2nd proofs

:o8 Norval Smith

Maurer, Philippe. 1995. Langolar: Un crole afro-portugais parl So Tom. Hamburg: Helmut
Buske.
Price, Richard. 1983. First time. Baltimore MD: e Johns Hopkins University Press.
Schumann, Christian L. 1778. Saramaccan-Deutsches Wrter-Buch. (Ms, Moravian Brethren,
Bambey, Surinam).
Schumann, Christian L. 1783. Neger-Englisches Wrter-Buch. (Ms, Moravian Brethren,
Paramaribo, Surinam. Reproduced in Kramp, Andr A., Early creole lexicography: A study
of C. L. Schumanns manuscript dictionary of Sranan. Unpublished D.Litt. dissertation,
State University of Leiden).
Smith, Norval. 1999. Pernambuco to Surinam 16541665? e Jewish slave controversy. In
Huber, Magnus & Parkvall, Mikael (Eds), Spreading the word: Papers on the issue of diusion
of Atlantic creoles. London: University of Westminster Press. 251298.
Smith, Norval. 2001. Voodoo Chile: Dierential substrate eects in Saramaccan and Haitian. In
Smith, Norval & Veenstra, Tonjes (Eds), Creolization and contact [Creole Language Library
23]. Amsterdam: John Benjamins. 4280.
Smith, Norval. 2004. A case of creole substrate allophony 350 years on: Nasality in Saramaccan.
(Paper given at the Society for Pidgin and Creole Linguistics meeting, Boston).
Smith, Norval. Forthcoming. English-speaking in early Surinam? In Van den Berg, Margot,
Cardoso, Hugo & Selbach, Rachel (Eds), On gradual creolization: Studies celebrating Jacques
Arends [Creole Language Library]. Amsterdam: John Benjamins.
Smith, Norval & Cardoso, Hugo. 2004. A new look at the Portuguese element in Saramaccan.
Journal of Portuguese Linguistics 3: 115147. (Special issue on Portuguese creoles edited by
Norval Smith).
Smith, Norval & Haabo, Vinije. 2007. e Saramaccan implosives: Tools for linguistic
archaeology? Journal of Pidgin and Creole Languages 22.1:101122.
Smith, Norval & van de Vate, Marleen. 2006. Population movements, colonial control and vowel
systems. In Plag, Ingo & Bhatt, Parth (Eds), e structure of creole words: Segmental, syllabic
and morphological aspects [Linguistische Arbeiten 505]. Tbingen: Niemeyer. 5982.
Sommier Discours. 1638. Sommier Discours over den Staet () van Brasil. [14/1/1638.] [Edited
by S. Carpentier, Bijdragen en Mededelingen van het Historisch Genootschap te Utrecht 1979:
257317.]
ornton, John. 2005. Not Imbangala. e Angolan background of early Palmares, 15791648.
(Paper delivered at the conference Slavery from within: Comparative perspectives and
legacies in the Atlantic world. Roosevelt Study Center, Middelburg, Netherlands).
Williamson, Kay & Wakama, Carol. 2001. A slightly mixed language in Okrika. Paper presented
at the 30
th
Colloquium on African Languages and Linguistics, Leiden University.
Encoding path in Mauritian Creole
and Bhojpuri
Problems of language contact
*
Sibylle Kriegel
Laboratoire Parole et Langage, CNRS
Universit Aix-Marseille 1
Ralph Ludwig
Martin-Luther-Universitt Halle-Wittenberg
Fabiola Henri
Universit Paris 7
Tis article explores the question of language contact between
Mauritian Creole and the Indo-Aryan language Bhojpuri. Historically,
contact between these two languages came about as a result of the
massive immigration of Indians to Mauritius, dating from the second half
of the 19th century, afer the abolition of slavery. Tis contact situation
has lasted up until present times, since Bhojpuri remains, afer Mauritian
Creole, the most widely used spoken language in Mauritius. Diachronically,
there are at least two stages of infuence: (i) Bhojpuri had some infuence
on Creole from the 1830s onwards, and (ii) the two languages continue to
interact until today, the result of which is the possibility of even more
evident diversifcation in the diastratic and diatopic architecture of the
creole. Te more general issue is narrowed down to an analysis of a
particular semantic-grammatical category: ablative Path expression.
In this article, we defend the hypothesis that the use of depi as an ablative
marker is essentially due to Bhojpuri infuence. Additionally, certain kinds
of uses of depi as an ablative marker function as distinctive markers of the
variety of Mauritian Creole spoken by Indians. Tis study draws on data
from recent research carried out in Mauritius, as well as the analysis of a
corpus of written texts and spoken discourse, recorded and transcribed
by the authors.
*
We are grateful to Shelly Ann Meade for very useful comments on this paper and help in
translation. We also would like to thank our Bhojpuri speaking friends Mme Gajadhur, M Ram-
dhanee and Mme Shibduth for thorough discussion of our examples.
2nd proofs
:o Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
:. Introduction
In this paper, we will examine the interaction between the expression of the notion
of path, notably the ablative relation, and certain tendencies which characterize
the linguistic ecosystem on the island of Mauritius. Tis is a multilingual envi-
ronment where French-lexifed Mauritian Creole is the main spoken language,
and exists alongside various other languages. Te present paper will focus on one
variety of Mauritian Creole which is heavily infuenced by Mauritian Bhojpuri, an
Indo-Aryan language.
i. e ecosystem: Mauritian Creole and Bhojpuri
i.: Multilingualism in Mauritius
Legally, the omcial status of languages is not stipulated in the Mauritian Constitu-
tion. But Article 49 of the 1992 Constitution treats the language of Parliament and
specifes that the omcial language of the Assembly is English, but all members
may address the presidency in French (Leclerc 2007). It should be noted that both
written and spoken English are largely restricted to the political arena, and that
English plays only a minor role in the everyday life of Mauritians. According to an
omcial census in 2000 (Atchia-Emmerich 2005: II, Appendix), the main language
spoken at home is Mauritian Creole with 826,152 speakers (70.08%) and Bhojpuri
with 142,387 speakers (12.07%), out of a total of 1,178,848 people (100%). If one
takes into account those people who declare that they have two home languages,
the fgures are 949,270 (80.52%) Creole speakers and 150,032 (12.72%) Bhojpuri
speakers. Te Ethnologue (Gordon 2005) gives a fgure of 336,000 Bhojpuri speakers,
apparently based on less recent data. According to the similar fgures in Neer-
puth (1986) and Baker & Ramnah (1985), approximately a third of the population
speaks Bhojpuri.
1
Although, according to the 2000 census, Bhojpuri is losing im-
portance, it still plays an important role in the spoken register. In this article, we
focus on the interaction between Creole and Bhojpuri, the main languages used in
the spoken register.
:. Tese accounts do not specify whether Bhojpuri is spoken as L1 or as L2. Te lack of dis-
tinction between L1 and L2 concerns all the quoted fgures and may be due to the fact that the
distinction is not specifed in the omcial censuses in which people were simply asked to specify
(1) the language usually spoken at home, and (2) the language of their forefathers. Concerning
the dimculties of interpreting the omcial censuses, see Baker & Ramnah 1988.
Encoding path in Mauritian Creole and Bhojpuri ::
But frst, we wish to make some observations on the Creole spoken by all or
the majority of Mauritians on a day-to-day basis: Recent developments reveal that
the status of Creole has changed as compared to a decade ago. We are observing an
increased use of Creole in formal contexts: a marked increase in the use of Creole
in television programmes and daily newspapers, as well as a change in attitudes
towards the introduction of Creole in education.
2
However, the use of written
Creole remains problematic: the writing system (in itself) presents an obstacle.
Te use of Creole in the written medium would seem to be considered by the
majority of the language users as going beyond a cultural taboo. Nevertheless, the
Graf-larmoni (harmonized writing system) proposed by Vinesh Hookoomsing
upon request by the Ministry of Education and Scientifc Research, was approved
by the government in September 2004 and may prove to be the solution to this
problem (Hookoomsing 2004).
i.i Bhojpuri, its varieties, and its relation to other Indian languages
Bhojpuri
3
is an Indo-Aryan language of the Eastern Zone (Shukla 1981)
4
and is
in no way a dialect or corrupted form of Hindi, as is ofen believed. It is mainly
spoken in the western part of the state of Bihar, in the eastern part of Uttar Pradesh
and in some regions of Nepal. In India alone, there are approximately 25 million
speakers (Gordon 2005). Mesthrie (1991: 242) points out the immense regional
variation, the absence of a standard form of the language, and the constant infu-
ence from closely related varieties like Awadhi in the west, Magahi and Maithili in
the east, and more importantly, standard and regional Hindi.
