ArticlePDF Available

The creoleness of Unserdeutsch (Rabaul Creole German): a typological perspective

LLM Special Issue 2017 ISSN: 0023-1959
Journal of the Linguistic Society of Papua New Guinea
ISSN 0023-1959
Special Issue 2017
Péter Maitz & Craig A. Volker (eds.):
Language Contact in the German Colonies:
Papua New Guinea and beyond
LLM Special Issue 2017
Siegwalt Lindenfelser & Péter Maitz
University of Augsburg
In this paper, we discuss to what extent the German-based contact language
Unserdeutsch (Rabaul Creole German, cf. Volker 1982) matches the category
‘creole language’ from both a socio-historical and structural perspective. As
a point of reference, we will use typological criteria that are widely supposed
to be typical for creole languages. It is shown that Unserdeutsch fits fairly
well into the pattern of an ‘average creole’, as has been suggested by data in
the Atlas of Pidgin and Creole Language Structures (Michaelis et al. 2013).
This is despite a series of atypical conditions in its development that might
lead us to expect a close structural proximity to the lexifier language, i.e. a
relatively acrolectal creole. A possible explanation for this striking
discrepancy can be found in the primary function of Unserdeutsch as a
marker of identity as well as in the linguistic structure of its substrate
language Tok Pisin.
This contribution is an extended and modified version of Maitz & Lindenfelser
(submitted). Underlying research has been funded by the Deutsche Forschungsge-
meinschaft / German Research Foundation (MA 6769/1-1). We wish to thank Craig Alan.
Volker and Susanne Klohn for their corrections and stylistic advice on our English.
The creoleness of Unserdeutsch (Rabaul Creole German)
Creole languages, German-based, language universals, language structure,
linguistic typology, Rabaul Creole German, Unserdeutsch
Unserdeutsch (Rabaul Creole German) developed among mixed-race
children at a Catholic mission station in Vunapope (near Rabaul) at the
beginning of the twentieth century (cf. Volker 1982; Maitz 2017). At that
time, the whole Bismarck Archipelago was under German colonial rule,
including the Gazelle Peninsula in north-eastern New Britain, which is the
cradle of Unserdeutsch. Today, only about 100 elderly L1 speakers of
Unserdeutsch, the only German-lexified creole language in the world, are
still alive (cf. Maitz 2016; Maitz & Volker 2017). Apart from its unique
lexifier language, Unserdeutsch ought to be of particular interest for creolists
in several aspects.
First, it developed amidst an exceptionally sharp contrast between a
strongly isolating extended pidgin language (Tok Pisin) as its main substrate,
and a highly inflected Germanic superstrate language (German). Second, to
mention only the two aspects that are specifically relevant here, the socio-
historical and socio-communicative profile of Unserdeutsch, on the one hand,
displays a range of similarities to other creole languages in the world, but, on
the other hand, shows a considerable number of very atypical characteristics.
Regarding the high amount and various kinds of atypical characteristics, one
would probably expect a language structure that is atypical for a creole
language. The question arises, whether or to what extent the grammatical
structure of Unserdeutsch reflects its socio-historical and socio-
communicative peculiarities. To put it differently: To what extent does
Unserdeutsch meet into the cross-linguistic or typological definitions of
creole languages (cf. Arends, Muysken & Smith 1995; Bartens 2013;
Michaelis et al. 2013; Velupillai 2015; Holm & Patrick 2007)?
Until lately, it would have been difficult to answer this essential question,
since the language data available were not sufficient for such a purpose. Only
a very small amount of the material collected by Craig Volker in 1979 and
1980 survived, the main part having been lost since then (cf. Götze et al., this
LLM Special Issue 2017
issue). In addition, a considerable amount of the remaining data is actually
Standard German rather than basilectal Unserdeutsch, since many of the
speakers at that time were proficient in the lexifier language as well.
However, only basilectal Unserdeutsch, the pole most distant from the
lexifier, can be considered when it comes to investigating the creoleness of
the language.
With the new documentation project launched at the
University of Augsburg in 2014, there are now enough data suitable to reach
firm conclusions.
In this article, we shall pursue the question of the creoleness of
Unserdeutsch by describing its fundamental, typologically relevant structural
features based on new language data and by confronting them with results of
research in creole universals. We will begin by briefly addressing typical and
atypical features in the genesis and the socio-communicative profile of
Unserdeutsch from a creolistic point of view (section 2). This is followed by
methodological reflections on the empirical basis of the structural analyses
as well as the point of reference used in determining creole typicality (section
3). Afterwards in section 4, basilectal Unserdeutsch will be located among
the world’s creole languages by means of fundamental typological
(phonological, morphological, and syntactic) criteria.
We conclude in
section 5, by summarising and discussing the results of the analyses.
The ontogenesis of Unserdeutsch has been described in detail elsewhere (cf.
Volker 1982; Maitz 2017). Therefore, the typical and atypical conditions in
the genesis of Unserdeutsch will only be outlined in broad terms here,
referring to the papers mentioned above for a more detailed overview.
For some further details on the creole continuum of Unserdeutsch see Maitz (2017).
Further information about the interviewed speakers and the fieldwork methodology used
for data collection can be found in Götze et al. (this issue).
In that section, we are bound to draw on main differences between Standard German and
Unserdeutsch in order to set both apart. A detailed description of typological features of
Standard German, however, is far beyond the scope of the present study. For basic
linguistic aspects of Standard German, please refer to works such as Hawkins (1986),
König & Gast (2009) and Wiese (1996) as well as the literature mentioned at relevant points
in the text.
The creoleness of Unserdeutsch (Rabaul Creole German)
On the one hand, Unserdeutsch can be regarded as a typical representative
among the world’s creole languages for at least the following reasons:
a) The language emerged at a missionary station, thus in a German colonial
settlement environment (cf. Mufwene 2009).
b) Being a colonial variety, its development is a result of linguistic and
social violence (cf. Arends, Muysken & Smith 1995: 4).
c) It constitutes a (fully) nativised, restructured contact variety.
d) Its birth depicts an abrupt, catastrophic development scenario (cf.
Bickerton 1988; Thomason 2008: 251).
e) It developed in a contact situation including (at least) one local language
and one European language. Its grammar is strongly based on the local
language (substrate language, in this case Tok Pisin), while its lexicon is
derived predominantly from the European language (superstrate
language, in this case Standard German) (cf. Thomason 2008: 243; Tryon
& Charpentier 2004: 5).
On the other hand, at least the following characteristics can be argued to make
Unserdeutsch appear as a rather atypical case among the creole languages of
the world:
a) Unserdeutsch is a boarding school creole. It thus belongs to the small set
of creole languages that emerged in a school context rather than among
slaves in the environment of plantations. Therefore, children and youths
did not only act as language ‘regulators’ in this case, but also as
‘innovators’ (to use these vivid, but problematic terms).
b) The children amongst whom Unserdeutsch developed had full access to
the lexifier language: they indeed acquired an expanded oral and literal
competence of Standard German. This was due to the enforced
acquaintance and usage of Standard German in the mission school from
the very beginning. This circumstance runs contrary to the assumed
typical development of pidgin and creole languages including restricted
access to the lexifier language at an early stage (cf. Lefebvre 2004: 89).
c) Unlike most other pidgin and creole languages (cf. Bakker 2000: 48;
Romaine 1988: 24), Unserdeutsch served as a means of horizontal in-
group communication instead of vertical out-group communication even
LLM Special Issue 2017
before its nativisation (cf. Maitz 2017). This explains why Unserdeutsch
could already become an exclusive means of familiar everyday
communication among the first speaker generation within the small,
close-knit and strictly endogamic mixed-race community. As a
consequence and in contrast to classical pidgins, Unserdeutsch was
already significantly expanded in its usage and its functions before the
process of nativisation set in.
d) Due to the small size of the language community and its close-knit social
networks (particularly resulting from the forced intragroup marriages),
the language could stabilise remarkably quickly. Within only one
generation, it was largely established and the process was almost
completed already in the second speaker generation.
From the perspective of a sociolinguistically grounded language typology
(cf. Trudgill 2011) we must assume that social structures or functions of
language are reflected in the structure of a language. In other words, the
socio-communicative conditions of the Unserdeutsch genesis should be
reflected within the structural design of the language, according to our
starting assumption. Several aspects of the macro-sociolinguistic context
regarding the development of the language and its use are counted among
those that have been identified as complexity retaining or complexity
increasing factors in the light of recent research. This means that these social
aspects are considered to retain or increase irregularity, syntagmatic and/or
paradigmatic redundancy, and/or morphosemantic intransparency according
to sociolinguistic typology (cf. Kortmann & Szmrecsanyi 2009, Maitz &
Németh 2014, and Trudgill 2011). In the case of Unserdeutsch, the following
aspects suggest the retaining of structural complexity:
a) the speakers unrestricted access to the lexifier language and their
competence in it,
b) the high prestige of Standard German and the low prestige of
Unserdeutsch caused by a standard language ideology, which is typical
for colonial contexts (cf. Lippi-Green 2012: 235247) and which was
widespread in the social environment of the language community (as seen
in emic language names such as Kaputtene Deutsch ‘broken German’ or
Falsche Deutsch ‘wrong German’),
The creoleness of Unserdeutsch (Rabaul Creole German)
c) the small, socially isolated and therefore closed community with its dense
social networks (cf. Trudgill 2011),
d) the fact that Unserdeutsch served as an emblematic in-group code, an
esoteric language (in the sense of Thurston 1987), and
e) the fact that Unserdeutsch was used in expanded contexts of everyday
Against this background, we hereafter want to pursue the issue of whether
the typological profile of Unserdeutsch actually deviates from the profile of
other creole languages or not. More precisely, we will approach the question
whether Unserdeutsch does in fact show a noticeable structural complexity
in the context of creole languages (cf. the claim of creole grammars being the
world’s simplest grammars, McWhorter 2001).
When it comes to determining the structural creole typicality of
Unserdeutsch, this is naturally only feasible by means of a point of reference.
In this respect, one soon comes across the widely debated question
concerning the existence of structural creole universals. It is not our present
task to give a full reflection of the entire discussion here. In a simplified
manner, the different research positions can basically be differentiated into
two opposing factions: On the one hand, there is the uniformitarian position.
Its advocates strictly reject the idea of structural creole universals, arguing
against the backdrop of the postcolonial conviction that creoles are full-
fledged languages no different from non-creoles (cf., e.g., Mufwene 2000;
DeGraff 2005). Not least, they argue with the fact that until this day no
relevant structural features occurring either in all creole languages or
otherwise only in creole languages could be identified. On the other hand,
there is the exceptionalist position (cf., e.g., Bakker et al. 2011; McWhorter
2000; 2001). Its proponents hold the view that creole languages share certain
typological similarities that set them apart from non-creoles, mainly because
of the particular sociohistorical context of their emergence. Such similarities,
backed by some statistical evidence, are not necessarily seen in specific
grammatical features, but rather in a cluster of co-occurring features or in the
LLM Special Issue 2017
absence of certain fundamental features. The unexceptional (non-)occurrence
of features is not as relevant as statistic evidence concerning the appearance
of certain typological features (cf. especially Bakker et al. 2011).
By addressing the issue of the creole typicality of Unserdeutsch, we are
automatically positioned within an exceptional, or rather distinctional,
framework (cf. Bakker et al. 2011: 35). We thus hold the view that the
particular sociohistorical characteristics of creole languages may justify
certain typological convergences. These relevant characteristics may be seen
in (1) the relatively young age of creoles in comparison to non-creoles, (2)
the crucial role of universals of second language acquisition in the
development of creoles (reflected in L2 simplifications), and, last but not
least, (3) the relatedness of common superstrate or substrate languages.
In this study we examine the occurrence and realisation in Unserdeutsch
of three fundamental typological variables, each on the level of phonology,
inflectional morphology and syntax. The features chosen are those that are
mentioned as typical most frequently in the relevant literature and, if
available, supported by statistical evidence. All of them are fundamental and
generic typological characteristics, since we are not concerned with specific
individual grammatical features or categories.
As far as it is possible and seems reasonable, our main basis of
comparison is the data of the Atlas of Pidgin and Creole Language Structures
(Michaelis et al. 2013). These data depict the currently most reliable picture
of an “average creole”.
Especially in cases where the typological feature
refers to the relationship between the creole and its lexifier, we will compare
the Unserdeutsch data to the system of Standard German. Further, preferably
empirically oriented, literature will be consulted if it seems reasonable in the
interest of a more differentiated interpretation of the findings.
