter. 1979. Vari-
ability in the English of Two Indian Communities. Arlington, VA: Center for
Applied Linguistics.
Wolfram, Walt and Clare Dannenberg. 1999. Dialect identity in a tri-ethnic con-
text: the case of Lumbee American Indian English. English World-Wide 20(2):
179–216.
Wolfram, Walt, Clare Dannenberg, Stanley Knick, and Linda Oxendine. 2002.
Fine in the World: Lumbee Language in Time and Place. Raleigh, NC: North
Carolina State University.
Wolfram, Walt and Jason Sellers. 1999. Ethnolinguistic marking of past be in
Lumbee Vernacular English. Journal of English Linguistics 27(2): 94–114.
c h a p ter 6
Bequia English
James A. Walker and Miriam Meyerhoff
1 Introduction
1
Bequia is a small island located in St Vincent and the Grenadines in
the Eastern Caribbean whose sociolinguistic situation has been shaped by
two main geographic factors: its relative isolation and its small size (see
Map 6.1 ). The largest and northernmost of the Grenadines, Bequia lies
14 km (8 miles) south of St Vincent, separated by a deep channel. Until
the airport was built in 1992, the only way of reaching the island was by
ferry from St Vincent. Bequia is roughly hook shaped, approximately 11 km
(7 miles) from north to south and 8 km (5 miles) at its widest east–west
point, and split between its leeward (west) and windward (east) coasts
by a low mountain ridge running roughly north–south. Given the size
of the island and its relatively shallow timeline of settlement, the current
population of about 5,000 people features a surprising degree of diversity
in the varieties of English and English-based creole they speak.
1 The French gave Bequia its current name, which is said to derive either from the Carib name becouya
‘island of clouds’ or from the French word béquille ‘crutch’ (from the island’s shape). The current
pronunciation [ˈbɛkwe:] seems to be an English reading of the French spelling.
128
Bequia
Martinique
St Lucia Caribbean
Mount
Sea
Pleasant
St Vincent Industry Bay
Hamilton
Barbados
Spring Bay
Port Elizabeth
Admiralty
Atlantic
Grenada Lower Bay Bay
Ocean
Friendship Bay
3.1 Phonology
3.1.1 Short vowels
The front vowels kit and dress are generally short and lax ([ɪ] and [ɛ]),
though they occasionally occur as tense vowels ([i] and [e]). The kit vowel
is sometimes lowered to something close to [ɛ] (e.g. miracle sounds like
[mɛrəkl]). The trap vowel normally occurs as a low central [a], though
we have occasionally heard a more fronted [æ]. This system is mirrored in
the back vowels to some extent. The foot and lot vowels are generally
short and lax ([ʊ] and [ɑ]), though they are sometimes lowered to [ɔ] and
[a]/[ɑ], respectively. The strut vowel normally occurs as a central [ʌ], but
we have heard it lowered and backed to [ɑ] or [ɔ] (e.g. cup [kɔp]).
3.1.3 Diphthongs
The price and mouth diphthongs occur as [aɪ] and [aʊ], respectively,
though occasionally their onsets are raised to [ə] or backed and rounded to
[ɔ] (e.g. price [pɹɔɪs]). The choice diphthong usually occurs as [ɔɪ] but its
onset is often lowered, fronted and unrounded to [a] (e.g. choice [tʃaɪs]).
3.1.5 Prosody
The prosody of Bequia English is characteristically Caribbean, in that
there is a wide pitch range and unstressed vowels often receive their full
value. The latter pattern may be related to a rightward stress shift noted by
Wells (1982: 572) (emphatically or phrase-finally). There is a great deal of
variability in the realization of unstressed vowels: unstressed front vowels
tend to occur as [ɪ] or [i], while unstressed low vowels may occur as [a],
[ɑ] or [ə]. Unstressed vowels with an underlying following /r/ may occur
with or without r-colouring.
3.1.6 Consonants
Bequia English has the full set of English stops: /p b t d k g/. Word-
initially, /t/ and /d/ have more dental articulations and /k/ and /g/ tend to
be palatalized (e.g. Coast Guard [kj o:s gj a:d]). Word-final /t/ is often fully
released and is sometimes deleted, especially in function words, such as
but [bʌ] and about [əbaʷ]. Word-medially, /t/ generally occurs as an oral
stop (i.e. not flapped) or as a glottal stop, so that after may be pronounced
as [aftə] or [afʔə]. The fricatives are / f v θ s z ʃ (ʒ)/. Word-initial /h/
is variably present, normally omitted in function words such as here and
him. The interdental fricatives are most often realized as the alveolar/dental
stops [t] and [d] (e.g. think [tɪŋk], there [dɛ]), though there appears to be
some variation according to style or topic (Ng 2009). The nasal consonants
are /m n/ and word-final /ŋ/. After back vowels, word-final /n/ tends to be
velarized, so that Hamilton sounds like [hamɪltɔŋ]. The sonorant and glide
consonants are /w j l r/. The lateral /l/ is typically ‘light’ rather than ‘dark’.
Generally, /r/ is realized as a retroflex [ɹ], though postvocalic rhoticity is
variable.
Consonant clusters involve a number of processes. There is widespread
stridentization of the first element in [str] clusters (so that industry is
pronounced [ɪndʌʃtri]), and palatalization in /tr/ clusters [tʃ] (Partridge
Bequia English 133
2009). Final clusters may be devoiced (so that /dz/ in kids is realized as [ts])
and final stops in clusters are usually deleted, especially /t/ and /d/ (so respect
is realized as [rispɛk]). There is systematic pronunciation of ask as [æks],
and common metathesis of [sp] in some words (e.g. crisp pronounced as
[krɪps] and crispy as [krɪpsi]), though metathesis may be lexically restricted,
since mask undergoes final deletion [ma:s].
3.2 Morphosyntax
3.2.1 Plural formation
Plurality is indicated in several ways. Nouns may be marked with the
Standard English plural -s suffix and irregular forms (1) or with the regional
Caribbean strategy of postnominal (and) them (2). Unmarked (bare) nouns
also occur in contexts where the meaning is clearly plural (3).
(1) Hear, the children these days, they live in a bed of roses. (Speaker 36)
(2) a. When the stagaboys and them, which is the bugs them, coming
out to catch you, you want catch them quick. (Speaker 5)
b. Well, I hear old time people say turn you pocket an dem on the
wrong side. (Speaker 36)
(3) I still think the teacher used to try they best, eh? (Speaker 29)
3.2.2 Pronouns
Bequia English varies between the Standard English pronominal system,
distinguishing subjects (I, she, he, we, they) and objects (me, her, him,
us, them) with different forms for every pronoun except you, as well as
possessive pronouns (my, your, his, her, our, their), and a system in which
the same forms may be used for subjects, objects and possessives (4). With
reflexive pronouns speakers either combine the general pronoun with -self,
giving forms such as meself, weself, sheself, youself etc. (5), or use the Standard
English forms, myself, herself etc.
(4) a. He (done) ask me what me want, (but ?? I) tell him I want nothing,
(Speaker 6)
b. After me teach she, he come for L. G. [proper name]. (Speaker 6)
c. Not for we to live. (Speaker 306)
d. All of we descendants is from there. (Speaker 306)
e. Shoes for help they foot when the sun hot. (Speaker 12)
f. You put it over your shoulder and you get you corn and all thing
in it. (Speaker 26)
134 james a. walker and miriam meyerhoff
g. So who mother ain’t go have they young children and see um
something pick it up and go away with it? (Speaker 24)
(5) a. I’m a person who sit down by meself. (Speaker 6)
b. Everybody enjoy theyself. (Speaker 301)
c. When it’s youself. (Speaker 23)
There is also a gender-neutral third-singular pronoun e [i], which can be
used in all of the cases where Standard English uses it, as well as to refer to
human beings whose sex is known (6).
(6) a. E say it does vomit she. (Speaker 6)
‘She says it makes her vomit.’
b. It na sound like e break. (fieldnotes)
‘It didn’t sound like it [the glass] broke.’
Pronouns may be emphasized using one (7), though we have only heard
this variant in first person singular.
(7) a. Q: So you alone live here?
A: Yeah. I mean [name] and dem doz come and go, but is me one
doz sleep and everything. (Speaker 10)
b. Is not me one, me and a cousin [go fishing together]. (Speaker 17)
c. Sometime me one me drop sleep. (Speaker 9)
Null subjects also occur in Bequia English, at higher rates than in Standard
English, but much less frequently than in typical ‘null subject’ languages
such as Spanish.2 Subjects are most likely to be omitted in first person
singular or second person (8), though other topical subjects may also be
omitted (9).
(8) a. I said, you know, Ø came from St Vincent. (Speaker 20)
‘I said, you know, [I] came from St Vincent.’
b. Brad! Ø Call you! Come here! (grandmother calling to grandchild
from a window, fieldnotes)
‘Brad! I am calling you! (from: me call you) Come here!’
(9) Q: What would you say it [whale meat] taste like?
A: Well, Ø taste like beef. (Speaker 20)
2 Based on a subsample of eighteen speakers from our corpus, balanced across Hamilton, Mount
Pleasant and Paget Farm, the rate of null subjects in affirmative declarative clauses is 5 per cent
[471/8820], while the rate in Standard English is estimated between 1 and 2 per cent in non-
coordinated clauses (Meyerhoff 2000).
Bequia English 135
3.2.3 Tense and aspect
Tense and aspect are marked in Bequia English through an array of
morphosyntactic variants, though unmarked (bare) verbs are the most
common realization across all communities, for both present and past
reference (10).
(10) a. Even my boy child go out. (Speaker 14)
b. When we go round we play a ring song. (Speaker 1)
Present-tense verbs may be inflected with -s (11a) across all persons and
numbers, although very infrequently in all communities (Walker 2010).
The present progressive may also be used, occurring either with an overt
or null auxiliary (11b–c). In addition, the verb may occur with will (12) or
doz (13) in contexts of habitual aspect, although both are very infrequent.
(11) a. Of course, I pays my bonds. (Speaker 101)
b. We are not talking about the grave. (Speaker 23)
c. You Ø going to dance, the same Ø going on. (Speaker 1)
(12) These things will never show up on any test. (Speaker 27)
(13) But people doz hamper what your children they does. (P14:79)
In the past tense, verbs are variably marked with a number of pre- and
post-verbal morphemes. Weak verbs are generally unmarked for past tense
(14b), but inflection with the [t, d] suffix (14a) occurs at low rates across
all communities (though more frequently in Mount Pleasant). Strong and
semi-strong verbs are also variably marked for past tense, through stem
changes and/or final [t, d] (14c, 15). The irregular verbs be and have and
modal verbs are typically overtly marked for past tense (16).
(14) a. School stopped at age fifteen. (Speaker 313)
b. That was a night we always look forward to. (Speaker 20)
c. Yesterday I sit down here . . . (Speaker 13)
(15) a. When I was sixteen, me brother send for me in Georgetown.
(Speaker 11)
b. In those days I built myself a boat, it was ten feet long. (Speaker 7)
(16) a. That was what she eating everyday, corn coocoo, and fish, bush
water tea. (Speaker 19)
b. But my we weren’t the fastest but we want to be the king.
(Speaker 13)
c. I had somebody so and I lose them. (Speaker 24)
136 james a. walker and miriam meyerhoff
d. Macintosh wouldn’t need all that food for all of them. (Speaker 13)
e. I couldn’t stay out late when I was a teenager. (Speaker 24)
Two creole-like preverbal markers are used in Bequia English: bin (17)
and done (18). When combined with stative verbs, bin seems to convey
simple past, as in (17a), but with dynamic verbs the interpretation may be
remote past (where bin alternates with did ) (17b–c). Bin may also mark
irrealis mood (17d). Bin is largely restricted to Hamilton, and even there
occurs at a very low rate. The preverbal marker done indicates that an action
is completed and in some ways corresponds to the Standard English perfect
(which occurs very rarely in Bequia English) (18). We found occurrences
of done only in our Hamilton and Southside interviews, and at low rates.
(17) a. Them not bin have no engine. (Speaker 1)
b. When I bin going to school . . . (Speaker 1)
c. The time I telling you, he bin there. (Speaker 6)
d. If rain na bin come, I bin going today by a lady name Miss F.
(Speaker 1)
‘If it hadn’t rained, I would have gone today to visit a lady called
Miss F.’
(18) a. I could deh here now; one minute you pass off, you done dead.
(Speaker 19)
‘I could be here now; the next minute you pass away and you’ve
died.’
b. They done call me already. (Speaker 314)
‘They had called me already.’
c. I done accustom to home here.(Speaker 5)
Aspectual markers other than those we have already discussed may be
broadly divided into recurring events (habituals) and ongoing events (pro-
gressives). Habituals are marked with a broad range of pre- and postverbal
elements (19): deh, do(z), ah, V-ing, will/would and used to.
(19) a. All day you outside [when I was a kid], you deh running around.
(Speaker 102)
b. We doz close at ten. (fieldnotes)
c. It ah burn, did stop, but it used to burn me. (Speaker 17)
d. All them we na bin cussing. (Speaker 1)
e. We ain’t using no gun in the argument, you know, to settle it.
(Speaker 2)
f. Right now I don’t think I’ll walk alone in the dark. (Speaker 2)
Bequia English 137
g. Because if the fowls go and eat it, it would kill them. (Speaker 29)
h. We used to start Christmas as it were, say, from Nine Morning.
(Speaker 304)
Aspectual markers can combine with tense forms to yield more complex
distinctions: bin and V-ing together express past habitual (20), and did +
V may also express past habitual (21).
(20) When I bin going to school. (Speaker 1)
(21) There’s a lot of people did spoil [the] lobster industry. (Speaker 304).
3.2.5 Existentials
There are three main types of existential construction in Bequia: the
dummy subject there followed by some form of the verb be, as in Standard
English (be-existentials); or the dummy subject it, either with some form
of have (have-existentials) (30a–c), or with got (30d) (got-existentials).
3.2.7 Negation
Negation in Bequia English takes several forms, the most frequent being
not/-n’t, which is also found in Standard English, and ain’t (which may be
pronounced as [ɛ̃ː], [ɛn], [ɛnt]), found in many nonstandard varieties of
English and in English-based creoles. Other options are na and don.
(33) a. They en really have the oldest in our day [at school] (Speaker 306)
b. Oh God look he dead reach and nobody na know. (Speaker 6)
‘Oh God, look, he’s fallen dead and nobody knew.’
c. And when he is old, he is not depart. (Speaker 2)
d. Not even the moon. (Speaker 303)
e. If it hasn’t have eggs, all well. (Speaker 304)
In addition, never and na can be used to express ‘at no time; never’ and
‘not a’ as shown in the following examples.
3.3 Lexicon
In addition to the common vocabulary inherited from English, Bequia
English contains many words that do not occur widely in the English-
speaking world, although much of this vocabulary is found in other
Caribbean English varieties (or at least, in the eastern Caribbean). For exam-
ple, boilin is a local word for a soup made from fish heads and dumplings,
which is used elsewhere in the Caribbean to refer to fish-head soup (with-
out dumplings). Coastal trees in Bequia (and many other islands of the
Caribbean) are called sea grapes (coccoloba uvifera), because their fruit resem-
bles bunches of grapes. There are a number of words of likely African origin,
such as dukuna ‘corn pudding’, which may be derived from Gã-Adangme
doko na ‘sweeten (vb) mouth’ (Allsopp 2003: xxxiv, 207). In Bequia English,
when people are referring to the bad luck brought on by attention from
others, they talk of maljo(u), a term that occurs in varying forms through-
out the Caribbean and perhaps derives from French mal d’yeux ‘bad eyes’
(or Spanish mal de ojo or a Portuguese equivalent) (Allsopp 2003: 364).
Some words and phrases (like comess ‘gossip’ and jumbie ‘ghost’) are
also found in other Caribbean English varieties, though some are more
preferred in Bequia or Bequians use different words to refer to the same
thing. For example, what is known as suck-teeth throughout the Caribbean
(the ‘kissing’ noise made by intaking breath over closed teeth, a sign of
irritation or disapproval) is also called chups in Bequia English. Similarly,
we have recorded pickney ‘children’, which seems to be a more recent
adoption from elsewhere in the Caribbean. Finally, some of the words and
phrases that people use in Bequia English may be used in ways similar to
lesser-known dialects of English, such as to vex ‘to annoy, bother’ and to
teef ‘to steal’ (< thief ).
4 Conclusion
Although Bequia is a lesser-known island of the eastern Caribbean, it rep-
resents an interesting situation of dialect and language contact contained
within a relatively small geographic space. The diverse linguistic varieties
spoken in Bequia show contributions in phonology and morphosyntax
from nonstandard varieties of English (including British and Irish English,
as well as ‘white’ Caribbean English) and from English-based creoles,
with additional lexical contributions from other languages such as French
and Portuguese. Yet, despite over 150 years of intensive contact, the
geographically proximate communities of Bequia have managed to retain
142 james a. walker and miriam meyerhoff
unique ways of speaking, in the presence or absence of some features, and
in the distribution and conditioning of others.
The speakers we sampled in our research represent the older genera-
tion in Bequia, who acquired their way of speaking before the island was
opened up to outside influence and greater mobility. The tourism industry
in Bequia continues to expand, with more and more outsiders from North
America and Europe building homes and resorts on the island. As a result
of increasing access to higher education and exposure to other, more main-
stream linguistic models, there is some evidence that younger people are
losing some of the more distinctive features or have begun to use them in
ways that differ from that of their parents and grandparents (Daleszyńska
2012).
References
Adams, Edgar. 1996. Linking the Golden Anchor with the Silver Chain: A His-
torical and Socio-economic Perspective on Shipping in St Vincent and the
Grenadines. Kingstown, St Vincent and the Grenadines: R&M Adams Book
Centre.
Allsopp, Richard. 2003. Dictionary of Caribbean English Usage. Kingston, Jamaica:
University of the West Indies.
Daleszyńska, Agata. 2012. Variation in past tense marking in Bequia creole:
apparent time change and dialect leveling. PhD dissertation, University of
Edinburgh.
Meyerhoff, Miriam. 2000. Constraints on Null Subjects in Bislama (Vanuatu):
Social and Linguistic Factors. Canberra: Pacific Linguistics Publications.
Meyerhoff, Miriam and James A. Walker. 2007. The persistence of variation
in individual grammars: copula absence in ‘urban sojourners’ and their
stay-at-home peers, Bequia (St Vincent and the Grenadines). Journal of
Sociolinguistics 11: 346–66.
2012. Grammatical variation in Bequia (St Vincent and the Grenadines). Journal
of Pidgin and Creole Languages.
2013. Bequia English. Westminster: Battlebridge Publications.
Ng, Zoë. 2009. A social dialect study of (th) stopping in Bequia English. MA
thesis, University of Edinburgh.
Partridge, Andrew. 2009. Mapping the vowel space in Bequia creole. MSc disser-
tation, University of Edinburgh.
Price, Edward. 1962. The Redlegs of Barbados. Journal of the Barbados Museum
and Historical Society 29: 47–52.
Price, Neil. 1988. Behind the Planter’s Back: Lower-Class Responses to Marginality in
Bequia island, St Vincent. London: Macmillan.
Rochefort, César de. 1666. A History of the Caribby Islands. Rendered into the
English by John Davy.
Bequia English 143
Sheppard, Jill. 1977. The Redlegs of Barbados: Their Origins and History. Millwood,
NY: KTO Press.
Van Herk, Gerard. 2000. The question question: auxiliary inversion in early
African American English. In Shana Poplack, ed., The English History of
African American English. Oxford and Malden, MA: Blackwell, 175–97.
Walker, James A. 2010. Looking for agreement in the Eastern Caribbean: evidence
from Bequia. Paper presented at the annual meeting of the Society for Pidgin
and Creole Linguistics, Baltimore, MD.
Walker, James A. and Miriam Meyerhoff. 2006. Zero copula in the eastern
Caribbean: evidence from Bequia. American Speech 91: 146–63.
Walker, James A. and Jack Sidnell. 2011. Inherent variability and coexistent systems:
negation on Bequia. In Lars Hinrichs and Joseph Farquharson, eds., Variation
in the Caribbean: From Creole Continua to Individual Agency. Amsterdam and
Philadelphia: John Benjamins, 39–55.
Wells, J. C. 1982. Accents of English, vol. 3: Beyond the British Isles. Cambridge
University Press.
c h a p ter 7
Saban English
Jeffrey P. Williams and Caroline Myrick
1 Introduction1
Saba, like Bequia described by Meyerhoff and Walker in Chapter 6, is a
very small island in the Caribbean Sea whose sociolinguistic landscape has
been shaped by its size and isolation. Located in the Leeward chain of the
Lesser Antilles, Saba is approximately 45 kilometers south of St. Martin.
Formerly, Saba had been part of the Netherlands Antilles, a political unit
that was dissolved as an autonomous country within the Kingdom of the
Netherlands in 2010. Saba, along with Sint Eustatius and Bonaire, were
united as special municipalities of the kingdom.
With a total land area of 13 square kilometers, Saba is the smallest
inhabited island in the archipelago of the Leeward Antilles and one of the
smallest inhabited islands in the West Indies (see Map 7.1 ).2 The island
consists of the remnant of a volcanic cone and reaches a maximum elevation
of 887 meters at Mt. Scenery. Saba’s physiography is one of steep peaks
and facing cliffs dissected by ’guts’ (ghat or ghaut from Hindi) that lead
the extensive rainwater to the sea. The rocky, almost entirely impenetrable
coastline and the lack of any natural harbor also discouraged ships from
setting anchor at the island.3 The completion of the Leo A. Chance Pier
in 1972 enabled larger vessels, as well as smaller tourist vessels, to visit the
island. Prior to that, ships had to anchor offshore and passengers and crew
would take a dinghy to shore. Saba’s interaction with the outside world was
greatly enhanced by the construction of the Juancho E. Yrausquin airstrip
1 Fieldwork from which this chapter derives was conducted on several occasions over a period of just
over thirty years. Williams spent a month in the village of Windwardside in 1982 and revisited the
island in 2002 and 2004. Myrick spent a month doing fieldwork in Hell’s Gate, Windwardside, and
The Bottom in 2012 and 2014.
2 Petit Martinique – a dependency of Grenada – is the smallest inhabited island in the region, with
only 586 acres.
3 In a recent archaeological study, Espersen (2009) has shown that the now abandoned village of
Mary’s Point (also known as Palmetto Point) had a tide-dependent beach landing that facilitated
boat contact between the village and what were the Danish Antilles (US Virgin Islands).
144
ATLANTIC OCEAN
Green
Torrens Island
Flat Point
Torrens Bay
Point Airfield
Mary’s Point
Well’s Mtn 585 m Cove Bay
Bay Hell’s
Sandy Cruz Gate Spring Bay
Middle Island
Mt Scenery
Ladder (887 m)
Bay Rendezvous Core Gut Bay
Windwardside
The Gap
The Level
The Bottom
Booby Hill
St John’s
Giles Quarter
Fort Bay
0 800 m
Great Level Bay
0 0.5 mile
CARIBBEAN SEA
4 The pre-approved aircraft are limited to STOL (short take-off and landing) aircraft, which seat a
maximum of fifteen people.
5 On Williams’ first visit to the island in 1982, he spoke with older Saban women who had never left
their villages of birth.
6 Euro-Sabans are a relatively homogenous social group in opposition to Euro-descendants on other
anglophone islands where we find strong local differences between “clear-skinned,” “red,” and other
social classificatory groupings.
Saban English 147
white speakers in the historically white villages of Windwardside and Hell’s
Gate. Data from the village of The Bottom are from both black and white
speakers.
7 Williams wrote a Master’s thesis on “white” Saban English in 1984 and has provided information on
the Windwardside dialect embedded in various publications on the Euro-Caribbean varieties.
148 jeffrey p. williams and caroline myrick
strongly suggests that the anglophone population that arrived on Saba in
the time prior to 1659 had been transported to other islands in the West
Indies as indentured servants. The harsh conditions of indenturage, the
increasing dependence on African slave labor, as well as land shortages
forced many servants to seek refuge on the small, isolated islands of the
Caribbean Basin. In short, the English-speaking settlers had come to Saba
from various parts of the British Isles via St. Kitts, Barbados, and Antigua
during the exodus of indentured servants in the middle of the seventeenth
century. They had come to Saba to avoid the hardships and possible fatal
outcomes of indenturage. Due to its isolation and lack of colonial infras-
tructure, Saba was an ideal location for relocation; one where there was
little chance that the escapees would be caught and subject to harsh pun-
ishment and re-indenturage. The social circumstances that these escaped
servants found themselves in was identical to those that escaped slaves also
found themselves in during the height of forced labor transportation in the
Americas. These escaped servants, not unlike their African counterparts,
formed discrete, isolated communities and developed distinctive forms of
speech.8
When the English took possession of the island in 1665, Henry Morgan
removed the seventy Dutch-speaking colonists to Sint Maarten.9 In the
archival accounts, a population record from 1699 indicates that almost all
of the population was from the British Isles (Johnson 1979: 15). While we
have not seen the original report, we can venture to speculate that the
account might not be completely accurate since many of the previously
Dutch surnames were anglicized early in the settlement history of the
island.
In spite of several dispossessions by England and periods of English
rule, Saba has remained part of the Hollandish sociopolitical sphere since
the mid seventeenth century. Saba finally became a permanent Dutch
possession in 1816. As the linguistic evidence shows, Dutch rule has not
translated into Dutch linguistic or sociolinguistic influence on Saba. As part
of the 2010 Acts that changed the political status of Bonaire, St. Eustatius,
and Saba (Wet op de Openbare Lichamen Bonaire, Sint Eustatius en Saba),
the Kingdom of the Netherlands made the use of English by residents of
Saba and St. Eustatius for official business and governmental transactions
formally recognized.
8 For further discussion that draws parallels between escaped African slaves and escaped European
servants, see Beckles (1986).
9 Additionally, 102 African-descended slaves were transported to Jamaica at the same time.
Saban English 149
Een ieder kan de Nederlandse taal gebruiken in het verkeer met de in artikel
4b, eerste lid, bedoelde organen en personen.
2. Een ieder kan:
a. het Papiaments gebruiken in het verkeer met de organen van het openbaar
lichaam Bonaire;
b. het Engels gebruiken in het verkeer met de organen van het openbaar lichaam
Sint Eustatius of Saba.
[Anyone can use the Dutch language in communications with the bodies
referred to in Article 4b, first paragraph, and people.
2. Anyone:
a. can use Papiamentu with official public offices and persons representing
those offices in Bonaire;
b. can use English with official public offices and persons representing those
offices in Sint Eustatius and Saba.]
10 The opening of a medical school in the early 1990s has changed the sociolinguistic profile of the
island and increased the resident population by approximately 30 percent.
11 The village of Palmetto (or Mary’s) Point was relocated between 1920 and 1934. Oral history provides
one set of explanations for the relocation while archaeological evidence provides another. Espersen
(2009) provides a detailed historical archaeology for the site, with a sophisticated nuancing of
oral histories and comparative information regarding other isolated, enclave communities in the
colonial Americas.
150 jeffrey p. williams and caroline myrick
were seemingly based on both ethnicity and color within a colonial social
framework.12
Like I said, in each village ’tis different. Also, less than 1000 people . . . St.
John’s, those people they talkin’ more like Irishmen. In the Bottom, well
they have different. They have-the old white ones, those who are English,
they spoke London English, or tried to. [Windwardside, W/♂/ma/1982]13
This local exegesis on ethnodialectology remains unchanged as the follow-
ing transcription of a recording over thirty years later shows.
Well . . . The Bottom I mean talks a little different to the people over
here . . . Yeah. The St. John’s people, they talk a little different to The Bot-
tom people too. Yeah, you can listen to – you can hear it, I mean. Well the
Hell’s Gate and Windwardside, you know, it’s the same thing, you know?
Difference. [Hell’s Gate, W/♂/ea/2012]
The varieties spoken in the historically Euro-descended villages of Wind-
wardside, Hell’s Gate, St. John’s, and Mary’s Point belong to the Euro-
Caribbean Anglophone Linguistic Area (ECALA) as defined in Williams
(2012). The hallmark of these varieties is their koineized nature, exhibit-
ing a structured integration and complex variation of forms drawn from
regional source input dialects. Euro-Saban varieties are no different in these
regards. Color has been a key social marker in Saban identity and it has
governed settlement and marriage as it has throughout the West Indies
throughout the colonial and neo-colonial periods.
It is evident that in the founder period of settlement of Saba, a great deal
of dialect and language contact took place. Plantations never developed on
Saba and the social economy remained at the founder stage of Mufwene’s
development scheme. Minimally, the monolithic varieties of Dutch and
English were in contact as well as an unknown number of African languages
12 In his study of sojourners in the West Indies and the northeast coastal United States, Karras (1992)
discusses the pattern of ethnic segregation that was prevalent during the late colonial period.
13 Each recorded example provides the following information on the speaker: [village, ethnicity/sex/age
group/year recording was made]. The following abbreviations and conventions are used:
ad: adolescent, approximately between the years of 14 and 20
c: child, approximately between the years of 0 and 9
e: elder, approximately beyond 66 years of age
ma: middle age, approximately between 36 and 50 years of age
oa: older age, approximately between 51 and 65 years of age
pread: preadolescent, approximately between 10 and 13 years of age
w: “white” speaker; identifies as an individual of European descent
ya: young adult, approximately between 21 and 35 years of age
♂: male speaker
♀: female speaker
Saban English 151
that were brought to the island with the slave population that came during
the latter part of the seventeenth century. The Dutch were not a social
factor for long in the sociolinguistic history of the island. However, the
English-speakers increased in number and importance.
The outcome of the dialect contact that occurred involved such factors
as ethnicity, kinship, locality, and demographic proportions. Each village,
being isolated developed its own sense of identity: a composite drawn from
those individuals who had come together to make up that community.
Euro-Sabans have continued to see themselves as descended from set-
tlers from the British Isles, and not from the Dutch as the Netherlands
administration has attempted to portray them as.14 In point of fact, when
Saba became a permanent possession of the Netherlands in 1816, there was
formal stipulation that all official documents be translated into English
since there were no speakers of Dutch on the island (Hartog 1975: 22). This
act of identity has had linguistic consequences for the Euro-Sabans.
Although all Sabans are taught Dutch in school from the first grade on,
they do not use it among themselves in ordinary conversation because it is
associated with an identity that is distinctly non-Saban. The use of Dutch
on the island is restricted primarily to the classroom, to interaction with
native speakers of Dutch, and to some governmental and official events.
Since Dutch was the official language of the Netherlands Antilles until
the addition of Papiamentu and English in 1984, up until that time birth
records, passports, and the only available newspaper were in that language.
Even though Sabans have been very competent speakers of Dutch and
make use of opportunities to use the language in their interactions with
Netherlanders, the language itself would not have survived for the time that
it has were it not for the political system that mandated primary instruction
in Dutch.
With the development of a Saban national identity within the larger
framework of West Indian identity has come the genesis of a Saban
linguistic identity as well.15 This has brought about the development of
another focused variety of Saban English that can be identified, particu-
larly by non-Saban West Indians as a discrete variety. This development has
been accompanied by Sabans’ increased interaction with other Caribbean
14 See Johnson 1979 for a full discussion of this point as well as recent discussions in The Saba Islander –
a blog/newspaper published by Mr. Will Johnson of Saba.
15 Euro-Saban varieties of English might have influenced Bermudian English. After the outbreak of
the Boer War in South Africa, the English established prison camps on the island of Bermuda.
In 1902, over one hundred Sabans were employed at the camps. Many of those chose to remain
on Bermuda, and in the 1970s there was a population of nearly 200 Sabans residing on Bermuda
(Hartog 1975: 68).
152 jeffrey p. williams and caroline myrick
nationals, the recent opportunities for social and geographical mobility, and
the constantly increasing number of tourists who visit the island annually.
Sociolinguistic focusing that had taken place at the level of regional, i.e.
village, accents and dialects is in the process of becoming diffuse through
the process of the interaction of those factors. Many Sabans that we have
spoken with over the years have told us how the distinctive dialects were
being lost by the younger Sabans due to their extended periods of residence
off the island. In some regards, the dense local networks have been in a state
of decay, giving way to the creation of new, less dense networks that span
the geographical and social diversity of the island, and may even connect
individuals to another island such as Curacao, Aruba, or St. Eustatius, or
even another nation outside the Caribbean such as Canada or the United
States.
dress
Saban English shows lowering of the dress vowel, with bet being realized
as low as [draes], or somewhere in between [ɛ] and [ae].
trap/bath
The trap/bath set is lowered and backed, occupying the lowest vowel
space in Saban English. In the speech of the older residents of Hell’s Gate,
foot
Saban English demonstrates lowering of the foot vowel. In fact, the foot
vowel is realized lower than the boat vowel. There is evidence of additional
centralization of the foot vowel in the Hell’s Gate variety, causing most
lexical items with the foot vowel to be realized with the strut vowel.18
We find examples such as took being realized as [tʌk]. In the village of
Windwardside, the strut vowel is lowered to bought space, with tough
realized as [tɔf].
strut
Acoustic analysis of Saban English shows absence of a central vowel. It is
noteworthy that Saban English speakers do tend to centralize unstressed
syllables as schwa, supporting Williams’ (2012: 147) position that "[n]on-
weakening of vowels in unstressed syllables is not a feature of the Euro-
Caribbean Anglophone Linguistic Area."
lot
For those speakers who do not show a lot – thought merger, the lot
vowel is realized as /ɑ/ For many speakers, this vowel shows raising and
backing, and is close to merged with the thought vowel (realized as
/ɔ/). Saban Englishes differ from the General American lot – thought
mergers by approximation, in which the thought vowel tends to lower
towards lot.
3.1.2
Diphthongs
face
The face vowel is monophthongal; in older Hell’s Gate speakers, pre-nasal
fleece is realized as the pre-nasal face vowel, so that mean and main are
homophonous.
near
The near and square vowels are close/merged, so that fear and fair are
homophonous.
17 Vowels in pre-nasal position are not treated any differently in Euro-Saban English. In other words,
we do not find pre-nasal tensing of trap/bath vowel like we would see in North American English.
18 This pattern is reminiscent of the put–putt merger we find in Scottish English.
154 jeffrey p. williams and caroline myrick
price
price is dipthongized; the nucleus is fronted to either [ʌ] or [ɛ], so that
price is realized as [prʌɪs] or [prɛɪs]. The nucleus of the price vowel
occurring before voiceless consonants (e.g. tight) is higher than when
before voiced consonants (e.g. tide).
choice
The nucleus of the choice vowel is lowered, so that void is [vʊɪd], and
sometimes fronted, so that void is [vʌɪd].
goat
goat [ou] is monophthongal; usually lengthened (duration), e.g. most as
[mo:st], or with an offglide of [ə], e.g. most as [moəst].
(1) whole road ➔ [ho:l ɹo:d]
(2) boat ➔ [boət]
mouth
In the mouth set, the nucleus is backed and raised to /ɔ/ so that house is
realized as /hɔʊs/.
goose
goose is monophthongal and very backed for older speakers. In younger
speakers, there is evidence that goose is moving forward.
nurse
In rhotic words with stressed nuclear [r], the nurse vowel merges with the
force vowel (Windwardside) or the strut vowel (Hell’s Gate). When
r-less, the vowel is realized as [ʌ], e.g. nurse as [nʌs].
start
The start vowel is realized as [ɑr] and is homophonous with the north
vowel.
Saban English 155
force/north
The force/north distinction/split (i.e. horse/hoarse distinction) has
been preserved on Saba, most strongly in the variety spoken in the village
of Hell’s Gate, but it appears to be undergoing merger by approximation.