Bhojpuri was exported to diferent parts of the world from the 1830s onwards
through the emigration of Indian workers as indentured labourers, frst to Mauri-
tius (from 1834), then to Guyana (from 1839), Trinidad (from 1845), South Africa
(from 1860), Surinam (from 1873) and Fiji (from 1879).
Tus indentured labourers who embarked mainly in the port of Calcutta,
but also Bombay and Madras, arrived in masses in Mauritius from 1835, the
i. See Ludwig & Henri & Bruneau-Ludwig, forthcoming.
. Te name Bhojpuri is derived from Bhojpr, once an important old town in western Bihar.
Bhoj denotes the warrior clan of rulers of the city in earlier times the Ujjaini Bhojas, while pr
is the word fortown, common throughout the country. (Mesthrie 1991: 29f).
|. Bhojpuri is ofen classifed in the Magadhan group of Indo-Aryan (Domingue 1971: 24,
Neerputh 1986: 6, for a summary of classifcation criteria of the Indic languages, see Mesthrie
1991: 9f).
2nd proofs
:i Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
omcial year of the abolition of slavery.
5
Toussaint (1974) writes that the efect of
this movement was a defnitive Indianization of Mauritius. He underlines that up
until the end of Indian immigration in 1909, the island received almost 450,000
indentured labourers, the majority of whom did not return to India. Tese in-
dentured labourers were mostly Bhojpuri speakers (Neerputh 1986: 9f; see also
Mesthrie 1991: 26), which explains the importance of this language in Mauritius.
However, it should be emphasized that there was a considerable population
of speakers of other languages from South Asia, including Dravidian languages
like Tamil and Telugu, languages which a minority of todays speakers still declare
as being their language usually spoken at home,
6
sometimes alongside Bhojpuri
(see the Population Census of the Republic of Mauritius 2000).
7
As far as the relationship between Indian Bhojpuri and the variety spoken in
Mauritius is concerned, Baker & Ramnah (1988) carry out a detailed comparison
between modern Mauritian Bhojpuri and texts of twelve diferent geographical
varieties of Indian Bhojpuri published by Grierson (1903), concluding that the
standard Magahi variety comes closest to Mauritian Bhojpuri overall but that
it is apparent that Mauritian Bhojpuri does not derive from this alone but is a
blend of several varieties of Indian Bhojpuri.
Domingue (1981) lists grammatical features where simplifcation is evident,
namely the system of honorifcs (see below) and the notion of gender. Te femi-
nine marker on the verb only continues to exist in the second person singular
past tense of certain archaic registers of Mauritian Bhojpuri. Moreover, according
to Domingue (1981: 7) Mauritian Bhojpuri has lost the optional adjective agree-
ment which exists in Indian Bhojpuri. Domingue claims, moreover, that these pro-
cesses of simplifcation are not a result of contact with Creole, but they appear
to be due to the move itself, to the geographical and historical distances which
have severed the links between the transplanted variety and its origins (1981: 52).
Baker & Ramnah (1985: 230) conclude that Mauritian Bhojpuri is not a radically
,. For details concerning the immigration of populations from the Indian sub-continent to
Mauritius in the 18th century (thus before the arrival of indentured labourers), see Vaughan
(2005).
o. Although Tamil, Telugu and various other languages of India are returned as language
usually spoken in the home in the Mauritian censuses, this does not necessarily mean that there
are any homes in which these are usually spoken (Philip Baker, personal communication).
. According to Atchia-Emmerich (2005), nowadays Bhojpuri is spoken by the majority of
Indians in Mauritius, even those who in the census indicate that they have kept their ancestral
languages, like the Tamils, the Marathis and the Telugus.
Encoding path in Mauritian Creole and Bhojpuri :
restructured form of Indian Bhojpuri and has certainly not undergone anything
comparable to the process of creolization.
8
It seems that Mauritian Bhojpuri, to
speak in terms of markedness theory, is afected by certain processes of marked-
ness reduction.
Let us consider an example on the morphological level: the system of personal
pronouns in Indian and Mauritian Bhojpuri (as part of the system of honorifcs).
Te simplifcation processes become evident if one compares both varieties of
Bhojpuri:
(1) System of personal pronouns
Indian Bhojpuri
9
Mauritian Bhojpuri
10
Mauritian Creole
1 I= ham ham mo
2 I= formal: a:p tou ou
informal: tu: to
familiar: t
I= formal: wan ou li
informal: u:
1 FB hamman (ke) hamni nu
i FB formal: a:p lo:g tou zot
informal: tuhan (ke)
FB formal: u: lo:g holog zot
informal: onn
h
an (ke)
When compared to Indian Bhojpuri, Mauritian Bhojpuri is characterized by a
system in which forms of the 2nd person and the 3rd person singular and plural
are reduced. We will return to this point in 2.3.2. Mauritian Bhojpuri is also dis-
tinguished from Indian Bhojpuri by various phonetic and morphophonological
processes/modifcations.
8. See also Mesthrie (1991: 93f) who speaks of simplifcation in the transplanted varieties.
Like Baker & Ramnah (1985) for Mauritian Bhojpuri, he underlines that the structural simplif-
cation processes in the South African variety are not evidence of pidginization and creolization
(1991: 104).
. See Shukla (1981: 76). Where several forms are noted, the frst one indicates the formal
form, the second the informal and the third, familiar.
:o. Te information in this column was compiled using data from our own studies, where
we transcribed Mauritian Bhojpuri using the writing conventions proposed by Hookoomsing
(2004). It corresponds largely to Neerpuths (1986: 27) table, apart from some divergences in the
spelling.
2nd proofs
:| Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
Concerning the question of mutual intelligibility between Hindi and Mauritian
Bhojpuri,
11
it is important to specify that many Mauritian Bhojpuri speakers are
also speakers of or have a (passive) competence in Hindi,
12
especially since younger
generations learn Hindi as heritage language at school and a majority of Mauritian
Bhojpuri speakers have access to televised programs in Hindustani (Hindi/Urdu).
Tus, there is mutual comprehension, at least among certain speakers, and equally
a permanent contact between Hindi and Bhojpuri in Mauritius.
i. Contact between Mauritian Creole and Bhojpuri
i..: Creole inuences on Bhojpuri
Generally speaking, the tendency to replace marked features by unmarked ones
(mentioned above) can be attributed either to the infuence of Creole or to
the laws of orality which show a tendency towards generalizing the unmarked
form and which also govern (historical) creolization and the use of Bhojpuri in
Mauritius.
13
However, the infuence of Creole on Bhojpuri is clearly evident in the area of
lexis. As will be shown in the following examples, Mauritian Bhojpuri makes use
of several Creole borrowings.
14
(2) a. Mauritian Bhojpuri
Bazaar se pom anye.
market 78B apple bring
::. Cf. Domingues (1981: 52) comment concerning linguistic awareness: Mauritian Bhojpuri
is so diferent from Continental Bhojpuri that its speakers are not really sure that they are, indeed,
speaking that language. In fact, most of them believe that they speak some patois of Hindi, very
much like Mauritian Creole speakers believe that they speak a patois of French. If this analysis
remains true for many older speakers, the attitudes of younger generations of speakers have
nevertheless changed insofar as they draw a distinction between Bhojpuri language of the older
generation for many, and Hindi, prestige language learnt at school.
:i. Our informants clarify that those who studied Hindi at school up to CPE (end of primary
education) have passive competence, i.e., they understand but do not speak Hindi. Tose who
study Hindi in high school have an active competence.
:. Tus Baker & Ramnah (1985: 219 f) mention that the function of the unmarked form of
the noun in Mauritian Bhojpuri corresponds exactly to that of the Creole Prototype. However
they are unable to determine whether this parallel is truly due to a Creole infuence.