The primary data that have been collected via semi-structured narrative
interviews as part of several field trips to Papua New Guinea and Australia
between 2014 and 2017 (cf. Götze et al. 2017, this issue, and Maitz, König
& Volker 2016) serve as our empirical basis. Only the basilectal part of the
data is considered here, since the creole character of a given variety is
naturally lower towards the acrolectal pole (cf. Maitz 2017). The corpus is
By using the term ‘average creole’ in this paper, we do not intend to add another
theoretical construct to the ongoing debates about shared profiles of creole languages. We
use the term at this point only to refer to a statistical mainstream regarding certain
typological features of creole languages as reflected in the APiCS data.
The creoleness of Unserdeutsch (Rabaul Creole German)
currently under construction so that the whole data have not yet been
collected and prepared in a form appropriate for use in corpus research.
Therefore only the data already transcribed and hence accessible for
systematic analyses could be included here, forcing us to forego quantitative
statements. Although in principle this lowers the resilience of our data, in
reality, however, the data already evaluated and all experience from the field
clearly show that especially basilectal Unserdeutsch shows at most a very
small variation with regard to the typological features examined here. This
variation may, since not explicitly addressed, be disregarded insofar as the
following analysis does not aim at a detailed presentation of grammatical
facts, but merely intends to show typological tendencies. Moreover,
individual deviations from the rule, or their realisations in high frequency
environments, have been excluded, as these are obviously lexically stored
constructions (chunks) that are not rooted in the system of the language.
Finally, we disregard all sorts of individual occurrences, because, as
mentioned before, only recognisable patterns are relevant in this context.
4.1 Phonology
Three articular phonological features are most frequently mentioned and of
course discussed in creolistic universals research (cf. Velupillai 2015: 53
54, McWhorter 2001, Klein 2006, etc.). These are (1) the absence or
depletion of typologically unusual, marked vocals and consonants of the
lexifier language, (2) a relatively small phoneme inventory, at least in
comparison to the lexifier language, and (3) simple syllable structures.
regard to these typological features, it is said (in a somewhat simplified
manner) that creole languages tend to be phonologically less complex than
non-creoles (cf., e.g., McWhorter 2001 and Parkvall 2008). Even though
these claims have been questioned over the last years, with cross-linguistic
What is also frequently named among the central phonological features of creole
languages is the absence of (lexically or grammatically distinctive) tone (cf. McWhorter
2000: 8690; Maurer & APiCS Consortium 2013c). Like all pidgin and creole languages
in the Pacific region, Unserdeutsch does not have distinctive tones. This is hardly
surprising, since Tok Pisin as well as Standard German are not tone languages.
LLM Special Issue 2017
(counter) evidence leading to an increasingly nuanced view (cf., e.g., Klein
2006 and Velupillai 2015), one can clearly state that all three claims apply to
Unserdeutsch (for a more detailed description cf. Maitz & Volker
4.1.1 Phoneme inventory
The issue of phoneme inventory and the absence of marked phonemes shall
be treated jointly, as these aspects are related to some extent.
Taken as a whole, it can be seen that the phonology of Unserdeutsch is
largely based on its substrate language, Tok Pisin (cf. Laycock 1985). The
most obvious aspect is the vowel system, which essentially corresponds to
the system of Tok Pisin. Basilectal Unserdeutsch displays a five-unit vocal
system similar to Tok Pisin, consisting of the five short vowels /i/, /ɛ/, /a/, /u/,
/o/. With the exception of /ɛ/, these are qualitatively identical with the vowels
of Tok Pisin. As can be seen by this, there is a clear tendency for the long
vowels of the lexifier language to be shortened in basilectal Unserdeutsch, as
in (1) (for a phonology of German cf. Wiese 1996). Similarly, Standard
German umlaut vowels, which are regarded as typologically highly marked,
tend to be represented by their delabialised equivalents, as in (2). The
Standard German reduction vowels [ə] and [ɐ] in unstressed syllables are also
absent and replaced usually by [ɛ], as in (3).
(1) a. SG groß [gro:s] ‘big’ UD [gros]
b. SG lieben [li:bən] ‘love’ UD [libɛn]
c. SG stehlen [ʃte:lən] ‘to steal’ UD [ʃtɛlɛn]
(2) a. SG Hügel [hy:gəl] ‘hill’ UD [higɛl]
b. SG Frühstück [fry:ʃk] ‘breakfast’ UD [friʃtik]
c. SG hören [hø:ʁən] ‘hear’ UD [hɛrɛn]
(3) a. SG aber [a:bɐ] ‘but’ UD [abɛ]
b. SG alle [alə] ‘all’ UD [alɛ]
c. SG Teller [tɛlɐ] ‘plate’ UD [tɛlɛr]
The creoleness of Unserdeutsch (Rabaul Creole German)
This five-unit vowel system is significantly smaller than the one previoiusly
described by Volker (1982). The difference relates to Volker assuming two
rows of short vowels with distinct degrees of opening, thus resulting in nine
vocal phonemes. However, in the light of the data, his description seems to
be untenable. Admittedly, vowels may be realised with a differing degree of
opening, but this variation does not seem to be phonologically distinctive;
the variants appear in the same distribution. The correction in this matter is
of major relevance for our investigation, especially since nine-unit vowel
systems are counted among the most complex and rare vowel systems in
creole languages worldwide (cf. Haspelmath & APiCS Consortium 2013a;
Klein 2006). A five-unit vowel system with three vowel heights, on the other
hand, clearly classifies Unserdeutsch as an average creole language in that
regard (cf. Haspelmath & APiCs Consortium 2013a).
Tendencies similar to the vowel system can be observed in the consonant
system of Unserdeutsch, which is apparently based on the phoneme system
of Tok Pisin. The parallels result from the depletion of the marked consonants
of the lexifier language and a reduced phoneme inventory in contrast to the
lexifier language (for a more detailed account cf. Maitz & Volker
forthcoming). In Unserdeutsch, only three consonant phonemes of the
lexifier language are systematically preserved that are not part of the core
phoneme inventory of Tok Pisin: the unvoiced fricatives /f/ and /ʃ/, as well
as the affricate /tʃ/, the latter only playing a marginal role in Standard German
as well as in Unserdeutsch. All other consonant phonemes and allophones,
which are considered to be marked from a cross-linguistic perspective,
present in Standard German, but absent in Tok Pisin, are completely or at
least partially depleted or substituted in Unserdeutsch. This applies to [ç], [χ],
[pf], [ts], [ʀ]/[ʁ]
and [z]:
The correction of Volker’s phonological interpretation is important, not least because his
results have found their way into creole and typological literature (cf. Velupillai 2015: 125
126; Klein 2006) and have been used as evidence against the postulate of the phonological
simplicity of creole languages. For example in Klein’s sample of 23 creole languages (cf.
Klein 2006), the nine-unit vowel system described by Volker (1982) is ranked as the most
complex one, whereas the five-unit vowel system described here would sort Unserdeutsch
into the group of creole languages with the smallest vowel inventory in Klein’s sample.
In Standard German, three main free-variant allophones of /r/ are predominantly
distributed on a regional level. There is strong linguistic and extralinguistic evidence for
assuming that the superstrate language of Unserdeutsch was a predominantly Westphalian-
Rhenisch coined spoken Standard German, thus originating from the northwest or central-
west region of Germany (cf. Maitz & Lindenfelser forthcoming).
LLM Special Issue 2017
SG [ç] UD [h] / Ø SG Kirche ‘church’ UD [kirhɛ]
SG [χ] UD [h] / Ø SG lachen ‘to laugh’ UD [lahɛn]
SG [pf] UD [f] SG Pflanzung ‘plantation’ UD [flansuŋ]
SG [ts] UD [s] SG zusammen ‘together’ UD [susamɛn]
SG [ʀ]/[ʁ] UD [r] SG trinken ‘to drink’ UD [triŋkɛn]
SG [z] UD [s] SG diese/dieser/dieses ‘this’ UD [disɛ]
To summarise, Unserdeutsch shows a profile typical for creoles with regard
to the size and type of its phoneme inventory.
4.1.2 Syllable structure
If we look at the complexity of syllable onsets, we see that creole languages
hardly seem to differ from non-creoles, so that most pidgin and creole
languages in the world show complex syllable onsets (cf. Maurer & APiCS
Consortium 2013a). Unserdeutsch belongs to this group, as it allows complex
syllable-initial consonant clusters, like its lexifier and its substrate language.
Only in very rare cases are initial consonant clusters simplified in
Unserdeutsch (as in 4ab). By contrast, the complexity of syllable codas
seems to be an important criterion of differentiation from a creolistic point of
view. It is often argued that creole languages prefer CV structures and thus
open syllables (cf. Velupillai 2015: 54; Kaye & Tosco 2001: 76). This strong
and general claim has been refuted by cross-linguistic evidence over the last
years (cf. Maurer & APiCS Consortium 2013b; Velupillai 2015: 304).
Nevertheless, it can reasonably be concluded that the vast majority of pidgin
and creole languages do not tolerate complex syllable codas like those
appearing in Standard German (cf. Maurer & APiCS Consortium 2013b).
Basilectal Unserdeutsch fits into this description, as it displays a marked
preference for less complex syllable codas, in contrast to its lexifier language,
Standard German, but similar to its substrate language, Tok Pisin. Yet, whilst
in Tok Pisin a major role is given to vocal epentheses (cf. Smith 2008: 203
204), there seem to be no epenthetic vowels in Unserdeutsch. Instead,
Unserdeutsch displays a strong tendency to delete syllable-final consonants,
leading to a weakening of syllable codas and a clear tendency towards CVC
and CV structures (as in 4). With regard to cluster complexity, Unserdeutsch
The creoleness of Unserdeutsch (Rabaul Creole German)
clearly falls into the category of less complex pidgin and creole languages
(cf. Maurer & APiCS Consortium 2013b).
(4) a. SG bist [bɪst] / [bɪs] ‘(you) are’ UD [bis]
b. SG nicht [nɪçt] / [nɪç] ‘not’ UD [ni]
c. SG Tag [taχ] ‘day’ UD [ta]
d. SG vielleicht [filae
̯çt] maybe UD [filae
e. SG Abend [a:bənt] ‘evening’ UD [abɛn]
4.2 Inflectional morphological features
Morphological simplicity is considered a structural main characteristic of
creole languages (cf. Crowley 2008: 75). Although the amount of simplicity
remains a controversial issue, as it varies considerably from case to case,
there is at least agreement that earlier postulates of a complete absence of
inflectional morphological substance in creoles are untenable (cf. Bartens
2013: 92). Nevertheless, in contrast to its particular lexifier language, the
tendency of creoles towards morphological simplicity is apparent. This
simplicity manifests itself in the absence of complexifying (redundant or
irregular) categories and markers on the one hand, as well as in a general
preference for transparent, linear structures on the other hand. The following
section is limited to the consideration of inflectional morphology. Three main
features that are backed empirically and frequently put forward in the
relevant literature shall now be analysed with regard to their occurrence in
Unserdeutsch in contrast to Standard German: (1) inflectional poverty, (2)
the absence of marked grammatical categories, and (3) minimal allomorphy.
4.2.1 Inflectional poverty
Creole languages tend to be isolating languages from a typological
perspective (cf. Lefebvre 2004: 217), hence preferring analytical ways to
encode grammatical information (if required at all) over synthetic strategies.
However, the obvious conclusion that the low level of syntheticity might be
balanced out by a high level of analyticity, has been proven wrong. Creole
languages are by no means necessarily more analytic than non-creoles and
are only significantly less synthetic (cf. Siegel, Szmrecsanyi & Kortmann
LLM Special Issue 2017
2014). As statistical evidence from recent typological research has shown
repeatedly, these two indices are not mutually exclusive. In actual fact, the
correlation between them may even be positive (cf. Kortmann &
Szmrecsanyi 2009 and Maitz & Németh 2014). A low level of syntheticity,
i.e. inflectional poverty in the narrow sense, is one of the creole features cited
most prominently. With this in mind, Unserdeutsch is expected to display
considerably fewer word-internal grammatical markers than Standard
A series of categories marked synthetically in Standard German is either
marked analytically in Unserdeutsch or not marked at all, i.e. generally
a. Absence of synthetic markers on nouns
In Standard German, nominal plurality is marked by means of suffixation
and, partially, an additional change of the stem vowel (umlaut). In
Unserdeutsch, however, plurality is generally marked by analytic means
through adding the prenominal plural word alle, as described earlier by
Volker (1982: 31). This is obviously based on the way of marking plurality
in Tok Pisin (cf. Tok Pisin ol haus ‘houses’):
(5) er mal-en
alle plan fi bau-en alle haus.
3SG.M draw-V PL plan for build-V PL house
SG: ‘Er hat die Pläne für den Bau der Häuser gezeichnet.’