Words with post-back vowel /r/ are typically r-ful in Saban English (see
/r/ discussion below); some lexical exceptions include the words more and
farm, which are typically r-less.
3.1.4 Consonants
3.1.4.1 Initial /h/
Word initial h-dropping is common in Euro-Saban dialects. When initial
h is present it is realized as [j] as in the pronunciation of Hugo as "You-go."
This pattern is most common when there is a following consonant.
3.1.4.2 Despirantization
The processes of despirantization, where interdental fricatives are real-
ized as voicing equivalent plosives, is common in all varieties of Saban
English. Intervocalically and phrase finally, most are glottalized, e.g. math
as [maʔ] and birthday and [berʔdei]. In the section below we elaborate on /t/
glottalization.
3.1.4.5 /r/
Saban English is not classifiable as purely rhotic or non-rhotic due to a
two-way split for /r/. Post-vocalic /r/ is disfavored in unstressed syllables
and when following a front vowel, while /r/ is favored following a nuclear
/r/ or a back vowel (with the exception of lexical items more and farm).
Based on mixed-effects models run in Myrick (2014), Saban English thus
156 jeffrey p. williams and caroline myrick
shows the following hierarchy of /r/ favorability: unstressed syllable < post-
front vowel < nuclear /r/ post- back vowel.
(4) BA’ELY any CARS [Windwardside, W/♀/ma/2012]
(5) the YEA’ I was BORN [Windwardside, W/♀/ma/2012]
(6) in the sand UNDA’ WATA’ [Hell’s Gate, W/♂/ma, 2012]
While /r/ vocalization can occur in word-final position when followed
by a consonant or a vowel, a following consonant favors r-lessness over a
following vowel.
Speakers with higher education (i.e. beyond the seventh grade) tend
to be more rhotic overall than speakers with lower education, supporting
Williams’ (2010) suggestion that level of education correlates negatively
with degree of rhoticity in speakers of Euro-Caribbean Englishes.
3.1.4.6 Metathesis
Like other varieties of Euro-Caribbean English, Euro-Saban English evi-
dences lexicalized metathesis in forms such as (7) through (9). The feature
is common across communities and ethnicities on Saba.
(7) ask ➔ [æks]
(8) sistern ➔ [sɪstrn̩]
(9) pattern ➔ [pætrn̩]
3.2 Morphosyntax
3.2.1 Pluralization
Not unlike other Euro-Caribbean varieties (see Williams 2010), pluraliza-
tion is variably marked in Euro-Saban English varieties. Plural nouns are
variably marked with the Standard English -s suffix as in (10) through (13)
below.
(10) The bananas is put to come ripe. [Windwardside, W/♂/ch/1983]
(11) I never lock no doors. [Windwardside, W/♀/ma/1983]
(12) While we to do the lightin’ of the candles. [Windwardside,
W/♀/ma/1983]
(13) All them bones is broken. [The Bottom, W/♂/oa/2012]
Saban English does, however, use plural -s absence for count nouns
(e.g. mile, year, minute) that follow a quantifier.
(14) sixteen year I spent there [Hell’s Gate, W/♀/oa/2012]
(15) half mile wide and six mile long [Hell’s Gate, W/♂/oa/2012]
(16) she figure me ten cent [The Bottom, W/♂/oa/2012]
As noted in Williams (2010), Saban English does not make use of postpo-
sitional dem for pluralization. Undoubtedly, this is due to lack of develop-
ment of a Creole-speaking population on the island.
3.2.2 Pronouns
The pronominal systems of Saban varieties of English make use of the
pronouns found in Standard English but with different case realizations
with some exceptions. Older speakers, in general, make use of subject
pronouns in all positions.
(17) Them cost plenty money. [Windwardside, W/♂/ch/1983]
(18) It was hard for he. [Hell’s Gate, W/♂/ma/2012]
(19) Took we down. [The Bottom, W/♂/oa/2012]
(20) . . . leave she in charge of the church. [The Bottom,
B/♀/oa/2012]
In addition to the pronouns found in Standard English, Saban English
incorporates a second-person plural pronoun, aayu, stemming from an
158 jeffrey p. williams and caroline myrick
assimilated version of “all you.” It is used equally by all speakers (i.e. it
appears to be a stable feature), and typically appears in both interrogative
and declarative sentences, although only in subject form. Aayu can address
a specific pair or group of interlocutors, or refer to an abstract group (e.g.
all Americans).
(21) Aayu has a Social Security thing [Hell’s Gate, W/♂/oa/2012]
(22) Aayu know when you . . . boil custard? [Hell’s Gate,
W/♂/ma/2012]
(23) Aayu don’t have this in America? [The Bottom, B/♂/ch/2012]
3.2.3 Possession
Object-form for possessive-form occurs with first-person singular posses-
sion, shown in examples (25) and (26), and subject-form for possessive-form
occurs with third-person singular possession, shown in example (24).
(24) . . . had a problem with he hips [Hell’s Gate, W/♂/oa/2012]
(25) He’s me family anyhow [The Bottom, W/♂/oa/2012]
(26) . . . with me heart [Hell’s Gate, W/♀/oa/2012]
19 It could be argued that what is absent in the phrase is the past-tense marker, rendered in Standard
English as “He never answered me.” The semantics of the phrase in Euro-Saban English is that the
individual did not answer a specific question and the time frame remains open for him to provide
an answer.
Saban English 159
third-person present indicative -s is reallocated as one grammatical resource
to signal habitual aspect.
3.2.4.2 Aspect
Aspect is a diagnostic feature of the anglophone Caribbean and has three
exponent features in the region: habitual, progressive, and completive.
Euro-Saban English, not unlike other Euro-Caribbean varieties, evidences
multiple exponence in some categories.
3.2.4.2.1 Habitual
Euro-Saban English has multiple exponence in the habitual category that
is most likely a product of earlier sociolinguistic patterns that have been
blended through village interactions over time.
(28) Them be’s helping their father with so much. [Hell’s Gate,
W/♂/oa/2012]
(29) I don’t be with the hot water. [Hell’s Gate, W/♂/ma/2012]
(30) That how come I be without a walk. [The Bottom,
W/♂/oa/2012]
(31) It don’t be nice. [The Bottom, B/♀/oa/2012]
Verbal -s One predominant feature of Euro-Saban English is the use of
verbal -s in all persons to signal habitual aspect as the examples which
follow demonstrate.
3.2.5 Copula
The copula is typically not present in Euro-Saban English. Copula absence
may occur before the following environments, shown in the examples
below: progressive verb (48), adjective/adjective phrase (49), noun phrase
(50), preposition/prepositional phrase (51-52), and quotative like (53).
(48) I Ø going with her [Hell’s Gate, W/♀/c/2012]
(49) You Ø like an old woman! [The Bottom, B/♂/c/2012]
Saban English 161
(50) Who Ø your family? [Hell’s Gate, W/♂/ma/2012]
(51) He Ø from Saba too [The Bottom, W/♂/oa/2012]
(52) I know where you Ø from [Windwardside, W/♂/ma/2012]
(53) So I Ø like, “Mommy? Where Ø we going?” [Windwardside,
B/♀/c/2012]
(54) While we to do the lightin’ of the candles. [Windwardside,
W/♀/ma/1982]
(55) I’s the onliest one here. [Windwardside, W/♀/ma/1982]
3.2.6 Questions
In older and more vernacular speech, questions are marked by intonational
patterns and not through subject–auxiliary inversion. Younger generations,
however, show a higher preference for subject–auxiliary inversion.
(56) What his name is [Hell’s Gate, W/♀/c/2012]
(57) You have been to see it [Hell’s Gate, W/♂/oa/2012]
3.2.6 Negation
Negation in Saban English has two main forms, the most common being
ain’t ([ɛː̃ ] [ɛn] [ɛnt]) that is found in many of the nonstandard and
lesser-known varieties of English as well as in English-based creoles. There
is frequent use of ain’t in the vernacular for singular and plural number as
well as being in place of didn’t and haven’t. Standard English preverbal not
is also used in less vernacular contexts, where either social circumstances
or less dense network ties influence the shift.
(58) There ain’t too many bad things [Hell’s Gate, W/♂/oa/2012]
(59) That ain’t true [Hell’s Gate, W/♂/c/2012]
(60) He ain’t send me nowhere. [The Bottom, W/♂/oa/2012]
(61) I ain’t say nothing [Hell’s Gate, W/♀/oa/2012]
(62) She ain’t decide yet. [Windwardside, W/♀/oa/2012]
3.2.7 Prepositions
Euro-Saban English varieties evidence prepositional usage that is not found
in Standard Englishes. Locative prepositions tend to be omitted before
macro locations, such cities, countries, other islands, as well as the villages
162 jeffrey p. williams and caroline myrick
on Saba. Examples (63) and (64) show examples of omissions before island
and state names.
(63) The other two was born Ø Saba [Hell’s Gate, W/♂/oa/2012]
(64) I went Ø Sint Maartens for school [Windwardside,
W/♀/oa/2012]
When referring to micro locations, Saban English speakers typically employ
by or to for "at." These locations include homes, stores, restaurants, and
landmarks. The generalization of by to mean both "by" and "at" may be
a substrate effect from Dutch (as Dutch, like German, has only one word
to mean both "by" and "at"), supported by evidence of the same substrate
effect seen in the English of the Pennsylvania Germans (Wolfram, personal
communication, April 11, 2013) and Yiddish speakers (Benor 2012).
(65) The party was by my house. [Windwardside, W/♀/c/2012]
(66) there by Saba Treasures (the restaurant) [The Bottom,
B/♂/c/2012]
(67) the reception was kept by her parents’ home [Hell’s Gate,
W/♂/oa/2012]
(68) She teaches me to school. [Windwardside, W/♂/c/1982]
(69) (the house) right over to the end [Hell’s Gate, W/♀/oa/2012]
(70) Down here to the harbor [The Bottom, W/♀/oa/2012]
(71) She teaches me to school. [Windwardside, W/♂/c/1982]
(72) We just shared it to the neighbors. [Hell’s Gate, B/♀/ma/2012]
3.2.8 Complementizers
Euro-Saban English also makes extensive use of for to complementizer
constructions.
(73) They is ready for to come ripe. [Windwardside, W/♂/ma/1982]
3.3 Lexicon
Vocabulary of Saban English varieties exhibits similarities to other
Caribbean English varieties, both creole and non-creolized.
(74) drop ‘to say’
(75) smokey ‘foggy’
Saban English 163
(76) remind ‘remember’
(77) from ‘since’
(78) molest ‘annoy or bother’
(79) ground ‘field’
(80) awoy/‘woy ‘a greeting’
4 Conclusion
The varieties of Saban English are some of the more robust in the Caribbean
English sociolinguistic landscape, even in spite of the island’s small size.
Relatively stable populations and favorable economic conditions have sup-
ported the maintenance of the local populations, although emigration for
education and employment has impacted males to a higher degee than
females. An emerging Saban identity has developed throughout the colo-
nial period and contributed to a projected Saban sociolinguistic identity,
which has valued the vernacular.
Throughout its European historical period, isolation has played a key role
in the trajectories of Saban Englishes. Isolation, in the Caribbean context,
is not an absolute. Sabans we have spoken with described interactions with
other islanders, both on Saba and on other islands. The expanses of sea that
separate Saba also join it to other sociolinguistic communities. Large-scale
external impacts from tourism or immigration have not taken place on
Saba, as the island still does not have the capacity to receive larger aircraft,
with the world’s shortest commercial runway at just under 400 meters
in length. These social conditions have coalesced so that Saban English
varieties are not highly endangered, which also sets them apart from many
other lesser-known Englishes worldwide.
References
Beckles, Hilary. 1986. From land to sea: runaway Barbados slaves and servants,
1630–1700. In Gad J. Heuman, ed., Out of the House of Bondage: Runaways,
Resistance and Maroonage in Africa and the New World. London: Frank Cass,
97–4.
Benor, Sarah. 2012. Becoming Frum: How Newcomers Learn the Language and
Culture of Orthodox Judaism. New Brunswick, NJ: Rutgers University Press.
Crane, Julia. 1971. Educated to Emigrate: The Social Organization of Saba. Amster-
dam: Van Gorcum.
164 jeffrey p. williams and caroline myrick
1987. Saba Silhouettes: Life Stories from a Caribbean Island. New York: Vantage
Press.
Eliason, Eric A. 1997. The Fruit of Her Hands: Saba Lace History and Patterns.
Saba: Saba Foundation for the Arts.
Espersen, Ryan. 2009. From folklore to folk history: contextualizing settlement
at Palmetto Point, Saba, Dutch Caribbean. Master’s thesis, University of
Leiden.
Hartog, Jan. 1975. History of Saba. Saba: The Saba Artisan Foundation.
Hickey, Raymond. 1999. Ireland as a linguistic area. In James P. Mallory, ed.,
Language in Ulster. Holywood, Ireland: Ulster Folk and Transport Museum,
36–53.
Johnson, Will. 1979. Saban Lore: Tales from My Grandmother’s Pipe, 3rd edition.
Published by the author.
Karras, Alan L. 1992. Sojourners in the Sun: Scottish Migrants in Jamaica and the
Chesapeake, 1740–1800. Ithaca: Cornell University Press.
Myrick, Caroline. 2014. Putting Saban English on the map: A descriptive analysis
of English language variation on Saba. English World-Wide 35(2): 161–92.
Williams, Jeffrey P. 1984. White Saban English: a socio-historical study. Master’s
thesis, University of Texas at Austin.
2010. Euro-Caribbean English varieties. In Daniel Schreier, Peter Trudgill, Edgar
Schneider, and Jeffrey P. Williams, eds., The Lesser-Known Varieties of English:
An Introduction. Cambridge University Press, 136–57.
2012. English varieties in the Caribbean. In Raymond Hickey, ed. Areal Features
of the Anglophone World. Mouton: de Gruyter, 133–60.
c h a p ter 8
1 Introduction
St. Eustatius is an English-speaking island in the Dutch Caribbean. This
dialect displays a handful of correspondences with other Englishes spoken
in geographically proximate areas, but what is most noteworthy about this
dialect is that so much of its grammar is significantly different from many
of those same nearby varieties. Historical, linguistic, and ethnographic
data are interwoven to make the case that Statian English sounds different
from most other Englishes of the Caribbean basin because the colonizing
and settlement patterns of the island differed from plantation societies
focusing on the production of cash crops. St. Eustatius was a commercial
center instead, offering an entrepôt for goods (and, at times, slaves) for sale
to customers from the eastern rim of the Americas. In this import–export
context, English as a lingua franca of trade emerged with its own distinctive
features.
St. Eustatius English is a dialect that has never been the focus of a single
piece of published research until Aceto (2006), even if it has been alluded
to in other works (e.g. Hancock 1987; Holm 1988–9: 452–5). Williams
(1983: 95) notes that there are no published sources on the contemporary
Englishes of the Windward Netherlands Antilles (i.e. Saba, St. Martin,
and St. Eustatius), and this research characterization has not changed
much in the last nearly thirty years. Hancock (1987) presents fifty kernel
Statian phrases and sentences (displayed among thirty-three other English
varieties). Holm (1988–9: 452–5) contains a short section on Statian. Parsons
(1933–43) includes St. Eustatius in her wide range of West Indian folklore
presented.
This chapter briefly examines the grammatical features of English on
St. Eustatius. The language data is based on several weeks of fieldwork
I would like thank the people of St. Eustatius who were generous in sharing their time and language
with me. Any errors or shortcomings are, as always, mine alone.
165
166 michael aceto
conducted on the island in the summer of 1998. The historical presentation
and linguistic data are more expansive in Aceto (2006). The present chapter
summarizes Aceto (2006) and then provides a discussion that explores the
role that Statian English played in the Americas and what dialect and creole
studies can learn from this English variety.
2 Background
St. Eustatius lies on the eastern edge of the Caribbean Sea approximately
40 miles to the south of St. Martin. Today, the residents of St. Eustatius
are, for the most part, people of African descent, though historically this
was not always the case (as is discussed below). There is some immigration
from Papiamentu-speaking islands such as Aruba and Curaçao that are also
part of the wider Dutch Antilles as well as from geographically proximate
English-speaking areas such as Nevis and St. Kitts. Only one or two flights
a day in small propeller planes connect Statia with St. Martin, which is
one hub for air flights in the eastern Caribbean. Before this service was
established in the last fifty years, one could only travel to the island by
boat.
In 1933 Dutch became the official language of the public school system
in St. Eustatius; in 1976 English was made the language of public educa-
tion. However, Dutch is again the language of education, though some
English language instruction unavoidably and undoubtedly occurs since,
for those who are locally born, Dutch is no one’s native language. Statians
have changed their minds at the voting booth several times in the twenti-
eth century about which language of instruction they would prefer general
classes to be held in, either English or Dutch. I was surprised by how
strongly locals felt that they should know some form of Dutch since (as
was explained to me) they are part of the Dutch Antilles. One consultant
even referred to the Netherlands as the mother country of Statia. Promis-
ing students who are more fluent in Dutch often go to the Netherlands
for study. Statian English is most folks’ native language, except for some
Papiamentu speakers, who also know at least some of the local vernacular,
and possibly several administrators and educators who originally arrived
from the Netherlands.
4 Demographics
From the population peak of 8,124 persons (approximately 63 percent of
the population as slaves) in 1790, the number of Statia’s residents began to
dwindle. Once the economic base deteriorated at the end of the eighteenth
century, few immigrants moved to the island and most folks of European
descent moved away. After emancipation in 1863, many of the African-
descended males began to move about the Caribbean and the Americas
in general (often as sailors on whaling vessels; Crane [1999: xxiii]) in
search of work. This itinerant labor-related pattern continues to this day.
The following demographics corroborate this characterization (all figures
are from Hartog 1976: 102, 127, 134, except as indicated). In 1818, 2,668
3 Keur and Keur (1960: 40) assert that the population of St. Eustatius in 1780 was between 20,000
and 25,000, a figure that seems hard to reconcile in terms of housing and infrastructure for anyone
who has visited this small island; furthermore, such a large population on a relatively small island
would surely have been mentioned by Schaw in her journal (see Andrews and Andrews 1934).
4 For comparison, in the same year, nearby Dutch St. Martin contained 4,230 slaves and 1,290
Europeans and freed persons (Hartog 1976: 52).
170 michael aceto
Table 8.1 Results of 1974 Census in St. Eustatius
according to place of birth
5 Some of the social contexts of card playing and dominoes for my audio recordings involved at least
four and as many as six informants being recorded at once.
6 However, another consultant insisted on two occasions that her English variety was no different
than mine even though, from a descriptive perspective, her variety of English exhibited nearly all of
the general features presented in this chapter.
172 michael aceto
his research on the Sea Coast Islands, one Gullah speaker told him, “You
call [our vernacular] Gullah; we call it English.”
There exist relatively little linguistic data on Statian English. The work
of Elsie Clews Parsons (1933–43: 376–86) reveals some folk tales from Statia
which largely corroborate the data I gathered on the island. From both
my fieldwork and Parsons (1933–43) Statian English emerges as a dialect of
English, similar in regards to many features to African American Vernac-
ular English (AAVE). In fact, one may notice that Statian English often
seems like a Caribbean version of AAVE. That is to say, there are few
features of Statian English that are so-called “creole” features though there
are at least three that may be considered so: unmarked verbs interpreted
as past, the preverbal future marker go, and the post-nominal pluralizer
dem. In the last ten or so years, it has been asserted that so-called creole
languages manifest structural characteristics different from non-creole lan-
guages (McWhorter 1998, 2000; Parkvall 2001),7 but many researchers have
remained unconvinced that these assertions/alleged diagnostic structural
features are exclusive to the group of languages researchers call creoles (Plag
2001). Rizzi concludes “creoles do not look different from other natural
languages in any qualitative sense” (1999: 466). Many creolists would prob-
ably agree with Mufwene (2000, 2001) that creolization is a social process
and not a structurally defined one. I will make reference below to Parsons’s
ten pages of tales to corroborate my own data and/or to demonstrate that
she found features that I was unable to record in my naturally occurring
discourse or interviews.8
At least two language-external factors have greatly impacted the for-
mation and emergence of Statian English and illustrate why this variety
may appropriately be considered as a dialect of English (if not all so-called
English-based or -derived creoles as well; see Mufwene 2008).9 First, there
seems to have been a significant segment of the local island population who
were first-language speakers of what was emerging in the eighteenth century
as North American English varieties (despite the fact that Statia was a Dutch
7 I have always found McWhorter’s (1998) analysis less than compelling because it does not discuss fea-
tures that these languages display but rather features they lack; thus this sort of analysis is predisposed
to finding so-called creole languages comparatively deficient or lacking. Thus, subsequently desig-
nating them as “simpler” than other human languages seems only a step away from first presenting
them as “lacking” specific features.
8 It is important to remember that folk tales deliberately performed in “story-telling mode” may tend
to invoke features not normally associated with everyday vernacular language.
9 However, the term creole is still a useful one to designate the sociohistorical contexts in which many
of these languages emerged (Mufwene 2000).
St. Eustatius English 173
colony). Furthermore, a significant segment of the European-derived pop-
ulation was also comprised of second-language speakers of English (i.e.
Africans and their descendants were not the only persons acquiring a vari-
ety of English). Lastly, the demographic history of St. Eustatius reveals a
less disproportionate ratio between those of African and European descent
than was often found in other more typical creole-speaking plantation
areas such as Jamaica, Antigua, and even Suriname. That is, in Statia,
Africans and persons of African descent in subsequent generations had
a greater probability of hearing first- and second-language varieties spo-
ken by Europeans, Africans, and slaves born locally than in many other
anglophone Caribbean locations. Africans and persons of African descent
only comprised as much as one half of the population (ranging to a high
of approximately 65 percent) at various times in the island’s history. It is
often an overlooked fact in creole studies that many Englishes emerged
in the Americas, among peoples of both African and European descent,
that are structurally closer to dialects of the lexifier, e.g. varieties spoken
in the Bahamas (see Childs, Reaser, and Wolfram 2003), the Turks and
Caicos islands (see Cutler 2003), Anguilla (see Williams 2003), probably
the Cayman islands, and AAVE just to name a few of the likeliest candi-
dates (see below and Aceto 2003 for a discussion of what he calls dialect
creole varieties).
Hancock (1987) provides one of the few printed sources of Statian lin-
guistic data. Those data reveal a prevalent occurrence of the suppletive “to
be” form in Statian English: [wɑz] “was,” e.g. [tri ə hi frɛn waz dɛʌ] “Three
of his friends were there” (283), [ɪf yu wɑz stɪl di li:də] “If you were still
the leader” (321), [a jɔs wɔz čatɪn] “I was merely chatting” (322), which
was corroborated by my data and Parsons (1933–43: 376, 378, 381–3, 386) as
well. Common constructions such as completive aspect done (e.g. [ɑi dʌn
šat mai han]) “I already played out my hand (in cards)” and possessive con-
structions with successive noun phrases without inflectional morphology
(e.g. [de kaa] “their car”) are heard on the island, but these constructions
are present in a number of other Englishes as well (e.g. general Southern
American English and AAVE). Many other so-called “creole” features are
not found in this variety, as is discussed below. The following is a short
presentation of features found in Statian English based on my fieldwork,
Parsons (1933–43), and Hancock (1987).10 The following presentation does
10 Crane (1999) transcribed a series of autobiographical sketches for her informants. She then sent the
individual chapters back to the informants for their approval and corrections. Though it is possible
to glean some vernacular usages from this work, most of the stories are presented in a variety close
to general standards of written English.
174 michael aceto
not pretend to be exhaustive. For more a more complete data set, consult
Aceto (2006).
12 Parsons (1933–43: 376) reveals that one of her informants for St. Eustatius, though locally born,
lived in St. Kitts for forty years; another informant was born in St. Kitts and lived there for some
unspecified amount of time before moving to Statia.
178 michael aceto
according to the 1974 census (see above). Past progressive constructions are
typically indicated by was + Verb-in.
(15) [ai goɪn nau]
“I’m going now.”
(16) [a train tu sii]
“I’m trying to see.”
(17) [a jʌs pɪkɪn ə fait nau]"
“I’m just picking a fight now.”
(18) [hu a bɛt]
“Who’s betting?”
(19) [das wai am lukɪn ta tɛl ya wan taim]
“That’s why I’m looking to tell you once.”
(20) [a tɛl ya, de ə du ratɪn θings tu ya, don de]
“I told you, they are doing rotten things to you, aren’t they?”
(21) [we de də luz ɑniθɪŋ ya wan]
“What are they losing? Anything you want.”
(22) [a woz waakin ɪn di rod]
“I was walking in the road.”
5.2.1.4 Habituality
The use of preverbal doz to indicate habitual aspect is often considered diag-
nostic of anglophone eastern Caribbean varieties (see Holm 1988–9: 158–
60). However, it appears to be absent from Statian English.13 The geograph-
ically proximate island of Saba, which shares a similar Dutch/anglophone
13 Habitual doz is generally thought to be absent from Jamaican Creole English, but I have documented
it in speakers from Bastimentos, Panama, whose ancestors were largely immigrants from Jamaica
(as well as from Providencia and San Andres) in the late nineteenth and early twentieth centuries.
St. Eustatius English 179
history, also appears to lack this feature. In Statia this aspectual distinction
is indicated by bare or unmarked present tense verbs (e.g. [ši sii ši brʌdə]
“she sees her brother (on weekends)” (Hancock 1987: 288) as it is in many
varieties of English in the Americas. Additionally Statian English may add
adverbials such as aal taim “all the time” or aalweiz “always” to indicate
habitual, continuous, or repetitive actions. Past habitual actions are indi-
cated by used to [yuustu], which is confirmed by Parsons (1933–43: 383).
However, as is common in many Englishes of the Americas, progressive
and habitual aspectual strategies often semantically overlap; consequently,
it seems that habituality in Statian English can also be signaled by aspectual
forms such as preverbal a and the suffix -in as the following data illustrate:
(23) [a tɛl ya, de ə du ratɪn θings tu ya, don de?]
“I told you, they are doing rotten things to you, aren’t they?”
(24) [yu hav siin ɪr laŋ taim]
“Have you been seeing her for a long time?”
5.2.1.5 Futurity
Futurity is indicated by preverbal [go] (or one of its reflexes, e.g. [ga]) or
present forms in some instances. This feature is one of three or four features
associated with so-called creole languages that may distinguish this variety
from what are traditionally considered dialect varieties of English spoken in
the western hemisphere. Two other such features are the preverbal negator
no and the post-nominal pluralizer dem. Regarding most other grammatical
features Statian English resembles English dialects spoken in the Americas.
In the western Caribbean, gwain is a common future-tense marker but
was rejected by all interviewed informants in Statia even though goin, which
is diachronically the origin of gwain, is commonly heard on the island;
forms related to gonna are spoken as well (e.g. [gɔn]).14 No occurrences
of gwain were recorded in Statia. In at least two instances, (25) and (27),
futurity was indicated by the present-tense verb form and some reference
to time (e.g. two o’clock, next time, etc.) instead of any overt grammatical
marker. Hancock (1987) presents goin (290) (see discussion of progressive
aspect above), gon [gõ] (301, 302), and go (304) as future markers, all of
which were recorded in my data as well. Parsons (1933–43) presents many
instances of goin as a future marker (e.g. 277, 380, 386), but no occurrences
of either go or some reflex of going to (e.g. gonna).
14 The potential future tense marker gwain, which is so common in the western Caribbean, was
rejected in other anglophone eastern Caribbean research locations such as Barbuda (see Aceto
2002b) and Dominica (Aceto 2010) as well.
180 michael aceto
(25) [ai kʌm daun di rod tu oklak]
“I’ll come down the road at two o’clock.”
(26) [aim gʌnə kɛč ɪm . . . gʌnə kɛč ɪm]
“I’m going to catch him . . . going to catch him.”
(27) [a sii ya neks taim sʌmwier]
“I’ll see you next time somewhere.”
(28) [a ga rez yu mo do]
“I’m going to raise you more though.”
(29) [mo go aks ya]15
“Am I going to ask you?”
(30) [wi go fain yu]
“We’ll find you.”
(31) [a wan tɛl ya wa a ga gi ta yu]
“I want to tell you what I’m going to give to you.”
(32) [ya go an drink]
“Will you have a drink?/Are you going to drink?”
(33) [yu no go pe]
“You’re not going to pay?”
(34) [a gɔn wʌk tumaro]
“I’m going to work tomorrow.”
5.2.4 Questions
Interrogative forms show the common vernacular pattern in which there
is no inversion of subjects and verbs or “do support” in questions lacking
modal verbs. However, Parsons (1933–43) reveals at least two instances of
subject–auxiliary verb inversion for interrogatives: “Is it dis?” (380) and
“what is dat?” (385). Hancock (1987) reveals no instances of subject–verb
inversion for questions.
(49) [hau yu ple]
“How did you play?”
(50) [ʏa taak wa mi du]
“Are you talking about what I do?”
(51) [wat ya min]
“What do you mean?”
(52) [we i go]
“Where did he go?”
(53) [wa da gem]
“What is the game?”
(54) [we yu drink]
“What do you drink?”
(55) [hu gɛt kaa]
“Who’s got a car?”
(56) [ši goɪn skul]
“Is she going to school?”
(57) [ši aks ya fa mi]
“Did she ask you for me?”
5.2.5 Negation
Verb phrases are negated by a range of typical strategies found among
vernacular varieties spoken in the anglophone Americas. Preposed before
the main verb of a predicate are the following options in Statian English:
St. Eustatius English 183
en (< ain’t), don (< don’t), or no. Some auxiliary verbs have a postposed
clitic-like nasal -n, e.g. shouldn “shouldn’t,” didn “didn’t.” The modal can’t
is indicated by [kaan]. These strategies are confirmed by Hancock (1987)
for en (287, 300, 302, 303) and don (299). He reveals no instances of the
common creole language negator no for Statian English and neither does
Parsons (1933–43). Parsons reveals only instances of don’t and not as negators
as well as nasal clitics attached to the end of auxiliary verbs.
(58) [ɑ en gat tu rid]
“I don’t have to read.”
(59) [ɑ no seɪn]
“I’m not saying.”
(60) [a ein ga no fʌkɪn θɪŋ man]
“I don’t have a fucking thing, man.”
(61) [mebi ši don kʌm]
“Maybe she won’t come.”
(62) [ɑ dɪdn no]
“I didn’t know.”
(63) [ɑ no čit]
“I didn’t cheat.”
(64) [a dont gotə gi ya aal]
“I don’t have to give you everything.”
(65) [a don no di tɛknɪk]
“I don’t know the technique.”
(66) [de kʌmɪn . . . yu en kʌm]
“They’re coming; you’re not coming?”
(67) [a kaan go . . . pe yo mʌni]
“I can’t go. Pay your money.”
(68) [ɑ en gat pe]
“I don’t have to pay.”
5.2.6 Plurality
Statians have several options for indicating plurality in their language.
The common anglophone eastern Caribbean plural marker [an dɛm] (see
Aceto 2002b) is heard in St. Eustatius but it is less common than simple
184 michael aceto
postnominal [dɛm]. In at least one instance below the data reveal a com-
mon dialect construction with prenominal [dɛm] that indicates not only
possession but plurality as well. In instances of this nature, a (redundant)
postnominal plural marker is rarely if ever heard. Hancock (1987) only
reveals postnominal dem. Parsons (1933–43) does not indicate any of the
aforementioned strategies; she only indicates the plural explicitly with the
metropolitan English pattern of suffixation by a bound inflectional mor-
pheme. Again, it’s puzzling since the strategies documented below are heard
so robustly on the island.
(69) [di poskaad dɛm]
“The postcards.”
(70) [wai ya taakɪn manwɛl dɛm]
“Why are you talking to Manuel and his buddies?”
(71) [hi rid . . . a no i rid ɛm . . . i bai dɛm bʊk]
“He reads. I know he reads them. He buys books.”
(72) [de gat tʌbako an dɛm fram kyuba]
“They have/get tobacco and other related things from Cuba.”
5.2.7 Possession
In Statian English possession is indicated by the common vernacular pat-
tern of juxtaposing the possessor noun phrase before the possessed noun
phrase, e.g. [de kaa] “their car” (Hancock 1987: 291) and [mʌɪ dadi haus]
“my father’s house” (284). This pattern is corroborated many times in Par-
sons (1933–43), e.g. wife bosom, bull milk (383). I encountered no instances
of possession indicated by bound inflectional morphology in Statian
English.
5.2.8 Infinitives
All infinitive verbs were indicated by to only. There were no instances
in my data of an infinitive marked by any reflex of for as is common
in many Englishes in the western hemisphere. Hancock (1987) and Par-
sons (1933–43) corroborate the previous characterization. One informant
rejected infinitivals marked by fi/fu as not local forms.
(73) [trai tu kɛč mi]
“Try to catch me.”
(74) [ʏa haf tu ple daimʌn]
“You have to play diamonds.”
St. Eustatius English 185
Table 8.2 Pronouns in Statian English
Singular
5.2.9 Pronouns
Table 8.2 presents the pronominal forms which are heard in St. Eustatius.
All forms should be considered to have multiple functions as subject,
object, and possessive pronouns unless otherwise indicated.
In most ways, the pronominal system of Statian English is quite sim-
ilar to metropolitan English with some phonological differences (more
below). In my Statian English data there were few instances of [mi] as
a first-person subject pronoun; Hancock (1987) and Parsons (1933–43)
record none.16 However, the first-person possessive pronoun was usu-
ally [mai] or [mʌɪ], though a few instances of [mi] occurred as well.
Parsons also presents several instances of [mi] as a possessive pronoun
(e.g. 381). Statian English always distinguishes gender differences between
the third-person singular pronouns. Statian English also lacks the object
16 One informant insisted that mi as a first-person singular subject pronoun was not a local form.
Though I recorded several instances, it could possibly have been imported from one of the nearby
anglophone islands such as St. Kitts where it is heard more regularly.
186 michael aceto
pronoun om that is so common in much (though not all) of the eastern
Caribbean.17
Many eastern Caribbean varieties, as reported in Hancock (1987: 298),
lack the second person plural form [unu] or any of its reflexes that are
so common in anglophone western Caribbean varieties.18 Instead, eastern
varieties mostly reveal the common regional form [aayu] or [alyu] or some
reflex of those forms. However, in Statian English, the only second-person
plural pronoun attested in my data and the printed sources is [yu]. Accord-
ingly, Statian English also does not reveal the common eastern Caribbean
first-person plural pronoun [aa(l)wi]; instead it exhibits only [wi]. Both of
these features resemble those found in metropolitan varieties of English.
The pronominal system of Statian English is similar to other restructured
varieties of English spoken in Anguilla (Williams 2003) and the Turks and
Caicos islands (Cutler 2003). These forms may be related to forms spoken
earlier in the history of St. Eustatius during its period of commerce in the
late eighteenth century rather than assuming them to be recent changes
due to purported “decreolization.”
19 Hartog (1976) reports that when a Dutch bishop visited the island in 1836, he could speak no English
and, since the Governor of the island could speak little Dutch, communication was difficult.
188 michael aceto
eighteenth century, “St. Eustatius is the rendezvous of everything and
everybody meant to be clandestinely conveyed to America. It is easy to
get oneself carried thither, and military adventurers of all nations have
congregated at the island.” This intense multilingual setting seems to have
contributed to the emergence of some English variety as the lingua franca
of the island (see Baker [2000: 48] for what he terms “a medium for
interethnic communication”), specifically when the recipients of many
of Statia’s goods were English-speaking sailors and merchants headed for
colonies in North America and later the USA.