:|. Baker & Ramnah (1985: 226) estimate some 1000 Creole words in Mauritian Bhojpuri,
cf. also Neerputh (1986: 11). Mesthrie (1991: 153) reports examples of the transfer of Creole
lexemes to South African Bhojpuri via Mauritian Bhojpuri.
Encoding path in Mauritian Creole and Bhojpuri :,
b. Mauritian Creole
Amenn enn pom depi bazar.
bring ?D:< apple 78B market
Bring an apple from the market.
Tese processes of borrowing include the use of entire phrases which seem to
become complex lexical items in Bhojpuri:
(3)
15
a. Mauritian Bhojpuri
Ham danbwa se nikla thain.
1I= forest 78B come.from 7KN.7IF.1I=
b. Mauritian Creole
Mo sorti dan bwa.
1I= come.from BE9 forest.
I come from the forest.
However, in the area of lexis, the extent to which Bhojpuri words are replaced by
Creole equivalents varies according to geographical (urban versus rural areas)
and age diferences. Evidently, urban and younger speakers use Creole words
more frequently than rural or older speakers. Tus, two sociolinguistic factors
are shown to be important when considering the evolution of the linguistic eco-
system of Mauritius: urbanization and diagenerational variation.
16
Here, we will
merely make reference to these few examples concerning lexis and will return to
the possible reciprocal infuences on morphosyntax in the following paragraph, as
well as in 3.
i..i Bhojpuri inuence on Creole
It has been shown that Bhojpuri has, for its part, lef several lexical traces on Creole.
Baker & Ramnah (1985: 226) cite a fgure of more than 300 words of Indo-Aryan
origin in Mauritian Creole, of which the majority are probably from Bhojpuri
(Kriegel & Ludwig & Henri forthcoming).
During our surveys, informants brought to light the fact that certain Bhojpuri
words have been integrated into Creole, some examples being:
17
karay (frying
:,. We are grateful to Kumari Issur, who provided us with this example and who also drew our
attention to the fact that dan-bwa as a single word translates as forest and can even be replaced
by lafore where Bhojpuri is used in contexts calling for a higher register. It would seem, then,
that Bhojpuri borrows lexical items (lexemes) from various Creole registers.
:o. For more detailed analysis of these factors, see Daz & Ludwig & Pfnder (2002).
:. See Baker (1982: 314 f.) for a methodical summary in which some of these words are
found.
2nd proofs
:o Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
pan), deksi (cooking pot), pagla (crazy/mad), douk (problem), gounga (mute/
dumb), tamasa (party).
According to Tomason & Kaufman (1988), infuence on morphosyntax is only
found in cases of relatively intense language contact. More specifcally, the pronom-
inal system of Mauritian Creole shows, even though indirectly, a possible piece of
evidence of Bhojpuri infuence.
18
Mauritian Creole diferentiates between the polite
and informal forms of the 2nd person singular, i.e. the opposition ou vs. to (see (1)).
Te French pronoun vous did not become the generalized form, expressing both
distance (vous) and proximity (tu), as it did in several French-based Creoles.
19
(4) Mauritian Creole
a. To/ou pou vini demin?
2I= CE: come tomorrow
Will you come tomorrow?
b. Ki ou/to kone ou/twa? ou/to pe rakont mwa zistwar!
what 2I= know 2I=.E8@ 2I= 7IF tell 1I=E8@ story
What do you know about this? Youre joking!
(Lit. you are telling me stories)
c. To/ou pran sa sime la to/ou al drwat mem.
2I= take :;C way :;< 2I= go straight
Turn on this street and go straight ahead.
In the context of language contact in Mauritius, the preservation of this opposi-
tion from French could be related to Bhojpuri. At frst it would seem that this
observation is paradoxical since, as has been shown in (1), Mauritian Bhojpuri has
lost the formal/informal distinction in the 2nd person singular personal pronoun.
:8. Another possible infuence from Bhojpuri is discussed in Kriegel & Michaelis (2007): Te
polyvalent item ar in modern Mauritian Creole may be a trace of the conjunction marker ar/
aur in Bhojpuri (Kriegel & Michaelis 2007: 121f). According to Corne (1983: 77 f), certain pro-
cesses of reduplication in Mauritian Creole could be indicative of another Bhojpuri trace. Corne
demonstrates that, where reduplication of numerals as in (i) is concerned, it is more likely that
this resembles Bhojpuri and Hindi strategies (see also Baker & Corne 1987: 80f), but not a result
of Malagasy infuence or Bantu languages (see Baissac 1880: 90).
(i) Li don en broke dilo kat-kat dimun.
He gives a jar of water to each of four people.
Additionally, this should not be likened to the universally widespread strategy of using redupli-
cated numbers adverbially.
:. It should however be pointed out that this opposition is not always attributable to Bhojpuri
infuence: it also exists in other French-lexifed Creoles e.g., Louisiana Creole or Guyanese
Creole and is attested in Antillean and Haitian Creole up to the early 19th century.
Encoding path in Mauritian Creole and Bhojpuri :
In fact, Mauritian Bhojpuri has the pronoun tou for the 2nd person singular,
20

used in informal as well as formal registers. Terefore phonetic proximity between
tou and to does not constitute a satisfactory explanation: If Mauritian Creole had
generalized to (< French tu) while eliminating ou (< French vous), then this ex-
planation would be plausible. However, Mauritian Creole is one of the few creoles
that preserved the opposition between the two. How should there be any relation-
ship with Bhojpuri given that this language eliminated this very opposition? We
suggest, nonetheless, that there is a possible connection: Upon closer examination,
one fnds that Bhojpuri did lose the formal pronominal form of the 2nd person
singular, but the formal-informal opposition is perfectly maintained in the verbal
system:
(5) Mauritian Bhojpuri
a. Tou kahan djat hawe?
2I= FH;F go.7IF 7KN.2I=?D<EHC7B
Where are you going? (informal)
b. Tou kahan djat hawa?
2I= FH;F go.7IF 7KN.2I=<EHC7B
Where are you going? (formal/polite)
as well as for possessives:
(6) Mauritian Bhojpuri
a. Tora sak ha?
FEII.?D<EHC7B bag :;C
Is this your bag? (informal)
b. Tohala sak ha?
FEII.<EHC7B bag :;C
Is this your bag? (formal/polite)
Te distinction between formal and informal is still marked, not on personal
pronouns, but on the verb, where a fnal -e is used for informal register and -a
for formal (see Neerputh 1986: 27f and 50f; Sewtohul 1990: 126 f). For a more
detailed conjugation table, see Baker & Ramnah 1985: 223). Tis distinction
also occurs with possessives, referred to as the genitive pronoun in the literature
(Shukla 1981: 168).
To sum up, one could state that the two languages maintain the distinction
between formal and informal forms although, in terms of morphology, this is
expressed at diferent levels.
io. Neerputh (1986: 27). Tis form also exists in Indian Bhojpuri, see Shuklas (1981) table.
2nd proofs
:8 Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
We have already mentioned the fact that several French-lexifed Creoles, notably
Seychelles Creole, which is similar to Mauritian Creole in many respects,
21
general-
ized the form derived from French vous to formal and informal registers. We could
also consider a noteworthy observation concerning another variety of exported
Bhojpuri from the context of indentured labour: according to Mesthrie (1991: 100),
South African Bhojpuri lost the distinction between formal and informal both on
the verb and on the pronoun. Tus, two varieties (other French-lexifed creoles and
South African Bhojpuri) which are similar to the two Mauritian languages lost the
formal/informal distinction. If the distinction is maintained in Mauritian Creole
as well as in Mauritian Bhojpuri, although this is manifested at diferent levels, we
propose that this could be due to the reciprocal infuence and interaction between
Mauritian Creole and Mauritian Bhojpuri. We therefore hypothesize that this is a
case of convergence between Mauritian Creole and Mauritian Bhojpuri.
. e expression of Path
.i Te semantic categories: Ablative, allative and essive
Let us frst examine the semantic categories involved in a local situation. For this
purpose, we will adopt the conceptual and terminological suggestions of Lehmann
(1992) and Talmy (1985). A local situation presents, according to Lehmann (1992),
the following structure (see also Michaelis (this volume) for its application to
creole languages): A moving or located object or Figure
22
is involved in a situation
and locally related or oriented with respect to a local region of a reference object,
which can also be called Ground. It is important to distinguish between the local
region or place of the reference object and the orientation or Path (Talmy 1985).