EN: ‘He drew the blueprints for the construction of the houses.’
The syntactic position of alle may alternatively be filled by another word
indicating plurality, such as an indefinite pronoun or an adjective, as in (6):
Verbs in basilectal Unserdeutsch systematically end on -en, except for a small group of
high-frequency verbs (see below) and they are not inflected in person and number.
Furthermore, these invariant verb forms are temporally unspecified to a large extent, as
they may represent past, present and future actions likewise. On these grounds, {-en} seems
to be a suffix indicating word class, i.e. a verb marker. Another reason for this analysis is
the fact that {-en} is also attached to borrowed verb stems, and only to verb stems, as in
ringen jeman ‘to call somebody’ (SG jemanden anrufen) or riden fahrrad ‘to ride a bike’
(SG Fahrrad fahren).
The creoleness of Unserdeutsch (Rabaul Creole German)
(6) a. einige von mei cousine war da.
some of 1SG.POSS cousin.F COP.PST there
SG: ‘Einige meiner Cousinen waren da.’
EN: ‘Some of my cousins were there.’
b. du hat drei monat.
2SG have three month
SG: ‘Man hatte drei Monate (Zeit).’
EN: ‘You had three months.’
Remnants of mostly lexeme-bound synthetic forms, partially linked with an
additional analytic marker, are rare. They may be interpreted as irregular
plural forms that are stored holistically and have only remained because of
their high frequency, as in (7).
(7) a. zeit fi die jetz su hat kind-er.
time for 3PL now to have child-PL
SG: ‘Es ist jetzt Zeit für sie, um Kinder zu kriegen.’
EN: ‘It is now time for them to have children.’
b. alle frau-en muss näh-en alle kleider.
PL woman-PL must sew-V PL clothes
SG: ‘Die Frauen mussten Kleidung nähen.’
EN: ‘The women had to sew clothes.’
Such remnants are therefore not relevant at this point, as it is obvious that the
productive and unmarked way of plural marking in Unserdeutsch is analytic.
The use of plural markers, which is rare in European languages, is considered
common in creoles (cf. Haspelmath & APiCS Consortium 2013b).
While the marking of number only happens outside the word boundary,
case inflection is, apart from single, holistically stored constructions,
completely absent; the category as such is omitted (see below). Therefore, it
can be noted that in basilectal Unserdeutsch the rich synthetic noun inflection
of its lexifier language has, apart from single occasional, fossilized forms,
largely been depleted.
LLM Special Issue 2017
b. Absence of synthetic markers on verbs
In Standard German, person and number are generally marked directly on the
verb by means of suffixation and, partially, additional stem inflection
(umlaut), regardless of the presence of further analytic markers. In basilectal
Unserdeutsch, the synthetic marker is generally omitted, so that the verb
remains uninflected. The marking of person and number is shifted to the
subject, usually a pronoun or a noun phrase, as in (8).
(8) ich sag-en sie: du wart-en fi wenn du hat
1SG say-V 3SG.F 2SG wait-V for when 2SG AUX.PST
de kin son ge-krie!
ART.DEF child already PTCP-get
SG: ‘Ich sagte ihr: Warte, bis du das Kind bekommen hast!’
EN: ‘I told her: Wait until you gave birth to the child!
The only exception is the auxiliary and copula sein ‘to be’, which is usually
inflected (cf. Volker 1982: 36); however, among some basilectal speakers,
even this highest-frequency verb remains uninflected with bis (obviously
derived from the Standard German 2SG form bist) used across all persons and
numbers, as in (9).
(9) a. mama du hör-en i bis deutsch am spreh-en!
mum 2SG hear-V 1SG COP German PROG speak-V
SG: ‘Mama, hörst du, ich spreche Deutsch!’
EN: ‘Mum, do you hear me, I am speaking German!’
b. die bis von vunapope.
3PL COP from Vunapope
SG: ‘Sie sind von Vunapope.’
EN: ‘They are from Vunapope.’
Hence, the verbal paradigm of Unserdeutsch consists of only a single
invariant form. This basic form is normally identical to the Standard German
infinitive form; the only exception to this rule are high- and very high-
frequency verbs such as geht ‘go’, komm(t) ‘come’, muss ‘must’, will ‘want’,
The creoleness of Unserdeutsch (Rabaul Creole German)
weiss ‘know’, and hat ‘have’, as in (10), which are formed based on the
Standard German third person singular indicative present form. In respect to
these general and exception rules, Unserdeutsch exactly matches the pattern
of Portuguese and Spanish based creoles (cf. Bartens 2013: 100).
Regarding the marking of infinite participles, Unserdeutsch does not have
a present participle. The past participle, which is formed by means of a
variable circumfix (depending on the inflectional class) and partially through
an additional change of the stem vowel (ablaut) in Standard German, is
present in Unserdeutsch. The simplified formation rule in Unserdeutsch is
[ge- + basic form], the result of a reanalysis: gemahen ‘made’ (SG gemacht),
gekrie ‘gotten’ (SG gekriegt), geligen ‘lied’ (SG gelogen), etc. The formation
of the participle does in large parts correspond to general tendencies in Black
Namibian German (Küchendeutsch) and other L2 varieties of German (cf.
Deumert 2003: 584587). In Unserdeutsch, the morphological structure of
the participle can be described as the basic verb form plus the prefix ge-,
whereby the basic form consists of the verb stem and the verb marker -en.
The absence of the category mood and the loss of the preterite contribute to
the far-reaching loss of syntheticity in Unserdeutsch, as the synthetic mood
and preterite markers of the lexifier language are omitted entirely (see
c. Absence of synthetic markers on adjectives
Adjectival comparison in Standard German is generally achieved by adding
a suffix, whereby the stem vowel may change additionally in some cases
(umlaut). In Unserdeutsch, the comparative is usually formed in an analytic
way, except for high-frequent, lexicalised forms:
(10) a. wi hat ferti mehr snell.
1PL have finished more fast
SG: ‘Wir werden schneller fertig.’
EN: ‘We get finished more quickly.’
An interpretation of the suffix -en as an infinitive suffix is out of the question here, since
the verb in Unserdeutsch does not inflect for person and number (see footnote 5).
LLM Special Issue 2017
b. er wid arbeit mehr stark fi uns.
3SG.M AUX.FUT work more hard for 1PL.ACC
SG: ‘Er wird härter für uns arbeiten.’
EN: ‘He will work harder for us.’
The formation of the superlative appears less straightforward. While
especially high frequent adjectives follow the Standard German synthetic
pattern in this respect (de älteste brudä ‘the oldest brother’, sein jüngste sohn
‘his youngest son’), the synthetic form tends to be avoided with other
adjectives. It is mostly replaced by a periphrastic construction with the
intensifier particle ganz ‘very’, functionally resembling the elative: Peter
laufen ganz schnell (direct translation from the English stimulus Peter runs
Compared to its lexifier language, the synthetic conjugation of the
adjective is also largely omitted or simplified in Unserdeutsch. In accordance
with the clear tendency of creole languages to cope without agreement within
the nominal phrase (cf. Maurer & APiCS Consortium 2013d), Unserdeutsch
adjectives indicate neither gender, number nor case. Only one adjectival
inflectional suffix is retained in Unserdeutsch: {-e}. It has been reanalysed
as a uniform and invariant attributive marker, however. In attributive use it
obligatorily accompanies the stem of the adjective (such as the suffix {-pela}
in Tok Pisin, cf. Volker 1982: 41), whereas it is omitted in predicative and
adverbial use:
(11) du ni fihl-en kalt, du hat ein gut-e leben;
2SG NEG feel-V cold 2SG have ART.INDF good-ATTR life
heiß-e zeit du kann immer geht in salzwasser.
hot-ATTR period 2SG can always go in sea
SG: Man friert nicht, man hat ein gutes Leben; wenn es heiß ist, kann
man immer ins Meer gehen.’
EN: ‘You don’t feel cold, you have a good life; when it is cold, you
can always go into the ocean.
The creoleness of Unserdeutsch (Rabaul Creole German)
d. Absence of synthetic markers on pronouns
The forms of the indefinite, demonstrative and possessive pronouns, which
are inflected in Standard German, always remain uninflected in basilectal
Unserdeutsch. Here, either the Standard German form without an ending is
used across the entire paradigm (alle mein sahen ‘(all) my stuff’; dein frau
‘your wife’; alle sein kinder ‘(all) his children’, ganz viel cousine ‘quite a lot
of cousins’) or with the pronouns ihre ‘her’ and diese ‘this’, the Standard
German form with the suffix {-e} (das war ihre leben ‘that was her life’;
heiraten diese mensch ‘marry this man’). As the form is stable for every
single pronoun and across speakers to the greatest extent, there is no doubt
that the pronominal ending -e cannot be a suffix in Unserdeutsch.
In conclusion, it is apparent that the syntheticity of Unserdeutsch is
drastically reduced in comparison with Standard German: With only a few
exceptions, there are no synthetic markers across word classes. Since the
presence of some individual synthetic elements is not at all uncommon even
for creole languages (cf. Velupillai 2015: 328329), it can definitely be stated
that Unserdeutsch fits the structural typological design of creole languages
with regard to the criterion of inflectional poverty.
4.2.2 Absence of marked grammatical categories
The reason for the inflectional poverty of most creole languages cannot be
ascribed only to the increased use of analytic means, but also to the complete
omission of the grammatical categories of its lexifier language. This relates
in particular to such categories considered as marked from a cross-linguistic
perspective, especially categories ‘conditioned by syntax and devoid of
‘meaning’’ (McWhorter 2014: 95), such as case and gender.
In basilectal Unserdeutsch, a number of these categories of the lexifier
language are dropped. Only in one subsystem of its language system can an
increase in categorical complexity be observed: in the system of personal
pronouns, which shows reflections of the highly complex pronominal system
of Tok Pisin. In general, the pronominal system of Unserdeutsch is based on
the system of Standard German. However, with regard to personal pronouns,
the system is extended by an exclusive-inclusive distinction of the first
person plural pronoun, which can be traced back to substrate transfer from
LLM Special Issue 2017
Tok Pisin (cf. Verhaar 1995: 354355 and Mühlhäusler 1985: 343). The
pronoun uns expresses the inclusive function, as in (12), while wi represents
the exclusive meaning, as in (13).
(12) uns beide am spreh-en so schön, uns zwei
1PL.INCL both PROG talk-V so lovely 1PL.INCL both
am spreh-en unserdeutsch.
PROG talk-V Unserdeutsch
SG: ‘Wir beide unterhalten uns so schön, wir beide sprechen
EN: ‘The two of us are talking so lovely, we both speak
(13) wi tanz-en wenn wi hat musik; wi alle
1PL.EXCL dance-V when 1PL.EXCL have music 1PL.EXCL all
tanz-en, sauf-en, dann nächst-e ta wi kaputt.
dance-V tipple-V then next-ATTR day 1PL.EXCL exhausted
SG: Wir tanzen, wenn wir Musik haben; wir tanzen alle, wir saufen
und am nächsten Tag sind wir dann erschöpft.’
EN: We dance, when we have music; we all dance, drink (tipple), and
the next day we are all exhausted.
The description of Volker (1989a: 3132) concerning this matter suggests a
systematically distinct use of both forms. The analysis of the recent data,
however, indicates a rather unsystematic and inconsistent use of both forms,
which might be attributable to an erosion of the system. A relatively
consistent use of the inclusive pronoun uns seems to be restricted to the use
in dual contexts, as in (12).
Apart from this phenomenon, basilectal Unserdeutsch displays a clear
and strong tendency towards the omission of the grammatical categories of
its lexifier language. Beyond the complete loss of the Standard German
inflectional classes, this can be seen particularly in the absence of several
further grammatical distinctions of Standard German.
The creoleness of Unserdeutsch (Rabaul Creole German)
a. Absence of gender marking
For a creole language, the presence of the gender category, like all kinds of
nominal classifier systems, would be highly unusual (cf. Holm 2000: 216).
In line with this, the tripartite gender system of Standard German is
completely eliminated in Unserdeutsch. The Standard German definite
articles der (masculine), die (feminine), and das (neuter) are therefore, as in
English, merged into a single standard article de, as in (14):
(14) a. whether de mensch lieb-en de frau.
whether ART.DEF man love-V ART.DEF woman
SG: ‘ob der Mann die Frau liebt.’
EN: ‘whether the man loves the woman.’
b. de tür war weg von de klein-e haus.
ART.DEF door was away from ART.DEF small-ATTR house
SG: ‘Die Tür der Toilette hat gefehlt.’
EN: ‘The toilet door was missing.’
