Jeff Siegel, ed (2000); Processes of Language Contact: Studies from
Australia and the South Pacific. Fides (University of Montreal Press),
paperback ISBN: 2-7621-2098-5, xi+326pp.
Reviewed by Claire Bowern, Harvard University and Australian National
Publisher's announcement: http://linguistlist.org/issues/12/12-759.html
Previous review: http://linguistlist.org/issues/12/12-1548.html
Purpose of Book and Overview
Four of the papers in this book were presented at a symposium on
language contact ("Language contact and change: When languages meet"),
held in conjunction with the Australian Linguistic Institute in
Brisbane, 1998. The remaining seven papers were either presented in
other activities associated with ALS'98 or solicited from participants
in these events.
The papers in the volume all examine the processes involved in the
formation of pidgins and creoles, in the context of the contact
languages of Australia and the South Pacific. Jeff Siegel, in his
Introduction, classifies these processes into six types:
5. language shift
Related to these processes are the notions of substrate and superstrate
influences. The first four papers in the book (Koch, Crowley, Siegel et
al and Jourdan) deal with various instances of substrate influences.
Harold Koch traces two features of Tok Pisin and Melanesian Pidgins to
Australian Aboriginal grammar. Terry Crowley's first paper in the
volume, however, shows that Bislama predicate marking cannot be solely
attributed to substrate influence, and language-internal factors have
also been significant. Jeff Siegel et al look at why some substrate
features and not others have ended up in Tayo (a French-based creole
from New Caledonia).
Chapters 5, 6 and 7 are concerned with the notion of 'simplification'
in language contact. Bresnan's paper is a formal account of the
prevalence of free pronouns (as opposed to bound forms) in pidgins. Ian
Malcolm discusses simplification in relation to Aboriginal English,
particularly in the phonological and inflectional systems. He places
this in the context of the depidiginisation of the original
interlanguage and the features of the pidgin which are found in modern
Aboriginal English. Terry Crowley's second paper looks in more depth at
the notions or simplicity and complexity in grammatical change, and he
shows with examples from Vanuatu that exoteric and esoteric languages
need not correspond with 'simpler' and 'more complex' respectively.
The next two chapters look at the diffusion of language contact
varieties in Australia. Jane Simpson examines the role of "Afghan"
cameleers in the spread of Aboriginal pidgins in the 19th Century,
while Jennifer Munro shows that Kriol (the North Australian creole)
originated at Roper River mission and spread from there, rather than
the generally assumed view that Kriol evolved separately in several
locations and converged to a koine.
The final two chapters deal with depidginisation and decreolisation;
what happens when the pidgin or creole comes into extensive contact
with its lexifier? Geoff Smith examines this question in relation to
Tok Pisin, and Chris Corne in relation to Tayo.
Description of Contents
1. The Role of Australian Aboriginal Languages in the Formation of
Australian Pidgin Grammar: Transitive Verbs and Adjectives (Harold
Harold Koch focuses on two grammatical features of the pidgins and
creoles spoken historically and currently in Australia and the Pacific
region. Australian Pidgin (AP) was not only the ancestor of the creoles
spoken in Northern Australia, but also, through Queensland plantation
workers, of Early Melanesian Pidgin, the ancestor of the currently
spoken English-based creoles of the Southwest Pacific, such as Bislama
and Tok Pisin.
Koch examines two important features of Pidgins of the area: the -im/it
suffix which marks transitive verbs, and the so-called adjective marker
-fela/-pela. He shows explicitly how the marking of objects as verbal
suffixes in the Aboriginal languages around Sydney how have facilitated
the reanalysis of the unstressed pronoun; rather than analysing the
pronoun as the direct object, it is analysed as a part of the verb, an
agreement marker, and the object is assumed to be pro-dropped (a
feature of many Aboriginal languages).
Koch's paper could also be considered as a study of what one has to do
to show that there is substratum influence from a particular
construction in the substrate language; the phonological, morphological
and syntactic triggers of both the lexifier and the substrate language
must be taken into consideration.
2. "Predicate Marking" in Bislama (Terry Crowley)
In this paper, Terry Crowley examines the i- 'predicate marker' in
Bislama. He gives a summary of previous treatments of the construction
and its relationship to the subject pronouns. The predicate marker
shows interesting distributional restrictions; it occurs with the third
person pronoun em but not in all focus and subject limitation
constructions. Despite its virtual grammaticalisation as a subject
predicate marker, it retains some properties of a pronoun. Crowley
makes the point that the problem with analysis is probably expecting
the construction to fall neatly into one type, especially since it
probably has a number of different sources.
