Contact Linguistics. Chap

fana ‘shoot’ fana-si ‘shoot’ (tr.)

b. Solomons Pidgin (luk)luk ‘look’ luk-im ‘see’

sut ‘shoot’ sut-im ‘shoot’ (tr.)

(23) a. Kwaio mou ‘be broken’ mou-si ‘break it’

b. Solomons Pidgin birek ‘break’ (intr.) birek-em ‘break it’

This feature already existed in New South Wales Pidgin, but CEO substrate influence reinforced its use and productivity in MP.

The features described here represent only a small sample of the many parallels in grammar between MP and CEO languages. The pervasiveness of the latters’ influence led Keesing (1988:96) to observe that the grammar of MP “incorporates in a simplified way the core grammatical system more or less common to…mainly (Eastern) Oceanic languages.”


6.4.2. Internal innovation in the elaboration of MP.

The case of MP illustrates the kinds of interplay between internally-driven development and substrate influence that are typical of the elaboration of pidgin grammar. The transitivizing suffix –Vm apparently had its source in the English pronoun (h)im (and perhaps (th)em) which was often attached to verbs in the F.T.E that Europeans used with Pacific Islanders in the early 19^th century. CEO speakers reanalyzed the form as a transitivizing suffix equivalent to those in their L1’s. The functions of –Vm then expanded to match those of its equivalents in the substrates.

A similar collaboration between internal forces and L1 influence is seen in the development of MP’s pronominal system. The inventory of English-derived pronouns in early Pacific pidgin consisted only of a few forms: mi ‘I’, yu, and i (< he). Later, ol(geta) ‘they’ (< altogether) emerged, first attested in Queensland between 1842 and 1858 (Baker & Mühlhäusler 1996:558). This limited system was first expanded through internal processes of compounding, yielding new forms such as yumi (< you and me) as well as various compounds with –fela (< fellow), which was becoming increasingly productive as a derivational morpheme. Thus we find new pronouns like mi-fela, and yu-fela coined by analogy with established compounds like dis-fela (< this fellow), dat-fela (<that fellow), etc. These innovations are first attested in Queensland during the period of the labor trade, and were very probably created by CEO speakers. Eventually, this process of expansion and restructuring, employing the pidgin’s own lexical resources, resulted in an MP pronominal system that was very similar to that of CEO languages. The modern MP system distinguishes singular, dual, trial and plural numbers, and has an inclusive/exclusive opposition in the first person non-singular paradigm, just like CEO languages.
Siegel (1999:16-17) provides the following comparison of the pronominal systems of Tangoa (Vanuatu) and Bislama. Note that in Tangoa, rua and tolu are free forms meaning ‘two’ and ‘three’ respectively.

(24) The pronominal system of Tangoa (Camden 1979).

singular dual trial plural

1^st person inclusive -- eñrarua eñratolu eñra

1^st person exclusive enau kamamrua kamamtolu kamam
2^nd person egko kamimrua kamimtolu kamim
3^rd person enia enrarua enratolu enra/enira

(
25) The pronominal system of Bislama (Siegel 1999:17)

singular dual trial plural

1^st person inclusive -- yumitu(fala) yumitrifala yumi

1^st person exclusive mi mitufala mitrifala mifala
2^nd person yu yutufala yutrifala yufala
3^rd person (h)em tufala trifala olgeta
Several other features of MP syntax can be attributed more or less directly to internal grammaticalization processes, sometimes triggered or reinforced by substrate influence. These include strategies of embedding such as the use of sapos (< suppose) ’if’ to introduce conditional clauses, a se (< say) complementizer to introduce complements of verbs of assertion, etc., an all-purpose locative preposition long, also used to introduce clauses of purpose, and so on. Also of interest is the grammaticalization of tense-aspect markers such as Future bai (< by and by), Past been, Completive finis/pinis (< finish), etc. More will be said about the interplay between internal processes and substrate influence in the elaboration of pidgin TMA systems in the following chapter.
The brief overview provided in this chapter so far, suggests that the origins and development of pidgin grammar involve the same basic processes of simplification, L1 retention and internal innovation found in IL development in SLA. The difference is that the role and contribution of these three processes become increasingly attenuated in the course of SLA, as learners achieve more mastery of TL lexicon and grammar. In pidgin elaboration, by contrast, the full lexical and grammatical resources of the lexifier language are not available to learners. Hence they draw increasingly on their own L1 knowledge and on creative internal restructuring to expand the resources of the pidgin.
As Baker & Mühlhäusler (1996:578) note, this pattern of elaboration is true of many other contact situations in which contact varieties have been acquired as additional languages to serve as media of interethnic communication. This includes situations in which so-called “creoles” arose. In the following chapter, we will see that “extended pidgins” and (prototypical) creoles share much in common with regard to the circumstances and processes of their creation.