With regard to the indefinite article and the pronominal system, Unserdeutsch
has likewise retained only one gender-invariant form of the Standard German
paradigm. This form may be identical with the Standard German basic form,
as with the indefinite article and most pronouns, or with a suffixed form, as
mentioned above. Example (15) shows how not only is grammatical gender
absent in Unserdeutsch, but also the congruency of natural gender between
articles and pronouns and their antecedents, marked in Standard German, is
(15) ein frau un ihre herrgemahl.
ART.INDF woman and 3SG.F.POSS husband
SG: ‘Eine Frau und ihr Ehemann …’
EN: ‘A woman and her husband …’
LLM Special Issue 2017
b. Absence of case marking
The four-part case system of Standard German is basically not retained in
Unserdeutsch (cf. 16). Apart from some few lexicalised phrases (guten aben
‘good evening’), sporadic remnants of the case system are, similar to other
contact varieties of German (cf. Boas 2009: 204210), present only within
the paradigm of personal pronouns, as in (17).
(16) de schwester wokabaut herum mit ein groß-e
ART.DEF sister walk around with ART.INDF big-ATTR
kanda in ihre hand.
cane in 3SG.F.POSS hand
SG: ‘Die Missionsschwestern sind mit einem großen Bambusstock in
der Hand herumgegangen.’
EN: ‘The missionary sisters walked around with a big cane in their
(17) die hat ge-mah-en ihm ein chief.
SG: Sie haben ihn zum Anführer gemacht.
EN: They appointed (made) him chief.
However, even for personal pronouns, case differentiation does not exist to a
great extent at the basilectal level of Unserdeutsch, as in (18).
(18) wenn du zahl-en i de zahlung du geb-en
if 2SG pay-V 1SG ART.DEF payment 2SG give-V
de weiße, orait, i arbeit fi du.
ART.DEF whites all_right 1SG work for 2SG
SG: Wenn du mir den (gleichen) Lohn bezahlst, den du den Weißen
gibst, in Ordnung, dann arbeite ich für dich.’
EN: ‘When you give me the (same) payment you give the whites, all
right, then I’ll work for you.
The creoleness of Unserdeutsch (Rabaul Creole German)
c. Reduced tense system
In the verbal phrase, the marking of categories is a bit more sophisticated.
Most relevant here is the TMA system and grammatical voice. Firstly, the
complex Standard German tense system is substantially reduced in
Unserdeutsch. Secondly, the category tense generally seems to be less
grammaticalised, since it is obvious that the marking of tense is optional in
the basilect (cf. Volker 1982: 43). The temporally unspecified basic form
does formally correspond to the Standard German infinitive (see above).
Usually, these basic forms are used, and temporal meaning is solely
transferred to the verb by context. The occurrence of preterite forms is
restricted to a small, closed class of high-frequency verbs (modal verbs and
auxiliaries): war ‘was’, wollte ‘wanted’, musste ‘had to’, konnte ‘could’,
wusste ‘knew’. There is, however, an analytical past tense form in basilectal
Unserdeutsch, formed by the rule [hat + past participle], as in (19):
(19) meine vatä hat ge-sterb-en neunzehnunseksi.
1SG.POSS father AUX.PST PTCP-die-V nineteen_and_sixty
SG: ‘Mein Vater ist Neunzehnhundertsechzig gestorben.’
EN: ‘My father died nineteen-sixty.
Remnants of the Standard German past perfect tense with [war + past
participle] are apparently restricted to a small, closed class of main verbs such
as war gekommen ‘had come’, war geboren ‘was born’, war gestorben ‘had
died’. However, past tense meaning is typically either not indicated at all (as
in 20), or it is marked by lexical means, e.g., by using temporal adverbs, as
in (21):
(20) dann i geht zurück arbeit, dann i heirat-en, hat
then 1SG go back work then 1SG marry-V have
ein tochter …
ART.INDF daughter
SG: ‘Danach bin ich wieder in die Arbeit gegangen, habe dann
geheiratet, hatte eine Tochter …’
EN: ‘Then I went back to work, then I married, had a daughter …’
LLM Special Issue 2017
(21) sie flieg heraus gestern abend.
3SG.F fly out yesterday evening
SG: ‘Sie ist gestern Abend weggeflogen.’
EN: ‘She flew away yesterday evening.
All in all, it can be noted that the past tense forms of Standard German are
present in Unserdeutsch only either in a weakly grammaticalised manner or
in a few remnants, which can be understood as holistically stored and
partially reanalysed constructions.
The two future tenses of Standard German are merged into one single
form in Unserdeutsch, which, in being formed by the rule [wid + basic form],
is based on the pattern of the Standard German Future I, as in (22). Similar
to Standard German, the marking of future tense is not obligatory, as in (23):
(22) diese jahr die wid hat ni ein tanz.
DEM year 3PL AUX.FUT have NEG ART.INDF dance
SG: ‘Dieses Jahr werden sie keinen Tanz veranstalten.’
EN: ‘This year they won’t have a dance.’
(23) morgen sie flie su kokopo.
tomorrow 3SG.F fly to Kokopo
SG:Morgen fliegt sie nach Kokopo.’
EN: ‘Tomorrow she flies to Kokopo.
d. Reduced mood system
The verbal paradigm of Unserdeutsch has no imperative. The verb forms of
imperative clauses are formally identical to the verb forms of declarative
clauses. The same applies to word order, which is identical in imperative and
declarative clauses, in contrast to Standard German, as in (24):
(24) du ni denk-en dass i war ni angs!
2SG NEG think-V that 1SG COP.PST NEG afraid
SG: ‘Denk nicht, dass ich nicht Angst hatte!’
EN: ‘Don’t think I wasn’t afraid!’
The creoleness of Unserdeutsch (Rabaul Creole German)
There is nothing left of the Standard German subjunctive mood in
Unserdeutsch. The synthetic forms (present subjunctive and past subjunctive)
do not occur in basilectal Unserdeutsch, apart from single, separate
lexicalised constructions. The Standard German periphrasis with würde
‘would’ is not used either. The only grammaticalised way to indicate the
irrealis exists in the use of the wid-construction, which can appear in temporal
(see above) as well as in aspectual (see below) and modal function, as in (25).
This polyfunctional use of irrealis markers is considered typical for creole
languages (cf. Holm & Patrick 2007, Feature 6). In Unserdeutsch, the
construction is optional in its modal meaning, too.
(25) du wid sa was?
2SG AUX.IRR say what
SG: ‘Was würdest du sagen?’
EN: ‘What would you say?
e. Grammaticalised aspect system
Almost all creole languages indicate verbal aspect (cf. Maurer & APiCS
Consortium 2013e). The assumption of Bickerton (1981) that creole
languages are limited to one single aspect marker (indicating progressive or
a related kind of aspect), has turned out wrong from an empirical perspective
since then, as many creole languages (additionally) indicate further kinds of
aspect (cf. Velupillai 2015: 398), especially habitual and perfective aspect
(cf. Bartens 2013: 101ff.). With regard to the so-called am-progressive in
Standard German, one can assume the existence of a grammaticalised aspect
in the spoken domain (cf. Gárgyán 2013: 196), even though there is still a
lack of agreement regarding the classification of this phenomenon in the
grammar books. From a typological point of view, Standard German is
considered a non-aspect language by tradition (cf. Dahl & Velupillai 2013);
with leading grammars (still) avoiding a description of German as an aspect
language (cf. Gárgyán 2013: 151156).
The marking of grammatical aspect in Unserdeutsch is obligatory to a
large extent. Assuming that Standard German may not (yet) be described as
an aspect language, this means the formation of a new category and thus
grammatical complexification. At first glance, this seems to contradict the
LLM Special Issue 2017
creole feature formulated above, as it conversely postulates the absence of
certain categories. We should, however, consider, that cross-linguistically the
presence of the category aspect is regarded as unmarked, especially with
respect to creole languages. By way of exception, Unserdeutsch has in fact
come closer to the pattern of an ‘average creole’ by adding, or at least
expanding a category in this case.
Unserdeutsch has two different constructions to indicate aspect
grammatically. The first construction does formally correspond to the
Standard German am-progressive; it indicates either progressive, as in (26),
or habitual meaning, as in (27):
(26) de ganz-e tach sein mun is so voll
ART.DEF whole-ATTR day 3SG.M.POSS mouth COP.3SG so full
wenn er is am aufpass-en alle swarz-e labour.
when 3SG.M COP.3SG PROG take_care-V PL black-ATTR labour
SG: Er hatte den ganzen Tag den Mund voll [mit Betelnuss],
wenn (während) er auf die schwarzen Arbeiter aufpasste.’
EN: ‘Every day his mouth was full [with betelnut] when he
was looking after the black labourers.
(27) jeden tach fi drei wohe i war am spreh-en
every day for three week 1SG COP.PST HAB talk-V
mit sie.
with 3SG.F
SG: ‘Drei Wochen lang habe ich jeden Tag mit ihr gesprochen.’
EN: For three weeks I was talking to her every day.
In these constructions, the copula sein ‘to be’ may be dropped, as shown in
(28) and (12); see also section 4.3.3.
(28) i weiss ni whether de zwei brudä am leb-en
1SG know NEG whether ART.DEF two brother PROG live-V
zusammen or die beide zank-en …
together or 3PL both argue-V
The creoleness of Unserdeutsch (Rabaul Creole German)
SG: ‘I weiß nicht, ob die beiden Brüder zusammenleben oder ob sie
gestritten haben …’
EN: ‘I don’t know whether the two brothers are living together or they
had a quarrel …’
The consistent marking of progressive and habitual aspect that can be
observed in the data, is not atypical for creole languages (cf. Bartens 2013:
The second construction is the wid-construction mentioned above, which
widely corresponds to the past habitual use of would in English, as in (29).
(29) sie wid bleib bis sonne will geht unten dann sie geht
3SG.F AUX.HAB stay till sun will go down then 3SG.F go
zurück zuhause un koh-en.
back home and cook-V
SG: ‘Sie ist (jeden Tag) bis zur Dämmerung geblieben, dann ist sie
nach Hause zurückgegangen und hat gekocht.’
EN: ‘She would stay till dawn, then she went back home and cooked.’
Tense, mood and aspect markers in Unserdeutsch are placed preverbally in
adjacent position. Around 80 percent of the pidgin and creole languages in
the world follow this pattern (cf. Maurer & APiCS Consortium 2013f).
f. Marginal voice marking
Typically, there are no overtly marked passive constructions in creole
languages (cf. Crowley 2008: 82). In the Unserdeutsch data, passive
constructions appear extremely rarely, and if they do, it is mainly in the more
elaborated varieties beyond the basilect. Hence, passive voice shows a very
low degree of grammaticalisation in Unserdeutsch. In contrast to its lexifier
language, the data displays only one type of construction, consisting of the
inflected auxiliary sein ‘to be’ together with the past participle of the main
verb, as in (30). An agent role may be attached optionally by using the
preposition von.
LLM Special Issue 2017
(30) a. die war ge-lern-t wie zu koh-en.
DEM.3PL AUX.PST.PASS PTCP-learn-PTCP how to cook-V
SG: ‘Ihnen wurde beigebracht, wie man kocht.’
EN: ‘They were teached how to cook.
b. vor fünfzehn jahr-e ein buch war
before fifteen year-PL ART.INDF book AUX.PST.PASS
ge-schrieb-en von mein mama-s vater.
PTCP-write\PST-V by 1SG.POSS mother-GEN father
SG: ‘Vor fünfzehn Jahren wurde vom Vater meiner Mutter ein
Buch geschrieben.’
EN: ‘Fifteen years ago, a book was written by my mother’s
This passive construction shows great similarity to the passive voice in
English. It thus may be traced back to secondary adstrate influence,
explaining its very low degree of grammaticalisation in Unserdeutsch.
Altogether, it can be noted that Unserdeutsch matches the typological
mainstream of creole languages in respect of its category inventory as well
as its elaborateness compared to its lexifier language. A number of Standard
German categories that are marked from a cross-linguistic perspective have
either been omitted in Unserdeutsch or only play a marginal role, i.e. are
marked optionally. Even the categories that are not dropped as a whole do
not distort the overall picture, as ‘weakly obligatory’ inflectional categories
are considered typical for creoles as well (cf. McWhorter 1998: 792).
4.3.2 Minimal allomorphy
A very small amount of allomorphy, i.e. the broad absence of morphological
irregularity and suppletion, is considered to be typical for creole languages
(cf. Crowley 2008: 77; Bartens 2013: 92). Such a tendency towards an
increased transparency compared to its lexifier language (cf. Leufkens 2013)
can similarly be observed in the case of Unserdeutsch.