3. Predicting Substrate Influence: Tense-Modality-Aspect Marking in
Tayo (Jeff Siegel, Barbara Sandeman and Chris Corne)
In this paper, the authors address the comment in Mufwene (1990:6),
that "the fact that no attempt has been made to suggest any principle
regulating ... a selection of substrate features is deplorable." They
make predictions about the tense/aspect/mood system of Tayo, a French-
lexifier creole spoken in the village of St-Louis (near Noumea in New
Caledonia). Tayo is a good language for such a study, since the creole
has developed over the last hundred years and we know a great deal
about the way it has evolved and the languages involved. The authors
present a description of the tense/aspect/mood system of the substrate
languages. They give an outline of the three factors the authors
believe are the biggest determiners of substrate influence, and see to
what extent their prediction are borne out by the data from Tayo. The
factors in determining transfer from substrate to pidgin are:
availability (there must be an equivalent morpheme in the substrate and
creole); reinforcement (broadly, the prestige principles and speaker
numbers, and typological similarity) and simplification. The authors'
predictions are largely borne out by the data. 4. My Nephew is My
Aunt: Features and Transformation of Kinship Terminology in Solomon
Islands Pijin (Christine Jourdan)
Christine Jourdan's article highlights the importance of considering
not only the possibilities of substrate influence on creoles but also
the situation in which the creole arises. She does this by examining
the kinship system of Pijin (the creole of the Solomon Islands and
particularly of the capital, Honiara) and the substrate languages. She
finds that the kinship system of Pijin does not match any of the
substrate languages. but rather resembles must more the lexifier
language (English) in placing emphasis on the nuclear family. Jourdan
rejects, however, the argument that the kinship terminology is a direct
result of the use of the English terms. Rather, she traces it to the
social structure of urbanised Honiara families and the sphere of use of
Pijin versus traditional languages. In the spheres of language use in
which Pijin developed and is used, there is less of a need to refer to
5. Aboriginal English: From Contact Variety to Social Dialect (Ian G.
Ian Malcolm concentrates on simplification, particularly in phonology
(127) and inflectional morphology (128-131). Malcolm considers the
continuities and discontinuities between features of early pidgin data
and current Aboriginal English. He finds that while there has been a
drift towards features of standard English (particularly in grammar),
there is still a strong maintenance of features of Australian Pidgin
within modern Aboriginal English, even in the speech of urbanised
Aborigines when they are speaking in the absence of non-Aboriginal
Australians. We see also that although restructuring has been taking
place in Aboriginal English along the lines of rules in the Standard
English system, the rules are applied in distinctive ways, with results
that do not mirror Standard English.
6. Pidgin Genesis and Optimality Theory (Joan Bresnan)
Bresnan looks at simplification and universals in relation to the
prevalence of free (as opposed to bound) pronouns in pidgins and
creoles. The first hypothesis is that creoles tend to have free
pronouns because most of the lexifier languages do. As Bresnan points
out, however, this does not explain why West African Pidgin Portuguese
uses the full stressed Portuguese pronouns while standard Portuguese
uses the cliticised pronouns much more. Another hypothesis is that free
pronouns represent the default parameter of universal grammar. This
hypothesis is also rejected, as creoles can contain typologically
highly unusual features.
Bresnan's hypothesis appeals to simplicity; that is, that pidgins tend
to eliminate high marked structures and so, if free pronouns are less
marked than bound pronouns, they will be preferred, unless there is a
reason to keep them (like, for example, strong influence from substrate
languages). While this argument seems fine, I was not convinced,
however, that free pronouns are generally somehow less marked than
bound pronouns. Consider, for example, the very strong tendency of
pronouns to be reduced and cliticised.
7. Simplicity, Complexity, Emblematicity and Grammatical Change (Terry
Of those papers concerned with simplicity and simplification, Terry
Crowley's is the only one to tackle what is meant by simplicity, and
how it relates to contact situations and group languages. The area of
study is the island of Erromango, in Southern Vanuatu. He shows that
the two concepts are not necessarily tied together. A language may
undergo structural simplification without contact (as Allen (1997) has
shown for English, and Crowley argues for Ura, for example) and that an
exoteric language may not be the simplest language available.