7. Simplified languages.

Our overview of pidgin formation would not be complete without a brief look at a class of contact languages which have traditionally been classified as pidgins, but which seem to be different from either prototypical or extended pidgins. These include Sango, Kituba and others in Africa, Yimas Pidgin and Hiri Motu in Papua New Guinea, and so on. On the one hand, these languages differ from prototypical pidgins in having more elaborate lexical and grammatical apparatus than the latter. On the other hand, they differ from extended pidgins in drawing far more of their grammatical resources from the lexifier language than the latter do.

The reasons for such languages being labeled ‘pidgins’ have to do with the fact that they are typically used as lingua francas for inter-group communication, and are structurally simpler by comparison with their source language. For example, according to Dutton (1997), Hiri (‘Trade’) Motu (formerly Police Motu) originated as a simplified form of Motu which native speakers of that language employed with other groups who came to visit or trade with them. It became the lingua franca of the Port Moresby area after about 1880, when groups of various ethno-linguistic backgrounds settled the area. It was also adopted for use by the first police force, established in 1890 (hence the name “Police Motu”). Later, it became the principal, though unofficial, language of administration in surrounding areas.
The fact that this language came to be employed in a broad range of functions explains why it displays a richer and more complex structure than prototypical pidgins. However, Hiri Motu is still quite simplified by comparison with Motu proper. It lacks most of the inflectional and derivational morphology of Motu, its distinction between alienable and inalienable possession, its irregular verbs, etc. in phonology, many of Motu’s phonemic oppositions have been lost, e,g,, /g/ and /V/ merge to /g/, /r/ and /l/ to /l/, etc. The vocabulary of Hiri Motu comes mostly from Motu, but some of it comes from other Papuan languages, or from pidgin English. There are also a number of non-Motu features in Hiri Motu, apparently due to substrate influence from the languages of its learners, but their sources are difficult to trace (Dutton 1997:32).
Languages like Hiri Motu are difficult to categorize because their structural characteristics, histories and use overlap with those of different types of contact vernaculars. The difficulty is compounded by the fact that we know so little about their developmental histories. Did they originate as prototypical pidgins that later became elaborated, like MP? Were they the result of “imperfect” second language acquisition, which would account for the fact that their grammar and lexicon come from one primary source? Or were they the result of deliberate simplification of their L1’s by speakers of the source language who wanted either to facilitate outsiders’ efforts to communicate with them, or to prevent them from fully learning their language? If so, how much input came from the Foreigner Talk version of the language, and what contribution did learners’ L1’s make to the grammar? All of the above explanations have been offered for one or the other of these contact languages. It may well be the case that different explanations apply to different outcomes, or that a combination of causes explains a particular outcome.
At any rate, some distinction must be made between these cases and those we have classified (on the basis of clearer sociohistorical and linguistic evidence) as either prototypical or extended pidgins. Henceforth, we will refer to languages like Hiri Motu, Yimas Pidgin etc, as “simplified languages”, that is, simplified forms of their respective source languages. Some simplified languages have names of their own. Thus Kituba (spoken in West-central Africa) is a simplified form of “ethnic Kikongo” – a label representing a cluster of Kikongo varieties such as Kiyómbe, Kimanyánga, etc. (Mufwene 1997:203, fn 3). Similarly, Sango (spoken in the Central African Republic) appears to be based primarily on Yakoma, an Ngbandi variety already established as a lingua franca of the area. There were also inputs from other Bantu languages including Kituba and Lingala (themselves ‘simplified languages’) as well as some West African languages, not to mention French.
All of these contact varieties differ from extended pidgins primarily in having far less substratal input to their grammars than the latter do. This seems to be due in part to the continuing access that learners had to (close approximations of) the TL. Higher TL input reduces the need for reliance on L1 knowledge and other kinds of creative restructuring. (See discussion in Chapters 7 and 9). This makes simplified languages somewhat more akin to cases of group SLA such as “indigenized” varieties than to extended pidgins such as Melanesian Pidgin. Indeed, Pasch (1997:260) argues that “Sango is the result of a series of significant changes, losses and innovations which occurred due to imperfect acquisition of Yakoma by second-language speakers.” Later in the same paper, however, she claims that “Sango is a creole language which has emerged out of a pidgin.” These two descriptions may not be as inconsistent as they appear to be at first glance. There are certainly no clear lines separating languages like Sango from creoles, or either of them from “imperfectly learned” second languages. As we shall see in chapter 9, some so-called creoles like Bajan and urban Guyanese are much closer approximations to their superstrate sources than more “radical” ones such as the Surinnamese creoles. The difference here also has much to do with degree of continuing access to varieties of the TL models as used by native speakers or more proficient learners. Simplified languages seem to occupy a position somewhere between intermediate creoles and second language varieties that closely approximate the TL. All of these contact vernaculars share many similar processes of development (See also the discussion in Chapter 7, Section II.7). However, that does not necessarily mean that attempts to classify them differently on both sociohistorical and structural criteria are futile.