In this paper, we are particularly interested in the coding of Path.
23
We distinguish
the following types of Path: essive (at rest, to be at), allative (motion-to, to go
to) and ablative (motion-from, to come from) and will pay special attention to
the coding of the ablative relation in intransitive movement.
i:. Te separation of Seychelles Creole from Mauritian Creole predates large-scale Indian
immigration in Mauritius.
ii. Concerning this point, we follow Talmy (1985: 61) who gives the following defnitions:
Te terms Figure and Ground are taken from Gestalt psychology but we give them a dis-
tinct semantic interpretation here: the Figure is a moving or conceptually movable object whose
path or site is at issue; the Ground is a reference frame, or a reference point stationary within a
reference-frame with respect to which the Figures path or site is characterized.
i. Te Path (with a capital P) is the course followed or site occupied by the Figure object
with respect to the Ground object.
Encoding path in Mauritian Creole and Bhojpuri :
.i Te unmarked expression
In French, the preposition de normally encodes the ablative relation and the prep-
osition encodes the allative (and essive) relation. Tese two prepositions have
been lost during creolization and have not been replaced by other prepositions
in most varieties of Mauritian Creole. In ablative and allative contexts, the verb
combines with the Ground without any Path-denoting preposition:
Examples for the ablative relation:
(7) Mauritian Creole
a. Mo sorti Vakwa.
1I= come.from Vacoas
Im coming from Vacoas.
b. Dadi sort lakanpagn.
Grandma come.from countryside
Grandma comes from the countryside.
c. Zak sorti kot Pol.
Jacques come.from BE9 Paul
Jacques is coming from Pauls (place).
Examples for the allative relation:
(8) Mauritian Creole
a. Mo pe al Vakwa.
1I= 7IF go Vacoas
I am going to Vacoas.
b. Zak pe al kot Pol.
Jacques 7IF go BE9 Paul
Jacques is going to Pauls (place).
Te zero-expression of ablative and allative in Mauritian Creole corresponds to
the strategy employed by other French-lexifed creoles. Here we will limit our
comparison to Lesser Antillean Creole, especially Guadeloupean Creole and
Dominican Creole. Examples for the ablative relation:
(9) a. Dominican Creole (Ludwig & Telchid & Bruneau-Ludwig 2001: 39)
St bagay ki t sti lt pyi kmisyon
<E9FIJ thing H;B FIJ come.from other country goods
ki sti lt pyi.
H;B come.from other country
Tese were things which came from other countries, goods which came
from other countries.
b. Guadeloupean Creole (Ludwig & Montbrand & Poullet & Telchid 2002: 296)
Ou sti lwen.
2I= come.from far
Youve come from far.
2nd proofs
:8o Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
Examples for the allative relation:
(10) Guadeloupean Creole (G. Hazal-Massieux 1989: 207)
Mwen kal Paris.
1I= ?F<L.go Paris
I am going to Paris.
We also refer the reader to Michaelis (this volume), who works with examples
from Seychelles Creole corresponding to our examples of unmarked expressions in
(7)(10) from Mauritian Creole and Lesser Antillean Creole. In Seychelles Creole,
as in our examples (7)(10), the ablative relation and the distinction between abla-
tive and allative relation is not grammatically encoded.
Seychelles Creole
(11) Mon al dan bwa.
1I= go in forest
Im going into the forest.
(12) Mon sorti dan
24
bwa.
1I= come.from in forest
Im coming from the forest.
According to Michaelis (this volume), many creoles behave like Seychelles Creole
in that they do not express Path overtly; the appropriate interpretation has to
be inferred from the semantics of the verb. Even if we suggest, for instance, that
in Lesser Antillean Creole serial verb constructions must be considered in the
expression of Path (see 3.3), Michaeliss main hypothesis is of great signifcance
to our paper: She claims that those constructions not conforming to the encoding
strategies in the European base languages are attributable to substrate infuence.
In this paper, we will not be looking at the possible substrate languages of Mauri-
tian Creole but we will focus on a construction type current in some varieties of
Mauritian Creole which are infuenced by Bhojpuri, which can be qualifed as being
an adstrate language for Mauritian Creole. Our results will point to the same trend
as that suggested by Michaelis (this volume) and Essegby (2005). In the domain of
expressing Path, language contact seems to play a crucial role.
. Te marked expression: Comparison with Lesser Antillean Creole
When it comes to the more explicit expression of directionality (Path), we observe
that Mauritian Creole favours diferent strategies from Lesser Antillean Creole
which, once more, is used here as a point of comparison.
i|. Note that dan refers to the local region of the reference object or Ground and does not
express Path.
Encoding path in Mauritian Creole and Bhojpuri :8:
Lesser Antillean Creole can make use of prepositions, as shown by the fol-
lowing examples from Guadeloupean Creole and Dominican Creole.
(13) Guadeloupean Creole (Ludwig & Telchid & Bruneau-Ludwig 2001: 78)
Manman-y t ka voy-y f konmisyon
mother-3I=.FEII FIJ ?F<L send-3I=.E8@ do errands
a py jous Twarivy.
on foot to Trois-Rivires
His mother used to send him to run errands on foot all the way to Trois-Rivires.
(14) Dominican Creole (Ludwig & Telchid & Bruneau-Ludwig 2001: 40)
pi yo ay mont jik anho kot mouch James.
and.then 3FB go ascend to at.the.top BE9 Mister James
And they went right (all the way) up to Mr Jamess house
In addition, a widespread method of expressing directionality in traditional
basilectal Lesser Antillean Creole is the predicative serial construction, as in (15)
(allative touch) and (16) (ablative + allative) (this construction does not really
form part of younger generations competence nowadays):
(15) Guadeloupean Creole (Ludwig 1996: 248)
I pwan vwati, touch Lapwent.
3I= take car touch La Pointe
He went (all the way to) La Pointe by car.
(16) I pati Bast riv Lapwent.
3I= leave Basse-Terre arrive La Pointe
He lef Basse-Terre and got to/got as far as La Pointe.
Serial constructions in Guadeloupean Creole and Dominican Creole may also
express directionality in relation to the speakers perspective, without including a
(diferent or explicit) point of departure or arrival.
25
An example is (17):
(17) Guadeloupean Creole (Ludwig & Montbrand & Poullet & Telchid 2002: 257)
I ka plr, menn-y al!
3I= iviv weep lead-3I=.E8@ go
He/She is crying, take him/her away!
In Mauritian Creole, serial constructions only play a minimal role (see Kriegel &
Ludwig & Henri, forthcoming), whereas they play an important role in several
basi lectal varieties of French-lexifed Creoles. As for prepositional markers, the relation-
ship is reversed: they play an important role in Mauritian Creole. Particularly noteworthy
i,. One could argue that Ablative includes this explicit expression of the point of departure;
nevertheless, it is clear that constructions like (17) semantically include the idea of a nonspecifed
reference point which constitutes the point of departure for an intentional directional move-
ment (in this case, the situation of communication).
2nd proofs
:8i Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
is the frequency of occurrence of depi from, which especially characterizes the
Bhojpuri-infuenced variety, as has already been mentioned by Baker (1972: 112, 117).
.| Diferent functions and uses of depi (< depuis) in Mauritian Creole
We will now focus on the Mauritian Creole preposition depi, derived from French
depuis. It is undoubtedly the most frequently used explicit path marker in Mau-
ritian Creole, serving a variety of functions which sometimes seem incompre-
hensible if one is unaware of the contact situation between Mauritian Creole and
Mauritian Bhojpuri. We will frst list the principal functions:
.|.: Local point of departure (ablative)
Te examples in (18)(21) illustrate the most basic local use of depi, the simple
ablative (from) use.
(18) Mo papa sort depi Sesel, li nn vini pu travay dan Moris.
FEII father come.from 78B Seychelles 3I= 7IF come for work BE9 Mauritius
My father comes from the Seychelles, he came to work in Mauritius ().
(Bord la Mer 82)
(19) Kot to pou ale depi isi?
where 2I= CE: go 78B here
Where will you go from here?
(Dev Virahsawmy Hamlet 2, http://pages.intnet.mu/develog/)
(20) Pyer fnn zet so bann vye soulye depi enn pon.