Regarding nominal inflection, the loss of the complex Standard German
plural allormophy with its lexeme-dependent nine different ways of plural
marking, including Ø (cf. Werner 1969: 93) is especially striking (see
The creoleness of Unserdeutsch (Rabaul Creole German)
above). Apart from a few remnant forms of high-frequent lexemes and in
mesolectal or acrolectal varieties, the system of Unserdeutsch only shows
one unified, analytic way of plural marking by means of the prenominal
plural word alle ‘all’.
In the area of verbal inflection, the dropping of verbal categories as well
together with the transfer of grammatical information to analytic markers
(see above) leads to the almost entire loss of stem alternation (ablaut, umlaut).
In participles such as gesprehen (SG gesprochen ‘spoken’), gegeht (SG
gegangen ‘gone’), and gesterben (SG gestorben ‘died’), the regularisation of
strong and irregular Standard German verbs can be seen. Only the verb sein
‘to be’ displays suppletion.
Since the comparison of the adjective tends to be done analytically in
Unserdeutsch, the vowel alternations of the synthetic Standard German forms
has been lost. However, the suppletive forms of some high-frequency
adjectives have been retained: gut ‘good ’, besser ‘better’, beste ‘best’, viel
‘much’, mehr ‘more’, meiste ‘most’.
Another tendency towards the reduction of allomorphy can be observed
in word formation: The Standard German umlaut, sometimes evoked by
derivation, is not applied in such cases in Unserdeutsch: SG Brüderchen
‘brother (diminutive form)’ > UD bruderhen; SG jüdisch ‘Jewish’ > UD
judisch. There is a clear trend in Unserdeutsch to eliminate or regularise
irregular and otherwise intransparent forms to unify paradigms.
In the light of what has been said about the morphosyntactic
characteristics of Unserdeutsch, we can summarise by saying that the
inflectional morphological profile of Unserdeutsch is marked by the presence
of features and trends that have been identified as typically creole structural
traits in the literature.
4.3 Syntactic features
On the syntactic level, as well, creole languages are said to display reduced
overt complexity compared to their lexifier language by tending towards
regularisation. The result, a comparatively greater amount of structural
homogeneity (cf. Bakker 2008: 140), is represented by an affinity with fixed
SVO word order and the adjacency of verbal elements. This trend towards
structural homogeneity merely refers to the basilectal end of a creole
LLM Special Issue 2017
language, as there is naturally a considerable amount of variation across the
creole continuum, especially on the syntactic level. Another relevant aspect
can be seen in a tendency towards the dropping of function words, i.e. the
loss of purely grammatical-functional elements “devoid of content”, such as
at the level of inflectional morphology, favouring juxtaposition instead. The
manifestation of these three features in Unserdeutsch shall be examined
4.3.1 Fixed SVO word order
The vast majority of creole languages follow a fixed SVO word order (cf.
Huber & APiCS Consortium 2013; Velupillai 2015: 438). The reason for this
may be partly due to the fact that the majority of substrate and superstrate
languages prefer this syntactic pattern as well (cf. Muysken 1988: 290).
Through the renunciation of syntactic permutations in creole languages, the
formal marking of sentence types usually does not apply. This applies also to
the Standard German distinction between main and subordinate clauses in the
surface structure.
In Standard German, SVO word order is restricted to unmarked
declarative main clauses and unintroduced subordinate clauses, whereas even
here only verb-second is obligatory, and the preverbal position may be filled
by other constituents than the subject. Standard German can therefore not be
categorised as a purely SVO language (cf. Roelcke 2011: 5760). This is
why, for example, the World Atlas of Language Structures does not ascribe
a dominant word order to German (cf. Dryer 2013). In Unserdeutsch,
however, SVO order is obligatory, independent of the sentence type. The
fixed SVO word order in Unserdeutsch corresponds to the typological
positioning of its substrate language Tok Pisin, which is described as
exclusively SVO (cf. Michaelis et al. 2013: 3).
Regarding the declarative main clause, SVO word order is firmly
established in Unserdeutsch to the point that it persists when topicalising an
adjunct (as in 31). This is similar to the syntactic pattern of English, but in
sharp contrast to Standard German, which does not allow verbs in the third
position of a sentence.
The creoleness of Unserdeutsch (Rabaul Creole German)
(31) a. einige mal sie arbeit in garten.
few times 3SG.F work in garden
SG: ‘Einige Male hat sie im Garten gearbeitet.’
EN: ‘A few times she has worked in the garden.
b. wenn du will, du kann geht.
If 2SG want 2SG can go
SG: ‘Wenn man wollte, konnte man gehen.’
EN: ‘If you wanted (to go), you could go.
Imperative sentences in Unserdeutsch usually retain the SVO surface
structure, contrary to both the Standard German and the English verb-first
pro-drop construction, as in (32). This word order aligns with the common
imperative pattern in Tok Pisin.
(32) a. du komm sitz-en in mein office!
2SG come sit-V in 1SG.POSS office
SG: ‘Komm, setz dich in mein Büro!’
EN: ‘Come, have a seat in my office!’
b. du wart-en, i frag-en [Name] ers!
2SG wait-V 1SG ask-V PN first
SG: ‘Warte, ich frage erst [Name]!’
EN: ‘Wait, I ask [name] first!
Interrogative sentences are formed in a less uniform way. Polar questions,
showing VSO word order in Standard German, remain in unmarked SVO
word order in Unserdeutsch, as in (33):
(33) a. du hat schon ge-spreh-en zu [Name]?
2SG have already PTCP-speak-V to PN
SG: Hast du schon mit [Name] gesprochen?’
EN: ‘Have you already spoken to [name]?’
b. du hat ge-hör-en von [Name]?
2SG have PTCP-hear-V of PN
LLM Special Issue 2017
SG: ‘Hast du von [Name] gehört?
EN: ‘Have you heard about [name]?’
For WH-questions, there are two possibilities in basilectal Unserdeutsch:
basically, one with and one without wh-movement. The latter, with the
interrogative in clause-final position, seems to be restricted to speakers at the
very basilectal end of Unserdeutsch. As metalinguistic comments show, this
construction is regarded as a salient feature of basilectal Unserdeutsch with
the emic name falsche Deutsch ‘wrong German’, thus evaluated as ‘bad
German’ among the speakers. Hence, this word order can be considered as
stigmatised, and apparently, it is consciously avoided. This clause type
follows the SVO principle, as in (34).
(34) a. du wid geht wo?
2SG AUX.IRR go where
SG: ‘Wohin willst (würdest) du gehen?’
EN: ‘Where would you go?’
b. i hat ge-mah-en was?
1SG have PTCP-do-V what
SG: Was habe ich gemacht?
EN: ‘What have I done?
The second possibility to form WH-questions occurs more frequently and
corresponds to the Standard German pattern: the interrogative is moved to
the clause-initial position (wh-movement). In contrast to Standard German,
however, the sequence SV is retained, even in cases of object topicalisation,
again resulting in the verb in the third position of the clause, as in (35). All
these formative patterns are also found in Tok Pisin (cf. Mühlhäusler 1985:
(35) a. fi was du muss sterb-en?
for what 2SG must die-V
SG: ‘Warum musst du sterben?
EN: ‘Why do you have to die?
The creoleness of Unserdeutsch (Rabaul Creole German)
b. was du mein-en?
what 2SG mean-V
SG: ‘Was meinst du?’
EN: ‘What do you mean?’
c. was du kann sa?
what 2SG can say
SG: ‘Wie kann man sagen?
EN: ‘How (what) can you say?’
Since a formal distinction between main and subordinate clauses is absent in
Unserdeutsch, subordinate clauses, in contrast to Standard German, follow
the same canonic SVO word order. This applies irrespectively of the type of
subordinate clause, be it an unintroduced subordinate clause, a conjunctional
clause, as in (36), or a pronominal clause, as in (37).
(36) viellei jetz wi ni geht messe fi was wi war schon
maybe now 1PL NEG go mass for what 1PL COP.PST already
satt von.
fed_up by
SG: Vielleicht gehen wir jetzt nicht mehr zur Messe, weil wir schon
genug davon hatten.’
EN: ‘Maybe now we don’t go to mass, because we we’re already fed
up (already had enough).
(37) ein mensch wo kann spreh-en englisch …
ART.INDF person REL can speak-V english
SG: ‘Ein Mensch, der Englisch sprechen kann …’
EN: ‘A person, who can speak English …’
All in all, basilectal Unserdeutsch clearly shows a fixed SVO word order and
therefore can be classified as a typical creole language in this respect. On this
point, the profile of Unserdeutsch displays a great typological distance from
its lexifier language, whereas the structural closeness to its substrate language
Tok Pisin is all the more obvious. The observation suggests a profound
syntactic substrate transfer here.
LLM Special Issue 2017
4.3.2 Adjacency of verbal elements
Verbal elements enclosing other constituents in a clause, so-called ‘bracket
constructions’, are a special typological feature of (Standard) German syntax
(cf. Roelcke 2011: 6567). In Standard German subordinate clauses, the
clause-initial dependent word and the mostly clause-final finite verb
constitute the so-called sentence bracket around all other constituents.
Since there is no sentence bracketing left in Unserdeutsch (as a
consequence of all sentence types displaying a strict SVO word order), we
shall focus on the so-called grammatical bracket and the lexical bracket in
the following section (for bracket types in German cf. Weinrich 2007: 41
60). As corresponding research has shown, such bracket constructions are
frequently eliminated in intense language contact settings (cf. Riehl 2004:
106). The adjacency of relating elements, such as parts of the verbal complex,
is favoured, since their distant positioning would mean discontinuity and thus
a loss of transparency (cf. Leufkens 2013: 341).
a. Depletion of the lexical bracket
The term ‘lexical bracket’ refers to the fact that the constituents of phrasal
verbs are separated syntactically in certain clause types in Standard German.
This happens when the verb functions as a finite verb in V1 or V2 position,
dividing the constituents of the phrasal verb between the left and the right
sentence bracket.
In basilectal Unserdeutsch the lexical bracket tends to be depleted. Thus,
both constituents shift towards each other, albeit retaining a transposed order,
as in (38):
(38) a. dann wi ma weg alle schale.
then 1PL take off PL peel
SG: ‘Dann machten wir die Schalen weg.’
EN: ‘Then we took off the peels.’
The creoleness of Unserdeutsch (Rabaul Creole German)
b. er geht zurick zu de kinese wo sack-im
3SG.M go back to ART.DEF chinese REL sack-TR 1SG
SG: Er ging zu dem Chinesen zurück, der mich rausgeworfen
EN: ‘He went back to the Chinese who sacked me.’
Some Standard German phrasal verbs, such as aufpicken ‘to fetch, collect,
pick up’ or aufpassen ‘to take care, watch’, have been reanalysed as
inseparable prefix verbs in Unserdeutsch and consequently do not form any
kind of bracket, as in (39):
(39) de selbe zeit er aufpass-en alle halbweiß-e
ART.DEF same time 3SG.M look_after-V PL half_white-ATTR
kind am aben.
child-PL PREP evening
SG: ‘Gleichzeitig passte er am Abend auf die halbweißen Kinder auf.’
EN: ‘At the same time he took care of the half-white children in the
Remnants of the Standard German lexical bracket are retained rather
infrequently in Unserdeutsch. In such cases, the typological middle-field (cf.
Zifonun et al. 1997/2: 14981505 for a field typology of German syntax) is
typically restricted to one single constituent; all other constituents, if existent,
move to post-field position, as in (40):
(40) i bring-en de schlissel zurick zu de pflanzung
1SG bring-V ART.DEF key back to ART.DEF plantation
SG: Ich brachte den Schlüssel zum Plantagenbesitzer zurück.’
EN: I brought the key back to the plantation master.
The verb form sackim represents a hybrid construction, consisting of the English verb
stem (to) sack and the Tok Pisin transitive marker {-im}.
LLM Special Issue 2017
b. Depletion of the grammatical bracket
The grammatical bracket in Standard German consists of at least two verbs,
whereby the finite V1 or V2 position is taken by a modal verb (modal
bracket) or an auxiliary (tense or passive bracket). The main verb shifts to
clause final position in that case, the verbal complex thus embracing the
clause constituents following the finite verb.
Since passive constructions are extremely rare in basilectal Unserdeutsch
(see above), we confine ourselves to modal and tense brackets. Both bracket
types are depleted in the majority of cases, thus finite and infinite verbs are
placed in contact position, as in (41):
(41) a. darum wi muss-te geht zu kirhe bevor mitterna.
therefore 1PL must-PST go to church before midnight
SG: ‘Deswegen mussten wir vor Mitternacht zur Kirche gehen.’
EN: ‘Therefore we had to go to church before midnight.’
b. die hat bleib in cairns.
