Contact Linguistics. Chap

Exercise:
Morrill (1997:352) argues that a pidgin “is a new linguistic creation that may not be placed onto any existing family tree.” He therefore denies pidgin status to Sango on the grounds that it is simply a “distinct variety of Ngbandi” to which it is genetically related in the normal sense of the term. Read Pasch’s (1997) and Samarin’s (2000) descriptions of Sango and use the facts presented there to argue for or against Morrill’s position.


8. Issues of classification revisited.

The lack of consensus on the classification of the various types of contact vernaculars surveyed in this chapter is reflected in the fact that different scholars (and sometimes the same scholar) apply different labels to the same or similar languages. Thus Samarin (2000) insists that Sango is a pidgin, based on a variety of (mostly structural) criteria. Among these are:

40. 
    The drastic simplificatory changes which distinguish it from its primary source (Ngbandi)
    
41. 
    The rapidity of these changes (p. 308), and
    
42. 
    Its highly reduced vocabulary of roughly 1000 words (p. 321).
    

For Samarin, Sango is a pidgin because it is a “new” language, and “remarkably reduced by comparions with its source language” (p. 320). At the same time, he concedes that it does not meet “all the definitions of pidgins”, and “it certainly fails to have what are claimed to be typical features of pidgins in general” (ibid.). Presumably by this he means absence of features such as bound morphology, a TMA system, embedding strategies, etc, all of which Sango possesses.

This contrasts with Pasch’s view, mentioned above, that Sango, though once a pidgin, is now a “creole.” Her primary reasons for this include its elaboration and extension of use as a primary vernacular. She equates this process of “vernacularization” with “creolization.” Mufwene (1996) uses the same argument to claim that Kituba is a creole. Finally, Morrill (1997) argues that Sango is just a variety of Ngbandi that arose through somewhat imperfect SLA. How do we reconcile such different positions? The only solution, it seems, is to recognize that there are no absolute structural or non-structural criteria by which so-called “extended pidgins”, “creoles” “simplified languages” and “imperfectly learned” 2^nd language varieties can be distinguished. They all share certain processes of change and restructuring in common. They differ primarily in the extent to which one or the other process applies in their creation and development.
For these reasons, students of Contact Linguistics must be wary of terms like “pidginization”, “creolization” and the like, which imply some unitary process of change and development leading to the emergence of pidgins, creoles, etc. We have tried to establish that pidginization is a cover term for a complex of interrelated processes. Hence there is no single dimension along which pidgins, extended pidgins and simplified languages can be compared for the purpose of placing them on a continuum reflecting degrees of “pidginization.” It may be possible to compare these outcomes in terms of particular processes – degrees of reduction and simplification (to the extent that we can define these), or degrees of substratal influence, internally motivated change, etc. Intuitively, it seems possible to establish a cline of outcomes involving, for example, degrees of simplification. Prototypical pidgins might occupy the end of the scale furthest from their lexifier language, with more successfully learnt 2^nd language varieties at the other extreme, and simplified languages like Sango somewhere in the middle. But this would depend on consensus about the processes involved and the structural criteria for defining each type.
Even so, such a scale could not easily accommodate extended pidgins and similar outcomes (e.g. creoles and some indigenized varieties) characterized by heavy substratal influence. In the following chapter, we will consider the process of creole formation in the light of what we have discovered in this chapter. We will also question the concept of a uniform process of “creolization” in much the same way as we did for “pidginization.” We will also postpone until then discussion of the principles and constraints relevant to the formation of extended pidgins, since these are similar to those involved in the formation of (prototypical) creoles.


Summary and conclusion.

This chapter has tried to address issues concerning the classification, origins and development of various kinds of contact languages to which the term “pidgin” has been applied. Our approach has been essentially a conservative one, which identifies a class of ‘prototypical’ pidgins distinguished from other contact languages by a well-defined set of structural and sociolinguistic attributes. The former include a highly reduced vocabulary and grammar, while the latter include severe restrictions in range of functions and use as a marginal second language between groups of different language background.