Pyer 9ECFB throw FEII.3I= FB old shoe 78B ?D:< bridge
Pyer threw his old shoes of a bridge.
(21) Mo kapav tir sa depi la?
1I= can remove :;C 78B here
Can I (re)move this from there?
Depi from and dan in may also occur together, as illustrated in (22)(24).
(22) Mo nn tir mo lakle depi dan mo sak.
1I= 9ECFB take.out FEII.1I= key 78B BE9 FEII.1I= bag
I took the key out of my bag.
(23) Mo sorti depi dan lafore.
1I= come.from 78B BE9 forest
I am coming out of the forest.
(24) Li tir so linet depi dan so pos.
3I= take.out FEII glasses 78B BE9 FEII pocket.
He took his glasses out of his pocket. (Virahsawmy, Dev: Linconnsing Finalay,
http://pages.intnet.mu/develog/teatmlf.htm)
Encoding path in Mauritian Creole and Bhojpuri :8
Tese combinations of depi and dan match the conceptual structure of a local situ-
ation perfectly: Te preposition dan refers to the local region of the reference object
or Ground, and the preposition depi refers to the Path. But language contact may also
have played a determining role in creating these patterns, as we will see in 4.2.2.
.|.i Local point of departure and arrival (ablative + allative)
As expected, the preposition depi may occur together with an allative expres-
sion marked by ziska to, to express the beginning and end of a path, as seen in
(25)(27). Again, both depi and ziska can combine with dan in (see 25).
(25) Depi dan kan lao li vini ziska dan kan anba.
78B BE9 camp JEF 3I= come 7BB BE9 camp bottom
Hes going from the higher camp right down to the one at the bottom.
(Bord La Mer 82)
(26) Depi Vakwa ziska Maybour Pol inn dormi dan loto.
78B Vacoas 7BB Mahbourg Paul 9ECFB sleep BE9 car
Paul has been sleeping in the car from Vacoas to port Louis.
(27) Li nn get mwa depi lao ziska anba.
3I= 9ECFB look.at 1I=E8@ 78B top 7BB bottom
He looked at me from head to toe. (He looked at me from top to bottom.)
But it is less expected that depi should also be used to code the allative relation in
such patterns:
(28) a. Depi lao depi anba, ena bokou pou marse.
78B top 7BB bottom 7KN much to walk
(To go) from top to bottom, theres a lot of walking to do.
b. Depi sannmars depi lagar ena trak.
78B champs.de.mars 7BB gare 7KN tramc
From Champs de Mars up to the station, the tramc is jammed.
Tis rather interesting use of depi instead of ziska to mark the allative, which is
especially characteristic of the Indo-Mauritian variety of the Creole, was already
noted by Baissac (1880):
Depuis, dipis. Depuis ici jusque-l, Dipis ici zousqu-l; mais le crole disait avant quil
connt zousqu ou zisqu, jusque, au lieu de: Jai saut depuis ici jusque l, Mo t saut
dipis l, dipis l, ce qui tait plus original. (Baissac 1880: 78)
26
In Baissacs old texts we fnd an attestation of this use, see example (42).
io. From, dipis. From here to there, Dipis ici zousqu-l; but the Creole would say before he
knew zousqu or zisqu, to, instead of: I jumped from here to there, Mo t saut dipis l, dipis
l, which was in fact more original.
2nd proofs
:8| Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
.|. Temporal beginning
Like its French source word depuis, depi may express a temporal beginning.
(29) Depi katrer granmatin gardyen vinn tap laport ek
78B four.oclock morning watchman come knock door with
so gro dibwa.
FEII big wood
Since four oclock this morning, the watchman came knocking at the door
with his big stick. (Bord La Mer, 82)
(30) Depi yer li nn al Porlwi.
78B yesterday 3I= 9ECFB go Port-Louis
Since yesterday he lef for Port-Louis.
.|.| Temporal beginning and endpoint
Te combinations depi-ziska and depi-depi, which we saw in 3.4.2, can also
express temporal beginning and endpoint:
(31) depi sizer ziska witer
78B six.oclock 7BB eight.oclock
from six oclock to eight oclock
(32) depi sizer depi witer
78B six.oclock 7BB eight.oclock
from six oclock to eight oclock
.|., Temporal endpoint
Depi may also express a temporal endpoint in the absence of an expression for a
starting point:
(33) Sa trin la pou gouvernman sa, li ed depi ler.
:;C train :;< to government :;C 3I= help 7BB hour
Tis train belongs to the government, it has been useful up to today.
.|.o Extended uses
Like ablative markers of other languages, depi may be used in a range of more ab-
stract (nonspatial, nontemporal) functions. One is a human source:
(34) Me anmemtan nou pe truv osi bann priz de pozisyon
but same.time 1FB 7IF fnd too FB statement of opinion
eklere dan lapres ki kapav amenn inpe progre pu langaz
enlightened BE9 press H;B can lead some progress for language
Encoding path in Mauritian Creole and Bhojpuri :8,
Kreol ki li depi Gilbert Ahnee, J.Claude de Lestrac, () e lezot.
Creole H;B 3I= 78B Gilbert Ahnee J. Claude de Lestrac and other
But similarly/at the same time we are witnessing enlightened statements of
opinion in the press which could mean progress for the Creole language be it
from Gilbert Ahnee, J.Claude de Lestrac, Mme. Alfred, () and others. (Lalit
17/02/2004 http://www.lalitmauritius.com/)
(35) Nou nn gagn enn papye depi gouvernman.
1FB 9ECFB get ?D:< papier from government
We (have) received a document from the government.
Another use is the delimitation of an extension of patients (this is similar to
local function):
(36) Depi twa, depi to mama zot tou pou kone ar mwa.
from 2I=E8@ from FEII.2I= mother 2FB all CE: know with 1I=E8@
You and your mother, youre going to get a good beating.
(lit. From you to your mother)
In (37), we see a more abstract point of departure, and in (38) depi denotes a parti-
tive relation or origin.
(37) Mo byen satisfe pou truv li disparet depi mo lavi ()
1I= very satisfed CE: fnd 3I=E8@ disappear from FEII life
Im content to see him disappear from my life
(38) Dimoun sorti depi diferan klas sosial, diferan group etnik
people come.from from diferent class social diferent group ethnic
ek kars, diferan okipasion pe monte-desan.
with caste diferent occupation 7IF go.up-go.down
People from diferent social classes, diferent ethnic groups and castes, diferent
occupations, go and come. (Virahsawmy Dev: Prezidan Otelo, http://pages.
intnet.mu/develog/index2.htm)
.|. Depi in old texts
Except perhaps the extended uses, all the functions of depi, depi-ziska, and even depi-
depi are attested in older texts. We will only cite a few examples of the local function:
(39) Lher la foul conne a, zot sivr li a pi
when :;< crowd know this 3FB follow 3I=.E8@ by foot
dipi tou zot la vil.
78B all 3FB.FEII town
Te people heard about it, and so they lef their towns and followed him by
land. (Matthew 14, 13, translation by Anderson 1885)
2nd proofs
:8o Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
(40) () soley va vine noar, la line na pa va
soleil <KJ become black moon D;= <KJ
donne so clart, ztoal va tomb dipi dan l ciel ()
give FEII clart and star <KJ fall 78B BE9 heaven
Soon afer the trouble of those days, the sun will grow dark, the moon will no
longer shine, the stars will fall from heaven (Matthew 24, 29, translation by
Anderson 1885)
(41) Avla ki lherla mem, voal dan le tampl ti dcir an d,
voil H;B moment ?DJ veil FH;F temple FIJ tear FH;F two
dipi laho ziska anba, ()
78B top 7BB bottom
Ten the curtain hanging in the temple was torn in two from top to bottom
(Matthew 27, 51, translation by Anderson 1885)
(42) Mais so cloisons laambe l napas dibois napas plances: dipis en
but FEII partition room :;< D;= wood D;= planks 78B
haut, dipis en bas toute loison nque ne grand grand laglace mme.
top 7BB bottom all partition just ?D:< big big mirror mme
But the partition in his/her room is not made of wood or boards: from top to
bottom the partition is a big, big mirror. (Baissac 1880: 56)
|. e contribution of French and Bhojpuri to
Mauritian Creole Path expressions
|.: French infuence
French draws upon two main prepositional formulas to express Path or direction-
ality (point of departure and point of arrival): de and depuis jusqu. Te
depuis jusqu construction marks duration, extent and continuity in a local,
temporal or more abstract sense. It should, however, be stressed that the temporal
uses of depuis are much more frequent in contemporary French. It was found that
in the oral corpus CORPAIX, only 5 out of 234 occurrences of depuis are used in
a local sense (Berthille Pallaud, personal communication). However, from a dia-
chronic perspective, the primary use is the local one (see Haspelmath 1997: 139).