3PL AUX.PST stay in Cairns
SG: ‘Sie sind in Cairns geblieben.’
EN: ‘They have stayed in Cairns.’
To some extent, more often than in case of lexical brackets, a reduced bracket
is retained from Standard German. In this case, the middle-field is again
restricted to one clause constituent, as in (42). The negation particle ni is
always placed between the verbal elements, as with not in English sentences.
Only in more elaborated varieties nearby the acrolectal end of the creole
continuum, may two and more elements regularly occupy the middle-field.
(42) a. i hat kein brief ge-krie fi er.
1SG AUX.PST no letter PTCP-get from 3SG.M
SG: ‘Ich habe keinen Brief von ihm gekriegt.’
EN: ‘I didn’t receive a letter from him.’
b. die wid viellei tet-en i.
3PL AUX.IRR perhaps kill-V 1SG
The creoleness of Unserdeutsch (Rabaul Creole German)
SG: ‘Die würden mich vielleicht töten.’
EN: ‘They would maybe kill me.’
As measured by its basic tendencies, basilectal Unserdeutsch can at most be
regarded as forming brackets on a very limited basis. Complex bracket
constructions from its lexifier language do appear strongly in a restricted or
a simplified manner. We may thus conclude that the creole feature discussed
above does apply to Unserdeutsch at least to a large extent.
4.3.3 Dropping of function words
It is known that classes of function words in creole languages are restricted
to relatively few lexemes in comparison to their superstrate languages (cf.
Hurford 2012: 433). The available grammatical elements accordingly feature
a greater semantic extension. The use of function words in creole languages
differs from their use in (European) non-creoles on the syntactic level as well.
On the one hand, a higher number of constructions, for example possessive
constructions or specific clause connections, seem to be formed without
using a function word at all in creole languages, thereby showing a trend
towards juxtaposition (cf. Sutcliffe 2015: 239). The quite common use of
serial verb constructions in creole languages likewise fits into this picture (cf.
Aikhenvald 2006: 1). On the other hand, the use of function words in creole
languages seems not uncommonly to be less obligatory than in their
superstrate language. In the following section, we will discuss the possibility
of dropping different function words in Unserdeutsch that are obligatory in
Standard German.
a. Partial pro-drop status
A feature ascribed to many creole languages is their partial pro-drop status
(cf. Nicolis 2008: 279290). While referential pronominal subjects usually
cannot be dropped, expletives do rarely appear; thus, formal subjects (and
formal objects as well) are uncommon in creole language (cf. Haspelmath &
APiCS Consortium 2013c).
Referential pronominal subjects are generally used in Unserdeutsch on
(cf., e.g., 38, 39, 40). The occasional occurrence of referential zero subjects
LLM Special Issue 2017
in elliptical constructions, particularly in compound sentences, is roughly
comparable with their occurrence in spoken Standard German, as in (43):
(43) i heirat-en, hat ein tochter, dann ferti von
1SG marry-V have ART.INDF daughter then done with
de mensch, dann fund-en ein andre mensch …
ART.DE man then find\PST-V ART.INDF other man
SG: Ich habe geheiratet, hatte eine Tochter, hatte dann genug von
dem Mann, habe dann einen anderen Mann gefunden …’
EN: ‘I got married, had a daughter, then I was done with the (this)
man, then found another man …’
There are neither marked subject nor marked objects forms in Unserdeutsch.
This is mainly due to the fact that the third person singular neuter pronoun es
‘it’, serving as expletive in Standard German, is completely absent in
basilectal Unserdeutsch (as well as the pronoun man ‘one, you’ and largely
the passive, thus Unserdeutsch basically has no grammaticalised means to
express something impersonal without naming the agent). In constructions
where the expletive es is obligatory in Standard German, it is consequently
omitted in Unserdeutsch, partially by using alternative constructions, as in
(44) a. Ø is etwas spät.
COP.3SG bit late
SG: ‘Es ist etwas spät.’
EN: ‘It’s a bit late.
b. heute is regen.
today COP.3SG rain
SG: ‘Heute regnet es.’
EN: ‘Today it rains.’
b. Partial omission of the copula
There is some controversy over the status of the copula in creole languages.
Predicative adjectives tend to be connected via zero copula (cf. Bartens 2013:
The creoleness of Unserdeutsch (Rabaul Creole German)
100). Predicative noun phrases, however, appear approximately equally
frequent with or without an overt copula (cf. Michaelis & APiCS Consortium
2013 and Velupillai 2015: 409410).
Copula constructions with an overt copula are used in Unserdeutsch on a
regular basis, as in (45), though the occurrence of the copula is far less
obligatory than in Standard German: see, for example, (12) and (28).
(45) a. otherwise du bis hungri.
otherwise 2SG COP.2SG hungry
SG: ‘Sonst bist du hungrig.’
EN: ‘Otherwise you are hungry.’
b. du bis riti ein lüchner.
2SG COP.2SG really ART.INDF liar
SG: ‘Du bist wirklich (richtig) ein Lügner.’
EN: ‘You are truly (really) a liar.
However, copulas tend to be dropped in conjunction with predicative
nominal phrases, as in (46), as well as with predicative adjectives, as in (13)
and (47), at the basilectal end.
(46) a. wegen du Ø ein gut-e manager fi uns.
because 2SG ART.INDF good-ATTR manager POSS 1PL.ACC
SG: weil du ein guter Manager für uns warst.’
EN: ‘because you have been a good manager for us.’
b. alle Ø ein gruppe und alle Ø mission, ja, alle
all ART.INDF group and all mission yes, all
familie da Ø ein familie da.
family there one family there
SG: Alle waren eine Gruppe und alle von der Mission, ja, alle
Familien dort waren eine Familie.’
EN: Everybody was part of the group and all were from the
mission, yes, all families there were one family.
LLM Special Issue 2017
(47) a. i wart-en bis die Ø etwas groß.
1SG wait-V till 3PL a_bit big
SG: Ich habe gewartet, bis sie etwas größer sind …’
EN: I waited till they grew up a bit …’
b. die arbeit bis Ø dunkel.
3PL work till dark
SG: Sie haben gearbeitet, bis es dunkel war (wurde).
EN: They worked till it got dark.
c. Partial omission of articles
The vast majority of creole languages feature a definite as well as an
indefinite article (cf. Haspelmath & APiCS Consortium 2013de; Velupillai
2015: 365366). Articles and the category of definiteness are therefore rarely
deleted in creoles.
Unserdeutsch functions likewise, since a (gender-neutral) definite and
indefinite article is retained, as in (48):
(48) de mutter hat ein stroke.
ART.DEF mother have ART.INDF stroke
SG: Die Mutter hatte einen Schlaganfall.’
EN: ‘The mother had a stroke.’
The article seems to be more obligatory in comparison to the copula in
Unserdeutsch, although it may also be dropped in certain cases, as in (49):
(49) a. is Ø gut-e familie.
COP.3SG good-ATTR family
SG: ‘Es ist eine gute Familie.’
EN: ‘It’s a good family.’
b. er war Ø jung-e kerl.
3SG.M COP.PST young-ATTR guy
SG: ‘Er war ein junger Kerl.’
EN: ‘He was a young guy.’
The creoleness of Unserdeutsch (Rabaul Creole German)
d. Partial omission of adpositions and junctions
Typically, only very few adpositions from its superstrate language are
retained in a creole language (cf. Bartens 2013: 122 and Boretzky 1983: 194).
In the partially creolised Tok Pisin, only three prepositions are in use, and in
older language forms it was sometimes actually only one (bilong, cf.
Mühlhäusler 1985: 366). Many creole languages, for example, do not require
a preposition between a verb of motion and the location in directional
constructions (cf. Holm & Patrick 2007, Feature 19.2).
There are no postpositions (cf. Volker 1982: 52) and no circumpositions
from Standard German retained in Unserdeutsch. From the numerous
prepositions of its lexifier language, basilectal Unserdeutsch has only
retained those that are frequent in spoken German; these are used in a similar
manner. The dropping of prepositions occurs rarely and is clearly marked, as
in (50):
(50) a. i will geht Ø rabaul.
1SG want go Rabaul
SG: ‘Ich will nach Rabaul gehen.’
EN: ‘I want to go to Rabaul.’
b. i hol-en alle kind geht Ø ufer odä geht Ø
1SG fetch-V PL child-PL go coast or go
andre platz mit ein jeep.
other place with ART.INDF jeep
SG: ‘Ich habe die Kinder mit einem Jeep zur Küste oder zu einem
anderen Ort gebracht.
EN: ‘I brought the children with a jeep to the coast or to some
other place.
Subordinate structures are believed to be rather atypical for creole languages
as well, with paratactic structures preferred instead (cf. Bartens 2013: 129).
Their propensity for syntactic coordination, using asyndetic connections to
some extent, results in the omission of junctions in many cases (cf. Boretzky
1983: 208).
LLM Special Issue 2017
The junction inventory of Unserdeutsch is noticeably reduced in comparison
to its superstrate language. Its size grows only with increasing distance from
the basilectal end. The use of junctions is relatively obligatory in
Unserdeutsch as well. Nevertheless, they may be dropped, following certain
structures from the substrate or adstrate language, as in (51):
(51) a. i will du aufpass-en de flanzung.
1SG want 2SG look_after-V ART.DEF plantation
SG: ‘Ich will, dass du auf die Plantage aufpasst.’
EN: ‘I want you to look after the plantation.
b. de letzte mal i war in rabaul …
ART.DEF last time 1SG COP.PST in Rabaul
SG: Das letzte Mal, als ich in Rabaul war …’
EN: ‘The last time I was in Rabaul …’
In summary, it has been shown that various function words are anchored in
the grammatical system of basilectal Unserdeutsch. However, their use is
in some cases quite considerably less obligatory than in Standard German.
Especially the partial pro-drop status quite obviously fits this creole
characteristic as postulated in the literature. Like most creole languages,
Unserdeutsch distinguishes an invariant definite as well as indefinite article.
The significantly lower obligatorisation of the copula and the dropping of
other function words, which is at least possible to a limited extend, also fits
into the overall picture of an increased optionality in comparison to its lexifier
language when it comes to the realisation of function words. This increased
optionality is not surprising in the case of creole languages: It can be
observed in numerous language contact settings as a consequence or
reflection of the rise and loss of grammatical categories (cf. Tamm 2012:
151). In addition, it is characteristic for scenarios of language acquisition (cf.
Parodi & Tsimpli 2005). Both aspects play a considerable role in the genesis
of creole languages, so it is not surprising that they leave their traces in terms
of typological tendencies in the systems of these languages.
In the present case, a wide range of structural-typological characteristics
supports a remarkable finding: Apparently, Unserdeutsch largely corre-
sponds to the postulated pattern of an ‘average creole’.
The creoleness of Unserdeutsch (Rabaul Creole German)
In the end of our analyses on the structural-typological design of
Unserdeutsch, let us return to the initial question. In section 2 we noted that
Unserdeutsch, on the one hand, complies with the key characteristics of
creole languages and is thus undoubtedly to be classified as such, but on the
other hand shows several characteristics in its genesis that are considered as
atypical in creolistic theory. Moreover, all these characteristics give reason
to expect a structural convergence towards the lexifier language and thus
greater structural complexity.
However, the analyses have shown that when it comes to the structural-
typological design of its basilect, Unserdeutsch can be qualified as a largely
typical representative of the creole languages around the world. These
findings seem to run contrary to the results of Mühlhäusler (1984: 3840 and
1997: 200202), showing that Unserdeutsch does not in any way correspond
to the twelve creole features postulated by Bickerton (1981). However, this
apparent contradiction can be easily resolved by recognising two aspects.
First, Bickerton’s approach is to be regarded as problematic in itself (for a
critical evaluation with further references cf. Veenstra 2008), while the
APiCS data are grounded on a large-scale database. Second, Mühlhäusler’s
analyses are based on the data and description of Volker (1982), which is not
in all aspects consistent with the present results on the basis of new data by
speakers who are no longer competent in Standard German.
Unserdeutsch matches the pattern of an ‘average creole’ despite its
extended language functions, unlimited access to its lexifier language, and
competence in the lexifier, as well as close-knit social networks within a
largely closed, small community. How can this apparent discrepancy be
In search of a reasonable explanation, two factors seem to be of central
significance. First, there is the primary function of Unserdeutsch as a marker
of solidarity and identity, which helped to strengthen group identity and
cohesion and to draw a line against the (hostile) environment (cf. Volker
1989b). This function was crucial, as Unserdeutsch developed within a small
group of uprooted and socially isolated mixed-race children who were caught
between two stools, neither really belonging to the white colonialists nor to
the indigenous people (cf. Maitz 2017 and Maitz & Volker forthcoming).