Religious affiliations in the eighteenth century provide further indication
of the multilingual nature of Statian society. Hartog (1976: 115) lists four
general religious practices: Lutheran, English Presbyterian, Anglican, and
Judaism. At least two of these Christian groups explicitly received religious
instruction in some English variety. Tombstones in the Jewish cemetery
on Statia, as confirmed by this researcher, reveal inscriptions written in
both Portuguese and Hebrew, suggesting that these languages were most
likely used in Jewish homes and religious observances. Many of these Jews
may have also spoken Ladino, a language common among Sephardic Jews
whose ancestors were expelled from the Iberian Peninsula beginning in the
late fifteenth century (see Aceto 1997). There are no Jews on Statia today,
though archeological remains in the form of a cemetery, and the ruins of a
ritual bath and synagogue testify to their earlier previous presence.20
The multilingual and multiethnic diversity of St. Eustatius in the eigh-
teenth century is indicated by Andrews and Andrews (1934: 136), who
present the diary of a Scottish visitor, Janet Schaw. In 1775 she wrote, “But
never did I meet with such variety; here was a merchant vending his goods
in Dutch, another in French, a third in Spanish, etc. etc. They all wear
the habit of their country, and the diversity is really amusing. The first
that welcomed us ashore were a set of Jews.” Hartog (1976: 40) states that
Turks, Greeks, and Levantines (i.e. presumably people from the modern
Middle Eastern area of Lebanon) were also present among the merchant
class.21
20 Most of the Jews on Statia were Sephardim whose ancestors derived from Spain and Portugal as
result of the expulsions in the late fifteenth century, which Schaw’s diary confirms (Andrews and
Andrews 1934: 136–7). In 1722, the Jewish population stood at 21 persons in total: six adult males
with their wives and families. In 1781, at the end of the great commercial period, there were 350
Jews in total, including 101 adult males, their wives and families. Rodney deported approximately
one-third of the adult male Jewish population of Statia when he finished plundering the island. By
1790, there were only 157 Jews remaining on St. Eustatius. In 1818, only five Jews remained, and in
1846 the last remaining member of this once thriving community died.
21 However, ethnic, linguistic, and commercial diversity were not always welcome on the island. In
1739, Commander Faesch complained to the Amsterdam Chamber that “both French and English
St. Eustatius English 189
There is no empirical or historical evidence that a restructured variety of
Dutch ever emerged in the Dutch Windward Antilles (as occurred nearby
on the island of St. Croix) but instead a dialect of English. Nor is there any
empirical evidence that a “deeper” Jamaican-like or Antiguan-like English-
derived variety was ever spoken on St. Eustatius. Why did an English dialect
emerge during the colonial period in the Dutch-controlled colony of St.
Eustatius (as well as on Saba, St. Martin)? In the Dutch Windward islands
it seems that European competition over the Caribbean necessitated that
they welcome colonists, settlers, and merchants from any location as long
as they aided in populating and settling the island. In the case of Statia,
the Dutch welcomed colonists from many ethnicities and nations who
spoke several different languages. English was one of several languages that
were spoken on the island as well as Dutch, French, Portuguese, Hebrew,
and presumably several African languages spoken among the first slaves.
Despite the fact that St. Eustatius was (and still is) a Dutch colony, the
largest speech community on the island was comprised of English speakers
(of both European and African descent). When considering the island
during its period of great commerce, English seems to have emerged as
an interethnic lingua franca in which commerce was facilitated by the
use of this language in a multilingual context. In other words, Statian
English might best be considered as a one of several “commercial Englishes”
of the eighteenth century. This English variety arose in an environment
where plantation slavery was not the norm, and instead an import–export
economy was the focus.
The few “creole-like” features heard in Statian English are perhaps
accurately and simply accounted for by assuming they are due to post-
emancipation immigration from islands whose English-derived varieties
exhibit these same features. Small boat traffic among the islands of the
Lesser Antilles continues today; it is informal, unofficial, and more intense
(and thus difficult to gauge since there are often no records) than gen-
erally acknowledged in Caribbean studies. Statians have historically and
contemporaneously made contact with St. Kitts and Nevis (and vice versa)
whenever they had access to maritime routes of transportation. This con-
tact with St. Kitts is most likely to be responsible for the three or four
“creole-like” features examined in this chapter.
are bringing European goods to the market here, and foreigners even set up shops and trade here”
(Attema 1976: 38). Goslinga (1979: 82) claims that the economic success of Statia and its dependence
on smuggling in the second half of the eighteenth century "brought many unsavory characters to
the island." He leaves unexplained what this assertion specifically means in terms of ethnicity,
culture, and language.
190 michael aceto
What seems unusual about the case of St. Eustatius is that for many non-
English-speaking Europeans as well as African slaves, an English dialect
emerged and was identified as a “target” language (I use the word “target”
here only to indicate that some variety of English was identified by residents
of the island as the most prestigious [at least in public life], economically
powerful, socially expedient, or common language spoken on the island). It
wasn’t only slaves that were grappling with English varieties heard around
them but many Europeans on the island who natively spoke languages
other than English as well. That is, some of the Europeans spoke some
variety of English as a native language, but they do not seem to have
been the largest part of the population at least in the early colonial history
of the island since speakers of European languages on the islands came
from many different speech communities. However, two hundred years
later, by the time of emancipation, most of the island’s population was
largely of African descent and mostly monolingual in the local variety of
English. I have no colonial artifacts that demonstrate conclusively that the
earliest Europeans and Africans spoke more or less the same variety (regular
language variation and change not withstanding), but the high degree of
second-language speakers of European descent suggests that the dynamics
of language creation, emergence, and shift was different on St. Eustatius
than on the typical Caribbean island in which the largest segment of the
European population was derived from the homeland of the colonial power
in question.
There are at least four features of Statian English generally associated with
so-called creole languages: postnominal plural marker dem, verbs unmarked
for past contexts, preverbal negator no, and preverbal future-tense marker
go. Of course, none of these features is heard exclusively among so-called
creole languages, and no individual feature is diagnostic of this group either.
I am excluding features such as the preverbal completive marker done since
it is found in a range of varieties that are uncontroversially considered
dialects of English. See Mufwene (2000, 2001, 2008), who argues that all
European-language-derived creoles may be considered as dialects of the
lexifier since creolization is a social, not a structural process.
One of the more popular ways for linguists to imagine creole varieties
without many so-called creole features is simply to insist (even in the
face of no artifacts) that varieties like Statian English must have once
“looked” more like, say, Jamaican or Antiguan, and that the variety in
question has subsequently “decreolized” since emancipation. This seems
too easy a way out of the challenge of trying to understand the different
social and linguistic circumstances (or language ecologies à la Mufwene
St. Eustatius English 191
[2001, 2008]) under which Englishes emerged in specific locations in the
Americas. Language shift undoubtedly occurs in what might be labeled
creole-speaking as well as non-creole-speaking communities. Purported
“decreolization” in its most coherent version simply describes the role and
effect of literacy, familiarity with institutional varieties of English (local or
otherwise), and the effect that varying degrees of fluency in that standard
dialect have had on specific idiolects and community-based dialects in
general.
Aceto (2010) describes Kokoy, an English “creole” or dialect spoken on
the island of Dominica. Older residents who mostly spoke Kokoy in their
youth insisted that Kokoy is still spoken today, even if the number of
speakers is declining. Many of these same key consultants reported that a
process of shift occurred fairly rapidly within one generation in the 1950s–
70s in which young people started identifying the institutional English
variety used mostly in schools as the most prestigious target (see Garrett
2003 for the role of institutional varieties on community-based dialects
in the Caribbean). Kokoy speakers exhibit almost no mixing of Kokoy
and intermediate English features. When motivated, one switches between
these varieties as if one were switching from one grammatically unrelated
variety to another such as from the local French Creole to English. For
some speakers, Kokoy is simply not part of their repertoires; for some it is,
and these folks can switch from Kokoy to another language variety in the
same manner as traditional codeswitching. The concept of decreolization
has been undergoing a process of re-examination and deconstruction in
the last ten years or more (see Aceto 1999; Satyanath 2006) and certainly
data from Kokoy do not validate this abstraction as having any linguistic
reality in Dominica.
In any event, the term “decreolization” is largely meaningless. All living
languages change. Just because a dialect or so-called “creole” spoken by
people largely of African descent changes doesn’t mean the language is
“decreolizing.” A speaker cannot undo the process of “creolization” (even
if that term could be coherently distinguished from regular cultural and
linguistic emergence, variation, and change – and even that doesn’t seem
possible); a speaker’s language will change since all grammars, all idiolects,
are fluid and in the flux of becoming. Why should change in the Caribbean
be labeled uniquely as “decreolization”? Just because it’s purportedly done
by people of African descent is not a sufficient motivation for such a term.
The term is problematic and should be carefully considered before use, if
used at all. Many of my students from eastern North Carolina can shift
between local rural community-based dialects and institutional varieties of
192 michael aceto
English, and some even make an effort to replace their vernacular with a
dialect variety containing more standard-like features.
Le Page (1998: 91) makes several important points: “I came to realize –
too late, unfortunately, to stop David DeCamp and others from taking
up the ‘continuum’ model – that it was a false representation.” Le Page
also writes that DeCamp wrote him that he “regretted the concept had
been taken up with such enthusiasm, since he [DeCamp] had never found
that it could provide an account of more than 30% of his Jamaican data”
(1998: 92). That is, 70 percent of the data still could not be accounted for
with the continuum concept. If subsequent researchers have cherry-picked
their data and then sorted it along a continuum so it looks tidy and neat,
then perhaps that says more about the practitioners of creole studies in
terms of methodology than it does about the purported “uniqueness” of
the speech communities in question. On the other hand, if one wants
to read how scientists have regularly tampered with data, sometimes even
with “successful” results, then read John Waller’s Fabulous Science (2006).
It seems it is a more common (albeit secret) practice than we were led to
believe by our mentors. Furthermore, creolists used to insist (in the 1970s,
1980s) that the alleged variation of the continuum was due to “decreoliza-
tion,” which was perceived as a unilateral force affecting all Englishes of
the Caribbean. When this strong view of decreolization was questioned
and criticized with data that showed change was not unidirectional, then
creolists of the 1990s abandoned decreolization but still kept the contin-
uum (even if decreolization was the explanatory factor) no matter how
flawed a concept it was. Now it is simply asserted in creole studies that
a continuum exists uniquely in the anglophone Caribbean (and why not
the francophone or hispanophone Caribbean as well?). Either all natural
human languages exhibit a continuum of variation or none do. Clearly all
living languages exhibit variation, but can this variation be arranged along
a continuum without forcing the data to fit the continuum model? This
insistence by creolists that something vaguely different or unique happened
in the anglophone Caribbean that was qualitatively different from the rest
of the language-speaking world (both past and present) is troubling.
What would prevent me from arranging data from English speakers
in eastern North Carolina along a continuum from vernacular to the
standard? Many long-term locals, even so-called “white” folks, regularly
reveal constructions like she mean with no explicit copula verb, or say
<with> with a word-final [f], [daet] with the word-initial stop (see above),
and verbal forms like my brother’s out a-walkin. They also say carry for
“take,” cuss for “curse,” reach for “arrive,” wait on for “wait.” Sure, those
St. Eustatius English 193
same words are heard in the Caribbean but they are heard in other English-
speaking regions beyond as well. I also have many students who are the first
ones in their families to go to college who manifest the aforementioned
forms as well as also more bookish forms associated with institutionalized
literacy or the standard. I could arrange the data on a continuum easily
from (left to right) vernacular forms to standard forms. Couldn’t this been
done in any speech community in which a standard form associated with
education and literacy has been institutionalized?
It is also worth remembering that in the early twentieth century, Danish
physicist Niels Bohr defined the nature of light as neither particle nor wave
but both (just not at the same time). That is, Bohr’s complementarity
principle states that photons of light (and electrons) could behave either
as waves or as particles, but it is impossible to observe both the wave
and particle aspects simultaneously. This example from another scientific
discipline wrestling with the dual nature of natural phenomena suggests
that the conclusions we draw from our data depend crucially on our
theoretical orientation and the questions we are trying to answer.
My research in the Caribbean reveals a situation that is not all that
different from any other location in which the local vernacular norms are
sufficiently different from the institutional norms of the lexically related
standard. However, “decreolization,” even if incoherently used by most
creolists, is only one type of externally motivated change and does not rep-
resent change in and of itself in creole-speaking communities (see Mufwene
[2000: 77] as well). However, it is my feeling that an over-reliance on this
purported phenomenon as an explanatory force has obscured the varied
details of language emergence or language ecology in specific anglophone
locations in the Americas.
Aceto (1999) pointed out that a reliance on the concepts associated with
the purported creole continuum and the concept of “decreolization” has
a tendency to view all changes in, at least, anglophone creole-speaking
Caribbean societies only in terms of whether the feature in question was
similar to or different from more standard varieties of English in terms of
form and function. The creole continuum terms “acrolect,” “mesolect,” and
“basilect” reveal little about linguistic or sociolinguistic processes invoked
by speakers involved in the creation, distribution, and maintenance of their
language, except for the assumption (which is undoubtedly true in some
specific well-defined cases) that some speakers are consciously or uncon-
sciously shifting their language toward norms associated with metropolitan
or institutional English and that these same speakers have competency in
a range of lects, as undoubtedly do many speakers around the globe who
194 michael aceto
have some familiarity with literacy and institutional varieties of dialects
that are historically related to their community-based vernaculars. These
terms “acrolect,” “mesolect,” and “basilect” only measure whether a given
feature (or bundle of features) appears more like its metropolitan or stan-
dard version of the lexifier language or not. From this perspective, the
common assumption in the use of these terms is that so-called “mesolec-
tal” varieties have simply “decreolized” under pressure from more standard
varieties of English and that “basilectal” creoles have not undergone this
same unidirectional change toward features associated with the lexifier. It
must be remembered that the variety of Jamaican found in Bailey (1966)
represented an abstract bundle of features (i.e. the bundle is abstract,
not any individual feature) associated with the so-called “basilect,” and
DeCamp (1971) considered this type of abstraction to be a necessary first
step in understanding variation in creole-speaking communities (the idea
of the so-called [post-creole] continuum originated with DeCamp 1971).
This abstraction suggests very strongly that, both from a diachronic and
synchronic perspective, there have always been few if any speakers of the
purported “basilect” revealing all of its associated features; in other words,
few individual so-called creole speakers can be said to have ever exclusively
displayed all the features of the “basilect.” The reification “basilect” is sim-
ply a compilation of all the features that are considered typologically the
furthest or most different from varieties of metropolitan English. Further-
more, any language, including so-called creole languages, can change in
ways left unexplored by the assumptions of the creole continuum, even
if the effects of internally induced change have been left largely unex-
plored by researchers studying creole-speaking societies (see Aceto 1999).
In addition, some types of change, whether they are externally or internally
motivated, may occur which do not resemble metropolitan varieties of
English. I am in complete agreement with Satyanath (2006), who writes,
“decreolization . . . is a working hypothesis and not an empirically tested
claim” (186).
The assumption in this chapter is that, not withstanding regular
diachronic change that all human languages exhibit everywhere, Statian
English sounds quite similar today to the colonial variety spoken 200–300
years ago. In other words, it is not necessary to reify decreolization and
the purported creole continuum to try to explain the synchronic shape of
Statian English. For a detailed discussion of the limitations of the creole
continuum abstraction, consult Aceto (1999), (2002a), and (2003). I am
in agreement with Mufwene (2008), who makes the case that so-called
“creole” languages are just the latest (i.e. in the last 500 years) stage of
St. Eustatius English 195
the dispersal of a subgroup of the Indo-European family of languages
that began 10,000–12,000 years ago. In other words, the English-derived,
French-derived, etc., varieties spoken in the Americas, Africa, and Asia
are dialects or “daughters” of their lexifiers in the same way French and
Spanish, for example, are daughters of Vulgar Latin spoken by the Romans
in the administration of their empire. However, every language has what
Mufwene calls its own specific language ecology (i.e. the history of who
came in contact with whom, the linguistic variants available, whatever social
factors are attached to those variants as well as the specific ethnographic
circumstances [2008: 54]). Therefore “[l]anguages evolve in non-uniform
ways” (Mufwene 2008: 14), without needing “decreolization” to account
for them. This chapter and Aceto (2006) describe the emergence of Statian
English.
References
Aceto, Michael. 1997. Saramaccan Creole origins: Portuguese-derived lexical cor-
respondences and the relexification hypothesis. In A. K. Spears and D. Win-
ford, ed., The Structure and Status of Pidgins and Creoles. Amsterdam: John
Benjamins, 219–39.
1999. Looking beyond decreolization as an explanatory model of language
change in Creole-speaking communities. Journal of Pidgin and Creole Lan-
guages 14: 93–119.
2002a. Going back to the beginning: describing the (nearly) undocumented
anglophone creoles of the Caribbean. In G. G. Gilbert, ed., Pidgin and
Creole Linguistics in the 21st Century. New York: Peter Lang, 93–120.
2002b. Barbudan Creole English: its history and some grammatical features.
English World-Wide 23: 223–50.
2003. What are creole languages? An alternative approach to the anglophone
Atlantic world with special emphasis on Barbudan Creole English. In
M. Aceto and J. P. Williams, eds.,Contact Englishes of the Eastern Caribbean.
Amsterdam and Philadelphia: John Benjamins, 121–40.
2006. Statian Creole English: an English-derived language emerges in the Dutch
Antilles. World Englishes 25: 411–35.
2010. Dominican Kokoy. In Daniel Schreier, Peter Trudgill, Edgar W. Schnei-
der, and Jeffrey P. Williams, eds., The Lesser-Known Varieties of English: An
Introduction. Cambridge University Press.
Aceto, Michael and Jeffrey P. Williams, eds. 2003. Contact Englishes of the Eastern
Caribbean. Amsterdam: John Benjamins.
Andrews, Evangeline Walker and Charles McLean Andrews. 1934. Journal of a
Lady of Quality; Being the Narrative of a Journey from Scotland to the West
Indies, North Carolina, and Portugal, in the Years 1774 to 1776. New Haven,
CT: Yale University Press.
196 michael aceto
Attema, Y. 1976. St. Eustatius: A Short History of the Island and Its Monuments.
Zutphen: De Walburg Pers.
Bailey, Beryl L. 1966. Jamaican Creole Syntax: A Transformational Approach.
Cambridge University Press.
Baker, Philip. 2000. Theories of creolization and the degree and nature of restruc-
turing. In Ingrid Neumann-Holzschuh and Edgar W. Schneider, eds., Degrees
of Restructuring in Creole Languages. Amsterdam: John Benjamins, 41–63.
Baker, Philip and Adrienne Bruyn, eds. 1998. St. Kitts and the Atlantic Creoles.
London: University of Westminster Press.
Bhatt, Parth and Ingo Plag, eds. 2006. The Structure of Creole Words: Segmental,
Syllabic and Morphological Aspects. Tübingen: Max Niemeyer.
Childs, Becky, Jeffrey Reaser, and Walt Wolfram. 2003. Defining ethnic varieties
in the Bahamas: phonological accommodation in black and white enclave
communities. In Aceto and Williams, eds., 1–28.
Crane, Julia G., ed. 1999. Statia Silhouettes. New York: Vantage Press.
Cutler, Cecilia. 2003. English in the Turks and Caicos Islands: a look at Grand
Turk. In Aceto and Williams, eds., 51–80.
DeCamp, David. 1971. Toward a generative analysis of the post-creole continuum.
In Dell Hymes, ed., Pidginization and Creolization of Languages. Cambridge
University Press, 349–70.
DeGraff, Michel, ed. 1999. Language Creation and Language Change: Creolization,
Diachrony, and Development. Cambridge, MA: MIT Press.
Garrett, Paul. 2003. An “English Creole” that isn’t: on the sociohistorical origins
and linguistic classification of the vernacular English of St. Lucia. In Aceto
and Williams, eds., 155–210.
Goslinga, Cornelis Ch. 1979. A Short History of the Netherlands Antilles and Suri-
nam. The Hague: Martinus Nijhoff.
Hancock, Ian. 1987. A preliminary classification of the anglophone Atlantic creoles,
with syntactic data from thirty-three representative dialects. In G. G. Gilbert,
ed., Pidgin and Creole Languages. Honolulu: University of Hawaii, 264–
334.
Hartog, J. 1976. History of St. Eustatius. Aruba: De Wit.
Holm, John. 1988–9. Pidgins and Creoles, 2 volumes. Cambridge University Press.
Jameson, J. Franklin. 1975. St. Eustatius in the American Revolution. In Charles
Toth, ed., The American Revolution and the West Indies. Port Washington,
NY: Kennikat.
Keur, John Y. and Dorothy L. Keur. 1960. Windward Children: A Study in the
Human Ecology of the Three Dutch Windward Islands in the Caribbean. Assen:
Royal Vangorcum.
Le Page, R. B. 1998. Ivory Towers: The Memoirs of a Pidgin Fancier. A Personal
Memoir of Fifty Years in Universities around the World. Society for Caribbean
Linguistics.
Le Page, R. B and David DeCamp. 1960. Jamaican Creole. London: Macmillan.
McWhorter, J. 1998. Identifying the creole prototype: vindicating a typological
class. Language 74: 788–818.
St. Eustatius English 197
2000. Defining “creole” as a synchronic term. In Ingrid Neumann-Holzschuh
and Edgar W. Schneider, eds., Degrees of Restructuring in Creole Languages.
Amsterdam: John Benjamins, 85–123.
Mufwene, Salikoko, S. 1996. The founder principle in creole genesis. Diachronica
13: 83–134.
2000. Creolization is a social, not a structural, process. In Ingrid Neumann-
Holzschuh and Edgar W. Schneider, eds., Degrees of Restructuring in Creole
Languages. Amsterdam: John Benjamins, 65–84.
2001. The Ecology of Language Evolution. Cambridge University Press.
2008. Language Evolution: Contact, Competition and Change. London and New
York: Continuum.
Parkvall, Mikael. 2001. Creolistics and the quest for creoleness: a reply to Claire
Lefebvre. Journal of Pidgin and Creole Languages 16: 147–51.
Parsons, Elsie Clews. 1933–43. Folklore of the Antilles, French and English, 3 vols.
New York: American Folklore Society.
Plag, Ingo. 2001. The nature of derivational morphology in creoles and non-creoles.
Journal of Pidgin and Creole Languages 16: 153–60.
Postma, Johannes Menne. 1990. The Dutch in the Atlantic Slave Trade 1600–1815.
Cambridge University Press.
Rizzi, Luigi. 1999. Broadening the empirical basis of universal grammar models:
acommentary. In Michel DeGraff, ed., Language Creation and Language
Change. Creolization, Diachrony, and Development. Cambridge, MA: MIT
Press, 453–72.
Satyanath, Shobha. 2006. English in the New World: continuity and change, the
case of personal pronouns in Guyanese English. In Bhatt and Plag, eds.,
179–99.
Waller, J. 2006. Fabulous Science: Fact and Fiction in the History of Scientific
Discovery. Oxford University Press.
Williams, Jeffrey. 1983. Dutch and English Creole on the Windward Netherlands
Antilles: an historical perspective. Amsterdam Creole Studies 5: 93–103.
2003. The establishment and perpetuation of anglophone white enclave com-
munities in the eastern Caribbean: the case of Island Harbor, Anguilla. In
Aceto and Williams, eds., 95–119.
c h a p ter 9
1 Introduction
There is a variety of English spoken in the port city of Gustavia on the
French island of St. Barthélemy in the Leeward Islands in the northeastern
corner of the Caribbean. It has the smallest population of speakers of
English in the Caribbean and is probably the most endangered variety.1
The speakers of the language do not have a name for their speech other
than English. I call this variety Gustavia English (GE) due to its association
with the town of Gustavia rather than with the entire island. Most of the
speakers of GE are Afro-European while most of the rest of the population
of the island is white and French-speaking.
The name of the island is also spelled St. Bartholomew or St. Barths/
Bart(s). In 2007 the island’s governance separated from Guadeloupe and
became the Overseas Collectivity of Saint Barthélemy. The 21 sq km
(8 sq. miles) island is within sight of St. Martin 19.5 km (12 miles) to
the northeast, and the islands of Saba, St. Eustatius (also called Statia),
and St. Kitts are about 48 km (30 miles) to the south and southwest. The
port town of Gustavia is the largest settlement on the island. The island’s
population of 2,332 (CIA 2012) is primarily supported by tourism.
Most of the previous linguistic and sociolinguistic research on St. Barths
has focused on the varieties of French spoken on the rest of the island.
My previous article (Decker 2004a) is the most comprehensive description
of GE to date. That article focused primarily on identifying the historical
roots of GE, partly through an examination of some linguistic features. I
proposed that English came to St. Barths around 1785.
This chapter will build on the previous article and give further descrip-
tion of the phonology and morphology. I believe that GE represents an early
stage of contact between Africans and Europeans. The contact occurred in
1 The English of Webster Yard, Anguilla, described by Williams (2003), may be considered more
endangered, but the people are simply shifting to another close variety of English.
198
The English of Gustavia, St. Barthélemy 199
a context in which slaves and owners lived and worked closely; a context
which offered sufficient opportunity for the slaves to acquire English. There
is no evidence of a reduction and expansion process. Most nonstandard
features can be traced to regional varieties of English.
3 Descriptive features
Gustavia English is worth studying and describing in as much as it is unique
and has features that distinguish it from other varieties of related speech.
In this section I will show that GE is a nonstandard, restructured variety of
English. There is no evidence of it having been a pidgin, nor having gone
through a creolization or decreolization process. There is possibly some
2 I use Cruttenden’s (2001) description of English as a model of SE. He includes comments on General
American (GA) and Received Pronunciation (RP).
202 ken decker
Table 9.1 Phonetic realizations of the lax
vowels of Gustavia English
kit /ɪ/ ɪ ̝ɪ ı̈ ɨ ə ɛ̝
dress /ɛ/ ɛ ɛ̝ ̞e
trap /æ/ æɐ
palm /ɑ/ ɑɐʌ
strut /ʌ/ ʌɵo
foot /ʊ/ ʊu
evidence of contact influence from St. Kitts Creole (SKC), but it is only
weak evidence at best. I have been able to identify possible sources in the
British Isles and North America for some of the unique features described
below.
In this section I will describe aspects of GE segmental phonology. This
description is based on two half-hour recorded conversations, each between
a male and a female resident of Gustavia, all over 55 years of age. I have
also used notes from five other unrecorded interviews I conducted, two
texts from Maher (1987, 1996), and nine short texts from Parsons (1933).
I have not included data from the Shrimpton (1994) texts from 1804.
My comparison of these texts with other evidence of GE (Decker 2004a)
indicates that they do not represent the speech of Gustavia.
dress /ɛ/
The dress vowel /ɛ/ is fairly stable in closed syllables. One informant
tended to produce a slightly higher vowel [ɛ̝ ]. In open syllables there
is more variation with a number of high-function words being pro-
nounced close to the face vowel /e/. For example, they was transcribed
as [dɛn dɛ̃ dɛ̝ ̃ d ̞e d ̞eɪ]. This represents variation from a more
creole-influenced nasal dress vowel [ɛ̃] to an acrolectal face vowel [ ̞eɪ].
Cruttenden (2001: 130) says that Middle English [aː] split in Early
Modern English to [eː] and [ɛː], which then became [eɪ] in Present RP
English. The [ɛː] has been preserved in many regional dialects in England
in words like make, take, and catch. In GE these words have the dress
vowel /ɛ/, as is typical in CECs.
trap /æ/
The trap vowel /æ/ is found in both closed and open syllables. In my
data there are a few occurrences of a more central variant [ɐ] in open
syllables. The presence of the trap vowel in GE is noteworthy since it is
204 ken decker
not found in basilectal SKC, but it is common in SE. If GE had developed
out of SKC, or Antiguan, this trap vowel would not be present. This is
one bit of evidence that there has not been restructuring in GE.
palm /ɑ/
In GE the palm vowel /ɑ/ is fairly stable. In open syllables it sometimes
varies as far forward as [ɐ]. There are examples of the strut vowel where
we usually hear the palm vowel, such as /wɑz/ pronounced as [wʌz] was.
One of the features of CEC phonology is a distinction between a length-
ened palm vowel /ɑː/ and an unlengthened vowel /ɑ/. This distinction is
not very prevalent in GE but there were a few examples. These may be the
result of contact with neighboring CEC varieties, but it doesn’t seem to be
sufficient evidence of an earlier creolization of GE.
strut /ʌ/
Unlike neighboring CEC varieties the strut vowel /ʌ/ occurs in GE,
but usually as a diaphoneme of other vowels. For example: cut was usually
pronounced by my informants, as [kot kɵt], but there are a few examples
of [kʌt]. Similarly, come is usually heard as [kõm], but can also be heard
as [kom̞ kʌm]. It may be possible that this variation is a result of an
incomplete foot–strut split process that began during the transition
from Middle English to Early Modern English.
The strut vowel can also be found in words usually pronounced with
the palm vowel, such as /wɑz/. Even though around and again are consis-
tently pronounced with the palm vowel as the first syllable, about is usually
heard with the strut vowel in the first syllable.
There are also a few words in which I believe the strut vowel is the
target pronunciation, not a diaphoneme. For example, us is not a pronoun
used in CEC but it is used in GE. There are only a few examples of it in any
of the data, but it is always pronounced with the strut vowel /ʌ/. Since
this is not a vowel found in CECs it has not been influenced by Caribbean
pronunciation.
foot /ʊ/
The foot vowel /ʊ/ occurs very infrequently in the data. It is a very stable
vowel occurring in closed syllables, except when the onset or coda of a
one-syllable word is dropped, as in /ʊd/ would or /kʊ/ could.
One language helper consistently produced [tʊ] for the preposition to,
while the other language helpers produced [tu]. Other examples of this
variation include: [ful fʊl] full, [ɡud ɡʊd] good, and [lukɪn lʊkɪn]
looking. Cruttenden (2001: 122) says that the pronunciation of some words
The English of Gustavia, St. Barthélemy 205
Table 9.2 Phonetic realizations of the tense
vowels and diphthongs of Gustavia English
fleece /i/ i
face /e/ e ɛ eɪ ɪɛ
goat /o/ o o̞ ɤ ʌ ɤ̞ ɵ ʊo
goose /u/ u
mouth /aʊ/ aʊ ʌo ou
choice /ɔɪ/ ɔɪ ɑɪ
price /ɑɪ/ ɑɪ ɐɪ ɔɪ
fleece /i/
The fleece vowel /i/ is quite stable, occurring in open and closed
syllables. It is more like North American varieties of English rather than
varieties found in the British Isles. Cruttenden (2001: 106) says this vowel
has been stable since 1500. A few words spelled with <ea> shifted from
[ɛː ɛːə ɛ] to [iː] by the end of the seventeenth century. There are
a few occurrences where the fleece vowel is lengthened in the stressed
syllable of a multi-syllable word, for example /tiːʧɐs/ teachers and /iːzj ɑ/
easier. This instability of the <ea> words in Early Modern English may
account for this feature in GE.
face /e/
In GA and RP English the face vowel is a diphthong /eɪ/, but in GE, and
some CECs, it is a monophthong /e/, as in /selɪn/ sailing. It is found in
both open and closed syllables. The /e/ pronunciation is similar to varieties
of English in lower northern and north Midland England, Ireland, and
Scotland. There are a few words for which I have samples with and without
206 ken decker
the diphthong /eɪ/, for example /mebi meɪbi/ maybe. (See comments in
the dress vowel section above concerning the [ɛː eː eɪ] variation.)
The pronunciation of the diphthong may be a result of convergence with
SE pronunciation learned through school instruction.
Two of my informants each said name as /nɪɛm/. The /ɪɛ/ diphthong
is the basilectal SKC pronunciation of the face vowel. This may be an
indication of convergence with SKC, but it is weak since it only occurs in
one word.
goat /o/
In GA and RP English the goat vowelis a diphthong /oʊ/ or /əʊ/ respec-
tively. In GE, and some CECs, the goat vowel is a monophthong /o/,
as in /nobɑdɪ/ nobody. The /o/ pronunciation is similar to some varieties
of English in lower north and north Midland England, Ireland, Scotland,
and Wales. Cruttenden (2001: 136) says that the shift from Old English
[ɔ] to present-day /əʊ/ in open syllables “reached a quality near [oː] in
the sixteenth and seventeenth centuries.” This may be an indication of the
divergence of this vowel from what became GA and RP.
In rapid speech I have examples of it becoming more central as [ɵ],
losing labial rounding becoming [ɤ], sometimes with a slight drop to [ɤ],̞
or dropping as low as [ʌ]. There are also examples of it becoming nasalized
when followed by a nasal consonant, as in /kõm/ come, or an implied nasal,
as in /dõ/ don’t.
Cruttenden (2001: 114) says that Present RP English /ʌ/ derives from
Middle English /ʊ/. In the nineteenth century this vowel shifted to [ɤ].
Possibly this provides an explanation for the variability in GE.
One of my informants pronounced boat as /bʊot/. The /ʊo/ diphthong
is the basilectal SKC pronunciation of the goat vowel. This may be an
indication of convergence with SKC, but it is weak since it only occurred
in one word.
goose /u/
The goose vowel /u/ is quite stable, occurring in open and closed sylla-
bles. The term /skuːnɑ/ schooner was the only word in which the goose
vowel was consistently lengthened by all informants. The unlengthened
pronunciation is more like North American varieties of English rather than
varieties found in the British Isles.
mouth /aʊ/
The mouth diphthong /aʊ/ is quite stable in GE, occurring in open
and closed syllables. GA and RP varieties of English also pronounce
The English of Gustavia, St. Barthélemy 207
the mouth vowel as /aʊ/. All CECs also have a mouth vowel diph-
thong that begins at a low or midpoint sliding to a high, back rounded
point. As part of the Great Vowel Shift, Middle English [uː] began
diphthongization in the fifteenth century. It reached the present pro-
nunciation in the eighteenth and nineteenth centuries (Cruttenden 2001:
137–8).
Cooper’s (1979: 65) description of SKC gives /ou/ as the pronuncia-
tion for the mouth vowel. In a few words one of my GE informants
raised the pronunciation as high as /ou/ when the diphthong was the
first syllable of a word. Another informant pronounced it as /ʌo/ when
preceding a velar nasal /ŋ/. It is possible that there may have been some
convergence with SKC that has created this variability, but it is weak
evidence.
choice /ɔɪ/
In many CECs the choice vowel /ɔɪ/ has become unrounded and merged
with the price vowel /ɑɪ/. There are only a few examples of this shift in
GE: /bɑɪz/ boys, /pɑɪnt/ point, and /wɑɪs/ voice. Cooper (1979:52) does not
list /ɔɪ/ as a basilectal diphthong in SKC, but he lists /chais/ choice as an
example of the /ɑɪ/ diphthong.
However, in some cases GE has also done the opposite. Throughout
the data there are examples of the price vowel /ɑɪ/ beginning at a low
rounded point, merging with the choice vowel /ɔɪ/. My older informants,
and samples from Maher and Parsons, tend to use a raised and rounded
beginning pronunciation /ɔɪ/. This results in pronunciations such as: /sɔɪd/
side, /ɔɪlɑn/ island, /wɔɪl/ while, and /fɔɪn/ find. The younger informants
used /ɑɪ/ or a slightly raised /ɐɪ/.
The Atlas of English Dialects (Upton and Widdowson 2006: 18) identifies
this pronunciation for find in parts of the West Country, from the West
Midlands to the Isle of Wight, Essex, and Kent. Wright (1898–1905) also
identified this pronunciation for time in Ireland.
Other than Gustavia, I have only observed the raised pronunciation /ɔɪ/
for the PRICE vowel in St. Maarten. Roberts (1988: 92) mentions this
as a feature found in Bajan. Neither Wells (1982), nor Holm (1988), nor
Hancock (1969, 1986) mention this as a feature of CECs.
price /ɑɪ/
Aside from the /ɔɪ/ > /ɑɪ/ and /ɔɪ/ > /ɑɪ/ shift described above, the price
vowel /ɑɪ/ occurs in many words, just as in SE. It is very stable, found in
open and closed syllables.