Tus, French, as the lexifer language during creolization and as a current
adstrate, explains:
- the basic semantic value of depi, even though depi seems unmarked when
compared to French
Encoding path in Mauritian Creole and Bhojpuri :8
- some extensions of meaning of depi-ziska (semantic contiguity in general)
- the uses of (a form of) depi for the partitive and possessive (French as an adstrate)
Te infuence of French, however, does not explain:
- the frequent use of depi as an ablative marker in Mauritian Creole; French also
acted as a lexifer language for other creoles in which, however, this marker
did not undergo extension of meaning and where serial constructions are used
more predominantly (cf. the example of Guadeloupean Creole)
- the neutralization of depi-ziska to depi-depi
- uses like depi dan
- the almost exclusive use of certain functions of depi in the Indo-Mauritian
variety of Creole (cf. 4.2)
|.i Bhojpuri infuence
|.i.: e hypothesis marking at the lectal level
We propose that Bhojpuri played a critical role in the generalization of depi as an
ablative and sometimes allative marker. Proof of the plausibility of this hypothesis
may frst be sought in lectal usages of depi:
i. Depi used as an ablative marker, side by side with the zero marker (cf. example
18: Mo papa sort depi Sesel vs. Mo papa sort Sesel) is a peculiarity of the
Indo-Mauritian variety of the Creole.
ii. As we briefy mentioned in 3.4, the tendency to replace ziska until by depi is
characteristic of the Indo-Mauritian variety of Creole, many of whose speakers
are also Bhojpuri speakers (see examples (28b) and (32)).
We will revisit this hypothesis of lectal marking in 4.2.3 afer examining data
which would explain the transmission of structural features from Bhojpuri to Creole.
|.i.i e particle se in Bhojpuri: contact with Creole
In Mauritian Bhojpuri, the postposition se (see examples (2) and (3)) is used as an
ablative marker; it is the range of functions and the frequency of use of this postpo-
sition which were the determining factors for the generalization of depi in Creole.
Te postposition se is a feature of Indian Bhojpuri and also appears in other
Bhojpuri varieties, for example South African Bhojpuri.
27
We will confne our
examples to cases where depi is used as an ablative marker.
i. For Indian Bhojpuri se, Mesthrie (1991: 262) shows a range of functions which cover a rela-
tively large semantic domain. However, illustrations are lacking: Te main postpositions are: ()
(b) se covering the roles source, comitative, instrument and force (to be designated 78B).
2nd proofs
:88 Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
(43) Indian Bhojpuri (Shukla 1981: 161)
Lalit pul
h
se ku: d-i:.
Lalit bridge from jump-2I=.C.<KJ
Lalit will jump from the bridge.
(44) South African Bhojpuri (Mesthrie 1991: 257)
Bgh jangal se nikral.
tiger jungle 78B emerge.3I=FIJ
Te tiger emerged from the jungle.
(45) Indian Bhojpuri
Ham bharat se ayal hai.
1I= India 78B come 7KN
I come from India.
(46) Indian Bhojpuri
Perre se ghir gal ba.
tree 78B escape.3I=.above go 7KN
He fell from the tree.
(47) Indian Bhojpuri
Bharat se ham Mauritius ayli.
India 78B 1I= Maurice go.1I=<;CF7IJ
I went from India to Mauritius Island.
Te marker se plays a central role in the Mauritian Bhojpuri system, and the
Creole translations (the b sentences in the following examples) clearly show the
functional equivalence with depi.
(48) a. Mauritian Bhojpuri
Ham ghar se awa thain.
1I= house 78B come 7KN.7IF.1I=
b. Mauritian Creole
Mo pe vini depi lakaz.
1I= 7IF come 78B house
Im coming from the house.
(49)
28
a. Mauritian Bhojpuri
Bazaar se pom anye.
market 78B apple bring
b. Mauritian Creole
Amenn enn pom depi bazar.
bring ?D:< apple 78B market
Bring an apple from the market.
i8. Example (49) repeats example (2).
Encoding path in Mauritian Creole and Bhojpuri :8
(50) a. Mauritian Bhojpuri
Hai sab informasion hamni ke internet se milal ba.
:;C all information 1FB E8B internet 78B get 7KN
b. Mauritian Creole
Tou sa bann linformasion la, nou nn gagn
all :;C FB information :;C 1FB 9ECFB get
li depi internet.
3I=.E8@ 78B internet
All this information we found on the internet.
Te followiing examples show point of departure and point of arrival in a single sen-
tence. Point of arrival may be expressed by the postposition le, whose use is optional.
(51) a. Mauritian Bhojpuri
Ham ego taksi leni Vacoas se sahar (le).
1I= ?D:< taxi take.1I= Vacoas 78B Port-Louis 7BB
b. Mauritian Creole
Mo nn pran enn taksi depi Vakwa ziska Porlwi.
1I= 9ECFB take ?D:< taxi 78B Vacoas 7BB Port-Louis
I took a taxi (to go from) Vacoas to Port-Louis.
(52) a. Mauritian Bhojpuri
Ronpwen se lagar (le) trak ba.
roundabout 78B station (7BB) tramc 7KN
b. Mauritian Creole
Depi ronpwin ziska lagar ena trak.
78B roundabout 7BB station 7KN tramc
From the roundabout up to the station, the tramc is jammed.
Bakers (1996) observations about the frequent confusion between depi and
ziska among Bhojpuri speakers concur with ours. Baker concludes: Bhojpuri, in
common with Hindi and other languages of northern India, has a single morpheme
corresponding to both depuis and jusqu. I thus think that the modern confusion
between these two words results primarily from interference with Bhojpuri. Our
data, however, (see examples (51) and (52)) do not correspond exactly with this
statement, given that our informants use the marker le for allative. We propose
that the confusion between depi and ziska is attributable to a much more indirect
infuence from Bhojpuri: le is phonetically similar to se and moreover, its use is
optional. Tese two factors could explain the frequent replacement of ziska by
depi, the equivalent of se, in the Indo-Mauritian varieties of Mauritian Creole.
Mauritian Bhojpuri also allows the combination of the ablative postposition se
with the locative postposition me, providing a parallel to the combination of depi
2nd proofs
:o Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
and dan that we saw in 3.4.1 (53b is very similar to 24). Here the local region is
expressed by me, and the ablative Path by se.
(53) a. Mauritian Bhojpuri
Ham apan sak me se lakle nikalni.
1I= FEII.1I= bag BE9 78B key take.out.7IF
b. Mauritian Creole
Mo nn tir mo lakle depi dan mo sak.
1I= 9ECFB take.out FEII.1I= key 78B BE9 FEII.1I= bag
I took the key out of my bag.
In 3.4.1, we explained the combined uses of depi dan in Mauritian Creole by the
conceptual structure, noting that dan refers to the local region of the reference
object and depi to path.
However, this type of use does not exist in French or, as far as we know, in any
other French-lexifed creole. Again, the combination depi dan seems to be due to
Mauritian Bhojpuri infuence.
Moreover, it can be shown, that se is also used for time reference, and that le is
optional, as when used in local contexts. Again, Bhojpuri se is completely parallel
to Creole depi.
(54) a. Mauritian Bhojpuri
age se
long.time 78B
b. Mauritian Creole
depi lontan
78B long.time
for a long time
(55) a. Mauritian Bhojpuri
chau baje se aat baje (le)
six hours 78B eight hours (7BB)
b. Mauritian Creole
depi sizer ziska witer
78B six.oclock 7BB eight.oclock
from six o clock to eight o clock
|.i. Testing our hypothesis: the use of depi in Creole not inuenced by Bhojpuri
In order to test our hypothesis, we carried out a small survey in La Gaulette, a village
known for its creoleness in all senses of the word.