With this in mind, the genesis and stabilisation of Unserdeutsch was an act
LLM Special Issue 2017
of identity, whereby a certain distance towards the target language, i.e the
Standard German of the missionaries, was obviously intended. In this way,
an intragroup language and thereby social exclusiveness could be created. It
is indeed a common strategy in the context of pidgin and creole languages to
shape and underline an in-group identity by using salient structures deviating
from the target language (i.e. the lexifier language) on purpose (cf. Higgins
Along these lines, the adolescent speakers of the first generation
consciously avoided structural proximity towards the target language when
using Unserdeutsch. We have to assume that this was even more the case,
given that the older children and young adults had by all accounts gained an
elaborated target language competence (cf. Maitz 2017) and thus could have
done better if they had wanted to.
The intended structural distance was, as the structure of Unserdeutsch
shows, primarily achieved by following the substrate Tok Pisin, which was
already spoken as L1 by most children when they entered the mission (cf.
Janssen 1932). In this way, the evolution of Unserdeutsch represents one of
the rare cases, in which a pidgin, i.e. an early version of Tok Pisin, served as
the substrate language for an emerging creole language. This might be the
second factor that could explain the structural-typological creole typicality of
Unserdeutsch, despite the conditions mentioned above.
With all this in mind, the emergence of Unserdeutsch may be regarded
as an act of linguistic dissociation and of subtle linguistic subversion. As in
hardly any other case, the words, by which Hofmann (2003: 282) summarised
a position of Glissant (1997), apply to Unserdeutsch: “Creole is not the result
of restricted input, but the product of strategies of resistance” (quoted by
Siegel 2007: 191).
1SG first person singular
1PL first person plural
2SG first person singular
3PL third person plural
3SG third person singular
INCL inclusive
INDF indefinite
IRR irrealis
M masculine
NEG negation
The creoleness of Unserdeutsch (Rabaul Creole German)
ACC accusative
ART article
ATTR attributive
AUX auxiliary
COP copula
DAT dative
DEF definite
DEM demonstrative
EN English
EXCL exclusive
F feminine
FUT future
GEN genitive
HAB habitual
PASS passive voice
PL plural
PN proper noun
POSS possessive
PREP preposition
PROG progressive
PST past
PTCP participle
REL relative
SG Standard German
TP Tok Pisin
TR transitive
UD Unserdeutsch
V verb
Aikhenvald, Alexandra Y. 2006. Serial verb constructions in typological
perspective. In Alexandra Y. Aikhenvald & Robert Dixon (eds.), Serial
verb constructions: A cross-linguistic typology, 168. Oxford: Oxford
University Press.
Arends, Jacques, Pieter Muysken & Norval Smith. 1995. Pidgins and creoles:
An introduction. Amsterdam & Philadelphia: Benjamins.
Baker, Philip. 2000. Theories of creolization and the degree and nature of
restructuring. In Ingrid Neumann-Holzschuh & Edgar Schneider (eds.),
Degrees of restructuring in creole languages, 4163. Amsterdam &
Philadelphia: Benjamins.
Bakker, Peter. 2008. Pidgins versus creoles and pidgin creoles. In Silvia
Kouwenberg & John Singler (eds.), The handbook of pidgin and creole
Studies, 130157. Oxford: Wiley-Blackwell.
Bakker, Peter, Aymeric Daval-Markussen, Mikael Parkvall & Ingo Plag. 2011.
Creoles are typologically distinct from non-creoles. Journal of Pidgin
and Creole Languages 26(1). 542.
Bartens, Angela. 2013. Creole languages. In Peter Bakker & Yaron Matras
(eds.), Contact languages: A Comprehensive Guide, 65158. Boston &
Berlin: de Gruyter Mouton.
LLM Special Issue 2017
Bickerton, Derek. 1981. Roots of Language. Ann Arbor: Karoma.
Bickerton, Derek. 1988. Creole languages and the bioprogram. In Frederick J.
Newmeyer (ed.), Linguistics: The Cambridge Survey. Vol. 2, Linguistic
Theory: Extensions and Implications, 268284. Cambridge: Cambridge
University Press.
Boas, Hans C. 2009. The Life and Death of Texas German. Durham: Duke
University Press.
Boretzky, Norbert. 1983. Kreolsprachen, Substrate und Sprachwandel.
Wiesbaden: Harrassowitz.
Crowley, Terry. 2008. Pidgin and creole morphology. In Silvia Kouwenberg
& John Singler (eds.), The Handbook of Pidgin and Creole Studies, 74
97. Oxford: Wiley-Blackwell.
Dahl, Östen & Viveka Velupillai. 2013. Perfective/imperfective aspect. In
Matthew S. Dryer & Martin Haspelmath (eds.), The World Atlas of
Language Structures Online. Leipzig: Max Planck Institute for
Evolutionary Anthropology. URL: <>; Date
accessed: 20.01.2017.
DeGraff, Michel. 2005. Linguists’ most dangerous myth: The fallacy of creole
exceptionalism. Language in Society 34(4). 533591.
Deumert, Ana. 2003. Markedness and salience in language contact and second-
language acquisition: Evidence from a non-canonical contact language.
Language Sciences 25(6). 561613.
Dryer, Matthew S. 2013. Order of subject, object and verb. In Matthew S.
Dryer & Martin Haspelmath (eds.), The World Atlas of Language
Structures Online. Leipzig: Max Planck Institute for Evolutionary
Anthropology. URL: <>; Date accessed:
Gárgyán, Gabriella. 2013. Der am-Progressiv im heutigen Deutsch. Neue
Erkenntnisse mit besonderer Hinsicht auf die Sprachgeschichte, die
Aspektualität und den kontrastiven Vergleich mit dem Ungarischen.
Dissertation Universität Szeged. Networx, No. 63. URL:
<>; Date acces-
sed: 20.01.2017.
Glissant, Édouard. 1997. Poetics of Relation. Translated by Betsy Wing. Ann
Arbor: University of Michigan Press.
The creoleness of Unserdeutsch (Rabaul Creole German)
Haspelmath, Martin & APiCS Consortium. 2013a. Vowel height distinctions.
In Susanne Maria Michaelis, Philippe Maurer, Martin Haspelmath &
Magnus Huber (eds.), Atlas of Pidgin and Creole Language Structures
Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.
URL: <>; Date accessed: 20.01.
Haspelmath, Martin & APiCS Consortium. 2013b. Expression of nominal
plural meaning. In Susanne Maria Michaelis, Philippe Maurer, Martin
Haspelmath & Magnus Huber (eds.), Atlas of Pidgin and Creole
Language Structures Online. Leipzig: Max Planck Institute for
Evolutionary Anthropology. URL: <
parameters/23>; Date accessed: 20.01.2017.
Haspelmath, Martin & APiCS Consortium. 2013c. Expletive subject of
existential verb. In Susanne Maria Michaelis, Philippe Maurer, Martin
Haspelmath & Magnus Huber (eds.), Atlas of Pidgin and Creole
Language Structures Online. Leipzig: Max Planck Institute for
Evolutionary Anthropology. URL: <
parameters/64>; Date accessed: 20.01.2017.
Haspelmath, Martin & APiCS Consortium. 2013d. Definite articles. In
Susanne Maria Michaelis, Philippe Maurer, Martin Haspelmath &
Magnus Huber (eds.), Atlas of Pidgin and Creole Language Structures
Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.
URL: <>; Date accessed: 20.01.
Haspelmath, Martin & APiCS Consortium. 2013e. Indefinite articles. In
Susanne Maria Michaelis, Philippe Maurer, Martin Haspelmath &
Magnus Huber (eds.), Atlas of Pidgin and Creole Language Structures
Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.
URL: <>; Date accessed: 20.01.
Hawkins, John A. 1986. A Comparative Typology of English and German.
Unifying the Contrasts. London: Croom Helm.
Higgins, Christina. 2015. Insults or acts of identity? The role of stylization in
multilingual discourse. Multilingua 23(2). 135158.
Hofmann, Sabine. 2003. Against monolingualism: Roots, rhizomes and the
conceptualization of creole language. In Gordeon Collier & Ulrich
LLM Special Issue 2017
Fleischmann (eds.), A Pepper-Pot of Cultures: Aspects of Creolization in
the Caribbean, 277285. Amsterdam & New York: Rodopi.
Holm, John A. 2000. An Introduction to Pidgins and Creoles. Cambridge:
Cambridge University Press.
Holm, John & Peter L. Patrick (eds.). 2007. Comparative Creole Syntax.
Parallel Outlines of 18 Creole Grammars. London: Battleridge.
Huber, Magnus & APiCS Consortium. 2013. Order of subject, object, and
verb. In Susanne Maria Michaelis, Philippe Maurer, Martin Haspelmath
& Magnus Huber (eds.), Atlas of Pidgin and Creole Language Structures
Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.
URL: < parameters/1>; Date accessed: 20.01.
Hurford, James. 2012. The Origins of Grammar: Language in the Light of
Evolution. Oxford: Oxford University Press.
Janssen, Arnold M.S.C. 1932. Die Erziehungsanstalt für halbweiße Kinder. In
Josef Hüskes (ed.), Pioniere der Südsee. Werden und Wachsen der Herz-
Jesu-Mission von Rabaul zum Goldenen Jubiläum 18821932, 150155.
Hiltrup: Missionare vom Hlst. Herzen Jesu.
Kaye, Alan S. & Mauro Tosco. 2003. Pidgin and Creole Languages: A Basic
Introduction. München: LINCOM.
Klein, Thomas. 2006. Creole phonology typology: Phoneme inventory size,
vowel quality distinctions and stop consonant series. In Parth Bhatt &
Ingo Plag (eds.), The Structure of Creole Words: Segmental, Syllabic and
Morphological Aspects, 321. Tübingen: Niemeyer.
König, Ekkehard & Volker Gast. 2009. Understanding English-German
Contrasts. Berlin: Schmidt.
Kortmann, Bernd & Benedikt Szmrecsanyi. 2009. World Englishes between
simplification and complexification. In Lucia Siebers & Thomas
Hoffmann. (eds.), World Englishes: Problems Properties Prospects,
265285. Amsterdam & Philadelphia: Benjamins.
Laycock, Don. 1985. Phonology: substratum elements in Tok Pisin phonology.
In Stephen A. Wurm & Peter Mühlhäusler (eds.), Handbook of Tok Pisin
(New Guinea Pidgin), 295307. Canberra: Dept. of Linguistics, Research
School of Pacific Studies, Australian National University.
Lefebvre, Claire. 2004. Issues in the Study of Pidgin and Creole Languages.
Amsterdam & Philadelphia: Benjamins.
The creoleness of Unserdeutsch (Rabaul Creole German)
Leufkens, Sterre. 2013. The transparency of creoles. Journal of Pidgin and
Creole Languages 28(2). 323362.
Lippi-Green, Rosina. 22012. English with an Accent: Language, Ideology and
Discrimination in the United States. London & New York: Routledge.
Maitz, Péter. 2016. Unserdeutsch. Eine vergessene koloniale Varietät des
Deutschen im melanesischen Pazifik. In Alexandra N. Lenz (ed.),
German Abroad Perspektiven der Variationslinguistik, Sprachkontakt-
und Mehrsprachigkeitsforschung, 211240. Göttingen: V & R unipress.
Maitz, Péter. 2017. Dekreolisierung und Variation in Unserdeutsch. In Helen
Christen, Peter Gilles & Christoph Purschke (eds.), Räume Grenzen
Übergänge. Akten des 5. Kongresses der Internationalen Gesellschaft für
Dialektologie des Deutschen (IGDD), 225252. Stuttgart: Steiner.
Maitz, Péter, Werner König & Craig A. Volker. 2016. Unserdeutsch (Rabaul
Creole German): Dokumentation einer stark gefährdeten Kreolsprache in
Papua-Neuguinea. Zeitschrift für germanistische Linguistik 44(1). 9396.
Maitz, Péter & Siegwalt Lindenfelser (forthcoming). Gesprochenes
Alltagsdeutsch im Bismarck-Archipel um 1900. Das Zeugnis regional
markierter Superstrateinflüsse in Unserdeutsch. In Alexandra Lenz &
Albrecht Plewnia (eds.), Variation Norm(en) Identität(en). Berlin &
Boston: de Gruyter.
Maitz, Péter & Siegwalt Lindenfelser (submitted). Wie kreol ist Unserdeutsch?
Zeitschrift für Sprachwissenschaft.
Maitz, Péter & Attila meth. 2014. Language contact and morphosyntactic
complexity: Evidence from German. Journal of Germanic Linguistics
26(1). 129.