208 ken decker
3.1.3 Consonants
There are several features that are generally considered representative of the
phonologies of CECs (Baker 1999, Holm 1988, Wells 1982). Some of these
features occur in GE, but only a few of them are as strongly descriptive of
GE as for the CECs.
TH-stopping
TH-stopping describes the merger of the interdental fricatives /θ/ and
/ð/ with /t/ and /d/ respectively. Wells (1982: 565) says that this feature
is so strong in CECs that the fricatives only occur in the speech of the
most educated. This feature is prevalent in GE, but not exclusive. In the
texts from Parsons and Maher, <th> does appear in a few lines.3 In my
recordings there are no examples of the fricatives. Cruttenden (2001: 184)
says that this merger often happens in southern Irish speech. As with many
African languages, the dental fricative sounds are rare in many languages
globally. Therefore, this feature does not need to be traced to Irish influence,
neither does it indicate substrate influence from Africa. The development
of this feature could simply be considered an innovation parallel to that of
the CECs.
Rhoticity
The loss of preconsonantal and postvocalic /r/ in the prestigious RP dialect
began in the fifteenth century, but other varieties retained it (Upton and
Widdowson 2006: 31).4 Rhoticity is a variable feature in GE. In most cases,
postvocalic /r/ has been dropped as in example (6). However, there are a
few examples of postvocalic /r/ and rhotic vowels /ɝ/, as in example (7).
(6) /ɑbɑʊt tɛn jɛɑz ɑftɑ/
about ten years after
(7) /deɪ ɑr lɝnɨŋ/
they are learning
The speech of Barbados is fully rhotic (Wells 1982: 570). The scarcity of
post-vocalic /r/ in GE may indicate the lack of any significant linguistic
input from Barbados.
V–W confusion
Some varieties of CEC have variability in the use of /v/ and /w/. (See Baker
1999: 329.) In GE there are examples of /v/ shifting to /w/, as in example
(8) below. There are also just as many examples of /w/ shifting to /v/, as in
example (9) below. My youngest informants did not exhibit this feature.
(8) /yu gɑt ɑ mɑn wɔɪs/
you have a man’s voice
(9) /vɛl vɪn di ɑmɛrɪkɪn siplen . . . /
well, when the American seaplane . . .
Holm (2000: 162) raises the possibility that this variation may result from
substratal influence from West African languages. Wells (1982: 568) says that
4 /r/ = [ɹ].
210 ken decker
it is uncertain whether this feature has been influenced by eighteenth- and
nineteenth-century London Cockney or “an African substratum lacking
/v/.” It is also possible that the absence of a consistent model in either the
superstrate or substrate resulted in a continued variability in GE.
3.2 Lexicon
Baker (1999), in his attempt to identify diagnostic features that identify
CECs, lists 138 features, of which 82 are lexical items. There are very few
lexical features that suggest any connection of GE with CECs. One of
Baker’s features is the use of various forms of /fi fo fa/ as an infinitive
marker. Holm (1988: 168ff.) proposed that this feature may be a result
5 Wells (1982: 572) says it is a strut vowel /ʌ/, but I have always heard it higher and with rounding
of the lips.
The English of Gustavia, St. Barthélemy 211
of convergence between British English and African languages. There are
several examples of fah in Parsons’ and Maher’s texts, but they all have the
same prepositional role of for. In my data there are a few examples of /fɪ/
used like a CEC infinitive marker; see examples (11) and (12).
3.3 Morphosyntax
The English of Gustavia has become a fairly stabilized koiné. There are
nearly eighty to ninety years’ separation between Parsons’ texts and mine,
but there is very little difference in the morphosyntax of the speech. How-
ever, as I pointed out in my previous article (Decker 2004a), there is some
variation in the speech. Furthermore, there is very little difference between
GE and SE sentence structure. In this section I will only discuss a few
variations between GE and SE.
Past tense
The CECs are known for the tense, mood, and aspect (TMA) structure
of the verb phrase. The so-called past tense in CECs, which Bickerton
(1975) called an anterior tense, functions differently from English past
tense. There is virtually no evidence of the CEC TMA structure in GE.
Most GE verbs are inflected as in SE. However, there are many verbs that
are not marked for SE past tense nor are they marked for CEC anterior
tense either.
To get a sense of the variation I counted all of the verbs in Parsons’
texts that could be marked for past tense as indicated by the context. I
excluded the copula form (was), modal verb (would have), and irregular
verbs that have a unique past participle (gone). There were 198 verbs that
could be marked with the -ed suffix for past tense, 65 (33%) were not
marked correctly for SE. There were 91 irregular verbs and 18 (20%) of
them did not use the proper past-tense form.
214 ken decker
According to Cooper (1979: 81) the “past reference in SKNC6 depends
more heavily on context than on the actual presence of the preverbal
formative bin.”7 Example (17) shows an SKC utterance with an unmarked
past tense.
(17) mi wok haad dis maanin aredi (from Cooper 1979: 83)
“I worked hard this morning already”
(18) When de moder return she found de pig (from Parsons)
“When the mother returned she found the pig”
Example (18) shows a GE sentence with one of the verbs unmarked for
past tense other than the context. It is possible that all of the unmarked
and uninflected verbs are functioning like SKC verbs. I don’t have a good
explanation for this variation. It might be as a result of influence from
neighboring CECs or there may be some sort of relexification.
There were several examples of the use of /dɪd/ as a past-tense marker,
as in examples (19) and (20). In my previous article (Decker 2004a) I
proposed that this might mark past perfect tense. The use of did + VERB
for simple past tense was common in seventeenth-century southwestern
English varieties (Winford 2003: 315). Allsopp (1966) lists several uses of
did as part of the past tense verb phrase in several CECs: simple past,
imperfect, and perfect. This only occurred in the speech of my oldest pair
of informants.
(19) /hi dɪd lɛn ɪt/
“he lent it he had lent it”
(20) /dei dɪd fʊl ɵp dɪ biʧ/
“they filled the beach they had filled the beach”
There were also several examples of the use of /doz/ as a habitual aspect
marker, as in example (21). Allsopp (1966) says this is found throughout
the eastern Caribbean. This only occurred in the in the speech of my oldest
pair of informants.
(21) /ʃi doz pronɔʊŋs ɪt ɡʊd/
“she pronounces it good”
Copula deletion
In some CECs (Decker 2004b: 106; Holm 1988: 175ff.) there is a feature of
copula deletion in descriptive clauses, when the subject is modified by an
adjectival phrase. (See example (22) from Belize Kriol.) There were several
examples of copula deletion in Parsons’ and my texts, see examples (23) and
(24). However, these are not all descriptive clauses with adjectival phrases.
Cooper (1979) says nothing about copula deletion in SKC.
(22) Di froot aal swibl op.
“The fruit is all shriveled up.”
(23) I (am) sure dat . . . (from Parsons)
(24) /ɑi no pɑlɪtɪʧɑ̃n/ (from my data)
“I (am) not a politician”
4 Conclusion
The history of Gustavia, St. Barths, was quite different from the devel-
opments on plantation islands like St. Kitts or Antigua. In Gustavia,
non-English speakers and non-SE speakers worked alongside one another.
There were probably many varieties of English, but within the confines of
the small geographic area there was sufficient opportunity for a koiné to
develop.
GE is a unique variety of English, but it is not as divergent as the creole
varieties found on some neighboring islands. There has been little change
216 ken decker
in GE in the last hundred years. This is best exhibited by the degree of
similarity between Parsons’ texts from the 1930s, Maher’s texts from the
1980s, and my texts from the 2000s. The data support Mufwene’s (2008:
84) thesis that, “One is hard pressed to find in creoles any grammatical
features that have not been selected from the nonstandard varieties of the
relevant vernaculars or in their substrate languages, although these have
not been replicated faithfully.” There are a few grammatical features that
are similar to CEC structures. However, they only occur in a few locations
in the speech of a small number of the informants.
I have suggested that the phonological variation in GE vowels can be
explained by the changes in the phonology of English during the seven-
teenth century. However, if GE developed in the late eighteenth century,
the developments in England would seem too early. I have also shown
how some pronunciations that developed in the seventeenth century are
still retained in some regional varieties to this day. So it is not impractical
to expect that in the eighteenth century there were migrants from rural
areas of England who traveled to the Caribbean and provided the building
blocks of GE.
I believe that GE represents an early stage of contact between Africans
and Europeans. The contact occurred in a context in which slaves and own-
ers lived and worked closely; a context which offered sufficient opportunity
for the slaves to acquire English. The presence of some SE features and
the absence of certain CEC features are further evidence that there has not
been a creolization or decreolization process. There are very few features,
if any, that might suggest African substrate influence. In each case there
would have been convergence between similar features in the substrate and
superstrate languages. This would be a typical phenomenon in any contact
situation and not a unique creolization process.
Gustavia English will probably only survive for a couple more decades
at best. Most of the remaining speakers are elderly and the middle-aged
speakers have more opportunities to use French on a daily basis. A fur-
ther comparison of GE with the English of Saba and Statia may reveal
similarities.
References
Aceto, Michael. 2007. Statian Creole English: an English-derived language emerges
in the Dutch Antilles. World Englishes 25: 411–35.
Allsopp, Richard. 1996. Dictionary of Caribbean English Usage. Oxford University
Press.
The English of Gustavia, St. Barthélemy 217
Baker, Philip. 1999. Investigating the origin and diffusion of shared features
among the Atlantic English Creoles. In Philip Baker and Adrienne Bruyn,
St. Kitts and the Atlantic Creoles. London: University of Westminster Press,
315–64.
Bickerton, Derek. 1975. Dynamics of a Creole System. Cambridge University Press.
1981. Roots of Language. Ann Arbor: Karoma.
Bolinger, Dwight. 1989. Intonation and Its Uses. Stanford University Press.
Byrne, Francis. 1984. Fi and fu: origins and functions in some Caribbean English–
based creoles. Lingua 62: 97–120.
CIA. 2012. The World Fact Book. www.cia.gov/library/publications/the-world-
factbook/geos/tb.html (accessed 12 November 2012).
Cooper, Vincent O. 1979. Basilectal Creole, decreolization, and autonomous lan-
guage change in St. Kitts-Nevis. Unpublished PhD dissertation, Princeton
University.
Cruttenden, Alan, ed. 2001. Gimson’s Pronunciation of English, 6th edn. London:
Arnold.
Decker, Ken. 2004a. Moribund English: the case of Gustavia English,
St. Barthélemy. English World-Wide 25(2): 217–54.
2004b. The Song of Kriol: A Grammar of the Kriol Language of Belize. Belize
City: Belize Kriol Project.
Fleischmann, Ulrich. 2005. Black Culture, White Discourse and Creole History:
A Study on Interpretations of American Slavery. www.larramendi.es/i18 n/
catalogo imagenes/grupo.cmd?path=1000198 (accessed 6 November 2012).
Hancock, Ian. 1969. A provisional comparison of the English-based Atlantic
creoles. African Language Review 8: 7–72.
1986. The Domestic Hypothesis, diffusion and componentiality: an account of
Atlantic anglophone creole origins. In Pieter Muysken and Norval Smith,
eds., Substrata Verses Universals in Creole Genesis: Papers from the Ams-
terdam Creole Workshop, April, 1985. Amsterdam: John Benjamins, 71–
102.
Holm, John. 1988. Pidgins and Creoles, vol. 1: Theory and Structure. Cambridge
University Press.
2000. An Introduction to Pidgins and Creoles. Cambridge University Press.
Jeffry, Daniella. 1997. Understanding the language situation in Saint Martin: the
historical perspective. Unpublished manuscript.
Köhler, Åsa and Malin Runsten. 2001. St. Barthélemy – Sveriges sist koloni.
www.ce.kth.se/aom/cies/tms/uland/U-rapporter/23%20%StBarthélemy.
pdf (accessed 10 December 2002).
Maher, Julianne. 1987. Transcription of an interview. Unpublished manuscript.
1996. Fishermen, farmers, traders: Language and economic history on
St. Barthélemy. Language in Society 25: 373–406.
2010. The roots of linguistic conservatism in St. Barthélemy. Paper presented at
the 18th Biennial Meeting of the Society for Caribbean Linguistics Confer-
ence, University of the West Indies, Cave Hill, Barbados, 9–13 August 2010.
218 ken decker
Mufwene, Salikoko. 2001. The Ecology of Language Evolution. Cambridge
University Press.
2008. Language Evolution: Contact, Competition, and Change. London:
Continuum.
Parsons, Elsie Clews. 1933–43. Folklore of the Antilles, French and English, 3 vols.
New York: American Folklore Society.
Roberts, Peter. 1988. West Indians and Their Language. Cambridge University
Press.
Shrimpton, Neville. 1994. A Preliminary Note on Some Afro-English (English Creole)
Texts from Saint Bartholomew. Umeå, Sweden: Umeå University, Department
of English.
Shrimpton, Neville and Philip Baker. 1995. Buddy Quow, St. Kitts and St. Barts.
In Philip Baker, ed., From Contact to Creole and Beyond. London: University
of Westminster Press, 81–96.
Upton, Clive and J. D. A. Widdowson. 2006. An Atlas of English Dialects. Oxford,
Oxford University Press.
Wells, J. C. 1982. Accents of English, vol. 3: Beyond the British Isles. Cambridge
University Press.
Williams, Jeffrey. 2003. The establishment and perpetuation of Anglophone white
enclave communities in the Eastern Caribbean: the case of Island Harbour.
In Michael Aceto and Jeffrey Williams, eds., Contact Englishes of the Eastern
Caribbean. Amsterdam: John Benjamins, 95–119.
Winford, Donald. 2003. An Introduction to Contact Linguistics. Oxford: Blackwell.
Wright, Joseph. 1898–1905. English Dialect Dictionary. London: Oxford University
Press.
ch a p ter 1 0
Anglo-Paraguayan English
Danae M. Perez-Inofuentes
1 Introduction
Paraguay is a land-locked country in the heart of South America bordered
by Argentina to the south, Bolivia to the north, and Brazil to the east.
Six and a half million Paraguayans live on 400,000 square kilometers.
Among linguists, Paraguay is of particular interest because of its societal
bilingualism. Its official languages are Spanish and Guarani, that is, Guaranı́
Paraguayo, a koiné of a Tupi-Guarani language that underwent considerable
restructuring due to language and dialect contact. Even though it was long
regarded as backward and a hindrance to progress, Guarani is spoken by the
majority of the Paraguayans today, and since its recognition as one of the
two national languages in the 1967 Constitution, it has become a symbol
of national identity (Melia 2011). Thus, while Spanish is principally used
for official purposes and secondary education, Guarani is the preferred
language of 60 percent of the Paraguayan households. Most Paraguayans
speak a variety of jopará, a language consisting of both Spanish and Guarani,
which offers a showcase of language mixing (Zajicova 2009).
The Paraguayan diglossia is rooted in history. Due to Paraguay’s relative
remoteness in the South American hinterland, the European immigra-
tion progressed slowly in colonial times. Instead of receiving immigrant
families, Paraguay was rather attractive to individual adventurers trying
their luck. Accordingly, most immigrants were single men, and their off-
spring with Paraguayan women resulted in a stable Guarani-speaking mes-
tizo population (Melia 2011: 425–7). The War of the Triple Alliance in
1870, however, halved the Paraguayan population to 336,000. In order to
overcome the subsequent economical stagnation, the Paraguayan govern-
ment intended to stimulate European immigration anew by offering tax
relief and fertile lands at low prices to immigrant communities. Hence,
many immigrant groups such as Mennonites, German imperialists, or
French agriculturalists sought a future in Paraguay around the turn of the
219
220 danae m. perez-inofuentes
century (Warren 1985). Given that some of these enclaves maintained their
heritage languages, developing new varieties of them, languages from all
over the world are now spoken natively in Paraguay. As a result, Paraguay’s
particular heteroglossic situation and societal bilingualism are unparalleled
worldwide.
One of these immigrant groups is particularly interesting for sociolin-
guists. Like most other Latin American countries such as Argentina and
Brazil, Paraguay experienced considerable immigration from the British
Isles during the nineteenth century. Important enterprises such as the rail-
way and other transportation companies as well as influential financial
institutes were British-run. Nevertheless, the Anglo-Paraguayan commu-
nity differed from other English-speaking communities in Latin America
due to its extraordinary inception. While it consisted of no more than a
hundred British subjects in 1884 (Warren 1985: 16), it grew significantly in
1893 when a group of over four hundred Australians arrived in Asunción.
They were a group of socialists who had left their country on the brink of
civil war in Queensland. In response to the great Shearer’s Strike in 1891,
they planned to set up a communist society without social stratification as
an example to show the world that capitalism was outdated. The country
they chose to put this project into action was Paraguay (Kellett 1997).
Even though these conditions seemed ideal for a new, contact-influenced
English variety to emerge, English was not maintained as a communal lan-
guage among the Australian immigrants in Paraguay. The utopian project
failed, and the colony soon disintegrated. Nevertheless, descendants of the
original settlers still live in Paraguay, and their individual relationships to
their heritage language vary considerably. This chapter is a first attempt to
describe this hitherto unknown English-speaking community in Paraguay.
After giving an account of the sociolinguistic history, I will describe some
aspects of the internal development of the English language in this unique
case of trilingual language contact, paying particular attention to contact-
induced phenomena. The data presented here are part of an ongoing
research project that started in 2011 with several weeks of fieldwork.
Each member contributed at least £60 to the common pool of the New
Australia Co-operative Settlement Association to buy land, cattle, and
tools, and an independent currency was used within New Australia.
The New Australian community consisted of two groups of people.
The majority of the settlers were the “bushmen,” that is, shearers and
other agricultural workers from Queensland and New South Wales. Lane
was convinced that these “courageous” and “rough looking” but “tender”
workers were ideal participants to live up to his ideals of socialism and
brotherhood and to face the privations and hard work awaiting them in
Paraguay (Kellett 1997: 11). Nearly all the bushmen were young bachelors
of little or no formal education. The other group of settlers consisted of
educated white-collar middle-class workers, such as journalists, teachers,
and tradesmen. Their common goal was to establish a settlement of at
least 1,200 families independent from the Paraguayan society – a goal that
was not achieved: by the end of the century, approximately six hundred
Australian and British settlers had joined the socialist project in eastern
Paraguay (Souter 1991: 282).
At the time when the Australian pioneers arrived, the sociolinguis-
tic situation in Paraguay offered conditions that seemed ideal to ensure
the maintenance of the community’s racial and linguistic purity. The
222 danae m. perez-inofuentes
communication with the Paraguayan administration was carried out in
Spanish and made possible by a number of interpreters. The community’s
immediate neighbors, however, spoke Guarani. The inferior status of the
Guarani language and people was expected to inhibit miscegenation. Thus,
even though Guarani and Spanish were indispensible to get in touch with
their immediate environment, the maintenance of the English language
seemed to be secured.
Apart from the communicative difficulties, the Australian settlers in
Paraguay encountered living conditions that challenged them additionally.
On the one hand, the clearing of the land was laborious, and three serious
droughts during the 1890s destroyed most of their first crops. Other incon-
veniences such as wild animals, insects, diseases, and untamed cattle made
life on the colony even more difficult. According to Henry Connelly,1
one of the original settlers, it was particularly the women who suffered
from the hardships and instigated quarrels among the settlers. On the
other hand, the group’s communistic enthusiasm was not to last. Lane’s
totalitarian leadership and commitment to abstinence from alcohol con-
sumption, mateship, and the maintenance of the color line were soon
subject to debate in New Australia, and the bachelors and bushmen, in
particular, pursued interests that diverged from those of the middle-class
families. In addition, some men soon bartered goods and alcohol with
the native families that inhabited their land. As a consequence, the colony
was divided after a few months. Lane’s loyalists, mainly educated middle-
class families, reinitiated the experiment in another location called Colonia
Cosme, whereas most of the bushmen and bachelors remained with the
cattle stock on the original site of New Australia (Livermore 1950). Today,
the two communities still exist, and none of them maintained English as
a communal language. However, the sociolinguistic developments of the
two villages diverge considerably. In what follows, I give an account of each
community individually.
3.2 Phonology
In what follows, I first provide an impressionistic account of the six infor-
mants’ speech by describing their use of Wells’ (1982) set of vowels. The first
items in Table 10.1 represent the more frequently used vowels in each case. I
then describe two consonant features that are highly distinctive in English
dialectology, rhoticity and intervocalic T. Finally, I point out some of the
contact-induced particularities of PAE to give a more detailed impression
of this unique trilingual setting.
The comparison of two different generational groups illustrates that
their accents differ considerably. The vowel system of second-generation
speakers of PAE is consistent except for the variation of the bath and
goose vowels. The variation of the bath vowel, however, concerns only
the item dance pronounced with /æ/, whereas all other words, such as master
or half, are consistently pronounced with /ɑ/. The regular use of diphthongs
indicates that second-generation PAE was non-rhotic. This assumption is
confirmed by Connelly’s manuscript, which contains words such as warter
“water,” conferdance “confidence,” or oppersition “opposition.” The spelling
with hypercorrect /r/ suggests that Connelly did not pronounce postvocalic
/r/. This observation is corroborated by Lane’s novel, which reproduces the
workingmen’s speech with words such as pus “purse.” Finally, it is noticeable
that intervocalic T is always voiceless among second-generation speakers.
This consistent pronounciation of second-generation speakers contradicts
the assumption that the founders’ varieties were heterogeneous. Rather, the
data analyzed here suggest that the Cosme community spoke a consistent
variety of English close to British Standard English with little dialectal
variation.
230 danae m. perez-inofuentes
Table 10.1 Anglo-Paraguayan English vowel system
Group i Group ii
Keyword (2nd generation) (3rd and 4th generation)
kit ɪ ɪi
dress ɛ ɛ
trap æ æɛ
lot ɒ ɒᴐ
strut ʌ ʌ
foot ʊ ʊ
bath ɑː æː ɑː æː
cloth ɒ ɒᴐo
nurse ɜ ɜəo
fleece i i
face eɪ eɪ
palm ɑː ɑː aː
thought ᴐː ᴐː
goat əʊ oʊ
goose uː ʊː uː
price aɪ aɪ
choice ᴐɪ ᴐɪ
mouth aʊ aʊ
near iə iɾ ɪɾ
square eə ɛɾ eɾ
start ɑː ɑɾ
north ᴐː ᴐɾ
force ᴐː ᴐɾ
poor ʊə ʊɾ ᴐɾ
happy ɪ ɪie
letter ə əɾ ə
3.3 Morphosyntax
The only features that lend themselves to morphological analysis at this
point of my research are verbal endings and agreement. The manuscript of
working-class New Australian Henry Connelly displays a high variation in
the present-tense paradigm, that is, final /s/ often occurs with all persons
such as women appeals to us and I loves them. This feature is also reproduced
in Lane’s novel in direct speech such as I says and we puts (Miller 2007:
130) and indicates that some of the original settlers probably stemmed
from northern England (Trudgill 2010: 40). This assumption is, in fact,
consistent with the origin of the velar fricative /x/ mentioned above. The
fact that none of the audio recordings of second-generation speakers con-
tains instances of generalized present-tense /s/ indicates, once again, that
the well-educated founders of Cosme spoke a more Standard-like vari-
ety of English with their offspring in spite of being speakers of different
vernacular dialects.
The speech of the third and fourth generations does not have one single
instance of regularized final /s/. There are, however, instances of irregular
verbal endings, be they from L2 acquisition or L1 influence. It has yet to be
determined if the omission of third-person /s/, as in that expression mean, is
due to the aspiration of final /s/ in Paraguayan Spanish or a typical feature
of an L2 variety of English.
3.4 Lexicon
Connelly’s written document gives evidence of the earliest Spanish and
Guarani items introduced in the PAE vocabulary. It contains a considerable
number of loanwords from both languages such as the term montie “bush,
scrub” stemming from Spanish monte, which soon became part of the
PAE vocabulary. Similarly, the item camp – also reported in Falklands
Island English, although with the meaning of “settlement” (Britain and
Sudbury 2010: 219) – is a calque from Spanish campo “field” and was used
in expressions such as in the camp “in the field/country.” Many words that
do have English counterparts appear in Spanish in Connelly’s document
such as stacion “train station” (noticeably hypercorrected without word-
initial /e/), informe “report,” or novillo “calf.” Also words referring to local
issues are typically Spanish and Guarani, for example, mate cocido “mate
Anglo-Paraguayan English 233
[morning] tea,” bombilla “drinking straw,” and lexical items that describe
the Paraguayan landscape as tuyu “swamp.”
At the same time, certain typical Australian words have survived over the
generations. Thus, the Anglo-Paraguayans use tucker (pronounced without
postvocalic /r/) and no worries conscient of the fact that these items identify
them as Australians. Last but not least, the descendants of New Australia are
proud of their family names since they distinguish them from the rest and
legitimize their identification with their heritage. Their pronounciation,
however, varies from the original English version, and Bates, Drakeford, or
Smith, for example, are realized as /bateh/, /raifo/, and /esmit/, respectively.
4 Conclusion
This chapter introduced a hitherto unknown English-speaking community
in Latin America, the Anglo-Paraguayan community, which had its origin
in the socialist New Australia project of William Lane. The community’s
extraordinary inception with the aim of setting up a white, Anglo-Saxon,
and socialist community in isolation offered ideal sociolinguistic condi-
tions for a new English variety to arise. Once in Paraguay, however, their
English heritage language was exposed to the unique Paraguayan diglos-
sia, which challenged its maintenance. After the settlement divided into
Nueva Australia and Cosme, each community evolved differently. Cosme
remained in isolation before it was deserted by the next generations. New
Australians, on the other hand, adapted quickly to their environment leav-
ing their heritage language behind. Only vestiges of the English language
are left nowadays in the two villages.
The English language itself has undergone considerable changes in the
course of the twentieth century. The description of the most outstanding
features of PAE outlined here shows that English has become an L2 variety
in Paraguay. Despite the heterogeneous nature of the founders’ varieties,
the second generation from Cosme seemed to speak a relatively homoge-
neous, standard-like variety of English. The third and fourth generations,
in contrast, display many L2 features of English as well as considerable
L2 influence. This development allows the conclusion that PAE did not
survive as a native language in Paraguay, but it did establish itself as an
L2 variety among the urban descendants of Cosme. In Nueva Londres,
where it is non-existent today, it might reappear as a L2 among fourth- and
fifth-generation descendants.
The Paraguayan case of language shift is of special linguistic interest
and has great potential for further research. The fact that English as a
234 danae m. perez-inofuentes
global and prestigious language disappeared at the cost of a low-prestige
indigenous language makes it unique and begs for further investigations
in ethnolinguistics and the ecology of language (Haugen 1972). English
became a means for social mobility in Cosme, a community with limited
financial resources, and contributed significantly to the abandonment of
the village. In prospering Nueva Londres, on the other hand, English was of
no use for the second generation and disappeared quickly. A comparison of
the two communities is therefore interesting as it sheds light on the decisive
factors that underlie the process of language shift. Given that in this context
the community shifted from the more prestigious international language
to the indigenous local language, concepts such as English as a “Killer
Language” can be viewed in a more differentiated manner as claimed by
Mufwene (2008: ch. 12), and general assumptions on language prestige and
the instrumental value of languages may be rethought.
References
Britain, David and Andrea Sudbury. 2010. Falkland Islands English. In Daniel
Schreier, Peter Trudgill, Edgar W. Schneider and Jeffrey P. Williams, eds.,
Lesser-Known Varieties of English: An Introduction. Cambridge University
Press, 209–23.
Cortés-Conde, Florencia. 1996. Is stable bilingualism possible in an immigrational
setting? The Anglo-Argentine case. In Ana Roca and John B. Jensen, eds.,
Spanish in Contact: Issues in Bilingualism. Somerville: Cascadilla Press, 113–22.
Crystal, David. 2000. Language Death. Cambridge University Press.
Jefferies, Julian. 2010. Anglo-Argentine English. In Daniel Schreier, Peter Trudgill,
Edgar W. Schneider and Jeffrey P. Williams, eds., Lesser-Known Varieties of
English: An Introduction. Cambridge University Press, 195–206.
Haugen, Einar. 1972. The Ecology of Language. Stanford University Press.
Kellett, John. 1997. William Lane and “New Australia”: a reassessment. Labour
History 72(2): 1–17.
Livermore, Harold V. 1950. New Australia. Hispanic American Historical Review
30(3): 290–313.
Melia, Bartomeu. 2011. Historia de la lengua Guaranı́. In Ignacio Telesca (coord.),
Historia del Paraguay, 3rd edn. Asunción: Taurus, 425–51.
Miller, John Maurice [William Lane]. 2007 [1892]. The Workingman’s Paradise.
Charleston: Bibliobazaar.
Mufwene, Salikoko. 2008. Language Evolution: Contact, Competition, and Change.
London: Continuum.
Souter, Gavin. 1991 [1968]. A Peculiar People: William Lane’s Australian Utopians
in Paraguay. Brisbane: University of Queensland Press.
Trudgill, Peter. 2010. Investigations in Sociohistorical Linguistics: Stories of Coloni-
sation and Contact. Cambridge University Press.
Anglo-Paraguayan English 235
Warren, Harris Gaylord. 1985. Rebirth of the Paraguayan Republic. The First Col-
orado Era, 1878–1904. University of Pittsburgh Press.
Wells J. C. 1982. Accents of English, 3 vols. Cambridge University Press.
Whitehead, Anne. 1997. Paradise Mislaid: In Search of the Australian Tribe of
Paraguay. Brisbane: Queensland University Press.
Zajicova, Lenka. 2009. El bilingüismo paraguayo: usos y actitudes hacia el guaranı́ y
el castellano. Madrid: Iberoamericana.
ch a p ter 1 1
Gullah West
Texas Afro-Seminole Creole
Ian Hancock
1 Introduction
The language of the Afro-Seminoles is an English-related creole whose
origins go back four hundred years to the west coast of Africa. Sharing its
origins with Sea Islands Creole (SIC, usually called Gullah or Geechee), it
continues to be spoken by a dwindling elderly population of fewer than
three hundred in south Texas, central Oklahoma and northern Mexico.1
Called Seminole (“shim-i-no-li”) in the Brackettville, Texas, community and
Mascogo in the sister community in Nacimiento, Coahuila, Afro-Seminole
Creole (ASC) dates to the time of separation from Florida, following that
territory’s being sold to the United States in 1821.
Most of its living speakers in Texas were born in the closed environment
of the Fort Clark Indian Reservation; since 1917 their families have been
living in the town of Brackettville (“Brackett”) in close contact with other
Americans, and few people younger than about sixty-five are fluent in the
language. But because of the independence of the Afro-Seminoles, and
their earlier geographical and cultural isolation from the larger society,
their language has preserved far more of its original character than has
Sea Islands Creole, spoken on the Atlantic seaboard, to which it is closely
related.
1 That Suzanne Romaine should claim (2001: 160) that the language is extinct is puzzling, since she
has not ever visited the community, and must have relied solely on my own publications for her
information about Afro-Seminole Creole.
236
Gullah West: Texas Afro-Seminole Creole 237
we have records of it, but on the basis of its modern descendants, and the
historical facts we have gathered, to some extent we are able to reconstruct
the circumstances of its origins, and even to know what it might have
sounded like.
There are records from 1553 onwards of groups of British sailors going to
live on the West African coast more or less permanently. Some of these were
criminals, some political exiles, and others were simply attracted by local
African life and preferred to stay rather than return to Europe. Whatever the
reasons, these sailors were all men, and nearly all between the ages of 15 and
30. They are referred to in modern writings as lançados, a Portuguese word
meaning men who were “thrown” from the ships – Portuguese because the
first lançados, and indeed the first Europeans, to settle in West Africa were
Portuguese. The lançados we are concerned with in this case came from
all parts of Britain and spoke a great many different dialects of English
(standard English was still emerging so no one spoke that); there were no
radios or newspapers; literacy and schooling were privileges of the wealthy,
and contacts with speakers of other dialects were few. Those joining a
ship in port for the first time would have presented some problems of
communication to their new shipmates.
In the course of time, the sailors developed amongst themselves a kind
of English they could all understand. They did this by keeping those words
and constructions they had in common, and discarding whatever extreme
dialect forms might have hampered communication. This process, called
leveling, is something like creolization, except that because of the kinds of
speech involved, which were all dialects of the same language instead of
totally different languages, the “common denominator” level they reached
was far less different from the grammars of their different English dialects
than it would have been in a truly multilingual situation.
This leveled English was even more distinctive because it was used on
board ship, and as a result had a strong nautical flavor. Each sailor speaking
this Ship English could of course also speak his natural home dialect, though
if he did, it might have been difficult for his fellow crewmen to understand
him properly. When these sailors settled down on the Guinea Coast of West
Africa, they married African women and in the lançado–African households
that they established, the Guinea Coast Creole English slowly developed.
This kind of social arrangement existed between about 1580 and 1630,
after which time the English started to get their slaves directly from Africa
instead of from the Dutch in the West Indies. When the English became
recognized as slavers, they were naturally no longer welcomed in the same
way, and had to build castles and forts to protect themselves if they wanted
238 ian hancock
to stay in West Africa. But by this time, a couple of generations of Afro-
Europeans – the first Creole people – had grown up, and their language and
new society had become well established. The sailors continued to speak
Ship English, adding to it from their home dialects because they were now
no longer at sea, and adjusting it to the speech of their wives, who probably
spoke Serer, or Wolof, or Mandinka, or Temne, and who were also learning
to speak like their men. They kept the vocabulary of the men’s speech even
though the Africans greatly outnumbered the Europeans mainly because
creolized English was useful all along the Guinea Coast from settlement
to settlement, while the African languages were spoken over fairly small
areas and made trading difficult outside the home area. There was also a
Portuguese creole spoken along the same coast, and it is still spoken today
in parts of West Africa. Even Africans from elsewhere who came to live and
work in the Creole communities (people called grumetes or grumettos or
laptots) learned to speak Creole, and since, in many cases, it was the Creoles
rather than the Europeans who kept the slaves imprisoned on the coast
before transportation, it was Creole, not English, which they also learnt
to speak. It was essential that they knew at least some of that language,
because whenever possible, slaves who spoke the same language were kept
apart from each other. Creole was all they had in common. The Africans
learnt Creole so that they could speak to each other, not so they could
speak to their captors.
In the early years of the slave trade, slaves were kept waiting on the
coast for a year or even longer before shipment across the Atlantic, and
then that voyage could last for many weeks. This gave the earliest arrivals
plenty of time to acquire a good knowledge of Creole, and even when
the volume and efficiency of the slave trade increased, so that the newly
arrived slaves would not have had time to learn it, they still learnt some
from the slaves they were put to work with once they reached the Amer-
icas. Since they also worked with indentured whites, especially in Bar-
bados and North America, bondservants who were usually Scottish or
Irish and who spoke their own regional kinds of British English, and
since metropolitanization was already having an effect on the creole taken
across the Atlantic, it is probably safe to say that Black English, which
that speech has become today, never did have a wholly creole origin.
Because of the geographical isolation of the Sea Islands, and the compara-
tive absence of whites there, and because of the continual (and illegal) arrival
of creole-speaking West Indian slaves in the area until scarcely more than
a century ago, and the fact that slaves arriving on the Atlantic seaboard
were not all sent to other parts of the United States, Gullah does not
Gullah West: Texas Afro-Seminole Creole 239
have the same history which has produced Black English elsewhere in the
country.