29
We recorded discussions
i. In Mauritius, Creole in the social sense, refers to people of racially mixed or exclusively
African descent, as distinguished from the Indo-Mauritians.
Encoding path in Mauritian Creole and Bhojpuri ::
of about an hours length, and got data containing the examples in (5663) of this
variety of Creole, in which infuence from Bhojpuri is minimal.
When referring to ablative contexts, the speakers from La Gaulette employ
almost exclusively the unmarked form:
30
(56) Mo sipoze sa bann zanfan la sorti Kaznwayal
1I= suppose :;C FB child :;< come.from Case-Noyale
sorti Lagolet, sorti Tirivier Nwar eksetera.
come.from La.Gaulette come.from Petite.Rivire.Noire etc.
I imagine that these children come from Case Noyale, from La Gaulette,
from Petite Riviere Noire etc.
Te equivalent in an Indo-Mauritian Creole variety would be sorti depi Kaznwayal,
sorti depi Lagolet, sorti depi Tirivier Nwar.
We do not fnd occurrences of depi used instead of ziska, as illustrated in
(5759).
(57) Mem ziska aster nou trouv tou tou tou les parties politiques
even 7BB now 1FB fnd all :;< party political
Even up to now, uh, we see uh all, all all the political parties ().
(58) Zanfan bizin al lekol ziska sezan.
child must go school 7BB sixteen.years
Children must attend school up to the age of sixteen.
(59) Si ou gete langazman pe ankor kontinie ziska ler.
if 2I=.<EHC7B look campaigning 7IF still continue 7BB now
If you look more closely at it, the campaigning is still going on to this day.
In Indo-Mauritian Creole, we would ofen fnd depi aster up to now, depi sezan
up to the age of sixteen, depi ler up to now.
Otherwise, speakers from La Gaulette employ depi when it is functionally
similar to French. In (60), depi expresses local point of departure:
(60) Mo papa li kouma li nn gagn sa ti kas
FEII father I= like I= 9ECFB get :;C small cash
bisron la ou tour li inn ni depans tou
woodcutter :;< iI= fnd sc.ov, 9ECFB fnish spend all
kot laboutik, li pe sante depi kot laboutik
BE9 shop sc 7IF sing 78B BE9 shop
o. In our recordings, we did not fnd a contradicting example. However, we certainly cannot
exclude the occurrence of a depi as an ablative marker; on the other hand, an occurrence of this
would not be representative of this variety.
2nd proofs
:i Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
laba mem li pe vini.
here ?DJ 3I= 7IF come
My father, once he gets his wages as a woodcutter, youd see him spend it
all in the shop, he sings from the shop over there, he arrives.
Depi may also express a temporal point of departure/beginning:
(61) ki li gagn sa depi so nesans li
H;B 3I= get :;C 78B FEII.3I= naissance 3I=
gagn li, li ne avek sa.
get 3I=E8@ 3I= born with :;C
() that he has it from birth, he has it, he was born with it
(62) ou nn konn enn sanzman ki nn, ki nn
2I= 9ECFB know ?D:< change H;B 9ECFB
koul pei la net depi lindepandans
drown country :;< completely 78B independence
(Its as if) you had realized a change which, which completely ruined the
country since Independence.
Depi may also have extended uses, such as point of departure and point of arrival
in a semantic continuum:
(63) Li fel depi fers segonn trwaziem ziska siziem li fel mem.
3I= fail 78B frst second third 7BB sixth 3I= fail ?DJ
He failed from frst, second, third, up to the sixth he failed.
It is clear that in these examples, paralleling French, we do not fnd any trace of
infuence from Bhojpuri, either diachronically or structurally. As (63) shows, this
group of speakers seems to be more infuenced by the French adstrate. Moreover,
it seems that they use lectal marking of their particular variety in a very conscious
way; in the same discussion, they complain (in a spontaneous manner) about the
insumcient social and political recognition of the group of Creoles in Mauritius.
,. Conclusions and hypotheses
In light of the foregoing, the following conclusions can be formulated while
afording us an attempt at certain theoretical perspectives.
,.: Te structural level
It has ofen been claimed that it is atypical for creole languages to maintain prepo-
sitions from their lexifer, and that instead, in a compensatory or functionally
Encoding path in Mauritian Creole and Bhojpuri :
complementary manner, creoles employ other techniques, for example serial con-
structions, to express certain semantic notions.
Tus, in Antillean Creole the ablative denotation is either morphologically
unmarked or expressed by serial constructions. Te use of prepositions in Antil-
lean Creole would seem to be either a more recent phenomenon, due to French
infuence, or a further step in the grammaticalization of serial constructions (as for
example is the case of ba in Guadeloupean Creole, derived from the verb bailler,
but now interpreted as a preposition).
In Mauritian Creole, however, the preposition depi has been an integral part
of the creole system since the 19th century at least, while serial constructions are
rather rare. If, therefore, a complementary relationship exists between serial and
prepositional expression of the ablative, Mauritian Creole would be closer to the
French model, a fact which means that it is thus distinguished from the creole
prototype (leaving aside the potentially problematic nature of this notion).
In Mauritian Creole, and above all in Indo-Mauritian varieties, there is a ten-
dency to replace the prepositional sequence depi-ziska, which marks points of de-
parture and arrival, by depi-depi.
,.i Te sociolinguistic level
Certain uses of depi (notably depi as an ablative marker when used with verbs of
movement, as well as the tendency towards the neutralization of depi-depi) are
essentially characteristic of Indo-Mauritian speakers. It is clear that Mauritian
Creole shows distinctions at the sociolectal level, and that the Indo-Mauritian
variety serves as an identity marker of a particular social group. In the case of
the neutralization which interests us, it would seem that this is viewed as a lectal
identifer of Indo-Mauritian; this explanation only allows one to really explain
the non-occurrence of the sequence depi-depi in the variety spoken by this social
group.
,. Te interpretive level: explaining language contact in oral contextuality
We hope to have demonstrated that the two structural characteristics considered
here are to a large extent or entirely due to the infuence of Bhojpuri:
i. the generalization of the preposition depi to express ablative relations (corre-
sponding to se in Bhojpuri)
ii. semantic neutralization which is evidenced by the sequence depi-depi (cor-
responding to the sequence se-le in Bhojpuri, where the second element is
optional and phonetically close to the frst)
2nd proofs
:| Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
However, this explanation based on language contact cannot be attributed to
a single cause:
i. Te generalization of depi as an ablative marker (as compared to the unmarked
strategy, employed, for instance, in Seychelles Creole) would seem to be due to the
following processes: a disruption of unmarking and (subsequent) creolization by/
through a structural analogy between spoken French and Bhojpuri, an analogy
which afects not only this Creole, but also Mauritian Bhojpuri.
ii. Although, the explanation for the pattern depi-depi, based on Bhojpuri
contact, is plausible, it nevertheless does not sumce as a satisfactory explanation.
Te neutralization of a binary opposition which results in the generalization of
one of the two expressions is following Trubetzkoys observations a frequently-
occurring process of unmarking in the context of oral speech; it is not necessarily
explained by language contact phenomena. We therefore suggest that it is a matter
of convergence, governed in this case of Mauritian Creole by the social functions
of the language: it is sociolectal needs of identity and alterity marking which seems
to prevent the generalization of this sequence to all speakers. Tis also applies to
non-Bhojpuri speakers.
Special abbreviations
7IF aspect
?DJ intensifer
CE: modality
References
Anderson, S.H. 1885. Lvangile slon S. Matthi (dan langaz crol maurice) (Te Gospel accord-
ing to St. Matthew (in Mauritian Creole)). London: British and Foreign Bible Society.
Atchia-Emmerich, Bilkiss. 2005. La situation linguistique lIle Maurice: Les dveloppements
rcents la lumire dune enqute empirique. Ph.D. Dissertation Universitt Erlangen.
Baissac, Charles. 1880. tude sur le patois crole mauricien. Reprinted 1976. Genve: Slatkine
Reprints.
Baker, Philip. 1972. Kreol: A description of Mauritian Creole. London: C. Hurst.
Baker, Philip. 1982. Te contribution of non-Francophone immigrants to the lexicon of
Mauritian Creole. Ph.D. Dissertation, School of Oriental and African Studies, Univer-
sity of London.