8. Camels as Pidgin-carriers: Afghan Cameleers as a Vector for the
Spread of Features of Australian Aboriginal Pidgins and Creoles (Jane
9. Kriol on the Move: A Case of Language Spread and Shift in Northern
Australia (Jennifer M. Munro)
These two papers, by Jane Simpson and Jennifer Munro, are about the
spread of contact languages in Australia. Simpson examines examples of
the speech of the "Afghan" (ie, predominantly North Indian, Afghani or
Pakistani) camel drivers who travelled much of inland Australia last
century. She found some notable pidgin features in the data, such as
the use of the -im transitive marker. Other non-standard features
appeared to be the result of learner English. Simpson shows that the
early cameleers were probably a significant vector for the spread of
Jennifer Munro argues for the monogenesis of Kriol (the North
Australian creole spoken in the Northern Territory and Eastern
Kimberley region). She argues that the language spread from Roper River
mission after its formation in 1908, rather than the more traditional
assumption that the Australian Pidgin was creolised in several
different places and later converged. Munro writes that the
establishment of the mission and its settlement by small numbers of
speakers of several mutually unintelligible languages created the ideal
conditions for the creolisation of the pidgin that had been spoken in
the area for about 30 years previously. At this point the mission
became a permanent home (because of the threat of murder from the
pastoral lease-holders in the area). She then shows that the language
could have spread as a lingua franca between Aborigines working in the
cattle industry, and as a neutral language in communities where many
traditional languages were spoken.
10. Tok Pisin and English: The Current Relationship (Geoff P. Smith)
Geoff Smith examines the influence of English on Tok Pisin (TP) from a
corpus of about 400,00 words, recorded from first language TP speakers
in areas where Tok Pisin has been a traditional lingua franca. He
examines the relationship between TP and its lexifier. At one end of
the scale, he finds whole phrases and sentences imported into TP
speech, some adapted into TP grammar, others not. He also found
evidence of semantic influence on TP works, such as the use of pasim
(TP close, fasten) to mean "pass an exam". Sith found rather little of
the 'classic' codeswitching of Myers-Scotton (1993), however. The
outcome of Smith's study is that while TP is receiving increasing
borrowings from stand English, even in the speech of those that have
had very little exposure to the language, it is not yet in the stage of
a post-creole continuum.
11. Na pa kekan, na person: The Evolution of Tayo Negatives (Chris
Chris Corne uses data from Tayo (the French-based creole of the St-
Louis) to examine the influences on that language from French. The work
on negation was part of a wider survey of French versus Kanak features
of Tayo syntax. Corne gives a summary of negation in the substrate
languages and compares it to the constructions in Tayo. The data reveal
influence from French negation strategies in process, and this is
perhaps evidence for decreolisation. Corne makes the important point
that decreolisation can be seen to embody the same sorts of processes
that characterise creolisation; that is, the reanalysis of structures
under contact. However, the result may be something different from both
the substrate language and the superstrate influence.
A great strength of this book is its grounding in data, without
ignoring theory. Papers such as Bresnan's show that one need not
discuss either "theory" or "data", but each influences the other. The
integration of data into a coherent framework of process types makes
this book valuable reading.
Another strength of "Processes of language contact" is the wealth of
different view points expressed. The same topics and processes (for
example, simplification) are the focus of several papers, but the way
they are discussed and the conclusions reached are different. The same
topic approached from different angles. Some might argue that this
detracts from the overall coherence of the book, and certainly the
reader will not end up with the feeling that the issues in the book
have been addressed to satisfaction. This is in my view an advantage,
for these are messy issues and to present a single viewpoint in the
book would have been misleading.
Several papers (e.g. Koch, Malcolm, Crowley) made an important point
which is worth both stressing and further investigating; that is, that
reanalysis and grammatical influence from one language to another can
be indirect and the results of transferring a rule or part of a system
from one language to another may be different from the effects of the
original rule in the donor language.
A background in Pacific languages and/or Pidgin and Creole linguistics
is necessary to get the most out of this book, but I would very much
recommend it as a source of data and debate.
Allen (1997) Middle English case loss and the 'creolization'
hypothesis. English Language and Linguistics. 1:63-89.
Mufwene, S. (1990). Transfer and the substrate hypothesis in
Creolistics. Studies in Second Language Acquisition. 12:1-23
Thomason, S and T. Kaufman (1988); Language Contact, Creolisation and
Genetic Linguistics. Berkeley: U California Press.
Claire Bowern is a 3rd year PhD student at Harvard University and
Visiting Fellow at the Australian National University, Canberra. She is
writing her thesis on the linguistic history of Bardi, a non-Pama-
Nyungan language of North-Western Australia. Besides historical
linguistics, her other interests include phonology, language contact
and the preservation of endangered languages.