3 Beginnings of Gullah
The British took most of their slaves to Barbados, which they settled in
1627, before distributing them to their other colonies. By 1795, well over
half of the c. 2,000 Africans in South Carolina, which was founded in
l670, were from Barbados, though after 1698 they were being brought in
more and more from Africa directly. South Carolina originally covered a
huge area, which even included much of what is today Florida. Georgia was
then Creek Indian country, and was considered to be free territory. When it
became a colony by charter in 1732, it immediately tried to prohibit slavery,
but because of pressure from South Carolina its attempt was unsuccessful.
Up until 1749, Georgia had been getting its own slaves from Carolina,
but after that date began to import them from elsewhere. Unlike Car-
olina, Georgia continued to bring slaves in from the West Indies, and
until a halt was drawn to the importation of West Indian and African
slaves in 1770, they were arriving from Jamaica, Antigua, Barbados,
St. Croix, St. Kitts, St. Martin, St. Vincent, Montserrat, Nevis, Martinique,
Guadeloupe, Grenada, and Cuba, a pattern of settlement quite different
from that in South Carolina.
It is possible that Barbados, because of its history and settlement, never
did develop its own dialect of Creole English or, if it did, it was a highly
anglicized variety. But the other islands had their own creoles, and slaves
from Nevis, St. Vincent, and elsewhere must have experienced little dif-
ficulty in communicating with each other on the North American plan-
tations. Gullah appears to have grown out of a leveling of all of these. In
some respects, the process of leveling, which produced Gullah, was not
unlike the leveling which produced Ship English.
Gullah has characteristics found in several of the Caribbean island cre-
oles, but it isn’t exactly like any single one of them. It also has features
in common with Guinea Coast Creole brought in later with West African
slaves, and which do not turn up in Afro-Seminole, which separated from
the main body of Gullah speakers at an earlier time before that happened.
The West African creole most closely associated with Afro-Seminole Creole
is Krio, spoken in Sierra Leone and Gambia. Krio is a direct descendant
of Guinea Coast Creole English, but like Gullah is also something of
a hybrid, having overlays from Nigerian Pidgin, Jamaican Creole, and
almost certainly Gullah too, brought into Sierra Leone with the resettled
240 ian hancock
Nova Scotians. The picture gets complicated, for while there are reasons to
suspect that some of the Nova Scotians who arrived in 1787 spoke Gullah
(though no actual proof ), there were also Krio (and Mende and Vai) speak-
ers coming into Carolina from Sierra Leone. Certainly Krio and Gullah
share far too many similarities for it to be merely coincidence.
2 Although the popular association of the word Seminole today is with the Indian population in
Florida, according to Giddings (1858: 3) it was first used to refer to the African escapees into that
region, and was only later applied by the Creeks to the Indian fugitives. “Seminole” has generally
been supposed to derive from a Native American word cima meaning “a type of wild grass,” but
more recently another etymology in the Arawak word sı́maran meaning “bow and arrow” has been
proposed by José Arrom (1986). The Indians themselves pronounced cimarrón as cimalon or cimanol
transposing the “m” and the “l.”
Gullah West: Texas Afro-Seminole Creole 241
severe. The migration of Africans to the fort in Florida had stopped by the
mid 1760s:
Spanish power in Florida, moribund for a score of years, had been extin-
guished . . . the British were at last in control and runaway Negroes from
South Carolina and Georgia could no longer find refuge under the walls of
St. Augustine. (Porter 1971: 171)
This was going on even into the early 1820s. In a letter written at that
time, Charles Pinckney (1757–1824), one of the drafters and signers of
the Constitution of the United States (Powers 1998), complained about the
numbers of slaves escaping from South Carolina into Florida, which by then
had become US territory. Thus the black maroons, or Afro-Seminoles, were
seeking refuge in Florida between about 1690 and the 1820s; that they were
mainly from Georgia during the earlier part of that nearly 140-year time-
span, and that most Georgian slaves were West Indian rather than directly
African, supports the argument for a Caribbean origin of Afro-Seminole
Creole. One clue to the early makeup of the Afro-Seminole population is
provided by the words Joo and Joomaican, who are remembered as having
been present during the early period.
At the time that Florida became US territory, slavery was still legal,
and raids to capture free Africans (as well as Indians) created considerable
problems for Governor Jackson in his efforts to develop the new territory,
including further bloody conflicts; in December, 1835 Major Francis Dade
and his troops were ambushed by 300 Seminole warriors near Fort King
(Ocala), starting the Second Seminole War, an episode leading to the mass
removal of Seminoles to Indian Territory in what is today Oklahoma. By
1834, 3,824 Indians had been removed to the west. The war lasted until
1842, by which time 4,420 Seminoles had surrendered and been sent west.
From 1855 to 1858 the Third Seminole War (also known as the Bowlegs’
War) took place, when Billy Bowlegs and his family were captured and
deported to Indian Territory. Only about 300 Seminoles – almost all of
them Indians – remained in Florida, where they had been granted five
242 ian hancock
million acres of land further south in the Everglades. The first Indian
Seminoles from British territory were Oconee people from Milledgeville,
Georgia, who moved into Florida in 1750, over half a century later than the
first African escapees. These were joined by the Muskogee (cf. Mascogo as
a Seminole ethnonym), and following them were the Apalchicola, Chiaha,
Hitichi, Sawokli, and Tamathli, all of whom lived in the Chattahoochee
River area in western Georgia, and all of whom spoke dialects of Hitichi.
In 1767 they were joined by the Maskogee-speaking Eufala from Alabama,
and in 1788 other Maskogee-speaking groups also joined them. Following
the Creek War in 1813–14, the number of Indian Seminoles tripled because
of new arrivals from Georgia and Alabama – the Yuchi from Georgia, the
Alabama (from Alabama), the Yamassee, and the Apalachee. Today, the
Indian Seminoles in Florida speak two quite distinct languages, both of
them Mushkogean: Muskogee and Mikasuki. Groups of Back Seminoles
left Florida for other places as well; some went to the Bahamas (Wood
1980; Howard 2002), some were reportedly in Guanabacoa, Cuba, as early
as 1820, and others were invited to stay with the Cherokee. Still others
decided to remain in Florida.
4.1 Oklahoma
In 1849, some of the Oklahoma settlers applied to the Mexican government
for permission to go and live there, possibly because they believed they
would be more at home in a Hispanic environment and perhaps could
speak Spanish, but particularly because almost as soon as they had arrived
in Indian Territory, the US government declared them legally to be slaves,
while slavery had already been abolished in Mexico some twenty years
before. A group of about 500 Black and Indian Seminoles left Oklahoma in
the late fall of 1849, crossing Texas where they were joined by two hundred
Kickapoo Indians in the Brazos river valley near Waco, and crossing into
Coahuila, Mexico, in July 1850. At first the Black Seminoles settled in
Moral, not far from the Texas border, while the Indian Seminoles settled
separately at La Navaja and the Kickapoo at Guererro. Later the Black
Seminoles moved a hundred miles further into Mexico to Musquiz, soon
after that moving a few miles away to El Nacimiento de los Negros, with
a few families going instead to Matamoros. The Kickapoo moved to the
nearby colony of El Nacimiento de los Indios, but practically all of the
Indian Seminoles decided to return to Oklahoma (Opala 1980).
Slave raids continued even in Nacimiento, however, led mainly by
US Army Captain Warren Adams who was especially concerned with
recapturing slaves who had escaped from Texas; by now some 3,000 were
Gullah West: Texas Afro-Seminole Creole 243
living as fugitives in the Sierra Madre mountains. While the effects of these
raids hurt the Seminoles, much greater losses resulted from a smallpox
epidemic brought back from an encounter with the Comanches in 1857,
which left 74 people dead.
During their move west the Seminoles also encountered other Indian
languages such as Cherokee and Biloxi; in Mexico and Texas they interacted
with speakers of Kickapoo, Lipan, and other languages; the word ma:skô:ki
(Haas 1940:49, Loughridge 1964) is the Creek self-designation; people of
African descent are called (s)tilûsti in that language. The Mascogos or Black
Seminoles today do not speak any Indian languages, although individuals
knowing some words and expressions were alive into the 1970s and have
been recorded (e.g. kokka-yenna “where are you going?” kwa-he “(I’m going)
home”). But the fact that they were employed as interpreters for the US
Army a century before that is evidence enough that they were familiar with
various Native American tongues.
4.2 Mexico
In Mexico, the Black Seminoles met another Creole-speaking group who
were already there. These were the Black Creek who, like the Afro-
Seminoles, were originally Africans, who had become acculturated to the
Indians they lived with without losing their creole language. They were the
Africans who lived with the Upper Creek in Georgia, and who had also
been sent west to Indian Territory. While the Afro-Seminoles, who lived
with the Lower Creek and others in Florida, left Tampa Bay by boat for
New Orleans and traveled to Indian Territory via the Mississippi River, the
Black Creek reached Oklahoma overland. They were brought to Coahuila
and left there by their Indian owners, who had been negotiating for land
for them since 1834. In addition to these two groups, the community was
also being joined by “state-raised” men and women escaping from slavery
in Texas via an underground railroad leading south into Mexico. Such fam-
ilies as the Gordons and the Shields descend from these fugitives. Although
members of the Brackettville and Nacimiento communities recognize their
various origins and are pretty well aware of which family is Black Creek and
which is Seminole or one of the smaller contributing groups, the common-
est designation used by everybody, especially with outsiders, is Seminole.
4.3 Texas
In 1870 following negotiations with Mexico, the American government
sent US Cavalry Captain Franklin Perry to Nacimiento to recruit the
Seminoles, because of their reputation as fighters and because of their
244 ian hancock
familiarity with Native Americans, to come and help the US Army drive
the Plains tribes out of west Texas so that settlement there would be less of a
problem for the whites. The Seminoles agreed, and garrisoned themselves
under the leadership of General Bullis in Fort Duncan at Eagle Pass in
Maverick County, and Fort Clark at Brackettville in Kinney County, in
south Texas. They were successful, and continued to serve the United
States until they were discharged in 1914. For three more years they lived
on their own reservation at Fort Clark, but this was taken from them,
and since 1917 they have lived across the highway in Brackettville. Some
returned to Nacimiento, and others have gone to live in the neighboring
towns of Del Rio, Eagle Pass, Ozona, and elsewhere. Some even live now
in California, Missouri, and New York, and still make trips from time to
time to Brackettville at New Year, on Juneteenth, and for Seminole Day in
mid September.
The Seminoles were never informed of their rights as American Indians,
and later attempts to be included on the Seminole Register and to obtain
land of their own were ignored. As Woodhull says,
General Bullis was greatly honored, and his name and fame are held in
reverence by the people of the Southwest frontier, but his scouts have been
disbanded and their families have been moved off the Reservation at Fort
Clark. They are not entitled to consideration as Indians, because they did
not register under some provision of Congress, of which they knew nothing,
and they get no consideration as negroes. (1937: 127)
– Ah lib’ don’ deah til ah wuz nin’te’n, and ah don’ niva’ go bock nuh
mo’. Too ho’d don’ deah.
– Duh youngun’ dem nuh lika’ we way; dem nuh lika’ we duh talk lika’
dis.
– Duh tarm we gib you a’ dese: you lay down arm and stop de war;
you sojas go back an’ stay in der fo’ts; we Indyen cross ober duh
Ouitaloochie [River]; an from dis time fort’ for ebber affer, we make
de Grand Ribber duh line o’ boundary atween de two. We promise lib’
in peace and good tarm wi’ all white neighbor. Dat all got say.
– An wuh fuh we submit? We not conquered! We whup you people one,
two, tree time. We whup you, damn, we keel you well too. Mek so
[why] e submit? We com’ heah gib’ conditions, not askum.
6.2 Grammar
6.2.1 Nouns
Table 11.1 lists the articles used in ASC. Nouns do not usually change for
plural by adding an -s at the end as in English; a few words like day and
Gullah West: Texas Afro-Seminole Creole 247
Table 11.2 Subject pronouns
in Afro-Seminole Creole
ah, me “I”
you, hunnuh “you” singular
e, him “he, she, it”
we “we”
hunnuh, yall “you-all”
dem “they”
ting sometimes take a final -s, and the word chile has its own plural chirren,
but the usual way to show that there is more than one of anything is to
follow the word with dem:
(3) Fo uh me frien’
“Four of my friends”
The same word dem, when placed after someone’s name, means that person
and his family or group of friends:
6.2.2 Pronouns
The word for “I” is nearly always Ah, but me is sometimes used in emphatic
constructions, and before negative nuh, especially in the expression me nuh
know “I don’t know.” E is also the commonest word for “he” or “she” or
“it,” but him is used for emphasis very frequently. (See Tables 11.2 and
11.3 for subject and object pronouns.)
248 ian hancock
Table 11.3 Object pronouns
in Afro-Seminole Creole
me “me”
you, hunnuh “you”
um, rum “him, her, it”
we “us”
hunnuh, yall “you-all”
dem “them”
The form of um [əm] with an r, i.e. rum [rəm], is only used when the
word before it ends in certain vowels; this is the same in West African Krio
and in Sea Islands Gullah:
(6) G’am ([gæm] = gi um) tuh rum
“Give it to her”
(7) Ah cyan’ membuh rum
“I can’t remember it”
6.2.3 Possession
Possession is shown through juxtaposition of nouns without any overt
marking on either the head or dependent.
(8) Pompey dahdy
“Pompey’s father”
(9) John Horse hoss
“John Horse’s horse”
(10) Me ahnty neighbuh Toyota
“My aunt’s neighbour’s Toyota”
Hunnuh is only a plural pronoun in most related creoles, but this
is not the case in ASC. Possessive pronouns (and demonstratives) (see
Tables 11.4 and 11.5) go before adjectives. If the possessive pronoun comes
at the end of a sentence, it is followed by own:
(11) Darra-dey cah duh we own
“That car is ours”
(12) E nuh look lukkuh e own
“It doesn’t look like his/hers”
Gullah West: Texas Afro-Seminole Creole 249
Table 11.4 Possessive pronouns in
Afro-Seminole Creole
dis “this”
dish-yuh “this,” close by
da “that” (dat when emphatic)
darra “that”
dem “those”
dem-yuh “these”
yanduh “those,” far away
The same word own can go with a few other words too:
(13) Duh who-dat own?
“Whose is it?”
(14) Dishyuh mus be somebawdy own
“This must be somebody’s”
6.2.4.1 Adjectives
These behave just like verbs, except that without a tense or aspect marker
they still can have a “present tense.” It’s hard to think of adjectives having
tenses, but it’s one way to explain the difference between dis leaf yalluh
and dis leaf duh yalluh; the first one means “this leaf is yellow,” a kind of
permanent state which includes the present since it is yellow while you
make the observation about it; the second one has duh which is the aspect
250 ian hancock
word for action in progress, so it would mean “this leaf is getting yellow,”
or “this leaf is yellowing.”
Adjectives can be used with the other tense and aspect markers too, just
like verbs.
(15) a. Dis leaf en yalluh
“This leaf will be yellow”
b. Dis leaf ennuh yalluh
“This leaf is going to turn yellow”
c. Dis leaf bin yalluh
“This leaf was yellow”
d. Dis leaf binnuh yalluh
“This leaf was turning yellow”
e. Dis leaf done yalluh
“This leaf has turned yellow”
Adjectives are made comparative by using the word mo in front of them,
or if they are just short words, by adding -uh to them. Sometimes both mo
and -uh are used together. The word for “than” is nuh:
(16) a. You ogliuh nuh me
“You’re uglier than I am”
b. You mo ogliuh nuh me
“You’re uglier than I am”
c. You mo tankful nuh me
“You’re more thankful than I am”
They are made superlative by using the word mos’ in front of them, or
if they are just short words, by adding -is to the end. Sometimes both mos’
and -is are used together:
(17) a. You duh de odis’ ooman
“You’re the oldest woman”
b. You duh de mos’ odis’ ooman
“You’re the oldest woman”
c. You duh de mos’ tankful man
“You’re the most thankful man”
3 Uh is the form of duh that is used after the tense word bin, listed below (binnuh = bin duh).
252 ian hancock
When the verb alone comes after the word fuh, and there is no subject
pronoun, the fuh means “to”:
(19) Ah bin too bex fuh talk tuh rum
“I was too angry to speak to her”
When a subject pronoun comes before fuh and a verb, then fuh means
“must” or “should”:
(20) Ah fuh talk tuh rum
“I should talk to her”
(21) Ah bin fuh talk tuh rum
“I should have talked to her”
(22) Hunnuh nuh fuh jrink da worruh
“You mustn’t drink that water”
ii. With tense marker bin:
(23) Meck e bin churray um?
“Why did he throw it away?”
(24) Dem bin nyus fuh talk Simanole
“They used to speak Seminole”
(25) Dem bin pit e dahdy een jail4
“They put his father in jail”
(26) Dem wale me
“I was beaten; they beat me”
(27) Dem tief e car
“her car got stolen; they stole her car”
iii. With tense marker en:
The future word en has several other forms, such as gwen, gwine, ennuh,
gwunnuh, and so on. The pronunciation without the g- seems to be the
most common, and probably existed in the creoles from very early on. In
Trinidad Creole the future word go has another form oh, and in Saramaccan
Creole spoken in South America, the only form now is oh. Even in American
Black English, “I’m gonna do it” has the variant pronunciation “I’m ’onna
do it” and even “I’m uh do it.”
4 This sentence would also be the translation of ‘his father was put in jail,’ because there is no passive
in Seminole.
Gullah West: Texas Afro-Seminole Creole 253
(28) Hunnuh en fin’ we deh
“You will find us there”
(29) De sperrit-dem en kyah you’way
“The spirits will carry you off ”
(30) Hunnuh gwine dead too
“You-all will die too”
iv. With aspect marker duh:
(31) Ah duh chry fuh do um
“I am trying to do it”
(32) Molly duh cratch e so foot
“Molly is scratching her sore leg”
(33) Ah fuh duh talk tuh rum
“I should be talking to her”
(34) Dem duh jouk um
“They’re teasing him”
(35) Him duh go tuh school
“He is going to school”
Notice that in the last example, “he is going to school” can have two
different meanings, as in English. It can be the answer to “where is that
boy on his way to now?” and also to "what is he doing these days?” Some
creole languages have different constructions for each of these.
v. With aspect marker done:
(36) You done bruck um fuh chrue now
“You’ve really broken it now!”
(37) Ah done tiyah fuh read
“I have become tired of reading”
Some verbs used with done can be translated with “become” as well,
when there is no object following.
(38) E done fix
“It has become fixed”
(39) E done cook
“It has become cooked”
254 ian hancock
Compare the forms in (38) and (39) with those in (40) and (41).
(40) E done fix um
“He has fixed it”
(41) E done cook da poke
“He has cooked that pork”
vi. Bin with done and duh:5
(42) All me peepil binnuh talk um
“All my people used to speak it”
(43) Dem binnuh shout een de chuch
“They were singing in the church”
(44) E bin done tell me bout you befo
“She had told me about you before”
vii. En with done and duh:
(45) By dis time tumorra hunnuh en done spen two whole week yuh
“By this time tomorrow, you’ll have spent two whole weeks here”
(46) Hunnuh en uh see me, nuh worry
“You’ll be seeing me, don’t worry”
6.2.6 Auxiliaries
Seminole has taken some other auxiliaries from English, including must,
could, and would, and their combinations mussa, coulda, and woulda.
(47) Ah nuh bin know seh ah could do um
“I didn’t know I could do it”
(48) Ah shonuff would like fuh go too
“I’d sure enough like to go too”
(49) Ah coulda tell you dat easy
“I could have easily told you that”
(50) E woulda spile fuh chrue
“It would really have spoiled”
Two other verbs with characteristic pronunciations in Seminole are ha
(“have”) and leh (“let”):
5 Bin duh is usually pronounced binnuh in ordinary speech
Gullah West: Texas Afro-Seminole Creole 255
(51) Dem chillen nuh ha nuttin fuh do
“Those children have nothing to do”
(52) E ha fuh git back fuh school
“She has to get back to school”
(53) Leh we go, bubbuh!
“Let’s go, sonny!”
6.2.7 The be verb
Be here covers all the different forms of that verb – is, am, are, was, were,
being, and so on. In Seminole, there are different ways of saying this.
i. Be between nouns is duh in the present tense, (gw)en be in the future,
and binnuh in the past:
(54) Him duh de o’des one aroun yeh
“He’s the oldest one around here”
(55) Mr. Toughtry bin duh lyer
“Mr. Toughtry was a lawyer”
(56) E bin wan’ fuh be lyer
“He wanted to be a lawyer”
(57) Duh da e en be
“That’s what he’s going to be”
Duh is also used as a “highlighter” when certain words in a sentence need
to be emphasized. It this case, they come at the beginning as in (58a–b) or
with question words as in (59a–c).
(58) a. We wan’ talk tuh John
“We want to speak to John”
b. Duh John we wan’ talk tuh
“It’s John we want to talk to”
(59) a. Duh wisseh hunnuh duh gwine?
“Where are you going?”
b. Duh who-dat bin call me name?
“Who called my name?”
c. Duh wuh dem bin tell hunnuh?
“What did they tell you?”
Unlike the other creoles (except SIC), Seminole grammar will not allow
for verbs to be brought forward in the same way; both West African Krio
256 ian hancock
and Jamaican Creole can say duh buy you buy im or duh tief you tief im?
(“did you buy it or did you steal it”), but in Seminole it would have to be
you buy um or you tief um?
ii. Be in the sense of “exist” or “be in a place” (like Spanish estar), is dey:
(60) Hunnuh book dem dey pun da cheer
“Your books are on that chair”
(61) Muskittuh bin dey ebbawey
“Mosquitoes were everywhere”
(62) Ah en dey een me room
“I’ll be in my room”
6.2.8 Negatives
There are many examples of negative sentences in the earlier pages. Negative
constructions are usually made by putting nuh (or no or nah) right after
the subject noun or pronoun:
(63) Me nuh sabby um
“I don’t know him”
(64) Me oncle nuh know
“My uncle doesn’t know”
(65) Me ahnty nuh bin wan’ fuh know
“My aunt didn’t want to know”
(66) En ah n’ en tell um
“And I’m not going to tell her”
When a sentence has two parts, i.e. a subject and an object, both are made
negative, so it is correct Seminole grammar to say we nuh see nuhbawdy en
we nuh bin eat nuttin, “we didn’t see anyone and we didn’t eat anything.”
The verbs could, would, coulda, woulda, and kin (“can”) don’t have
negatives with nuh; the negative forms of these verbs are couldn, wouldn,
couldna, wouldna, and cahn’ or cyahn’.
The aspect marker done, when made negative, is not ∗ nuh done but
nabbuh: E nabbuh shem “he hadn’t seen her.”
(71) Dishuh yaze wey de doctor bin fix still nuh right
“This ear that the doctor fixed still isn’t right”
(72) Dem piece uh ood wey dey onneet da stove en ketch fire ef you
nuh moobe um
“Those bits of wood that are under that stove will catch fire if you
don’t move them”
iii. Following a verb or an adjective:
how “how”
meck, meck-so “why”
wisseh, wey “where”
wuh-time, win “when”
tuhday “today”
turruh-day “the other day”
soon “soon,”
soon een de monin “early in the morning”
6.3 Tags
Tags are little words you stick on the end of a sentence to give it a particular
tone. Two common tags in Seminole are enty and nuh:
(80) Nuffuh peepil bin deh dey, enty?
“Plenty of people were there, weren’t there?”
(81) Dem en come back, enty?
“They’ll come back, won’t they?”
Enty can also come at the front of a sentence as in example (82) or used
in combination with nuh as in example (83).
(82) Enty dem yie bin shet?
“Weren’t their eyes closed?”
(83) Nuh loss um, nuh
“Don’t lose it, will you”
6.4 Lexicon
Aspects of the structure and lexicon of ASC have been described elsewhere
(Hancock 1977, 1980a, 1980b, 1986, 1993, 1998), but a few characteristics
of the language may be noted here. ASC lacks much of the Mende and
other African-derived lexicon found in SIC. Nevertheless, ASC contains
about forty words of African provenance, some half of which are traceable
to KiKongo/KiMbundu, the balance to languages of the Guinea coast.
On the other hand it has a number of words of American Indian and
Spanish origin not found in SIC. Some are given here; further examples
and discussion are found in Hancock (1998).
With matches in Bantu: oolah “bedbug,” pingy “cooking pot,” cootie
“stunted pig,” teemuh “dig a hole,” zoondoo “a hammer.”
260 ian hancock
With matches in Twi: Cuffy “male given name,” Cudjo “male given
name,” kunkie “a tamal.”
With matches in Upper Guinea languages: boontuh “buttocks,”
chikka-bode “teeter-totter,” tabby “mud daub,” chooklah “girlfriend,”
ninny “breasts.”
The English items match in the main those found in other anglophone
Atlantic creoles, pointing to both place (southwestern England) and time
(the eighteenth century).
From SW English dialects: weekaday “weekday,” mole “fontanelle,”
yeddy “hear,” leff “leave,” broke “break,” loss “lose,” ees “yeast,” ood
“wood” (see Hancock 1994).
From Scottish English dialects: pit “put,” snoot “snout,” wurrum
“worm,” graytuh “grate,” bresh “brush.”
From Spanish: banyuh “wash,” kwahah “make cheese,” matatty
“grindstone,” soakettuh “mud,” beeoleen “violin,” calpintero
“woodpecker,” treego “rice,” huckle “adobe hut.”
From Native American languages: suffki “corn porridge,” stammal
“ground corn,” poleyjo “hominy,” polijotee “a corn-based drink.”
There are words for which no etymology has so far been found, such as
babba “carry on the back,” or skiffy “vagina” (though cf. Krio bamba “carry
on the back” and Bahamian Creole skiff “young woman”).
Appendix 1: Texts
Goot eebnin, goot eebnin, how hunnuh, how hunnuh duh do? Ah hope
hunnuh duh do fine, Ah duh do awright . . . now ah des’ come yuh dis
262 ian hancock
eebnin fuh see how hunnuh duh do en fuh pay me respecks tuh de dead.
Da duh me frien Silas Hall, en e dead yeah, e dead, you know e bin fuh
dead. Now nuh lahf, nuh lahf, hunnuh en die too, hunnuh en dead one
o’ dese days, see de man right deh? E duh dead; ef you nuh believe me
go obuh deh en joog um, e n’en’tannup, e nen hulluh, e n’en say nuttin.
Hunnuh gwine dead too!
Sistuh Phyllis, duh one ting ah wan’ fuh know; duh wisseh you bin git da ole
deep baid out deh fum? D’ole baid so deep e jeh’ like uh man fuh leddown
een e grabe, e ha fuh stan’ up fuh tun obuh een um!
All a we binnuh lib yuh fum de fus time da Cunnel McKenzie call we
fuh come. Hunnuh mussuh bin yeddy seh duh we wey done clay-out dem
Injin; whey-ever dem bin dey, duh we bin clay dem off dis lahn.
One ole king bin yeddy seh dem bin fuh pit one mo nyunger king een
e place, en e biggin fuh sorry bout um. E wan’ fuh keep e trone, so e call e
sojuh-dem fuh go kill all de peepil een de nation, fuh nuh leh nubbawdy
nuh dey fuh show de nyew king how fuh do e wuck. So de sojuh-dem teck
dem gun en dem go kill all de peepil. Nubbawdy nuh bin leff fuh show
de nyung king how fuh do e job. Atuh dat de fus’ king sen fuh de odduh
one seh him fuh come tuh him house eebnin time, en e fuh bring one fat
bohog long wit um. Time de nyung king yeddy dis e sen answer back suh
steaduh him fuh go to de fus’ king house, him fuh come tuh him house,
en e en gi um de bohog fuh present.
References
Arrom, José Juan. 1986. ‘Cimarrón’: apuntes sobre sus primeras documentaciones y su
probable origen. Serie monográfica no. 18. Santo Domingo (DR): Ediciones
Fundación Garcı́a-Arévalo.
Dillard, Joe. 1972. Black English: Its History and Usage in the United States. New
York: Random House.
Drechsel, Emanuel. 1976. Pidginization and creolization in North American Indian
languages: Mobilian Jargon and Afro-Seminole Creole. Unpublished report
to the National Science Foundation.
Evans, Christopher. 1990. A scout’s honor. The Fort Worth Star Telegram, March
25, pp. 7–8.
Giddings, Joshua R. 1858. The Exiles of Florida: Columbus: Follett. Reissued Bal-
timore: Black Classic Press, 1997.
Gonzales, Ambrose E. 1922. The Black Border: Gullah Stories of the Carolina Coast.
Columbia: The State Co.
Haas, Mary. 1940. Creek vocabulary. Unpublished mauscript.
Hancock, Ian. 1975. Creole features in the Afro-Seminole speech of Brackettville
Texas. Caribbean Linguistic Society Occasional Paper, no. 3.
1977. Further observations on Afro-Seminole Creole. Caribbean Linguistic Soci-
ety Occasional Paper, no. 7.
1980a. The Texas Seminoles and their language. Working paper of the Afro-
American Studies and Research Center of the University of Texas at Austin,
Spring.
1980b. Gullah in Texas. In Joe Dillard, ed., Perspectives on American English.
The Hague: Mouton, 305–33.
1986. On the classification of Afro-Seminole Creole. In Michael Montgomery
and Guy Bailey, eds., Language Variety in the South: Perspectives in Black and
White. Tuscaloosa: Alabama University Press, 85–101.
1993. Mortars and metates. In Peter Seitel, ed., Festival of American Folklife.
Washington, DC: Smithsonian Institution, 59–61.
1994. Componentiality and the creole matrix: the south-west English contribu-
tion. In Montgomery, ed., 94–114.
264 ian hancock
1998. History through words: Afro-Seminole lexicography. In L. Fiet and
J. Becerra, eds., Caribbean 2000: Identities and Cultures. Rockefeller Founda-
tion Publication, University of Puerto Rico, 87–104.
Haynes, Lilith. 1976. Candid chimaera: Texas Seminole. Term paper, Department
of English, New Mexico State University, Las Cruces.
Howard, Rosalyn. 2002. Black Seminoles in the Bahamas. Gainesville: University
Press of Florida.
Littlefield, Daniel F. 1977. Africans and Seminoles. Westport, CT, and London:
Greenwood Press.
Loughridge, R. M. 1964. English and Muskokee Dictionary. Okmulgee: Baptist
Home Mission Board.
Montgomery, Michael. ed. 1994. The Crucible of Carolina: Essays in the Develop-
ment of Gullah Language and Culture. Athens, GA, and London: University
of Georgia Press.
Opala, Joseph. 1980. A brief history of the Seminole freedmen. Occasional paper
no. 3 of the African and Afro-American Studies and Research Center, Austin:
University of Texas.
Porter, Kenneth Wiggins. 1947. Freedom over me: the story of John Horse (Gopher
John, c. 1812–1882) – Seminole Negro chief and his people in Florida, the
Indian Territory, Mexico and Texas. Unpublished typescript.
1971. The Negro on the American Frontier. New York: Arnos Press.
Powers, Bernard E. 1998. A founding father and Gullah culture. National Parks
11/12: 26–9.
Romaine, Suzanne, 2001. Afro-Seminole Creole. In John Algeo, ed., The Cam-
bridge History of the English Language, vol. 6: English in North America.
Cambridge University Press, 160.
Sivad, Doug [Douglas Davis]. 1993. African Seminoles. In Ron Sakolsky and
James Koehnline, eds., Gone to Croatan: Origins of American Dropout Culture.
Edinburgh: The AK Press, 263–70.
Turner, Lorenzo Dow. 1949. Africanisms in the Gullah Dialect. University of
Chicago Press.
Wood, David E. 1980. A Guide to the Seminole Settlements at Red Bays, Andros,
1817–1980. Nassau: The Bahamas Government Printing Department.
Woodhull, Frost. 1937. The Seminole Indian scouts on the border. Frontier Times
17(3): 118–27.
part iii
1 Introduction
After 140 years of near-total isolation, the inhabitants of Palmerston Island,
a tiny atoll in the Cook Islands group, have developed a unique linguistic
and cultural identity that draws on both English and Polynesian back-
grounds. They consider themselves ‘English’ in many ways – ethnically,
culturally, and linguistically – yet also have strong ties to the rest of the
Cook Islands, and to New Zealand.
Palmerston, then uninhabited, was settled in the early 1860s by a small
group that included the Englishman William Marsters, his three Cook
Island wives, a Portuguese-speaking man named Jean-Baptiste Fernandez,
and a small group of other Cook Island men and women, who may or
may not have remained beyond the initial few years (see Hendery 2013
for more details of the island’s early history). In 1877 when the first mis-
sionary visited, he noted that there were around thirty people on the
island (Gill 1877). The inhabitants today each trace their ancestry back to
Marsters and one of his wives and are monolingual speakers of Palmer-
ston Island English. In July–August 2009 when I conducted my field-
work there the population consisted of 13 adult women, 13 adult men,
and 28 children. Five of the people currently on the island are not orig-
inally from Palmerston: this includes the teacher (from New Zealand,
now married to a local man), two Rarotongan women married to local
men, and the two missionaries, from another island in the Cook Islands
group.
The island has been very isolated throughout its history. There is no
regular transport to or from it, and it is 400 km away from the closest
other inhabited islands. Up until around eight years ago, there were no
This research was supported under the Australian Research Council’s Discovery Projects funding scheme
(Project number DP110103714). Rachel Hendery was the recipient of an Australian Research Council
Postdoctoral Fellowship.
267
268 rachel hendery
moorings, so the visitors were few, as it was dangerous to anchor for long.
During the twentieth century, sometimes many years passed without any
contact with the outside world. Nowadays around thirty yachts in total
visit each year, during the August–September ‘cruising’ season. Most of
these are visitors from the USA or Europe. They are permitted to stay no
longer than three nights, and they sleep on their boats, coming ashore for a
few hours each day, usually meeting only one or two Palmerston families.
There is no television signal on the island, and until 2011, only one satellite
telephone and internet connection, used mainly by the teacher. Recently
Telecom has provided internet connections in all homes, but it remains
to be seen whether these are used and maintained. Some families listen to
radio, and most watch DVDs that they obtain from the visiting yachts,
so there is some potential outside influence on the language through these
media.
The small population and the isolation of the island mean that it is
possible to (a) record all Palmerston Islanders, and (b) track all external
influences on the language, making Palmerston Island a wonderful oppor-
tunity for studying the development of a linguistic and cultural identity in
small mixed-origin communities. During four weeks of fieldwork in 2009,
I spoke with all of the then-inhabitants of the island except for one elderly
man, who is blind and deaf.
Previous descriptions of Palmerston Island English have appeared in
Ehrhart (1996), and Hendery and Ehrhart (2011, 2012). Ehrhart (1996)
was based on fieldwork she conducted in the early 1990s, and as discussed
in Hendery and Ehrhart (2012) and as will be seen later in this chapter,
there seems to have been some change in Palmerston English in the past
twenty years. Hendery and Ehrhart (2011, 2012) describe the morphology
and morphosyntax of Palmerston English, with particular attention to
contact influence and ‘angloversals’. This present chapter will provide a
more extensive description of the variety than previous work.
2 Sociolinguistic description
People on Palmerston Island today divide themselves into three groups,
tracing their descent through the patriline back to each of Marsters’ wives:
Akakaingara, Matavia, and Tepou. The island itself, and all of the unin-
habited islets around the lagoon, are also divided in three, each belong-
ing to one of the three ‘families’.1 Family membership determines land
1 The term family on Palmerston Island refers to one of these three groups; for the smaller nuclear
family, the term household is used.
Palmerston Island English 269
inheritance, council representation, hunting and gathering rights, mar-
riage possibilities, and to some extent, day-to-day socialization patterns.
Perhaps because the islanders require marriage partners to belong to dif-
ferent ‘families’ (meaning that children usually have parents that are each
from different groups, and women change family affiliation at marriage)
there are no detectable linguistic differences between the three groups.