Maitz, Péter & Craig A. Volker. 2017. Documenting Unserdeutsch: Reversing
colonial amnesia. Journal of Pidgin and Creole Languages 32(2). 365
Maitz, Péter & Craig A. Volker (forthcoming). Unserdeutsch (Rabaul Creole
German). In Hans C. Boas, Ana Deumert, Mark L. Louden & Péter Maitz
(eds.), Varieties of German Worldwide. Oxford: Oxford University Press.
Maurer, Philippe & APiCS Consortium. 2013a. Syllable onsets. In Susanne
Maria Michaelis, Philippe Maurer, Martin Haspelmath & Magnus Huber
(eds.), Atlas of Pidgin and Creole Language Structures Online. Leipzig:
Max Planck Institute for Evolutionary Anthropology. URL:
<>; Date accessed: 20.01.2017.
LLM Special Issue 2017
Maurer, Philippe & APiCS Consortium. 2013b. Syllable codas. In Susanne
Maria Michaelis, Philippe Maurer, Martin Haspelmath & Magnus Huber
(eds.), Atlas of Pidgin and Creole Language Structures Online. Leipzig:
Max Planck Institute for Evolutionary Anthropology. URL:
<>; Date accessed: 20.01.2017.
Maurer, Philippe & APiCS Consortium. 2013c. Tone. In Susanne Maria
Michaelis, Philippe Maurer, Martin Haspelmath & Magnus Huber (eds.),
Atlas of Pidgin and Creole Language Structures Online. Leipzig: Max
Planck Institute for Evolutionary Anthropology. URL: <http://apics->; Date accessed: 20.01.2017.
Maurer, Philippe & APiCS Consortium. 2013d. Gender agreement of
adnominal adjectives. In Susanne Maria Michaelis, Philippe Maurer,
Martin Haspelmath & Magnus Huber (eds.), Atlas of Pidgin and Creole
Language Structures Online. Leipzig: Max Planck Institute for
Evolutionary Anthropology. URL: <
40>; Date accessed: 20.01.2017.
Maurer, Philippe & APiCS Consortium. 2013e. Tense-aspect systems. In
Susanne Maria Michaelis, Philippe Maurer, Martin Haspelmath &
Magnus Huber (eds.), Atlas of Pidgin and Creole Language Structures
Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.
URL: <>; Date accessed:
Maurer, Philippe & APiCS Consortium. 2013f. Position of tense, aspect, and
mood markers in relation to the verb. In Susanne Maria Michaelis,
Philippe Maurer, Martin Haspelmath & Magnus Huber (eds.), Atlas of
Pidgin and Creole Language Structures Online. Leipzig: Max Planck
Institute for Evolutionary Anthropology. URL: <http://apics->; Date accessed: 20.01.2017.
McWhorter, John H. 1998. Identifying the creole prototype: Vindicating a
typological class. Language 74(4). 788818.
McWhorter, John H. 2000. Defining “creole” as a synchronic term. In Ingrid
Neumann-Holzschuh & Schneider, Edgar (eds.), Degrees of Restruc-
turing in Creole Languages, 85124. Amsterdam & Philadelphia:
McWhorter, John H. 2001. The world’s simplest grammars are creole
grammars. Linguistic Typology 5(23). 125166.
The creoleness of Unserdeutsch (Rabaul Creole German)
McWhorter, John H. 2014. Mesolect as the norm: semi-creoles revisited. Papia
24(1). 91110.
Michaelis, Susanne Maria, Philippe Maurer, Martin Haspelmath, Magnus
Huber & Melanie Revis (eds.). 2013. The Atlas of Pidgin and Creole
Language Structures. Oxford: Oxford University Press.
Michaelis, Susanne Maria & APiCS Consortium. 2013. Predicative noun
phrases. In Susanne Maria Michaelis, Philippe Maurer, Martin
Haspelmath & Magnus Huber (eds.), Atlas of Pidgin and Creole
Language Structures Online. Leipzig: Max Planck Institute for
Evolutionary Anthropology. URL: <
73>; Date accessed: 20.01.2017.
Mufwene, Salikoko S. 2000. Creolization is a social, not a structural, process.
In Ingrid Neumann-Holzschuh & Edgar Schneider (eds.), Degrees of
Restructuring in Creole Languages, 6584. Amsterdam & Philadelphia:
Mufwene, Salikoko S. 2009. The evolution of language: Hints from creoles
and pidgins. In James Minett & William Wang (eds.), Language
Evolution and the Brain, 133. Hong Kong: City University of Hong
Kong Press.
Mühlhäusler, Peter. 1984. Tracing the roots of Pidgin German. Language &
Communication 4(1). 2757.
Mühlhäusler, Peter. 1985. Syntax of Tok Pisin. In Stephen A. Wurm & Peter
Mühlhäusler (eds.), Handbook of Tok Pisin (New Guinea Pidgin), 341
421. Canberra: Dept. of Linguistics, Research School of Pacific Studies,
Australian National University.
Mühlhäusler, Peter. 1997. Pidgin and Creole Linguistics. Expanded and
revised edition. London: University of Westminster Press.
Muysken, Pieter. 1988. Are creoles a special type of language? In Frederick
Newmeyer (ed.), The Cambridge Survey. Vol. 2, Linguistic Theory:
Extensions and Implications, 285302. Cambridge: Cambridge
University Press.
Nicolis, Marco. 2008. The null subject parameter and correlating properties:
The case of creole languages. In Theresa Biberauer (ed.), The Limits of
Syntactic Variation, 271294. Amsterdam & Philadelphia: Benjamins.
Parkvall, Mikael. 2008. The simplicity of creoles in a cross-linguistic
perspective. In Matti Miestamo, Kaius Sinnemäki & Fred Karlsson
LLM Special Issue 2017
(eds.), Language Complexity: Typology, Contact, change, 265285.
Amsterdam & Philadelphia: Benjamins.
Parodi, Teresa & Ianthi-Maria Tsimpli. 2005. ‘Real’ and apparent optionality
in second language grammars: Finiteness and pronouns in null operator
structures. Second language research 21(3). 250285.
Riehl, Claudia Maria. 32004. Sprachkontaktforschung. Eine Einführung.
Tübingen: Narr.
Roelcke, Thorsten. 2011. Typologische Variation im Deutschen. Grundlagen
Modelle Tendenzen. Berlin: Erich Schmidt.
Romaine, Suzanne. 1988. Pidgin and Creole Languages. London: Longman.
Siegel, Jeff. 2007. Transmission and transfer. In Umberto Ansaldo, Stephen
Matthews & Lisa Lim (eds.), Deconstructing Creole, 167201.
Amsterdam & Philadelphia: Benjamins.
Siegel, Jeff, Benedikt Szmrecsanyi & Bernd Kortmann. 2014. Measuring
analyticity and syntheticity in creoles. Journal of Pidgin and Creole
Languages 29(1). 4985.
Smith, Geoff P. 2008. Tok Pisin in Papua New Guinea: Phonology. In Kate
Burridge & Bernd Kortmann (eds.), Varieties of English: The Pacific and
Australia, 188209. Berlin & New York: Mouton de Gruyter.
Sutcliffe, David. 2015. Prima facie evidence for the persistence of creole
features in African American English and evidence for residual creole. In
Michael Picone & Catherine Davies (eds.), New Perspectives on
Language Variety in the South. Historical and Contemporary
Approaches, 233253. Alabama: University of Alabama Press.
Tamm, Anne. 2012. Optionality: Social cognitive factors in changing linguistic
complexity in the dialects of Estonia. Lingue e letterature d’Oriente e
d’Occidente (LEA) 1(1). 151162.
Thomason, Sarah. 2008. Pidgins, creoles and historical linguistics. In Silvia
Kouwenberg & John Singler (eds.), The Handbook of Pidgin and Creole
Studies, 243262. Oxford: Wiley-Blackwell.
Thurston, William. 1987. Processes of Change in the Languages of North-
Western New Britain. Canberra: Australian National University Press.
Trudgill, Peter. 2011. Sociolinguistic Typology: Social Determinants of
Linguistic Complexity. Oxford: Oxford University Press.
The creoleness of Unserdeutsch (Rabaul Creole German)
Tryon, Darrell T. & Jean-Michel Charpentier. 2004. Pacific Pidgins and
Creoles: Origins, Growth and Development. Berlin & New York:
Mouton de Gruyter.
Velupillai, Viveka. 2015. Pidgins, Creoles and Mixed Languages: An
Introduction. Amsterdam & Philadelphia: Benjamins.
Veenstra, Tonjes. 2008. Creole genesis: The impact of the Language
Bioprogram Hypothesis. In Silvia Kouwenberg & John Singler (eds.),
The Handbook of Pidgin and Creole Studies, 219241. Oxford: Wiley-
Verhaar, John W. M. 1995. Toward a Reference Grammar of Tok Pisin: An
Experiment in Corpus Linguistics. Honolulu: University of Hawai’i
Volker, Craig A. 1982. An Introduction to Rabaul Creole German (Unser-
deutsch). Unpublished MLitSt thesis, University of Queensland.
Volker, Craig A. 1989a. Rabaul Creole German syntax. Working Papers in
Linguistics 21. Honolulu: University of Hawai’i at Manoa. 153189.
Volker, Craig A. 1989b. The relationship between traditional secret languages
and two school-based pidgin languages in Papua New Guinea. Horizons.
Journal of Asia-Pacific Issues 3. 1924.
Volker, Craig A. 1991. The birth and decline of Rabaul Creole German.
Language and Linguistics in Melanesia 22. 143156.
Weinrich, Harald. 42007. Textgrammatik der deutschen Sprache. Hildesheim:
Werner, Otmar. 1969. Das deutsche Pluralsystem. Strukturelle D iachronie. In
Hugo Moser (ed.), Sprache, Gegenwart und Geschichte: Probleme der
Synchronie und Diachronie. Jahrbuch 1968, 92128. Düsseldorf:
Wiese, Richard. 1996. The Phonology of German. Oxford: Clarendon Press.
Zifonun, Gisela, Ludger Hoffmann & Bruno Strecker. 1997. Grammatik der
deutschen Sprache. Vol. 2. Berlin: de Gruyter.
Full-text available
This study examines asymmetries between so-called inherent and contextual categories in relation to the morphological complexity of the nominal and verbal inflectional domain of languages. The observations are traced back to the influence of adult L2 learning in scenarios of intense language contact. A method for a simple comparison of the amount of inherent versus contextual categories is proposed and applied to the German-based creole language Unserdeutsch (Rabaul Creole German) in comparison to its lexifier language. The same procedure will be applied to two further language pairs. The grammatical systems of Unserdeutsch and other contact languages display a noticeable asymmetry regarding their structural complexity. Analysing different kinds of evidence, the explanatory key factor seems to be the role of (adult) L2 acquisition in the history of a language, whereby languages with periods of widespread L2 acquisition tend to lose contextual features. This impression is reinforced by general tendencies in pidgin and creole languages. Beyond that, there seems to be a tendency for inherent categories to be more strongly associated with the verb, while contextual categories seem to be more strongly associated with the noun. This leads to an asymmetry in categorical complexity between the noun phrase and the verb phrase in languages that experienced periods of intense L2 learning.
This paper provides a comparative sketch of two emergent creoles: St. Lucia Creole English and Dominica Creole English. With very similar circumstances surrounding their formation and emergence, it is expected that they will have many features in common. While this is generally the case, however, this paper also outlines some areas in which their grammars diverge. This research represents original data for varieties that are somewhat under‐documented and calls for further data collection and fieldwork to confirm the patterns found here.
Full-text available
Taking up some of the goals of the Contrastive Program formulated in the sixties and seventies of the last century, this book presents a comprehensive and fine-grained analysis of the major contrasts between German and English with the aim of showing how far the two closely related languages have moved apart and of providing new foundations for the study and the teaching of English from the perspective of German, and the study of German from the perspective of English. It is based on numerous contrastive studies and on a thorough exploration of the relevant literature on the two languages as well as linguistic typology in general. The fourth edition has added further elaborations of the introductory and final chapters on the basis of recent contributions.
Aims and Scope Pacific Pidgins and Creoles discusses the complex and fascinating history of English-based pidgins in the Pacific, especially the three closely related Melanesian pidgins: Tok Pisin, Pijin, and Bislama. The book details the central role of the port of Sydney and the linguistic synergies between Australia and the Pacific islands in the late 18th and early 19th centuries, the role of Pacific islander plantation labor overseas, and the differentiation which has taken place in the pidgins spoken in the Melanesian island states in the 20th century. It also looks at the future of Pacific pidgins at a time of increasing vernacular language endangerment. © Copyright 2004 by Walter de Gruyter GmbH & Co. KG, D-10785 Berlin. All rights reserved.