This characterization still allows for some degree of diversity among ‘prototypical’ pidgins. There may be differences in degree of input from one or another source language to the pidgin’s lexicon or grammar. Some pidgins may draw on one primary source language for both components, while others may draw their lexicon from one primary source and model their grammar on another. Many prototypical pidgins have arisen in multi-language contact situations through a process of “tertiary hybridization”, that is, through use as a medium of interethnic communication among groups speaking different ‘substrate’ languages. But others (Russenorsk, Chinese Pidgin English, etc.) have arisen in two-language contact situations, and display the same characteristics associated with any prototypical pidgin. In short, there is no single formula for pidgin formation, and no fixed or invariant blueprint for pidgin structure. Rather than attempting to fit pidgins into a single mould, our concern should be to explain how particular configurations of social and linguistic factors promote differences in lexical and grammatical input, and in the eventual outcomes of pidgin formation.
We also cautioned against unqualified use of the term “pidginization” to describe the changes that create pidgins out of their inputs. For one thing, the term implies that there is always a single source language that is “pidginized.” But we saw that not all cases of pidgin formation involve a single source. Even more problematic is the fact that the term implies a unidimensional and unidirectional process of change. But we noted that so-called pidginization is really a complex combination of different processes of change, including reduction and simplification of input materials, internal innovation and regularization of structure, with L1 influence also playing a role. Also problematic is the notion of “degrees of pidginization”, which suggests that languages can be “pidginized” to varying extents. We saw, in fact, that the processes involved in pidgin formation are also found in other cases of language change and restructuring. But not all such situations lead to the emergence of a pidgin. Of special interest are “simplified” languages such as Yimas Pidgin and Hiri Motu. Are these cases of “pidginization”, or do they just happen to share certain processes of change with prototypical pidgins? If we are to speak of “degrees of pidginization” we must clarify which of these processes is being referred to. We then face the problem of how to measure degrees of simplification, reduction, etc.
We highlighted in particular the structural and developmental similarities between prototypical pidgins and early IL systems. While acknowledging the differences between the two types of language creation, we cannot but recognize the similarities in the component processes that characterize them. Both involve input from an external source language, interacting with input and influence from learners’ L1’s, processes of simplification, and internally motivated innovations. The strategies and processes of grammar construction in both cases are regulated by the same principles and constraints. These guide learners in their attempts to maximize ease of processing and production, to achieve economy and to establish common ground.
Finally, we distinguished prototypical pidgins from two other (somewhat less clearly defined) types of contact language – “extended pidgins” and “simplified languages.” Seen as final products, the latter two are in some sense mirror images of each other. Extended pidgins begin as prototypical pidgins and are elaborated in both lexicon and grammar when called upon to fulfill the functions of a primary vernacular. This process of elaboration draws heavily on input from the L1’s of the groups that adopt the erstwhile pidgin as a language of wider everyday communication. Hence the grammars of extended pidgins are much closer to those of their substrates than to that of their lexifier language. By contrast, simplified languages have the appearance of outcomes to which certain processes of “pidginization” have applied in much smaller measure than in the case of prototypical pidgins. Hence they represent closer approximations to their lexifier language in both vocabulary and grammar.
It remains unclear, however, whether simplified languages and extended pidgins can be so clearly differentiated in their earlier stages of development. Do simplified languages also begin as prototypical pidgins that learners create, and which they subsequently elaborate by drawing more lexical and structural input from the TL? Or are they created in the round by speakers of the TL who simplify it for the benefit of the learners? Intuitively, it would seem that both explanations apply. Only detailed examination of the sociohistorical contexts and earlier stages of development, as represented in actual samples of the relevant languages, can answer these questions with certainty.
What we do know of the history of extended pidgins and simplified languages reveals that changes in the social ecology of a language can result in quite significant change in the language itself. As Baker & Mühlhäusler (1990:112) note, longitudinal studies of such contact vernaculars as Chinese Pidgin English, Melanesian Pidgin and others, reveal that they go through phases of stability interspersed with periods of fluctuation and sometimes drastic breaks. Thus CPE expanded significantly from aboout 1830 on, when its use first spread beyond trading to interactions between Europeans and Chinese in more fixed domestic master/servant contexts. Changes in the social circumstances of their use also led to drastic elaborative change in all varieties of Melanesian Pidgin, when they became the primary media of inter-ethnic communication in their home territories. Today, languages like Bislama and Tok Pisin are recognized as official languages in Vanuatu and Papua New Guinea respectively. They continue to expand their resources in response to the growing demands placed on them. They demonstrate, in all stages of their history, the ways in which social factors can shape the very character of a language.