Linguistic variation on the island appears to be much more individual, and
also to some extent determined by social networks, in particular the bush
people/beachfella distinction.
Palmerston Islanders refer to those who live in the cleared sandy main
settlement area at the north end of the island as beachfellas. Bush people,
on the other hand, are those who have built houses further south, among
the palm trees that cover the rest of the island. In practical terms, the
distance between the most distant houses is only a few hundred metres. In
social terms, however, distinctions are made between the two groups, and
each believes the other to have different cultural and linguistic practices.
My own observations bear this out. Counting the frequency of various
typical Palmerston Island features in my data (pro-drop, lack of formal
plural marking on semantically plural nouns, use of -s marked verbs with
non-third-person subjects, and use of bare verb forms with third-person
subjects), and controlling for interlocutor and genre, I found a small but
statistically significant difference in the frequency with which the two
groups’ speech display these features (p < 0.05).
In terms of their orientation to other varieties of English, Palmerston
Islanders see UK English varieties as having high prestige, but most of
their contact with other varieties is with New Zealand English (towards
which they seem to have a neutral attitude) and Cook Island L2 English
(towards which their attitudes are rather negative). These linguistic atti-
tudes are mainly due to their historical connections. Palmerston Islanders
view themselves as distinct from the rest of the Cook Islands, and are
proud of their association with England. Public holidays are Queen Vic-
toria’s birthday and ‘Duke’s Day’, commemorating the day the Duke of
Edinburgh visited the island.
The school and the church play central roles in island life. Prescrip-
tive norms are disseminated through the school, and as the teacher is a
New Zealander, the children are encouraged to use standard New Zealand
English in written and formal spoken English. Some households explicitly
discourage their children from using certain standard features they have
been taught at school (especially the use of interdental fricatives, which
are not normally present in Palmerson Island English). The missionaries
who run the church are native speakers of Cook Island Māori, and are
270 rachel hendery
making an effort to teach this language through classes held in the school,
and through its use in prayers and hymns. This is a recent development,
and it is possible it will lead to more passive understanding of Cook Island
Māori among the population, as well as perhaps a more positive orientation
towards the language and towards Cook Island L2 English.
There is a large amount of linguistic variation on the island, phonetic,
morphological, and syntactic. This includes both inter-speaker and intra-
speaker variation. While much of it is variation between more ‘standard’
English features and Palmerston Island English features, some is among
various nonstandard alternatives. Generally I will discuss such variation
below in the description of the relevant features. For the sake of charac-
terizing Palmerston Island English as a distinct entity, it is necessary to
abstract across the variation and to some extent ignore the use of standard
English features even when they are the most frequent variant for some
speakers. Where necessary, therefore, I will adopt the (for Palmerston rather
artificial) concept of basilect and acrolect. This allows abstraction of the
pronunciations and grammatical features that are furthest from those of
Standard Englishes. There are no speakers of Palmerston Island English
who use this basilectal variety all the time, however, and I have never wit-
nessed even a single conversation where these basilectal features were not
heavily mixed with more standard English features. It should also not be
assumed by the reader that the features described here are necessarily older:
there is in fact evidence that variation existed right from the earliest decades
of the island’s settlement.
3 Features
3.1 Phonology
3.1.1 Consonants
Palmerston Island English has, at a minimum, the consonants listed in
Table 12.1. In basilectal Palmerston English there is no /ð/, /θ/, no /v/ or /h/.
Standard English /ð/ and /θ/ correspond to Palmerston /t/ word initially
and /s/ word finally. Standard English /v/ corresponds to Palmerston /w/
in all positions. Standard English /h/ corresponds either to a glottal stop,
or to zero: the glottal stop is common, but apparently optional. Standard
English voiced stops and fricatives are devoiced in the usual Palmerston
pronunciation.
None of these correspondences is absolute: for any word there is a range
of pronunciations along a continuum from Standard English to basilectal
Palmerston Island English 271
Table 12.1 Minimal consonant inventory for
Palmerston Island English
Plosive p t k
Fricative s
Affricate ʧ
Nasal m n ŋ
Lateral l
Rhotic r
Glide w j
Palmerston English. This means that one can analyse e.g. the voiced stops,
fricatives, and /v/ as having marginal phonemic status. They appear in free
variation with their voiceless counterparts, but only in the set of words that
have the voiced sounds in Standard English. The same cannot be said for
/h/. Glottal stop, /h/, and zero onsets appear to be in free variation in all
words that begin with a vowel. The /r/ is trilled.
Palmerston Island English has a phonotactic constraint whereby con-
sonant clusters in syllable codas are avoided. Clusters that would appear
in Standard English tend to be reduced in Palmerston English to a single
consonant. The velar nasal is not limited to syllable-final position to the
same extent as it is in Standard English, but can be found word initially in
proper names and in words that have their origin in Cook Island Māori.
3.1.2 Vowels
Variation in Palmerston Island English vowels seems to be on a continuum
from a system that clearly has its roots in a Northern English dialect through
to something more like that of Cook Islanders who are L2 English speakers.
Several Palmerston Islanders have vowels closer to those of Standard New
Zealand English. There are often striking differences between the vowels
of speakers from the same household, who grew up together and who have
the same degree of contact with New Zealand.
3.1.2.1 Monophthongs
Length rather than height appears to be the main distinction between
the fleece and kit vowels. While fleece tokens on average are higher
than kit tokens, there is a great deal of overlap between the two sets,
and any individual token of either set could fall within the average range
272 rachel hendery
of the other. Length, on the other hand, always distinguishes the two
sets, with fleece tokens generally lasting approximately twice as long as
kit.
In ‘traditional’ Palmerston Island English there is no split between the
strut and the foot lexical sets: both have the vowel [ʊ]. This was pre-
sumably the pronunciation inherited from William Marsters, and suggests
he came from the Midlands or further north in the UK. This is a very
salient feature to the community. Speakers will always produce examples
such as /kʊm/ or /bʊt/ when asked about how their grandparents spoke.
Most speakers today have the strut/foot split, but a small number of
lexical items that have remained in the foot set while other varieties of
English have assigned them to strut: bucket is always one of these, and
depending on the speaker the exceptions can also, for example, include
come, but, and cup. For some speakers, comer ‘newcomer/outsider’ has the
foot vowel while come has strut. The phonetic realization of the foot
vowel for Palmerston Islanders today is [ɒ] rather than [ʊ], but they pro-
duce something closer to [ʊ] when giving examples of their grandparents
speech. Several older speakers have [ɒ] for all words in both the strut and
foot sets.
There is a bath/trap distinction [a]/[æ], and unlike the strut/foot
pair, no indication that the split is recent. This helps narrow down Marsters’
place of origin, as this split occurred in southern England. This supports
the widely accepted story that he came from the Midlands (probably
Leicestershire or Birmingham).
The nurse vowel in Palmerston Island English is much more fronted
than it is in many other varieties of English, usually realized as [ɛː]. Length
rather than height is the main distinction between this vowel and that of
the dress set.
3.1.2.2 Diphthongs
The diphthongs of Palmerston English that are most different from those
of other English varieties with which it is in contact (Cook Island English,
Australian English, and New Zealand English) are those of face, goat,
and mouth. Face sometimes receives a monophthongal pronunciation
[eː]. For most speakers, however, it is realized as [ɛɪ]. The goat set has
an open rounded vowel, usually followed by a schwa off-glide: [ɔːə ]. The
mouth diphthong has a great deal of variation, with some speakers reg-
ularly producing a very New Zealand-English-sounding close vowel, but
most using a pronunciation more like conservative RP English: [aʊ]. Even
in speakers who otherwise show strong New Zealand English influence
Palmerston Island English 273
in their vowels, there is no trace of the near/square merger that New
Zealand English has undergone in recent years.
3.1.3 Stress
Stress in Palmerston Island English sometimes differs from the stress pat-
terns of Standard English. All of the cases of this I am aware of are ones
where Palmerston English has shifted the stress to the penultimate sylla-
ble, sometimes from the final syllable such as in inˈtroduce or ˈcanoe and
sometimes from the antepenultimate syllable, as in ocˈtopus.
Standard English object pronouns me, him, her, us, them, and the various
Palmerston extensions of these (two of us, us lot, big lot of them, etc.), are
used very frequently used as subjects, as shown in (3).
3.2.2 Nouns
Formal marking of pluralization on nouns is optional in Palmerston Island
English. When nouns are marked as plural, -s is generalized to any noun
(and has no /z/ allomorph as it does in Standard English). (This allomorph
would not be expected, in any case, due to the devoicing of final stops
described above. A /əs/ allomorph does exist, however.) Nouns that have
irregular plurals in Standard English sometimes receive the -s plural in
Palmerston English as well as the irregular plural marking, so that we
find singular/plural alternations such as childchildrens, womanwomens,
manmens.
The form of the indefinite article is always /ə/, never /ən/. The definite
article is /də//tə/, often reducing to /dt/ before vowels. Definite and
indefinite articles do not seem to have the same distribution as in Standard
English. A full account of Palmerston Island English article usage remains
a matter for future research. A few generalizations that can be made at
this stage are that definite articles are often used with years (I left here in
the 1992); Palmerston English has some mass nouns that are count nouns
in Standard English, and vice versa, which affects article selection; and
quantifiers in general interact with articles differently from how they do in
Standard English.
3.2.4 Verbs
3.2.4.1 Verb agreement
In Palmerston Island English we find that all persons and numbers can
occur with verbs unmarked for agreement, or with verbs marked with -s, as
Palmerston Island English 277
Table 12.3 Present-tense verb paradigm for Palmerston Island English
Table 12.3 shows. Ehrhart (1996) has analysed the system as one in which
the bare verb is used for third- person singular, and -s is used for all other
person and number categories.
This may have been the general system in the early 1990s when Ehrhart
conducted her fieldwork, but it is not possible today to find speakers
who consistently follow this pattern. The speakers who use -s on non-
third-person-singular verbs do so, at most, 30 per cent of the time. Many
speakers never use -s for these. The distribution is not accounted for by
e.g. the Northern agreement rule, although this may have contributed to
the origin of the Palmerston system. The frequency of 30 per cent is much
lower than the frequencies of other Palmerston Island English linguistic
features, so it seems unlikely that this frequency could be accounted for
simply by register variation or accommodation.
It seems likely that the apparent inconsistencies are the result of compet-
ing systems: the traditional one described by Ehrhart competing with the
Standard English system. Because the two systems are exactly the inverse
of each other, this explains why any person/number seems able to appear
with or without -s. Interestingly, there is a third pattern that some speak-
ers follow consistently: bare verbs for all persons/numbers. These speakers
seem to have resolved the conflict between the two systems by removing
all agreement markers entirely. Ehrhart did not find this bare verb system
among the speakers she worked with in the early 1990s, so it appears to be
a recent development.
3.2.4.2 Tense
A bare or -s marked verb in Palmerston Island English can have past, present
or future interpretation depending on the context. Some examples of these
forms with past and future reference are given in (9).
(9) a. Yesterday we eat fish.
‘Yesterday we ate fish.’
278 rachel hendery
b. We want to make sure everything is ready for tomorrow morning:
just hop on the boat, pile everyone inside and off we goes to the
motu.
‘We want to make sure everything is ready for tomorrow morning:
we’ll just hop on the boat, pile everyone inside and off we’ll go to
the islet.’
Alternatively, future time reference can be expressed with will/’ll, going
to, or gonna. Past time reference can be expressed with the Standard English
simple or perfect past constructions, including a perfect construction with-
out an auxiliary, which I will discuss under aspect below.
Preterite forms are often slightly different from those of Standard
English. Some irregular forms have been regularized, and consonant cluster
reduction means that final -d is often not present. In what might be another
strategy for overcoming the phonotactic constraint – by resyllabification –
the past tense is sometimes doubly marked by the addition of the -ed mor-
pheme to verbs that already contain it. This seems to occur in both the
preterite and in participle forms. Some examples of this double marking
are shown in (10).
(10) a. A man passeded him with his goat.
‘A man passed him with his goat.’
b. This island is blesseded by God.
‘This island is blessed by God.’
The avoidance of a consonant cluster cannot be the only reason for this
form, however, as we also find, for example, tooked, shooked, and stucked,
among other forms, where the addition of the /t/ suffix creates a cluster
that would not exist in Standard English. The only examples I have of
these latter forms, however, occur before words beginning with a vowel,
allowing resyllabification.
Another common way to express past tense in Palmerston Island English
is by use of the preverbal marker been. This is a construction found fre-
quently in creoles and other Pacific English varieties.
(11) a. She been record you?
‘Has she recorded you?’
b. We been go in a group.
‘We went in a group.’
This construction seems to be most frequent in cases where standard
English would use a perfect aspectual form such as She has recorded you or
Palmerston Island English 279
Table 12.4 Standard English and Palmerston Island English verbal and
adjectival predicates
I have eaten one, but as (b) shows, this does not hold exclusively, which is
why I have included this form as a tense rather than an aspect.
A complication of this form is that there is also a corresponding present-
tense construction consisting of an inflected present-tense form of be plus
the bare verb, illustrated in (12).
(12) a. Yeah I’m use this New Zealand slang, eh?
‘Yeah, I use New Zealand slang, eh?’
b. He’s steal it first and aks after
‘He steals it first and asks afterwards.’
c. Glad these things is clean their place now.
‘I’m glad they have cleaned their place now.’
d. Both of them are look.
‘Both of them (would) look.’
These examples all have different aspectual and modal readings so it
is difficult to know what the function of this construction is. As the
Palmerston Island English lexicon is very multifunctional, it is possible
that this construction is simply a natural outcome of collapsing some of
the distinctions between adjectives and verbs. In Table 12.4 I show how the
verbal and adjectival predicates of Standard English differ from each other,
and how these differences have been levelled out in Palmerston Island
English, first by the acceptability of a copular-less adjective predicate, and
secondly by the existence of the be + bare verb construction. The parallel
between this and verbal predicates without be is enhanced by the degree to
which the present-tense verb in Palmerston English resembles the infinitive,
since agreement is optional even with the third person singular.
3.2.4.3 Aspect
The only innovative element with a clearly aspectual function in Palmerston
English is the clause-final finish, which as in many other Pacific Englishes
280 rachel hendery
and English-based creoles has a completive meaning. In Palmerston Island
English this element seems to be new – Ehrhart did not find it when she
conducted her fieldwork in the early 1990s. It is optional, and not very
common. While I heard it more frequently than this, it only appears four
times in my recorded corpus. One of the four instances is not prototypical,
as it seems to be a paratactic juxtaposition of two clauses: I’ll leave you and
is/it’s finish.2 This example is interesting as an indication of how the finish
construction could have arisen, through juxtaposition and then reduction
of the clause is finish. Alternatively, or in combination with this, there may
have been contact influence from other Pacific English varieties.
(13) a. I painted the boat finish.
‘I painted the boat.’
b. I went feed my pig finish, came sit here.
‘I fed my pig and then came to sit here.’
c. Tell a story. I’ll leave you is finish.
‘Tell a story. I’ll leave you (before you begin).’
The perfect of Standard English can be used without an auxiliary in
Palmerston Island English, for example done, seen, been in (14). These only
seem to occur with perfect aspectual readings, so cannot be simply analysed
as levelling of the preterite and participial forms.
(14) a. If you done something stupid, I would say to you shei.
‘If you had done something stupid, I would say to you shei.’
b. I never seen any of my grandparents.
‘I have never seen any of my grandparents.’
c. I been to Australia three time.3
‘I have been to Australia three times.’
The continuous is formed, as in Standard English with a participle in
-ing, (which, as in many varieties of English, can also be realized as -in). A
small number of verbs have a present participle in -ening. The prototype
for this is fishening, which is used by all speakers the majority of the time.
The form is found occasionally for other verbs, in particular singening,
although this is not as universal as fishening. It seems unlikely that the -en
2 As mentioned earlier, is and it’s are indistinguishable in Palmerston Island English due to consonant
cluster reduction. Is finish would also be acceptable as a full clause, however, because of pro-drop.
The form finish is expected in the sentence It is finish(ed), as consonant cluster reduction means that
finished does not fit Palmerston Island English phonotactics.
3 This construction should not be confused with the been + infinitive construction described above.
That construction does not have the same aspectual constraint.
Palmerston Island English 281
in these forms should be equated with the -en of the past participle, as
these verbs do not take a past participle in en, either in Palmerston English
or Standard English. It is perhaps more likely that they arose through the
addition of -ing to forms such as fishin’ and singin’, making it similar to
the double-marking of past-tense verbs such as blesseded etc., mentioned
above.
As we saw for the perfect construction, an auxiliary is optional in the
continuous construction too:
(15) Dog is up on the bottom, jump out of window, and the boy looking
dog.
‘The dog is up on the bottom (of the windowsill), jumps out of the
window, and the boy is looking at the dog.’
3.2.4.4 Mood
There appears to be a rather infrequent subjunctive use of be in Palmerston
Island English, shown here in examples from Ned Marsters in 1959 and
from one of my informants in 2009.
(16) a. He says they all be the same, his family and Mahuta’s (Ned
Marsters 1959)
‘He says they are all the same, his family and Mahuta’s.’
b. I try to do it the way she’s saying it,so when I speak to the children
it be the way that she’s wanting.
‘I try to do it the way she says it, so when I speak to the children
it is the way she wants it.’
The other noteworthy feature of Palmerston English with regard to
mood is the development of the modal verb must. This has apparently
become an adverb on the model of Standard English maybe. This is illus-
trated in (17).
3.2.5 Prepositions
Selection of prepositions in Palmerston Island English often differs from the
corresponding prepositions in Standard English. Some verbs that require
prepositional complements in Standard English select directly for noun
phrases instead in Palmerston Island English. Two of these are illustrated
in (18).
(18) a. And the farmer went inside, went to look the water.
‘And the farmer went inside, went to look at the water.’
b. I’m not talking all of the island.
‘I’m not talking about all of the island.’
Some uses of prepositions in Palmerston Island English suggest that dif-
ferent ways of conceptualizing space apply than in other English varieties.
For example, one falls off rather than out of a tree and climbs on a tree
instead of up it, suggesting that climbing trees is envisaged as a person
pressed against a trunk, rather than sitting or standing among the branches
or in the canopy. This makes sense when one considers that most trees
on Palmerston Island are coconut palms. One falls from a window, rather
than out of a window. Some set idioms also have different prepositions in
Palmerston Island English. These include, for example, to put one’s trust
into someone, and to be good on something. One sets something with fire
rather than setting it on fire.
3.3.2 Fronting
Fronting of the predicate is very frequent in Palmerston English, especially
in non-verbal clauses. This produces a word order similar to that of Cook
Island Māori, so this construction may be the result of substrate influence
or subsequent contact with Māori L2 English speakers.
(21) a. Too small, the table
‘The table is too small.’
b. Is hot, the sun.
‘The sun is hot.’
Fronting, usually with pronominal resumption, can also be used for
topicalization, as in (22). A sentence can have multiple fronted topics, as
in (22b).
(22) a. The yacht, is he going to full up?
‘Is the yacht going to fill up?’
b. I know in the island – in the islets – mosquito, I don’t know
how he breeds.
‘I know . . . I don’t know how mosquitos breed in the islets.’ (Ned
Marsters, Burland Interview 1959)
On or for can be used as a topic marker before a fronted constituent, as
in the following examples.
(23) a. On the cup, I heard about that.
‘As for the cup, I heard about that.’
b. For me, I say ‘um’.
‘As for me, I say “um.”’
284 rachel hendery
3.3.3 Questions
Yes/no questions in Palmerston Island English can have a standard do-
support construction, as in (24a), or simply use declarative word order
with interrogative intonation as in (24b).
(24) a. Does they eat it?
‘Do they eat it?’/ ‘Will they eat it?’
b. You like the fish?
‘Do you like the fish?’
Wh-questions also frequently have the same word order as declara-
tives. Interrogatives are optionally fronted, and do-support generally absent.
Examples of wh-questions are given in (25).
(25) a. Where you was?
‘Where were you?’
b. What Shirley was doing there when you went?
‘What was Shirley doing when you went there?’
c. What for you want that?
‘What do you want that for?’
Embedded questions usually resemble those of Standard English in
their word order. This means that Palmerston Island embedded questions
and non-embedded questions can be identical. Occasionally verb inver-
sion occurs in embedded questions, such as the first embedded question
in (26).
(26) You describe where is the dog and where you go to the fish.
‘You describe where the dog is, and where you go to get to the fish.’
3.4 Lexicon
A striking feature of the Palmerston Island English lexicon is the flexibility
of words with regard to their word class. Many items function as nouns
and verbs, some examples of which are given in (27).
(27) a. Then you dough it.
‘Then you knead it.’
b. You broom now: next month you broom again
‘If you sweep now, next month you’ll have to sweep again.’
Some elements function both as adjectives and verbs, as illustrated in
(28).
Palmerston Island English 285
(28) a. We off the generator at two.
‘We turn the generator off at two.’
b. She bright her eyes.
‘She opened her eyes wide.’
As shown in the discussion of verbs and the description of coordination
patterns above, the syntactic distinction between adjectives and verbs is in
any case less sharp in Palmerston English than it is in many other varieties
of English. It is unclear, however, whether this is a contributing factor to
the multifunctionality of these words or the result of it.
Reduplication exists, but is infrequent and does not seem to be transpar-
ent or productive. Some examples of reduplicated forms are kaikai ‘feast’,
sing-sing ‘sing-along’, and chuck-chuck ‘to do something enthusiastically’.
The latter is illustrated in (29).
(29) You hear Bob chuck-chucking it in church today?
‘Did you hear Bob singing enthusiastically in church today?’
Most Palmerston English words with Māori origin are in the domains of
flora and fauna, but there are a handful of other Māori words in common
use, including umu ‘ground oven’, motu ‘islet’, ‘enua ‘island’, ‘uapo ‘sing-
along’, orometua ‘missionary’, and the honorific titles mama and papa used
for the elderly and for religious leaders.
There are several words with unknown origin: these include /‘ʃɒpaki/
‘skinny’ (prototypically used for fish, but by extension sometimes used
for other animals and humans), /ʃɛɪ/ (an admonition used to children
who are misbehaving, perhaps from English shame) and /ʃə’lɒk/ (a similar
admonition used for adults who are doing something stupid, perhaps from
English should look).
Some English-origin words are rare or archaic in other varieties of
English, but common on Palmerston. These include fowl, eyeglass, shanty,
poorman ‘grapefruit’, lead (for ‘electrical cord’), and pitch (used for any dark
unidentified mark or dirt).
Others have undergone semantic shift. A t-shirt is known on Palmerston
Island as a singlet, and a sleeveless vest is a short-sleeved singlet. Stoppers of
any sort, including screw-top lids, are bungs. A coloured ink is a coloured
pen. The word drink implies a hot beverage, such as tea or coffee; a cold
drink must be specified as such, and the phrase hot drink is redundant
and not used. A cup is not an umbrella term that encompasses glasses, but
rather used for a specific subset of drinking vessels. A cup must have a
handle. A mug, on the other hand, is a very large vessel for pouring from,
286 rachel hendery
not for drinking out of. Slippers are what other English-speaking countries
call flip-flops, jandals, or thongs.
One picks birds rather than catches them, and spills a drink (into a cup)
rather than pours it. Directions are usually given as down (towards the
north, where the main settlement is), or up (to the south, away from the
main settlement). An acceptable answer to the question Where are you
going? is simply Going up, or Going down. The phrase thank you is used to
mean something like ‘I understand what you are saying’, and the general
greeting to all visitors to the island or to a house, even to non-Palmerston
Islanders visiting for the first time, is welcome home.
4 Conclusion
Palmerston Island presents an extremely valuable case study for linguists
interested in mixed languages, variation, new dialect formation, and the
linguistic effects of isolation. It provides an interesting parallel to the
development of New Zealand English, as it was settled not long after
the colonial settlement of New Zealand, and both cases involved very
similar inputs: UK English and Māori. Comparison of Palmerston Island
English with other Pacific English varieties, in particular Pitcairn-Norf’k
(as Tahitian is very similar to Cook Island Māori), may also be fruitful.
The similarities found across Pacific English varieties can lead us to a better
understanding of the mechanisms involved in mixed language formation,
creolegenesis, and contact-induced change, in particular the role played by
the typological characteristics of the input languages. Palmerston Island
English is a particularly useful addition to such comparative study because
its history is relatively short and well documented. We know exactly what
the inputs were, and the degree and type of contact that has occurred since
settlement, both of which are more limited than is the case for many other
Pacific English varieties.
The future of Palmerston Island English is uncertain. The island is, at its
highest point, three metres above sea level, rendering it extremely vulnerable
to climate change. Severe cyclones have in the past deforested the island
of coconut palms and damaged the reefs, leaving the population without
their most important food sources. With climate change, such storms are
likely to become more frequent, and rising sea levels will ultimately make
the island uninhabitable. Most Palmerston Islanders have close family in
Auckland, New Zealand, and it is more likely that they would resettle there
than in Rarotonga, where they also often have family connections but do
not speak the language. The Palmerston Island community in Auckland is
Palmerston Island English 287
said to maintain Palmerston English to some extent, but it seems doubtful
that this could continue indefinitely. Further research on Palmerston Island
English is therefore a matter of urgency.
References
Burland, John. 1959. Unpublished transcript of an interview between John Bur-
land, Ned Marsters and Peka Marsters. Wellington, New Zealand. Alexander
Turnbull Library. MSX-8809.
Ehrhart, Sabine. 1996. Palmerston English. In Stephen A. Wurm, Peter
Mühlhäusler and Darrell T. Tryon, eds., Atlas of Languages of Intercultural
Communication in the Pacific, Asia, and the Americas. Berlin and New York:
Mouton de Gruyter, 523–36.
Gill, W. W. 1877. Unpublished Letter to the London Missionary Society regarding
visit to outstations July 3–Aug 14. LMS archives, School of Oriental and
African Studies, London. CWM South Seas Journals Box 11, 1871–81.
Hendery, Rachel. 2013. Early documents from Palmerston Island and their impli-
cations for the origins of Palmerston English. Journal of Pacific History 48:
309–22.
Hendery, Rachel and Sabine Ehrhart. 2011. Palmerston Island English. In Bernd
Kortmann and Kerstin Lunkenheimer, eds., The Electronic World Atlas of
Varieties of English [eWAVE]. Leipzig: Max Planck Institute for Evolutionary
Anthropology.
Hendery, Rachel and Sabine Ehrhart. 2012. Palmerston Island English. In Bernd
Kortmann and Kerstin Lunkenheimer, eds., The World Atlas of Varieties of
English. Berlin and New York: De Gruyter Mouton, 628–42.
ch a p ter 1 3
1 Introduction
This chapter considers forms of English spoken by Polynesian peoples in
New Zealand. We start with a geographic description of the Pacific, situ-
ating Polynesia and its peoples and their economic and political positions
in the region and outlining the effect these positions have had on the
movement of people and on their language use. An increasing number of
Polynesian communities are located outside their home islands, in largely
anglophone nation states on the Pacific Rim, resulting in bilingualism and
language shift to English. To describe the forms of English that emerge,
we need an understanding of the similarities which bind together, and the
differences which distinguish, Polynesian people of different ethnicities,
ages and language backgrounds. As part of this process, we discuss the con-
cept of pan-ethnic identities and language varieties, which conceal cultural
differences whilst embracing similarities. Once the sociolinguistic context
has been described, the second half of this chapter provides a comprehen-
sive overview of the (limited) existing literature on the features of Pasifika
Englishes in New Zealand.
The ‘Pacific Islands’ is a cover term used to refer to the more than
20,000 islands located in a geographic area spanning many thousands of
kilometres across the South Pacific Ocean. Much of this area is known as
Oceania, and is comprised of three main groupings of islands: Micronesia,
Melanesia and Polynesia. Micronesia is the area closest to the equator,
and includes the islands of Guam, Palau, Nauru and Kiribati, amongst
others. Melanesia refers to the group of westernmost islands. The area
reaches to the islands of New Caledonia and Vanuatu, includes Papua
New Guinea and parts of Indonesia, with Fiji at its most easterly point.
This positioning means that Fiji is geographically closer to Polynesia than
much of Melanesia, and has strong links with its Polynesian neighbours.
While we do not cover Fijian English in this chapter, it is one variety of
288
Pasifika Englishes in New Zealand 289
English in the Pacific islands that has been well described (Tent and Mugler
2008).
The Polynesian islands cover the greatest geographic expanse, roughly
triangular in shape, with Aotearoa/New Zealand, Rapa Nui/Easter Island
and Hawai‘i as its three corners. The geographic positioning of Hawai‘i,
and its status as part of the USA, has resulted in links which are more east-
ward than westward in orientation, leading to differences between Hawai‘i
and its geographically distant Polynesian relatives. American Samoa is in
a similar situation. Most of the Polynesian islands have relatively small
populations which historically have been ethnically homogeneous. While
some countries, such as Samoa, Tonga and Tuvalu, are independent nation
states, others such as Niue and the Cook Islands are in free association with
New Zealand, with its citizens having rights similar to those of New Zealand
citizens, including free entry into the country, and access to local education
for those who wish to study in New Zealand. Historically, the Polynesian
island states were self-sufficient, but in today’s global economy they are
often classified as relatively poor. Some are also threatened by sea-level rises
through climate change.
As a consequence, from the second half of the twentieth century, Poly-
nesian peoples began immigrating in large numbers to countries within the
Pacific Rim in search of better economic opportunities, especially in New
Zealand, Australia and the United States. The flow of populations between
Australia, New Zealand, the USA and the island states is often multidirec-
tional as close-knit Polynesian families often move back and forth, visiting
their relatives in their home islands, forming a chain of migration across
the Pacific. The amount of travel from the Polynesian islands is, however,
unbalanced. While birth rates on the islands are often higher than in devel-
oped countries, population numbers on the islands have been in continual
decline over the past thirty to forty years. Indigenous languages spoken on
the islands are also increasingly in an uncertain relationship with English
as the latter reinforces its status as the dominant language of education and
tourism. English has increasingly become a prestige commodity amongst
many of the younger population (Besnier 2003).
For many of the countries there are now more residents in New Zealand,
Australia and the USA than on the islands themselves. Yet, throughout the
diaspora, Polynesian peoples form a minority of the population in host
countries, and many reside in communities of low socio-economic status.
At the 2006 New Zealand Census, Polynesians constituted 7 per cent of the
total population, numbering just under 266,000 people. The Polynesian
peoples form a much smaller proportion of the total population in Australia
290 donna starks, andy gibson and allan bell
and the USA. Polynesian populations in the diaspora are less ethnically
homogenous than in the islands, with peoples from many islands residing
in Polynesian enclaves. The largest area in New Zealand is in Manukau on
the south side of Auckland. In all three major diasporic contexts, the largest
population is Samoan. From the dozen cultural groups of Polynesian origin
in New Zealand, nearly half the population is Samoan. The Cook Islands
(22%), Tongan (19%) and Niuean communities (8%) follow (Statistics New
Zealand nd), with smaller numbers from especially Tokelau and Tuvalu.
The term ‘Pasifika English’ could potentially refer to all the varieties of
English spoken throughout the Pacific. However, it typically refers only to
Polynesia and its diaspora communities that have come into contact with
British, Australian, American or New Zealand English. In this chapter, we
focus our description on the varieties of English spoken by Pasifika commu-
nities in New Zealand, which can be expected to share features with other
varieties in the islands and Australia. Pasifika Englishes in Hawai‘i, Amer-
ican Samoa and the USA diaspora are likely to have differences which
are due to the rather different input varieties. We know of no sociolin-
guistic research on Pasifika Englishes in the USA, though there are good
descriptions of Hawai‘ian English and Hawai‘i Creole (Sakoda and Siegel
2008).
2.2 Pan-ethnicity
The similarity of the substrate languages, along with historical and eco-
nomic similarities, laid the foundation for a variety of English which is
292 donna starks, andy gibson and allan bell
similar enough across the various ethnic groups to suggest the possibility
of a pan-ethnic description. A pan-ethnic classification offers advantages,
enhancing a sense of community across groups and providing a means
of inclusion for people with multiple ethnicities. We discuss some of the
social and linguistic reasons for the use of this term, and the description of
this variety below.
Our first reason relates to the heterogeneity of those grouped under
‘Pasifika’. The label covers the diverse range of peoples from a number
of different Polynesian communities, each with their different social, cul-
tural and linguistic backgrounds. Yet, as mentioned above, within each
Polynesian community, there are varying levels of proficiency, different
views of home and culture, and different ties with their homeland and
New Zealand. Each community is varied and complex, with multiple
languages and cultures, all of which interact in ways that both separate
individual Polynesian communities and link them in a pan-ethnic whole.
There are also an increasing number of people of mixed ethnicity especially
amongst Polynesian youth, with over one-third of Polynesian peoples in
New Zealand identifying with more than one ethnic group (Tahu Consult-
ing 2008). This diverse community share Polynesian origins. The shared
Polynesian cultures allow for individuals to participate in mutual festivals
and events and have a sense of community. The similarities in their lan-
guages add to this. While many in the community claim that Niuean and
Tongan, and also Cook Islands Māori and New Zealand Māori, are to
some extent mutually intelligible, for other languages in their Polynesian
group, intelligibility is restricted to core vocabulary items. The shared set of
core pan-ethnic vocabulary includes words of cultural significance (e.g. kai
‘food’), which helps support a pan-ethnic variety of English. The shared
language repertoire extends beyond lexical similarities. Greetings in the
respective languages are the most obvious example, as Polynesian greet-
ings are known by practically all members of the New Zealand Pasifika
community even though the greetings are, at times, very distinct.
A final argument for a pan-ethnic variety is an important historical one.
As noted earlier, many of the original Polynesian settlers arrived during
the same time period, the 1950s, worked in the same factories and became
members of the same neighbourhoods and communities, alongside New
Zealand Māori people. Given their early interaction with New Zealand
Māori, it is likely that Pasifika Englishes would share features of New
Zealand Māori English – which is itself a variety of Polynesian English.
By the 1950s, most New Zealand Māori spoke a vernacular variety of New
Zealand English having already made the shift from Māori to English as
Pasifika Englishes in New Zealand 293
dominant language (see Benton 1991). This variety probably served – and
may continue to serve – as a model for Pasifika Englishes in New Zealand.
Descriptions of Māori English are more numerous and more thorough
than those of Pasifika English in New Zealand and are potentially the
basis for forming hypotheses about likely features of Pasifika varieties (see
Warren 1998, 2006 and Warren and Bauer 2008 for a description of Māori
English).1
We recognise, however, that a pan-ethnic categorization obscures the
distinctiveness of each of the ethnicities and some important similarities
(see MacPherson et al. 2001 for a further critique). Consider, for instance,
the position of Cook Islands Māori, which is the closest language to New
Zealand Māori. There are strong cultural and linguistic links between NZ
Māori and the Cook Islands communities, and although Cook Islanders
are often aligned with other groups from the Pacific region under the
label ‘Pasifika’, they also refer to themselves with the label ‘Māori’. In
many ways, the NZ Cook Island peoples fall inbetween the NZ Māori
and Pasifika classifications. In this ethnic mix, language varieties may share
pan-Pasifika features but also differ from one another, reflecting the diverse
ethnic groups that comprise the NZ Pasifika community.
A second complication is the place of the Fijian community. Indigenous
Fijians are Melanesians. They differ linguistically from their Polynesian
counterparts and have a different settlement history in New Zealand, yet
they are often placed, and also position themselves, within the Pasifika
community. They also appear to share some of the Pasifika features of their
Polynesian neighbours (see Biewer 2012; Tent and Mugler 2008), perhaps
attributable to their shared British and New Zealand founder dialects.