Baker, Philip. 1996. On the development of certain prepositional forms in Mauritian and other
French Creoles. In Vronique, Daniel (Ed.), Matriaux pour ltude des classes grammati-
cales dans les langues creoles. Aix-en-Provence: PUP. 4159.
Encoding path in Mauritian Creole and Bhojpuri :,
Baker, Philip & Corne, Chris. 1987. Histoire sociale et crolisation La Runion et Maurice.
Revue qubcoise de linguistique thorique et applique 6(2): 7187.
Baker, Philip & Ramnah, P. 1988. Recognizing Mauritian Bhojpuri. In Barz, Richard K. & Siegel,
Jef (Eds). Language transplanted: e development of overseas Hindi. Wiesbaden: Otto
Harrassowitz.
Baker, Philip & Ramnah, Amarnath. 1985. Mauritian Bhojpuri: An Indo-Aryan language spo-
ken in a predominantly creolophone society. Papers in pidgin and creole linguistics: Pacic
Linguistics A-72(4): 21538.
Corne, Chris. 1983. Substratal refections: Te completive aspect and the distributive numerals
in Isle de France Creole. Te Reo 26: 6580.
Daz, Norma, Ludwig, Ralph & Pfnder, Stefan. 2002. Procesos lingsticos en situaciones de
contacto: Parmetros y perspectivas. In Daz, Norma, Ludwig, Ralph & Pfnder, Stefan
(Eds), La Romania americana: Procesos lingsticos en situaciones de contacto. Frankfurt:
Vervuert. 389441.
Domingue, Nicole. 1971. Bhojpuri and Creole in Mauritius: A study of linguistic interference
and its consequences in regard to synchronic variation and language change. Ph.D. Dis-
sertation, University of Texas.
Domingue, Nicole. 1981. Internal change in a transplanted language. Studies in the Linguistic
Sciences 11(2): 15160.
Essegbey, James. 2005. Te basic locative construction in Gbe languages and Surinamese cre-
oles. Journal of Pidgin and Creole Languages 20(2): 22967.
Gordon, Raymond G.J., Jr. (Ed.). 2005. Ethnologue: Languages of the world, 15
th
Edn. Dallas TX:
SIL International.
Haspelmath, Martin. 1997. From space to time: Temporal adverbials in the worlds languages
[Lincom Studies in Teoretical Linguistics 3]. Munich: Lincom.
Hazal-Massieux, Guy. 1989. La grammaticalisation des connexions. In Ludwig, Ralph (Ed.),
Les croles franais entre loral et lcrit. Tbingen: Gunter Narr. 20112.
Hookoomsing, Vinesh Y. 1986. Langue et identit ethnique: Les langues ancestrales Maurice.
Journal of Mauritian Studies 1: 11737.
Hookoomsing, Vinesh Y. 2004. A harmonized writing system for the Mauritian Creole Language:
Gra-larmoni. (16 December 2007). <http://www.gov.mu/portal/goc/educationsite/fle/
propo.doc>
Hookoomsing, Vinesh, Ludwig, Ralph & Schnepel, Burkhard (Eds). Forthcoming. Multiple Iden-
tities in Action: Mauritius and some Antillean Parallelisms. Frankfurt a. M.: Peter Lang.
Kriegel, Sibylle, Michaelis, Susanne. 2007. Conjunction and ditransitives : some functional domains
covered by avec, et, and ensemble. In Baker, Philip, Fon Sing, Guillaume (eds), e making of
Mauritian Creole. Analyses diachroniques partir des textes anciens. United Kingdom and
Sri Lanka : Westminster Creolistics Series 9. 113132.
Kriegel, Sibylle, Ludwig, Ralph & Henri, Fabiola. Forthcoming. Les rapports entre crole et
bhojpouri Maurice: Contact de langues et actes identitaires. In Hookoomsing, Vinesh y.,
Ludwig, Ralph & Schnepel, Burkhard (Eds).
Leclerc, Jacques. 2007. le Maurice. In Leclerc, Jacques. L amnagement linguistique dans le
monde. Published online: http://www.tlfq.ulaval.ca/axl/.
Lehmann, Christian. 1992. Yukatekische lokale Relatoren in typologischer Perspektive. Zeitschri
fr Phonetik, Sprachwissenscha und Kommunikationsforschung 45: 62641.
Lehmann, Christian. 1995. oughts on grammaticalization. Mnchen: Lincom. (First publica-
tion: 1982. akup 48. University of Cologne).
2nd proofs
:o Sibylle Kriegel, Ralph Ludwig & Fabiola Henri
Ludwig, Ralph, Telchid, Sylviane & Bruneau-Ludwig, Florence (Eds). 2001. Corpus crole: Textes
oraux dominicais, guadeloupens, guyanais, haitiens, mauriciens et seychellois. Hamburg:
Helmut Buske.
Ludwig, Ralph, Montbrand, Danile, Poullet, Hector & Telchid, Sylviane. 1990/2002. Diction-
naire crole. 2nd, revised edn. Paris: Servedit-Maisonneuve and ditions Jasor.
Ludwig, Ralph, Henri, Fabiola & Bruneau-Ludwig, Florence. Forthcoming. Hybridation linguis- Hybridation linguis-
tique et fonctions sociales aspects des contacts entre crole, franais et anglais Maurice.
In Hookoomsing, Vinesh, Ludwig, Ralph & Schnepel, Burkhard (Eds). Forthcoming.
Mesthrie, Rajend. 1991. Language in indenture: A sociolinguistic history of Bhojpuri-Hindi in
South Africa. Johannesburg: Witwatersrand University Press.
Neerputh, Naving C. 1986. Le systme verbal du Bhojpuri de LIle Maurice. Paris: LHarmattan.
Population Census of the Republic of Mauritius 2000. URL: http://www.gov.mu/portal/ sites/
ncb/cso/methodo.htm
Sewtohul, Goswami K. 1990. Diksyoner Kreol Bhojpuri. port Louis: Ledikasyon pu Travayer.
Shukla, Shaligram. 1981. Bhojpuri Grammar: Washington DC: Georgetown University Press.
Talmy, Leonard. 1985. Lexicalization patterns: Semantic structure in lexical forms. In Shopen,
Timothy (Ed.). Grammatical categories and the lexicon, Vol. 3. Cambridge: CUP. 57149.
Tomason, Sarah G. & Kaufman, Terrence. 1988. Language contact, creolization, and genetic
linguistics. Berkeley CA: University of California Press.
Toussaint, Auguste. 1974. Histoire de Maurice. Paris: PUF. [Collection Que sais-je?].
Vaughan, Megan. 2005. Creating the Creole island. slavery in eighteenth-century Mauritius.
Durham NC: Duke University Press.
Corpus and websites:
http://www.lalitmauritius.com/
http://pages.intnet.mu/develog/ Bord la Mer PLHDWU, le Maurice
CorpAix, research group DELIC, Aix-en-Provence.
Old texts of Mauritian Creole by Philip Baker, electronic version.
On the principled nature of the respective
contributions of substrate and superstrate
languages to a creoles lexicon
:
Claire Lefebvre
Universit du Qubec Montral
and Max-Planck-Institut fr Evolutionre Anthropologie
is paper argues that the respective contributions made to creoles
by their substrate and superstrate languages are principled, and that
they can be derived from a sound theory of creole genesis. More specically,
it is shown that the division can be derived from the nature and extent
of the main cognitive process at work in creole formation, relexication/
relabelling, and from how this process applies in the contexts where creole
languages emerge.
:. Introduction
Because they have limited access to the superstrate language, the creators of
radical creoles those that are closest to their substrate languages tend to use
their knowledge of their own language in interpreting what they hear from the
:. is paper is part of my long-term research program on the genesis of creole languages. e
research on which this paper is based was funded by SSHRCC. e content of the paper was
presented at the Creole Language Structure Between Substrates and Superstrates Conference
held in Leipzig (Germany), June 35, 2005. I am grateful to Susanne Michaelis for inviting me
to present my views on the topic, and to the Max Planck Institute for Evolutionary Anthro-
pology for its nancial contribution to this enterprise. I would like to thank the participants in
the conference for their comments and questions about my presentation, and three reviewers
for their comments and questions on an earlier version of this paper. Finally, many thanks to
Isabelle errien for helping put the material together, and to Andre Blanger for formatting
the manuscript.