3.1 Phonetics/phonology
3.1.1 Consonants
The data for consonants summarized in this section come from two studies
based on the PLMP interviews (Bell and Gibson 2008; Starks, Christie and
Thompson 2007), from the animated sitcom bro’Town (Gibson and Bell
2010), from the reading passage data collected in a South Auckland primary
school (Starks and Reffell 2005, 2006), and from two additional datasets:
from recordings of children in primary school playgrounds including a
school with a majority of Pasifika students (Kennedy 2006), and from
interviews with hip-hop artists (Gibson 2010).
Table 13.1 shows some of the Pasifika English consonants that have been
documented as differing from general NZE. Some of the features shown
are used more by older or ESL speakers, while others are representative of
young speakers.
The consonantal inventory of Pasifika languages is smaller than that
of English. This may lead to features of Pasifika English differing from
general NZE. As in the substrate languages, initial stop consonants are
often less aspirated than in Pakeha English, particularly for older/ESL
speakers (Gibson and Bell 2010; Starks et al. 2007). Voiced consonants
(which are lacking in Polynesian languages) are often devoiced in final
position across ages and ethnic groups.
Pasifika Englishes in New Zealand 295
Table 13.1 Consonants of Pasifika English2
3.1.2 Vowels
The information on vowels presented in this section is drawn mainly from
three studies, two of which consider only Niuean speakers. Thompson et
al. (2009) and Starks (2008a) are both based on the reading data from
11- to 13-year-old children of NZ Māori and Pasifika ethnicities from a
school in South Auckland where 57 per cent of the students are Pasifika.
Thompson et al. (2009) present an acoustic analysis of the vowel space of
four of the Niuean males, while Starks (2008a) provides an impressionistic
analysis of the vowel realizations of all forty children in the study. Starks,
Christie and Thompson (2007) looked at vowels in the Niuean PLMP
interview data. These data are supplemented by impressionistic comments
Pasifika Englishes in New Zealand 297
on particular vowel sounds based on the PLMP interview with a young
Samoan male and on the Pasifika characters in bro’Town (Bell and Gibson
2008; Gibson and Bell 2010).
The short front vowels in Pasifika Englishes in New Zealand appear to
differ from those in general NZE, where kit is centralized, and dress
and trap are raised. The studies consistently report the occurrence of
a relatively close and front kit for Pasifika speakers, a realization that
is also regarded as characteristic of NZ Māori English (Bell 1997). The
dress vowel was described as raised, as it is in general NZE, in the
acoustic analysis of the reading passages and in the PLMP data. But Starks’
(2008a) analysis showed variation between ethnic groups, with fewer raised
variants in the Samoan and Tongan groups. Cook Islands children behaved
similarly to their NZ Māori counterparts, with raised dress vowels. trap
was also described as raised in the acoustic analysis of reading data. In
the PLMP data, however, trap is often not raised, and is occasionally
centralized. Starks (2008a) also found unraised variants in the reading
data, predominantly for Samoan, Tongan and Niuean children. Cook
Islands children patterned with NZ Māori, having a raised trap vowel
[ɛ]. Starks (2008a) also found a centralized trap, similar to realizations
of the /a/ vowel in Pasifika languages. Overall, then, there appears to be
a pan-ethnic Pasifika close front kit vowel, which is shared with some
NZ Māori English, together with indications that Cook Islands Māori
English dress and trap pattern with NZE, but the other ethnic varieties
do not.
Aside from the short front vowels, the studies consider a range of other
vowels. fleece and thought are both noted for being monophthon-
gal, unlike in general NZE. thought was also described as being higher
than expected, while foot was slightly lowered, and both rounded and
unrounded forms occurred for lot. In NZE, strut and start are pho-
netically similar in their quality, almost behaving as a short–long pair
(Warren 2006). In both the PLMP and Niuean datasets, strut was noted
to be qualitatively different from start, with the Niuean children having
a raised strut, while the PLMP data suggested that strut was fronted
in comparison to start. Fronted forms of the goose vowel (also found
in NZ Māori English) and fronted offglides for goat were noted in the
speech of the young Samoan male and in bro’Town.
Thompson et al. (2009) found a tendency for diphthongs to have less
movement than would be expected in general NZE. There was agreement
across studies that diphthongs were relatively short in length, with Starks
298 donna starks, andy gibson and allan bell
Table 13.2 Pasifika vowels as compared to General NZE (adapted from
Bauer and Warren 2008)
4 Conclusion
Pasifika Englishes are robustly alive, carrying increasingly positive index-
ical value, particularly for those young Pacific people who are less flu-
ent in their Polynesian languages. For these, Pasifika Englishes provide
a means of projecting cultural identity. It is likely that pan-ethnic Pasi-
fika identification will strengthen, as greater proportions of the com-
munity claim multiple ethnicities, a process which will also include
more and more identities whose pan-ethnicity extends to NZ Māori and
Pākehā.
The future of Pasifika Englishes in New Zealand is unclear. There are
a substantial number of speakers who are part of established and growing
immigrant communities. Although the variety is challenged by prescrip-
tive language attitudes from within and outside, the social context, at
least within the Pacific homelands and within New Zealand, is likely to
give vitality to it. As the English proficiency of all Pasifika communities
increases, the ESL-like varieties are likely to reduce in strength, and ver-
nacular features of the varieties increase. At present, English and Pasifika
languages are in something of a diglossic relationship in much of the com-
munity, but given the gradual shift to English in diasporic communities
it is possible that standard NZE and Pasifika Englishes may develop their
own diglossic existence.
At present, we have only the most rudimentary understanding of the
dynamics of Pasifika Englishes. An ethnographically grounded analysis of
Pasifika ways of speaking English in New Zealand would shed consider-
able light on the variety as a whole, and on its internal diversity. Beyond
that, there is huge scope for researching the relationships between Pasifika
Englishes throughout the Pacific islands and in other diasporic settings.
References
Bauer, Laurie and Paul Warren. 2008. New Zealand English: phonology. In Bur-
ridge and Kortmann, eds., 39–63.
Bell, Allan. 1997. The phonetics of fish and chips in New Zealand: marking
national and ethnic identities. English World-Wide 18(2): 243–70.
2000. Maori and Pakeha English: a case study. In Allan Bell and Koenraad
Kuiper, eds., New Zealand English. Wellington: Victoria University Press,
221–48.
Bell, Allan and Andy Gibson. 2008. Stopping and fronting in New Zealand Pasifika
English. University of Pennsylvania Working Papers in Linguistics: A Selection
of Papers from NWAV 36, 14: 42–53.
Pasifika Englishes in New Zealand 303
Benton, Richard A. 1991. Maori English: a New Zealand myth? In Jenny Cheshire,
ed., English around the World: Sociolinguistic Perspectives. Cambridge Univer-
sity Press, 187–99.
Besnier, Niko. 2003. Crossing genders, mixing languages: The linguistic construc-
tion of transgenderism in Tonga. In Janet Holmes and Miriam Meyerhoff,
eds., Handbook of Language and Gender. Oxford: Blackwell, 279−301.
Biewer, Carolin. 2008. South Pacific Englishes – unity and diversity in the usage of
the present perfect. In Terttu Negalainen, Irma Taavitsainen, Päivi Pahta and
Minna Korhonen, eds., Dynamics of Linguistic Variation: Corpus Evidence on
English Past and Present. Amsterdam: Benjamins, 203–19.
2009. Modals and semi-modals of obligation and necessity in South Pacific
Englishes. Anglistik 20(2): 41–55.
2012. Acrolectal English in the South Pacific: emerging standards? In Ray-
mond Hickey, ed., Standards of English: Codified Varieties around the World.
Cambridge University Press, 336–53.
Britain, David and Andrea Sudbury. 2002. ‘There’s sheep and there’s penguins’:
convergence, ‘drift’ and ‘slant’ in New Zealand and Falkland Island English.
In Mairi C. Jones and E. Esch, eds., Language Change: The Interplay of
Internal, External and Extra-linguistic Factors. Berlin: Mouton de Gruyter,
211–40.
Burridge, Kate and Bernd Kortmann, eds. 2008. Varieties of English: The Pacific
and Australasia. Berlin: Mouton De Gruyter.
Gibson, Andy. 2010. Non-prevocalic /r/ in hip hop: a comparison of New Zealand
and the USA. Paper presented at African American Language 2, San Antonio,
Texas, USA.
Gibson, Andy and Allan Bell. 2010. Performing Pasifika English in New Zealand:
the case of bro’Town. English World-Wide 31: 231–51.
Gordon, Elizabeth, Lyle Campbell, Jennifer Hay, Margaret Maclagan, Andrea
Sudbury and Peter Trudgill. 2004. New Zealand English: Its Origins and
Evolution. Cambridge University Press.
Jennifer Hay, Margaret Maclagan and Elizabeth Gordon. 2008. New Zealand
English. Edinburgh University Press.
Kennedy, Marianna. 2006. Variation in the pronunciation of English by New
Zealand school children. Unpublished MA thesis. Victoria University of
Wellington.
MacPherson, Cluny, Paul Spoonley and Melani Anae, eds. 2001. Tangata O Te
Moana Nui: Evolving Identities of Pacific Peoples in Aotearoa/New Zealand.
Palmerston North: Dunmore Press.
Pollitzer, Daniela. 2008. Grammatical features in Pasifika English. Unpublished
Master’s thesis, University of Regensburg.
Sakoda, Kent and Jeff Siegel. 2008. Hawai‘i Creole: phonology. In Burridge and
Kortmann, eds., 210–33.
Starks, Donna. 2008a. National and ethnic identity markers: New Zealand short
front vowels in New Zealand Māori English and Pasifika Englishes. English
World-Wide 29(2): 176–93.
304 donna starks, andy gibson and allan bell
2008b. Potential variables for the study of grammatical variation in Niuean
English. New Zealand English Journal 22: 36–44.
2010. Being Niuean or being Niue? An investigation into identity labels. Lan-
guage, Identity and Education 9(2): 124–38.
Starks, Donna and Hayley Reffell. 2005. Pronouncing your Rs in New Zealand
English? A study of Pasifika and Māori students. New Zealand English Journal
19: 36–48.
2006. Reading ‘TH’: vernacular variants in Pasifika Englishes in South
Auckland. Journal of Sociolinguistics 10(3): 382–92.
Starks, Donna, Jane Christie and Laura Thompson. 2007. Niuean English: initial
insights into an emerging variety. English World-Wide 28(2): 133–46.
Starks, Donna, Laura Thompson and Jane Christie. 2008. Whose discourse parti-
cles? New Zealand eh in the Niuean migrant community. Journal of Pragmatics
40(7): 1279–95.
Starks, Donna and Laura Thompson. 2009. There’s not many Niueans: singular/
plural agreement in existential constructions in the New Zealand Niuean
community. World Englishes 28(3): 319–35.
Statistics New Zealand. No date. www.stats.govt.nz/Census/2006CensusHome
Page.aspx.
Tahu Kukutai (Tahu Consulting), Statistics New Zealand. 2008. Ethnic Self-
prioritisation of Dual and Multi-ethnic Youth in New Zealand. Wellington:
Statistics New Zealand.
Taumoefolau, Melenaite, Donna Starks, Karen Davis and Allan Bell. 2002. Lin-
guists and language maintenance: Pasifika languages in Manukau, New
Zealand. Oceanic Linguistics 41(1): 15–27.
Tent, Jan and France Mugler. 2008. Fiji English: phonology. In Burridge and
Kortmann, eds., 234–66.
Thompson, Laura, Catherine Watson and Donna Starks. 2009. Finding one’s own
vowel space: an acoustic analysis of the English of Niuean New Zealanders.
In Thomas Hoffman and Lucia Siebers, eds., World Englishes – Problems,
Properties and Prospects. Amsterdam: John Benjamins, 83–92.
Warren, Paul. 1998. Timing patterns in New Zealand English rhythm. Te Reo 41:
80–93.
2006. Oops, I’ve done a futt: quality and quantity in a New Zealand vowel
contrast. Te Reo 49: 125–43.
Warren, Paul and Laurie Bauer. 2008. Maori English: phonology. In Burridge and
Kortmann, eds., 77–88.
ch a p ter 1 4
Palauan English
David Britain and Kazuko Matsumoto
1 Introduction
The Republic of Palau/Beluu e˛r a Belau is an independent nation state of
the Western Pacific, consisting of an archipelago of around 350 small islands
stretched across 400 miles of ocean. Its nearest neighbours are Indonesia
and Papua New Guinea to the south, the Federated States of Micronesia to
the east, Guam and the Commonwealth of the Northern Mariana Islands
to the northeast, and the Philippines to the west. The islands have a
population of around 20,000 of which over 60 per cent live in the largest
city and former capital of Koror (Office of Planning and Statistics 2006:
23). The capital, in 2006, was moved to Ngerulmud, in Melekeok State on
the main island of Babeldaob. For most of the twentieth century, Palau was
under colonial administration: by Spain (1885–99), Germany (1899–1914),
Japan (1914–45) and finally, the United States of America (1945–94). It
formally gained its independence in 1994.
Our chapter examines the emergence of an anglophone speech commu-
nity in Palau, and aims to do three things: firstly to set the emergence of
Palauan English into the context of the country’s complex colonial past.
Palau’s four colonial rulers have exercised control in different ways, with
different degrees of settler migration, different attitudes towards the func-
tion of Palau as a ‘colony’, and widely differing local policies, leading to
very different linguistic outcomes in each case. We focus, however, on the
American era and the path to Palauan independence. Secondly, in exam-
ining the development of English in Palau, we apply Schneider’s (2007)
‘Dynamic Model’ of postcolonial English formation to this anglophone
community. This model attempts to provide a holistic social, historical,
Our work on Palauan English has been supported by the Ministry of Education, Culture, Sports,
Science and Technology in Japan: Ref. No. 22682003 (2010–2013). We would like to thank all of those
in Palau who kindly cooperated with our research, as well as our research assistants Akiko Okumura,
Tobias Leonhardt, Dominique Bürki and Dorothee Weber for their invaluable contribution to the
progress of this research.
305
306 david britain and kazuko matsumoto
political and attitudinal as well as linguistic account of the process by
which a new English emerges in a colonial environment. As we will see,
the case of Palau is important, because few communities in which English
has emerged as a result of American as opposed to British colonialism have
been examined in the model to date. The final aim is to present, based on
analyses of recordings of informal conversations among Palauans, an initial
portrait of the main phonological, grammatical and lexical characteristics
of Palauan English.
1 In 1964, six Palauan students were integrated among the seventeen Americans, but in the early 1970s,
such an integrated class disappeared, since ‘there were too few American children any longer to
warrant such special classes’ (Shuster 1982: 204, 206).
2 For example, Palau received $357,200 in 1978, $841, 300 in 1979 and over one million dollars for 1980
(Shuster 1982: 213), for a population of just over 12,000 in 1980 (Office of Planning and Statistics
2006: 23).
Palauan English 317
that the US period brought to Palau were: (a) the expatriation of all earlier
Japanese settlers and (b) the importation of Filipino labour, employed by
Palauans thanks to the so-called ‘compact money’ from the US.
Overall, while the Japanese era saw a rather interventionist and integra-
tionist approach at engendering consent, the US period was characterised
by a distant hands-off stance. Thus, it is most likely that the intensive
Japanese administrative strategy in Palau was far more influential upon
Palauan society and language, with respect to infrastructure, demography,
economy, education, belief systems and lifestyle, than either the earlier
European or the later American domination.
However, given America’s longer period of control (i.e. for half a century)
and ongoing financial aid, there are, of course, a number of similarities in
the sociolinguistic consequences of these two most recent colonial periods
in the history of Palau. During each administration, recognition of Japanese
or English as the high-status language as opposed to Palauan as the low
language (i.e. diglossia) was established, while the usual linguistic compart-
mentalisation was reinforced so that the colonial languages were used in
the school, legal, administrative and written domains, while Palauan was
mainly spoken in the home and traditional domains. Moreover, Japanese–
Palauan or English–Palauan bilingualism became the norm, while the use
of Japanese or English borrowing and code-switching in Palauan conversa-
tion has come to function as a typical ‘in-group’ language behaviour among
the different generations. However, the crucial difference is that the use
of Japanese was not restricted to those official domains; on the contrary,
face-to-face interaction in Japanese was commonplace in everyday life – in
the neighbourhood, at work and in the marketplace (Matsumoto 2010a,
2010b and Matsumoto and Britain 2003a).
We now turn to apply Schneider’s (2007) Dynamic Model to the context
of American control of Palau and the emergence of Palauan English. During
the first foundation phase (roughly from 1945 to 1962), then, as we have
seen, the sociopolitical background was that English was brought to Palau
during the Pacific War when the US Navy conquered Micronesian islands
one by one, defeating Japanese army and navy bases there (Hazel and Berg
1980). At the end of World War II, Micronesia, including Palau, became
firstly part of the ‘US Navy occupational territories’ before entering the
UN’s TTPI under the control of the US (Hazel and Berg 1980: 497). The
largest demographic change during this early phase was the repatriation
of Palau’s Japanese residents. From October 1945 to May 1946, 104,213
Japanese and 31,619 Okinawans were deported from Micronesia as a whole,
and 34,773 Japanese were repatriated from Palau (Palau Community Action
318 david britain and kazuko matsumoto
Agency 1978: 426–7, 452). The repatriation of such a large number of
Japanese and Okinawans meant that Palau’s principal sources of labour,
consumers and capital suddenly disappeared, leaving the total population
of Palau as small as 6,184 in 1947 (Abe 1986: 230). On the other hand,
the US administration brought only a few administrators and military
personnel who were temporarily stationed on each island (Aoyagi 1977:
49); Micronesia was never to see the arrival of large numbers of American
workers and settlers.
In terms of identity constructions, the STL strand, if one could use that
label for the temporary and short-term US military and TTPI personnel
stationed in Palau, was clearly conscious that under no circumstances would
their homeland ever be Palau. Officially, of course, settlement was not one
of the assigned functions of their control over the islands – their task
was to prepare the islands for ultimate independence, knowing that full
independence was, actually, detrimental to US foreign and military policy.
So, they focused on their assigned work, such as providing US aid packages
including food and clothes and dealing with war compensation, without
conspicuous efforts to integrate into the local community. In the early
period of American rule, for example, in 1952, the American Dependent
School for the children of Americans employed in Palau was established
by American parents who supported materials and salaries for the teachers
(Shuster 1982: 200). American children had a greatly different school life
from the Palauans; government taxis transported them between home
and school, while an American teacher taught them in their own special
building, with access to an abundance of teaching materials (Shuster 1982:
202).
On the other hand, despite the fact that Micronesian educational issues
were largely ignored, as mentioned earlier, the IDG strand knew that
those who collaborated with the colonisers would benefit the most, since
they had already lived through three colonial administrations. There was
competition among Palauan villages to provide the best school and the best-
trained teachers (Shuster 1982: 182). While no English-speaking teachers
and inadequate funds were provided to Palau by the American adminis-
tration, nine new school buildings were constructed by Palauan craftsmen
with the co-operation of all the villagers, using local materials in 1946
(Honolulu Advertiser 1946 in Shuster 1982: 182). As the obvious readiness
of the Palauans to engage in American education indicates, the IDG seems
to have been willing to accept the STL language and, once again, a new
social order. However, access to the STL strand and their language was very
limited for ordinary Palauans.
Palauan English 319
Sociolinguistic conditions during the first phase did not encourage fre-
quent interaction between STL and IDG beyond the work domain, since
language contact between the STL and IDG strands occurred only among
a small group of local elite who received teacher training in Guam or
Truk as well as the few who were employed by the TTPI government;
ordinary Palauans scarcely had an opportunity to access English. This is
because, as mentioned earlier, the US planned to produce a ‘local elite
whose interests would thereby become linked with those of the US’ (US
Solomon Report 1963 in Anglim 1988: 10). The American government
targeted the traditional elite, such as the two high chiefs, the titleholders
and the matrilineal kin of the meteet (see above) whose prestige, power
and authority had been challenged and reduced during the Japanese colo-
nial administration (Vidich 1952: 272), and who were understandably keen
to regain their lost status. Consequently, this elite resumed their posi-
tion of control and governed Palau according to traditional norms and
‘native customary law’ (Vidich 1952: 298). However, a superficial divi-
sion of executive, legislative and judicial functions allowed the traditional
elite to determine what constituted ‘native custom’, one consequence of
which was a regular abuse of power (Vidich 1952: 297–8). Thus, backed
by slogans such as ‘Palau for the Palauans’ and ‘Democracy for Palau’,
the US government turned the traditional elite into the new compliant
local political leaders, to run the territory in the interests of the traditional
elite at the local level, and in the interests of the Americans supralocally
(Vidich 1952: 344). Consequently, the situation in Palau provides a clear
example of Schneider’s claim that, in the initial foundation phase, ‘in
some cases, the settlers . . . privilege members of the local elite by teaching
them their language . . . in any case, in the IDG strand, marginal bilingual-
ism develops, predominantly among a minority of the local population’
(2007: 34–5).
In terms of linguistic effects, and given the extremely limited STL strand,
no strong linguistic impact upon the IDG strand could be observed. English
as a lingua franca was used only among local elites when meeting represen-
tatives of other districts for TTPI meetings, the Micronesian languages of
the TTPI being mutually unintelligible. In the STL, government and mil-
itary personnel stayed temporarily, making few efforts to acquire Palauan.
But perhaps because of the lack of an STL, and perhaps partially because of
America’s officially ‘non-colonial’ trusteeship role in Micronesia, Palauan
toponymic borrowings were readily observable during this early period –
Palauan place and landmark names were used routinely in Palauan English,
and indeed few places ever acquired English names. Much later on, some
320 david britain and kazuko matsumoto
newly made parks in Koror received English names (e.g. Long Island Park
and Ice Box Park), the Koror dock built by the Japanese was renamed from
Japanese hatoba to English T-Dock, and some of the remoter islands and
landmarks of major attraction to foreign tourists and divers have acquired
English names (e.g. Seventeen Islands and Oolong Channel).
The second phase ‘exonormative stabilisation’ can be said to have begun
in around 1962 when the Cold War confrontation between the US and the
Soviet Union was intensified, reminding the US of the strategic importance
of Micronesia. Consequently, US President Kennedy began to dramatically
accelerate development in the Trust Territory, investing enormously, with
particular emphasis on education (Shuster 1982: 196–8). This means that
for the first time, a sizable number of Americans teachers and Peace Corp
Volunteers arrived in Palau together with American educational system
and teaching material. Thus, some form of social stabilisation under US
control was beginning to materialise.
At this time, international pressure for rapid decolonisation was increas-
ing (Roff 1991). Other UN trustees – Great Britain, Australia and New
Zealand – were preparing in the 1960s to terminate their trusteeships in
the Pacific, and a number of independent Pacific nations emerged as a
result – Nauru, Samoa and Fiji, for example. In 1965, therefore, the US
established the Congress of Micronesia, an assembly with representatives
from each of the six districts of Micronesia (i.e. the Marshall Islands,
Pohnpei, Chuuk, Yap, Palau and the Northern Mariana Islands), which
legislate only on local matters (Anglim 1988: 11). In 1967, the Congress of
Micronesia set up ‘a status commission’ to discuss the Territory’s right to
self-determination (Anglim 1988: 11). However, in 1976, the US reached a
separate agreement with the Marianas that approved their changed polit-
ical status to become the ‘US Commonwealth of the Northern Mariana
Islands’ (Anglim 1988: 11), a similar status to that held by Puerto Rico. This
led to the end of Micronesia as one political unit. Palau and the Marshall
Islands voted for their own independent political status, as the Republic of
Palau in 1976 and as the Republic of the Marshall Islands in 1977, respec-
tively (Kabayashi 1994: 213). The remaining districts in the TTPI, namely
Yap, Chuuk, Pohnpei and Kosrae formed the Federated States of Microne-
sia. In 1978, the US Secretary of the Interior abolished the Congress of
Micronesia.
In 1983, national referenda in both the Marshall Islands and the Feder-
ated States of Micronesia approved and signed a ‘Compact of Free Asso-
ciation’ (see below) with the US and, consequently, became independent
nations. These agreements between the US and the two new Micronesian
Palauan English 321
states and between the US and the US Commonwealth of the Northern
Marianas left Palau as the last UN Trust Territory in the world. The reason
for delay in the case of Palau was that some provisions of the proposed
constitution of Palau, particularly relating to its nuclear ban and its ban
on giving legal ‘eminent domain’ to non-Palauans, were deemed to be
‘in conflict with US ambitions for Palau’ through the Compact of Free
Association (Anglim 1988: 14; Roff 1991: 99). In return, the Palauan con-
stitutional authorities required the US to: (1) provide substantial financial
subsidies to the islands and (2) give islanders the legal status of ‘habitual
residents’, allowing them the unhindered right to live and seek employ-
ment in the US (Anglim 1988; Roff 1991). The required number of votes to
support the Compact of Free Association was not reached in seven Palauan
national referenda over ten years. Several attempts were made to revise
the constitution or revise the compact agreement in order to win Palauan
approval. A violent conflict between those in favour of and those against
the Compact of Free Association with the US intensified. Palauan politics
became embroiled in corruption – bribery was used to lend support to the
compact, blackmail was rife, schools were suspended (Anglim 1988; Roff
1991). A number of murders, shooting sprees, firebombings and the strafing
of boats and houses belonging to the chief justice and anti-compact leaders
and lawyers occurred, while the first two presidents of Palau died under
suspicious circumstances (Anglim 1988; Roff 1991). Economic pressure was
also used; the US reduced the level of its support to Palau, while 900
out of 1,300 government employees were laid off and directed to support
the compact campaign (Anglim 1988: 21–7). This continued for over ten
years while the majority of Palauans persistently opposed the compact. In
1992, however, through a manipulation of the Palauan constitution that
lowered the required threshold to a simple majority, and because the only
realistic choice was between the agreement and continued US colonial
rule, Palau entered into the Compact of Free Association with the US. The
US terminated the last UN trusteeship in the world, and subsequently in
1994, at last, Palau technically became a sovereign state, the Republic of
Palau.
The US persistently avoided portraying these territories as colonies –
they were initially ‘in trust’ to the US, and then later in a ‘Compact of Free
Association’. Independence has, however, brought little real evidence that
Palau can act independently on the international stage. It almost always
votes with the US in the UN, for example on issues where Israel or Cuba
are involved, with senior Palauan minister Sandra Pierantozzi recently
admitting that ‘we don’t want to jeopardise that relationship, because it
322 david britain and kazuko matsumoto
would affect Palau’s economic welfare’ (Pierantozzi 2009, 30 October). It
has also rehoused former inmates of Guantanamo Bay detention centre.
It even holds its presidential elections on the same day as those in the
US. So on 6 November 2012, while Barack Obama retained the American
presidency, Tommy Remengesau Jr defeated incumbent Johnson Toribiong
in the Palauan presidential vote. Its total financial reliance on US funds has
rather tempered ambitions to be fully independent.
Palau has actually ultimately served a military role for the US only
potentially. The US identified Micronesia as a key strategic site for military
purposes for two reasons. Firstly, it serves as a first line of defence for the
US (see above). Secondly, during the Cold War, the US saw Micronesia as
a conveniently isolated nuclear weapons laboratory due to its geographical
remoteness in the Pacific, and as a strategic aircraft carrier due to its location
(Lynch 1973: 133–4; Pomeroy 1951: 169 in Anglim 1988: 3). It turns out that
although the Compact of Free Association gives the US the right to have
military bases in Palau, it has not yet chosen to have any. However, the
importance for the US is that it reserves the right to use Palau for military
purposes if necessary, and to prevent other nations from using Palau for
such purposes.
In terms of identity constructions during the second phase, two dif-
ferent sets of relationships in Palau need to be considered; the first is
the relationship between the IDG and STL from 1962 onwards, while
the second is the relationships between the IDG Palauans and an ADS
adstrate strand from the mid 1980s until today. Firstly, as seen earlier, Palau
received a relatively large number of Americans for the first time during
the 1960s and 1970s. They were again temporary and short-term person-
nel, yet the volunteers were said to be different from the earlier American
officials (Abe 1986: 206; Shuster 1982: 208). First and foremost, the vol-
unteers were required to receive 350 hours of local language and cultural
lessons in Hawai‘i before departing to each island (Trifonovitch 1971: 1082;
Rachebei and McPhetres 1997: 251). These young, enthusiastic volunteers,
who did try to communicate in the local language, gave the islanders confi-
dence that their language and culture also deserved the respect and esteem
of the outside world (Shuster 1982: 208). Also, they tended to be more
sympathetic to the islanders, often criticising US administration policies
in Palau (Abe 1986: 206; Rachebei and McPhetres 1997: 251–2). Thus,
language contact between the STL and IDG strand during the second
phase was no longer restricted to the small local elite who had monop-
olised access during the first phase, but rather broadened to include a
broader spectrum of Palauans, so that ordinary islanders were likely to
Palauan English 323
have better access to the STL and their English at school and work than
before.
However, during the 1970s and 1980s when generous US funds sent many
Palauan high-school graduates to US colleges, it was often the children of
the local elite who gained selection, since they could afford to attend private
American high schools in Koror where English had been the medium of
instruction. On one hand, although the accessibility of ordinary Palauans
to an STL strand temporarily improved in Palau during the 1960s and early
1970s, there are no longer American teachers or Peace Corps Volunteers in
Palauan public high schools today. Once again, there was a good degree
of distance between ordinary Palauans and the STL strand. On the other
hand, many elite Palauan children had, and continue to have, daily face-to-
face interaction with Americans not only in Palau, but also on American
soil, thanks to the continuation of funds devoted to higher education.
Segregational elitism also characterises the use of English in Palau. The
local elite who, as a consequence of American education, have access to
American English take some pride in this and know that knowledge of
English and its associated different worldview ‘gives them an extra edge of
experience and competitiveness within their own native group’ (Schneider
2007: 37). Ironically, it is this elite who, at the local political level, have
been promoting Palauan as a national language, and increasing its visibility
in local schools, while sending their own children off to the US for an
education in English (Matsumoto and Britain 2003a).
Secondly, a relationship between the IDG and an ADS strand has been
developing in recent years. During the mid 1980s when flights between
Manila and Koror enabled many Palauans to travel, the Philippines began
to be viewed ‘as a more advanced metropolitan country’ than Palau where
higher education, medical care, shopping and recreational facilities were
more highly developed and more readily available (Alegado and Finin
2000: 361). However, during the same period, a stream of Filipino contract
workers, many with college degrees and professional experience, began
landing in Palau mostly to take up low-paying employment as service
workers under the supervision of Palauans (Alegado and Finin 2000: 361),
but also to take up professional roles as doctors and teachers. This has
led to the development of a community that amounts to roughly 20 per
cent of the whole population of Koror, the largest city (Office of Planning
and Statistics 2006: 71). Tensions have risen, however, between Filipino
migrants and Palauans, since young, unskilled and untrained Palauans
often felt that their potential jobs were being ‘stolen’ by Filipinos. The
term chad ra Oles (literally, ‘people of the knife’) came to be applied to the
324 david britain and kazuko matsumoto
Filipinos, because of their reputation for using knives as weapons when
fighting (Alegado and Finin 2000: 361). Although Filipinos’ views of their
treatment by Palauans is generally favorable, there have been incidents of
alleged mistreatment of foreign workers in Palau, including ‘physical and
verbal abuse, working overtime and on days off without pay, withholding
monthly salary, deductions from monthly salary for the amount of airfare
and substandard housing’ (Alegado and Finin 2000: 362).
On the other hand, Filipinos are active participants in culturally signifi-
cant activities in Palau (e.g. planting taro and reef fishing). Since domestic
helpers have intensive and close daily interaction with Palauan families, it
is said that ‘it is all but impossible to avoid some level of integration and
assimilation above and beyond that of a hired wage employee’ (Alegado
and Finin 2000: 365). This view was supported by (a) interview data which
show that most feel that they ‘have to varying degrees become members of
the extended family household in which they are employed’, (b) in census
data ‘from the Immigration and Customs Division, as of the mid 1990s,
at least one hundred non-Palauans, of whom approximately 50% were Fil-
ipinos, were married to Palauans’ and (c) the fact that many of them have
mastered a good command of spoken Palauan (Alegado and Finin 2000:
365). So, it appears that while the IDG find the STL superior, but distant,
their relationship with the Filipino ADS strand is much closer and more
intimate.
As far as sociolinguistic conditions during the second phase are con-
cerned, it is clear that Palauan–English bilingualism has increased due to: (a)
radical educational reform (i.e. the enormously expanded US funding pot
not only introduced both American education programmes and English-
speaking teachers to Palau but also sent Palauan high-school graduates to
US colleges), and (b) daily contact with Filipino migrant workers, usually
in English. Although some say that many domestic helpers have acquired
Palauan, some effects of English being used as the lingua franca between
Palauans and Filipinos are also observable: (a) even elderly Palauans, who
used to be bilingual in Palauan and Japanese, have begun to use English
at home, and (b) Palauan children have become fluent in English but, it
is claimed, deficient in Palauan (Pierantozzi 2000: 355). Furthermore, con-
cerns have been raised that since Filipino domestic helpers often take care
of Palauan children, they no longer acquire the traditional cultural values
that had previously been passed from generation to generation (Pierantozzi
2000: 355).
Schneider has argued (2007: 39–40) that this second phase is the ‘kick
off phase for the process which is linguistically the most important and
Palauan English 325
interesting one, viz. structural nativisation, the emergence of structures
which are distinctive to the newly evolving variety’. The linguistic char-
acteristics of Palauan English are presented in Section 3 below. In addi-
tion, code-switching and mixing are common, and Tagalog words have
begun to be borrowed into English, suggesting that everyday interaction
with the ADS strand is having a linguistic impact on Palauan English
too.
Has Palauan English entered the nativisation phase? This is difficult to
judge, given that only very preliminary steps have been made so far to
investigate the structure of Palauan English. As Schneider suggests:
in former exploitation colonies . . . the STL strand is often demographically
weakened or even almost completely removed after independence, but the
effects and attitudes generated by them linger on and remain effective. Fac-
tors like the appreciation of English, its persistent presence with important
functions, and the desire to maintain contacts with the former colonial
power and to participate in international communication have the same
effect as the physical presence of large numbers of English speakers . . . the
pressure to accommodate to English usually affects primarily the IDG strand
people, leading to widespread second language acquisition of English and
sometimes almost complete language shift or even language death. (2007:
42)
Palau has never had anything but an extremely sparsely present STL strand,
but certainly the economic benefits of English are widely recognised (as are
those of Japanese, especially in the tourist industry), it retains its diglossic
High status in Palauan government and administrative life, and is used to
an increasing extent as Palau participates more and more in regional and
global forums, political, but also cultural, sporting and environmental.
Matsumoto’s (2001) survey of language attitudes in Palau clearly showed
islanders’ awareness of the importance of English (and Japanese) in Palau,
as well as strong agreement for the need to maintain both languages. Nev-
ertheless, the local language Palauan was deemed to be the most impor-
tant on the islands, and the one most wanted to preserve. Palauan is,
at this point at least, still a robust, healthy and valued language on the
islands.
kit i goat oˑ - oʊ
dress e goose u
trap–bath ɛ - a, a price ai
lot a-ɑ choice oi
strut a-ɐ mouth aʊ - ɑə
foot ʊ near ɪˑ
cloth ɑ square eˑ
nurse ɤɹ - ɜˑ - ə start aɹ - a
fleece i north oɹ
face eˑ - ei happy i
palm a-æ horses ɛ
thought ɑ-ɒ comma a
– kit and dress are relatively close in PE, and this leads to considerable
overlap between close kit and a relatively short fleece vowel. A similar
pattern for PhilE is reported by Llamzon (1997: 46), McArthur (2002:
346) and Tayao (2004: 1051).
Palauan English 327
– /a/ is highly variable, both at the inter- and intra-speaker level. While
many words in the bath set are consistently realized as [a], and many
in the trap set are fronted and raised relative to [a], further closer
examination of /a/ is required to determine the exact phonological
definition of this split in Palauan English. Similar variability is noted
for PhilE by Llamzon (1997: 46).
– While some tokens of strut in our data were relatively mid-open, and
some tokens of lot relatively back, there is a good deal of overlap of
these two sets;
– nurse is highly variable. Many rhotic tokens have back mid close
vowels, while fleece and goose are parallel and especially close vowels
at the front and back. There are no signs of diphthongisation, and no
signs of the fronting of goose.
– face and goat are similarly parallel – the nuclei of each are mid-close
and short for each, with little if any glide. There are no signs of goat
fronting.
– thought is mostly, but not entirely unrounded, but nearly always
short. Rhotic forms (north) tend to have a relatively close back vowel.
– Generally price and mouth have fully open nuclei and a full glide
to close position at the front and back respectively, though there is
evidence of some glide reduction for mouth.
– Both start and palm can be short, and, in the case of palm, also very
front.
– near and square are both monophthongs.
– happy is consistently tense.
– Fully unstressed vowels are rare, and consequently the vowel of the
final syllable of, for example, horses is routinely realized as [ɛ] (see
Tayao 2004: 1050, who reports the same for PhilE).
– The final vowel of words in the comma class is almost always realized
as [a].
3.1.2 Consonants
/p t k/: The voiceless stops almost always lack aspiration, as they do,
according to Llamzon (1997: 46) and Tayao (2004: 1053) in
PhilE. There is a good deal of variation in the realisation of /t/.
Intervocalically, both across word boundaries and
word-medially, /t/ can be realized not just as a [ɾ] flap, but also
as a glottal stop [ʔ]. /kw/ clusters are often [k] (e.g. equipment
[eˈkipment]).
328 david britain and kazuko matsumoto
/b d g/: These are often devoiced, not only in word-final position, but
also initially and medially (e.g. gun [kan]; guilt [kilt]; digging
[dikiŋ]).3 Llamzon (1997: 46, 47) finds the same for PhilE. /d/
word finally is also, among some speakers, realized as [ð] or [θ].
Palauan English undergoes heavy -t/-d deletion, in past-tense
morphemes as well as monomorphemes, and before vowels as
well as preconsonantally.
/θ ð/: Variation reigns here – /θ/ can be realized as [t], especially in
/θr/ clusters, [d] or even [ð], as well as [θ]. /ð/ can be
pronounced as [d] as well as [ð] (see Llamzon 1997: 46, 48;
McArthur 2002: 346; Tayao 2004: 1053 for similar findings in
PhilE).
/z/: A voiced realisation is rare in Palauan English (and in PhilE –
Llamzon 1997: 46, 48; McArthur 2002: 346; Tayao 2004: 1054),
apart from among some very proficient speakers. Words such
as is and was are routinely pronounced [is] and [was];
/ʃ ʧ ʤ/: /ʃ/ is usually realized as [s] (e.g. fishing [fisiŋ], she [si]), or
occasionally [sj ]. /ʧ/ is mostly pronounced [ts] (e.g. taro patch
[taɾopats], each [its]). /ʤ/ is variably realized as [ts] or simply
[s]. Palauan English patterns like PhilE in these cases
(McArthur 2002: 346; Tayao 2004: 1054).
/ŋ/: Although in -ing morphemes /ŋ/ can be realized with an
alveolar or more dental nasal stop, the velar predominates.
/r/: Palauan English is semi-rhotic. Although further investigation
is needed, Palauan English is not, unlike some other
semi-rhotic varieties, strikingly less rhotic in unstressed
syllables, presumably as a consequence of the more
syllable-timed nature of the variety’s prosody, and the
consequent relative lack of clearly unstressed syllables. Palauan
English’s semi-rhoticity also has consequences not only as can
be seen in Table 14.1, for the vowel system, but also for the way
in which hiatus is resolved (see below). In prevocalic position
an alveolar tap is the most frequent realisation, as in PhilE
(Llamzon 1997: 47, 48; McArthur 2002: 346).
/h/: Palauan English realises /h/ consistently.
3 An interesting case of a hypercorrection has been noted in Palau’s newspaper Tia Belau: ‘The damages
and losses they are requesting amount to $485 per week since mid March for their inability to fish
or collect grabs, different shells, sea cucumber, and other marine species for sale at the market’ (Two
arrested for machete attacks (2012, 7 September) Tia Belau [Koror, Republic of Palau]. Retrieved 10
November 2012, from http://tiabelaunewspaper.com/?p=594; our emphasis).
Palauan English 329
/l/: Palauan English makes a distinction between [l] prevocalically
and [ɫ] non-prevocalically, though the latter is not especially
‘dark’.
/ju/: Palauan English deletes the palatal glide after coronals and /m/,
but retains it after oral labials, velars and /h/.
3.1.3 Prosody
The hiatus resolution system of Palauan English is variable. Although it is
not unusual to find the hiatus from a high front vowel to the following
vowel being resolved with a [j] glide (very angry [veɾijaŋɡɾi]) and that from
a high back vowel being resolved with a [w] glide (go away [ɡowawei ]),
the predominant consonant used to resolve hiatus in Palauan English is
a glottal stop (e.g. my older boy [maiʔoɫdɛboi]). Hiatus resolution after a
non-high vowel is also mostly accomplished with the glottal stop – linking
and intrusive /r/ are virtually but not entirely absent. The use of the glottal
stop as a hiatus breaker, furthermore, extends to the definite and indefinite
article systems – see below in Section 3.2.10.
Palauan English appears to have a syllable-timed rather than stress-timed
prosody. Heavily reduced syllables are rare and speakers, for example, tend
to utter full forms of many of the small function words that are often
highly reduced in Inner Circle varieties. Except in the grammaticalised
verbal forms wanna, gotta and gonna, to is almost always [tu] and not [tə];
furthermore, my is always [mai] and not [mə], of is [ɑv] and not [ə], etc.
This characteristic of PE is also widely reported for PhilE (Gonzales 1983:
155; Thompson 2003: 52; Llamzon 1997: 46; McArthur 2002: 346; Tayao
2004: 1055).
In light of the above, it is perhaps not surprising that assignment of stress
is variable and often patterns in ways unlike those found in Inner Circle
varieties. The examples below from our data provide evidence of both the
lack of reduced syllables and Palauan English’s distinctive stress assignment
for some words:
hoping [hoˈpiŋ]
grandchildren [gɾantsiɫˈdɾɛn]
exam [ˈeksəm]
open [oˈpen]
secretary [sɛkɾiˈtaɾi]
forget [ˈfoɾɡet]
permission [pɛɾmiˈsjɒn]
compost [kɒmˈpos]
330 david britain and kazuko matsumoto
discipline [disiˈplin]
complicated [kɑmpliˈkeiɾɪt]
Similar claims are made about PhilE (Thompson 2003: 53; McArthur
2002: 346; Tayao 2004: 1055–7).
3.2.3 Negation
The use of negative concord is such a frequently occurring feature of the
world’s Englishes that Chambers (2004) labels it a ‘vernacular universal’.
It appears, however, to be infrequent in Palauan English ((26); see further
(27) below):
(26) I’m not in favour of nobody
Furthermore, secondary contractions of already contracted negated aux-
iliary verbs – variants such as ain’t, not infrequent in Britain and North
America – are also rare in Palauan English, as in (27):
(27) there ain’t no big money
Palauan English rarely uses never as a negator with definite time reference,
as in (28):
(28) yeah somebody bungee-jumped and never came up
3.2.4 Adverbs
Generally, adverbs are formed from adjectives as in Standard American
English, but occasionally the adverbial and adjectival forms are identical,
especially in the case of good, as in (29) and (30):
(29) we’re handling our lives good
(30) so we don’t spell it good
3.2.5 Prepositions
As is often reported for Outer and Emerging Circle Englishes, Palauan
English can differ from Inner Circle Englishes in its choice of prepositions,
Palauan English 333
as in (31), (32) and (33), or indeed whether or not one is required, as in (34)
and (35) (see Bautista 2000: 152–4; Thompson 2003: 53 for PhilE):
(31) once you spank your kids, they’ll put you to jail
(32) I share stories to boys, my friends
(33) and I get really pissed off from her
(34) they knocked the door and my brother came out
(35) he wanted to go Guam
3.2.6 Plurality
The morphological marking of plurality is highly variable in Palauan
English, as in (36)–(39):
(36) fifteen more minute
(37) girls clean and they’re the one who even fix the shoes
(38) two hot tea, please
(39) wash your hand before you come too close to me!
Some lexemes that are routinely pluralised in Inner Circle Englishes are
often treated as singular in Palauan English, as in (40):
(40) you want sunglass, I’ll buy you one
Occasionally lexemes treated as mass nouns in Inner Circle Englishes are
treated as count nouns and attract plural morphemes in Palauan English
(and in PhilE – Thompson 2003: 53), as in (41):
(41) he’s selling fishing gears
If a noun is preceded by a determiner or (non-numeric) quantifier which
is semantically plural, the noun often lacks what is, in essence, redundant
plural marking, as in (42)–(48):
(42) they just left her in one of the room
(43) one of my brother become elder of the church
(44) there’s several long distance carrier
(45) we don’t have enough word in Palauan
(46) he was like one of my dad’s favourite son
334 david britain and kazuko matsumoto
(47) you have any classmate you know that’s single?
(48) some of my classmate, they turn on the candle after the light out
Sometimes the indefinite article can precede non-overtly morphologi-
cally marked plural lexemes, as in (49) and (50):
(49) that’s the Palauan term for a smart or intelligent people
(50) she cannot have a children.
Much, used with non-count singular nouns in Inner Circle Englishes, is
frequently used with count nouns in Palauan English, as in (51)–(54):
(51) there’s not really much Chinese stores
(52) there’s not that much rangers working during swing shift
(53) even I have so much kids, no one can replace one
(54) I like the olden days, now too much problems.
3.2.7 Pronouns
Pro-drop strikes us as being more frequent in Palauan English than one
would expect in Inner Circle conversational English, as in (55)–(57):
(55) a: eighty four? oh, not eighty five?
b: yeah yeah she’s eighty three now
a: no, gonna be eighty five
(56) compare price of fish . . . from here to Guam is very different . . . over
there is three ninety five a pound
(57) a: that’s Nolan . . . want to be perfect.
b: all right
a: want to do everything just right
On rare occasions, relative clauses can have resumptive pronouns, for
example (58) and (59):
(58) but those are the things that she think they’re important
(59) we are supposed to set the rules and tell the domestic helper that
these are the rules that they have to abide by them.
Equally rare are subject relative clauses with a zero pronoun, as in (60):
(60) you know we have this game was taught by Japanese
Palauan English 335
3.2.8 Comparison
A good number of varieties have ‘double comparison’ and use both the
inflectional ending (-er for comparatives and -est for superlatives) and the
appropriate analytic marker (more or most). Palauan English is no exception,
as in (61):
(64) I saw my mother sitting on the bench outside with the blood on the
dress
(65) my mother always go to taro patch
(66) I don’t want my candidate to lose election
Descriptive grammars suggest that for both the definite and indefinite
article systems, Standard (British) English is allomorphic, with different
forms before vowels than before consonants. Following consonants are
preceded by a and the [ðə], whereas following vowels are said to attract
indefinite an and definite the [ði]. Britain and Fox (2009) have observed
that actually few varieties – Inner or Outer Circle – totally adopt this
strict allomorphy, with many using the preconsonantal forms in prevocalic
position, and inserting a glottal stop to resolve the ensuing phonological
hiatus. Palauan English also only variably maintains this allomorphy, with
preconsonantal [ðə] + [ʔ] used before the vowels in (67), (68), (69), (70).
For examples of lack of vowel-sensitive allomorphy for the indefinite article,
see (71), (72) and (73):
336 david britain and kazuko matsumoto
(67) you listen to the instructor
(68) he’s gonna help me get the stuff from the airport
(69) he died at the age of 92
(70) because of the alcohol and the drugs and that
(71) there’s a old age center
(72) he’s a old man
(73) they are looking for a place in the Pacific region to put a importer
for regional research
3.4 Discussion
Palauan English appears, on the one hand, far more striking phonologically
than grammatically, but, on the other, incorporates many of the features,
both phonological and grammatical, that Schneider portrays as typical
of postcolonial Englishes in general: vowel shortening, loss of short-long
contrasts (2007: 72), shortening or ungliding of diphthongs, the stopping
of /θ ð/, the reduction of word initial aspiration (2007: 73), shifts in
stress (2007: 74) and syllable timing (2007: 75); absence of nominal plural
marking (2007: 83); third person -s omission (2007: 83); a lack of yes–no
question inversion (2007: 84); and ‘idiosyncratic’ article use (2007: 85).
Indeed several of these features are, furthermore, characteristics of English
as a lingua franca (Seidlhofer 2005: 92) – English used by speakers of
different first language backgrounds as a means of communication.
It clearly shares many characteristics especially with Phillipine English,
as we have seen above, but it is not straightforward to ascertain precisely
the cause of this similarity. On the one hand, there has been large-scale
migration of workers from the Phillipines, many of whom play key roles in
the linguistic socialisation of children in Palau – as childcarers and domestic
workers, but also as teachers. On the other hand, the indigenous languages
of Palau and the Phillipines are both members of the Malayo-Polynesian
subfamily of Austronesian, and so there could well be substrate effects
involved too. Furthermore, although today well over a quarter of the pop-
ulation of Palau stems from the Philippines, this is a relatively recent trend.
There have been migrant workers in Palau since the end of the Pacific War,
but in small numbers until the mid 1980s – significant Philippine influ-
ence on Palau does not spread over the entirety of the island’s anglophone
period. And the fact that Palauan English shares many features with Outer
4 Tia Belau, 5 November 2012 (Koror, Republic of Palau). Retrieved 10 November 2012, from http://
tiabelaunewspaper.com/?p=110.
338 david britain and kazuko matsumoto
Circle Englishes worldwide also leads us to be somewhat sceptical about
a specific and crucial role of the Filipino migrant worker. Investigations
of Micronesian Englishes that have witnessed much less Filipino immigra-
tion may ultimately help us more robustly evaluate the strength of different
contributions to the structure of contemporary Palauan English. Lexically,
as one might expect, Palauan English incorporates a number of borrowings
from both Palauan – some of which were originally from Japanese – and
Tagalog.
4 Conclusion
We have attempted to plot the linguistic consequences of Palau’s complex
path to independence over the past 110 years, and especially the latter half of
that journey. Different colonial rulers have policed Palau in very different
ways and left very different legacies. The most intensive intervention was
undoubtedly that during the Japanese era – for the Japanese, like no
other colonial ruler, Palau was a settlement colony, with Japanese migrants
outnumbering the locals. At no other time in Palau’s colonial history
has there been anything more than a trickle of colonial settlement. Not
surprisingly, then, it is Japanese which has had the most impact on the
local language, with numerous borrowings integrated, including for food,
culture and core terms expressing feelings and emotions (Matsumoto and
Britain 2003a; Matsumoto 2010b).
The American period, as we have seen, began otherwise, with almost
total indifference and little attempt to develop the islands or prepare them,
as in fact was their duty under the UN Trust mandate, for independence.
As political circumstances changed, so did American policy, somewhat,
but there was never a great influx of American migrants, and consequently,
and because of aid policy, most contact that Palauans had with Americans
was in the United States during periods of college training. Unlike many
of the cases in Schneider (2007), therefore, the colonial ‘motivation’ was
neither settlement, nor indeed exploitation in a typical sense – there were
few natural resources to exploit. And it was not even to establish military
bases. The US has never done so. But as we have seen, the US’s main-
tenance of Palau served (and continues to serve, given the Compact of
Free Association) important potential geopolitical functions, which may
become more important, for example, in the context of the increased
global and especially regional power of the People’s Republic of China. In
Schneider’s terminology, then, we can nevertheless argue that Palau served
as an exploitation colony: ‘the primary goal . . . is to secure . . . political
Palauan English 339
and military interests in a region, not to spread the English language
or . . . cultural influence’ (2007: 65). Certainly the US did little to spread
the English language, and rather than bringing the English language to
Palau, it brought (some) Palauans to the English language. Palauan English
seems currently to be at the second exonormative stabilisation phase
in the Dynamic Model, but the potential for nativisation seems great
given the continued lack of American (English) input to everyday life in
Palau.
References
Abe, G. 1986. An ethnohistory of Palau under the Japanese colonial administration.
Unpublished PhD thesis, University of Kansas.
Aldridge, B. and C. Myers. 1990. Resisting the Serpent: Palau’s Struggle for Self-
Determination. Baltimore, MD: Fortkamp Publishing Company.
Alegado, D. and G. Finin. 2000. Exporting people: the Philippines and contract
labor in Palau. Contemporary Pacific 12: 359–70.
Anglim, J. 1988. Palau’s strategic position places Palauan democracy at risk. Working
paper of the Peace Research Centre, Research School of Pacific Studies, Australian
National University 40: 1–42.
Aoyagi, M. 1977. –
[Culture changes and social organisation: the case of Yap and Palau in
Micronesia], Shien 37: 43–65.
Bautista, M. 2000. The grammatical features of educated Philippine English. In
L. Bautista, T. Llamzon and B. Sibayan, eds., Parangal cang Brother Andrew:
Festschrift for Andrew Gonzalez on His Sixtieth Birthday. Manilla: Linguistic
Society of the Philippines, 146–58.
Britain, D and S. Fox. 2009. The regularisation of the hiatus resolution system
in British English: a contact-induced ‘Vernacular Universal’? In M. Filp-
pula, J. Klemola and H. Paulasto, eds., Vernacular Universals and Language
Contacts: Evidence from Varieties of English and Beyond. London: Routledge,
177–205.
Chambers, J. K. 2004. Dynamic typology and vernacular universals. In B. Kort-
mann, ed., Dialectology Meets Typology: Dialect Grammar from a Cross-
linguistic Perspective. Berlin: Mouton de Gruyter, 127–45.
Chappell, D. 1994. Secret sharers: indigenous beachcombers in the Pacific Islands.
Pacific Studies 17: 1–22.
Engelberg, S. 2006. The influence of German on the Lexicon of Palauan and
Kosraean. In K. Allan, ed., Selected Papers from the 2005 Conference of the
Australian Linguistic Society. www.als.asn.au
Firth, Stewart. 1973. German firms in the Western Pacific Islands, 1857–1914.
Journal of Pacific History 8: 10–28.
Flora, M. 1974. Palauan phonology and morphology. Unpublished PhD disserta-
tion, University of California, San Diego.
340 david britain and kazuko matsumoto
Gonzalez, A. 1983. When does an error become a feature of Philippine English?
In R. Noss, ed., Varieties of English in South-East Asia. Singapore University
Press, 150–72.
Greenhill, S. J., R. Blust and R. D. Gray. 2008. The Austronesian Basic Vocabulary
Database: from bioinformatics to lexomics. Evolutionary Bioinformatics 4:
271–83.
Hall, R. 1945. English loan-words in Micronesian languages. Language 21: 214–19.
Hazel, F. X. 1984. Schools in Micronesia prior to American administration. Pacific
Studies 8: 95–111.
Hazel, F. X. and M. Berg. 1980. Micronesia: Winds of Change. Saipan: Omnibus
Program for Social Studies-Cultural Heritage, Trust Territory of the Pacific
Islands.
Imaizumi, Y. 1990. 1914–22 [Japanese naval admin-
istration in Micronesia 1914–22]. [Studies in International
Relations] 17: 1–18.
Josephs, L. 1975. Palauan Reference Grammar. Honolulu: University Press of
Hawaii.
1990. New Palauan–English Dictionary. Honolulu: University Press of Hawaii.
1997. Handbook of Palauan Grammar, vol. i. Koror: Bureau of Curriculum and
Instruction, Ministry of Education.
1999. Handbook of Palauan Grammar, vol. ii. Koror: Bureau of Curriculum and
Instruction, Ministry of Education.
Kabayashi, I. 1994. The US Confidential Paper & Termination of the UN Trusteeship:
Solomon Report, Independence of Micronesia. Tokyo: Toshindo.
Keate, G. 1788. An Account of the Pelew Islands, Situated in the Western Part
of the Pacific Ocean. Composed from the Journals and Communications of
Captain Henry Wilson, and some of his Officers, who, in August 1783, were
there Shipwrecked, in the Antelope, a Packet belonging to the Hon. East India
Company. York: Edward Peck.
Keesing, R. 1988. Melanesian Pidgin and the Oceanic Substrate. Stanford University
Press.
Kikumura-Yano, A., ed. 2002. Encyclopedia of Japanese Descendants in the Americas:
An Illustrated History of the Nikkei. Walnut Creek: Altamira Press.
Kluge, P. 1991. The Edge of Paradise: Americans in Micronesia. New York: Random
House.
Leibowitz, A. 1996. Embattled Island: Palau’s Struggle for Independence. Westport,
CT: Greenwood.
Lingenfelter, S. 1974. Administrative officials, Peace Corps lawyers and directed
change on Yap. In D. Hughes and S. Lingenfelter, eds., Political Development
in Micronesia. Columbus, OH: Ohio State University, 54–71.
Lipski, J. with P. Mühlhäusler and F. Duthin. 1996. Spanish in the Pacific. In S.
Wurm, P. Mühlhäusler and D. Tryon, eds., Atlas of Languages of Intercultural
Communication in the Pacific, Asia and the Americas, vol. 2.1: Texts. Berlin:
Mouton de Gruyter, 271–98.
Palauan English 341
Llamzon, T. 1977. The phonology of Philippine English. In M. Bautista, ed.,
English is an Asian Language: The Philippine Context. Sydney: Macquarie
Library, 41–8.
Lynch, D. 1973. United States policy towards Micronesia, 1945-1972. Unpublished
PhD dissertation, West Virginia University.
Matsumoto, K. 2001. Multilingualism in Palau: language contact with Japanese
and English. In T. E. McAuley, ed., Language Change in East Asia. London:
Curzon Press, 84–142.
2010a. The role of social networks in the post-colonial multilingual island of
Palau: Mechanisms of language maintenance and shift. Multilingua: Journal
of Cross-Cultural and Interlanguage Communication 29(2): 133–65.
2010b. Palauan language contact and change: a sociolinguistic analysis of bor-
rowing in Palauan. In J. Dobovsek-Sethna, F. Fister-Stoga and C. Duval,
eds., Linguapax Asia: A Retrospective Edition of Language and Human Rights
Issues, Collected Proceedings of Linguapax Asia Symposia 2004–2009. Tokyo:
Linguapax Asia, 36–52.
Matsumoto, K. and D. Britain. 2003a. Language choice and cultural hegemony in
the Western Pacific: Linguistic symbols of domination and resistance in the
Republic of Palau. In M. Dedaic and D. Nelson, eds., At War with Words.
Berlin and New York: Mouton de Gruyter, 315–57.
2003b. Contact and obsolescence in a diaspora variety of Japanese: the case of
Palau in Micronesia. Essex Research Reports in Linguistics 44: 38–75.
2006. Palau: language situation. In K. Brown, ed., Encyclopaedia of Language
and Linguistics, vol. 9. Oxford: Elsevier, 129–30.
forthcoming a. The vernacularity of Palauan Japanese. In K. Matsumoto and
D. Britain, eds., Diaspora Japanese, special issue of International Journal of
the Sociology of Language.
forthcoming b. The sociolinguistics of the Japanese Diaspora: An overview
of Japanese migration history and language contact. In K. Matsumoto and
D. Britain, eds., Diaspora Japanese, special issue of International Journal of
the Sociology of Language.
McArthur, T. 2002. Oxford Guide to World English. Oxford University Press.
McHenry, D. 1975. Micronesia: Trust Betrayed. Washington, DC: Carnegie
Endowment for International Peace.
McKinney, R. 1947. Micronesia under German rule 1885–1914. Unpublished MA
thesis, Stanford University.
Mühlhäusler, P. 1996. German in the Pacific area. In S. Wurm, P. Mühlhäusler
and D Tryon, eds., Atlas of Languages of Intercultural Communication in the
Pacific, Asia and the Americas, vol. 2.1: Texts. Berlin: Mouton de Gruyter,
345–51.
Mühlhäusler, P. and P. Baker. 1996. English-derived contact languages in the Pacific
in the 20th century (excluding Australia). In S. Wurm, P. Mühlhäusler and D.
Tryon, eds., Atlas of Languages of Intercultural Communication in the Pacific,
Asia and the Americas, vol. 2.1: Texts. Berlin: Mouton de Gruyter, 497–522.
342 david britain and kazuko matsumoto
Nan’yō-chō [South Seas Bureau] (1928, 1939, 1941, 1942)
[Annual Report on Micronesia]. Tokyo.
Office of Planning and Statistics. 2005. 2002–2003 Census of Population and Housing
of the Republic of Palau. Koror.
2006. 2005 Census, vol. ii: Census Monograph Population and Housing Profile.
Koror.
Palau Community Action Agency. 1977. A History of Palau: Traders and Whalers,
Spanish Administration, German Administration, vol. 2. Koror: Palau Com-
munity Action Agency.
1978. A History of Palau: Japanese Administration, US Naval Government, vol. 3.
Koror: PCAA.
Peattie, M. 1988. Nan’yō: The Rise and Fall of the Japanese in Micronesia 1885–1945.
Honolulu: University of Hawaii Press.
Pierantozzi, S. 2000. Palauans and guest workers: an opinion paper. Contemporary
Pacific 12: 349–58.
2009. Palau supports US against Cuba for aid. Interview posted
to ABC Radio Australia, 30 October. Retrieved 10 November
2012, from www.radioaustralia.net.au/international/radio/onairhighlights/
palau-supports-us-against-cuba-for-aid
Pomeroy, E. 1951. Pacific Outpost: American Strategy in Guam and Micronesia.
Stanford University Press.
Rechebei, E. and S. McPhetres 1997. History of Palau: Heritage of an Emerging
Nation. Koror: Ministry of Education.
Roff, S. 1991. Overreaching in Paradise: United State Policy in Palau since 1945.
Juneau: The Denali Press.
Schneider, E. 2007. Postcolonial Englishes. Cambridge University Press.
Seidlhofer, B. 2005. English as a lingua franca. In A. Hornby, ed., Oxford Advanced
Learner’s Dictionary of Current English. Oxford University Press, R 92.
Shuster, D. 1978. Major patterns of social change instituted in Micronesia during
Japanese colonial rule, 1914–1940. Unpublished term paper, University of
Hawaii at Manoa.
1982. Islands of change in Palau: Church, school, and elected government,
1891–1981. Unpublished PhD thesis, University of Hawaii at Manoa.
Täuber, I. and C. Han. 1950. Micronesian islands under United States trusteeship:
demographic paradox. Population Index 16: 93–115.
Tayao, M. 2004. Philippine English: phonology. In E. Schneider, K. Burridge,
B. Kortmann, R. Mesthrie and C. Upton, eds., A Handbook of Varieties of
English, vol. i: Phonology. Berlin: Mouton de Gruyter, 1047–59.
Thompson, R. 2003. Filipino English and Taglish: Language Switching from Mul-
tiple Perspectives. Amsterdam: John Benjamins.
Tkel, D. 2000. Conversational Palauan. Guam: Micronesian Language Institute,
University of Guam.
Trifonovitch, G. 1971. Trust Territory of the Pacific Islands. In T. Sebeok, ed.,
Current Trends in Linguistics, vol. 8: Linguistics in Oceania. The Hague:
Mouton, 1063–87.
Palauan English 343
Tryon, D., P. Mühlhäusler and P. Baker. 1996. English-derived contact languages
in the Pacific in the 19th century (excluding Australia). In S. Wurm, P.
Mühlhäusler and D. Tryon, eds., Atlas of Languages of Intercultural Commu-
nication in the Pacific, Asia and the Americas, vol. 2.1: Texts. Berlin: Mouton
de Gruyter, 471–95.
United Nations (n.d.a). The United Nations and Decolonisation: Trusteeship
Council. Retrieved 10 November 2012, from www.un.org/en/decolonization/
trusteeship.shtml
United Nations (n.d.b). The United Nations and Decolonisation: International
Trusteeship System. Retrieved 10 November 2012, from www.un.org/en/
decolonization/its.shtml
Vidich, A. 1952. The political impact of colonial administration. Unpublished
PhD thesis, Harvard University.
Wilson, H. 1972. The phonology and syntax of Palauan verb affixes. Unpublished
PhD dissertation, University of Hawaii at Manoa.
Data sources
Two arrested for machete attacks (7 September 2012) Tia Belau [Koror, Republic of
Palau]. Retrieved 10 November 2012, from http://tiabelaunewspaper.com/?
p=594
Villagers ask officials to fix damaged sewer system in Melekeok, pay compensa-
tion (5 November 2012) Tia Belau [Koror, Republic of Palau]. Retrieved 10
November 2012, from http://tiabelaunewspaper.com/?p=110
Index
accents, 22, 23, 42, 61, 149, 152, 229 Cheyenne, 2, 108, 109, 112, 115, 116, 122
Aceto, Michael, 195, 216 choice vowel, 154, 207
acrolect, 11, 21, 22, 35, 42, 43, 46, 193, 270 clear-skinned
African American English, 111, 113, 143 as a social category, 146
Afro-Seminole, viii, 236, 239, 241, 246, 263, 264 code-mixing, 42
ain’t, 134, 136, 137, 139, 140, 161, 183, 245, 260, 332 code-switching, 12, 15, 42, 62, 224, 317, 325, 336
Appalachian English, 109 commercial Englishes, 189
a-prefixing, 160 Compact of Free Association, 306, 315, 320, 321,
Arapaho, 2, 117 322, 338
archaisms, 4 completive
areal feature, 156 done, 159, 173, 178, 190, 280
aspect, 16, 48, 90, 107, 108, 110, 111, 112, 120, 135, continuum model, 192
156, 158, 159, 160, 173, 175, 177, 178, 179, 213, convergence, 44, 50, 76, 122, 206, 207, 211, 216
214, 249, 250, 251, 253, 256, 278, 279 copula, 100, 107, 111, 123, 125, 142, 143, 160, 180,
progressive, 159, 160, 177 192, 213, 215, 331
aspiration, 59, 82, 83, 231, 232, 327, 337 copula deletion, 215
Australia, xv, 6, 11, 43, 70, 220, 221, 222, 223, creole languages, xiv, 142, 195, 196
224, 225, 226, 227, 228, 233, 234, 280, 289, cure vowel, 132
290, 307, 320, 341, 342, 343
dark /l/, 17, 296
Bahamas, 170, 173, 174, 196, 242, 264 decreolization, 107, 177, 186, 191, 192, 193, 194,
Barbados, 5, 130, 142, 143, 147, 163, 167, 170, 209, 195, 201, 216
217, 238, 239 dialect contact, xi, xiii, 4, 13, 151, 219, 308
basilect, 11, 21, 193, 270 divergence, 5, 44, 50, 206
bath vowel, 153, 229 Dutch Windward Antilles, 189
Bequia English, vii, 5, 128, 132, 133, 134, 135, 136,
137, 139, 140, 141, 142, 152 epenthesis, 25, 26, 83, 85
bilingualism, xiii, 6, 13, 16, 219, 220, 288, 307, Euro-Caribbean English
317, 319, 324 dialects, 156
borrowing, xiii, 39, 273, 317, 336, 341 Euro-Saban English, 152, 153, 155, 156, 157, 158,
Britain, David, 94, 234, 303 159, 160, 161, 162
British English, ii, 4, 17, 19, 25, 44, 59, 105, 211, exploitation colony, 338
230, 231, 238, 290, 308, 339
face vowel, 23, 153, 203, 205, 206
calque, 39, 212, 232 First Nations, 99, 104, 105, 122, 123, 124
Canada, xv, 4, 11, 99, 101, 102, 104, 105, 106, 121, fleece vowel, 154, 205, 231, 326
123, 124, 152 force vowels, 132
Canadian English, 105 foot vowel, 153, 204
Caribbean English, 137, 141, 142, 156, 158, 162,
163, 164, 174, 203, 216, 217 goose vowel, 58, 206, 229, 297
centralization, 153 Grenadines, 5, 128, 130, 142
344
Index 345
Guarani, 6, 219, 222, 223, 224, 225, 227, 228, 231, palm vowel, 80, 204, 210
232 Paraguay, xiii, 6, 219, 220, 221, 222, 223, 224,
Gullah, viii, 6, 172, 236, 238, 239, 248, 263, 264 226, 228, 233, 234, 235
Gustavia English, x, 6, 198, 201, 202, 205, 216, poor whites, 130
217 postcolonial English, 305, 306
postvocalic, 59, 132, 175, 209, 229, 231, 233
habitual aspect, 90, 159, 178 preconsonantal positions, 17
Hamilton, 130, 132, 134, 136, 137, 138, 139 price vowel, 154, 207
Hancock, Ian F., 95 prosody, 61, 84, 110, 132, 329
H-dropping, 60
Hickey, Raymond, 48, 95 quotative, 35, 46, 160
Hopi English, 110
Redlegs, 142, 143
indentured servants, 5, 148 reflexive pronouns, 87, 133
Irish English, 4, 29, 50, 71, 74, 75, 76, 77, 78, 79, retroflex, 20, 132
80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, rhoticity, 19, 47, 82, 175, 209
92, 93, 94, 95, 141
Saba, ix, xiv, 5, 144, 145, 146, 147, 148, 149, 150,
kit vowel, 152, 203, 231, 297 151, 155, 156, 161, 162, 163, 164, 165, 170, 176,
178, 187, 189, 198, 199, 216
La Pompe, 130 Schneider, Edgar, 7, 49, 126
lenition, 82, 95 Scots, 147, 186
linking /r/, 296, 300 Sea Islands Creole, 6, 236, 246, 260
lot vowel, 24, 153 Shelta, 4, 70, 71, 72, 94, 95
St. Kitts, 147, 160, 166, 167, 168, 170, 177, 185,
Maori, 274, 275, 302, 303, 304 187, 189, 196, 198, 199, 201, 202, 214, 215,
Maroons, 240 217, 218, 239
Marshall Islands, 320 St. Martin, 5, 123, 144, 165, 166, 169, 170, 189,
merger, 58, 59, 60, 80, 153, 155, 174, 208, 273 198, 239
metathesis, 85, 133, 156 Standard English, x, 96, 133, 134, 136, 137, 138,
Micronesia, xiii, 288, 305, 310, 311, 312, 314, 315, 139, 140, 157, 158, 161, 201, 229, 270, 271,
316, 317, 319, 320, 322, 339, 340, 341, 342 274, 275, 276, 279, 284
Middle English, 77, 85, 203, 204, 206 start vowel, 132
Mount Pleasant, 130, 134, 135, 137 stress shift, 132
mouth vowel, 207, 210 strut vowel, 24, 77, 153, 154, 204, 210
Navajo English, 108, 109, 110, 114, 116 thought vowel, 24, 58, 81, 153
near vowel, 132 TH -stopping, 21, 60, 100, 108, 208
Netherlands Antilles, i, 5, 144, 151, 165, 196, trap vowel, 203, 297
197 Travellers, 70, 71, 72, 73, 74, 75, 77, 78, 79, 80,
new dialect formation, xiii, 121, 286 81, 85, 87, 89, 91, 92, 93, 94, 95, 96
Northern Mariana Islands, 305, 311, 320 Trudgill, Peter, 7, 50, 96, 126
nurse vowel, 131, 272, 296 Tupi-Guarani, 219
Old English, 206, 207 unstressed vowels, 47, 76, 132, 203, 327