|
In their development of a framework for studying language change, Sarah Grey Thomason and Terrence Kaufman (1988), and also Frans van Coetsem (1988), distinguish between two primary types of contact-induced language change. Both sets of authors were motivated by the use of the imprecise and ill-defined terms of 'borrowing' and 'substratum interference' pervading the literature on language contact. Both point out that different parts of linguistic subsystems tend to be transferred from one language to another depending on the circumstances of the contact, most notably on whether it is speakers of the source or target language who are the agents of language change. In my work, I will be using Thomason and Kaufman's terms: 'borrowing' and 'interference through shift' (Thomason and Kaufman 1988: 37-38).(footnote 6) They define borrowing as incorporation of foreign features into a group's native language by speakers of that language: the native, or recipient, language is maintained but is changed by addition of incorporated features. Typically, the first elements to be borrowed are lexical items. This is because "unlike other linguistic levels, lexis is an open system and the incorporation of a non-primordial word causes minimal disturbance in relations existing within the units of the subsystem it entered" (Rot 1991:31). The presence of Indian words like bungalow, chutney, and pundit in English illustrates borrowing at work (American Heritage Dictionary 1992:254, 344, 1468). Also note that only the lexical items, and not their phonology, have been borrowed: English speakers do not pronounce these words as Indians do, but impose their own phonology on them. So is it also with the borrowing of the French term déjà vu, where "many speakers of English pronounce the u ... as [u], which represents an adaptation of the French [ü] pronunciation" (Van Coetsem 1988:8). Interference through shift, on the other hand, results from imperfect group learning during a process of language shift, and in the final stage of this process, errors may then spread to the original speakers of the target language. Unlike borrowing, interference does not begin with vocabulary, but rather with sounds and syntax, and sometimes morphology as well, before words from the shifting speaker's original language appear in the target language. The English spoken in the Indian subcontinent is an example of interference, with Indian-influenced characteristics such as its use of 'progressives' instead of the plain present tense for some verbs, resulting in statements such as I am understanding it now, as well as the distinctive Indian accent with its retroflex t and d (Crystal 1988:258). Often, the target language does not adopt many words from the shifting language at all. This is only reasonable: the shifting speakers are giving up their language for another, so they are going to begin by learning that language's vocabulary. Rudimentary communication is possible using only the lexicon of a language -- this is not the case for any other subsystem. Using the target language's vocabulary, even with one's own phonology or syntax, will get the speaker's message across, however stilted or odd it may seem to a native speaker, while using the target's phonology or syntax with one's own vocabulary will be unhelpful in achieving communication. Pidgins are living illustrations of this, with their reliance on vocabulary and often highly-simplified syntax (Janzig 1999:15). The shifting speakers will then keep their own words only for things the target language has none for -- local specialties, species, and other culture-specific items. The 'substratum influence' explanation of language change is often received with suspicion -- it is so often used in a vague or careless catchall way, to explain changes for which the author has no real explanation. There are seldom stringent methodological guidelines laid down or followed. Thomason and Kaufman (1988) have tried to bring some control mechanisms into this area by setting up the following methodological limitations: "in order to make educated guesses in this area," they say, "we must be able to identify a substratum language or language group, [at least some of] whose speakers shifted to the target language at the relevant time period; we must have information about its structure; and we must have information about the structure of the target language before the shift" (Thomason and Kaufman 1988:111). We can meet these guidelines in considering Frisian-Hollands contact. The substratum language is Old Frisian, whose speakers in Noord-Holland shifted to Hollands during the Middle Dutch period. We have information about the structure of Old Frisian from a large corpus of mostly legal texts, and we also have information about the structure of Middle Dutch and even, to a limited extent, Old Low Franconian before it came into contact with Frisian. The question then arises: was it Hollands borrowing from Frisian in the 12th through 15th centuries or was it interference from the substratum of Frisian shifting to Hollands during this period that is responsible for Hollands, the dialect spoken in the province of Noord-Holland and, by its extension, the standard Dutch now spoken throughout the country? Thus formulated, however, the question is misleading. It implies that only one or the other process may be operative at a given time, while in fact both processes may be in effect simultaneously. For example, this appears to be the case with the speakers of Cushitic (a non-Semitic language) who shifted to Ethiopic Semitic (Thomason and Kaufman 1988:133-135). Many Cushitic structures can be found in the Ethiopic Semitic languages, especially those further south. Although the presence of some of these structures may be due to structural borrowing, most can be traced to the retention of learners' errors when the Cushitic speakers shifted to Ethiopic Semitic. In addition, there are a number of Cushitic loans in the Ethiopic Semitic lexicon, showing that the native Ethiopic Semitic language speakers were borrowing them from the originally Cushitic speaking population. This was also the case in England in the centuries immediately following the Norman invasion, as English borrowed words from Norman French, while the erstwhile invaders were shifting to English (Thomason and Kaufman 1988:69). And this is also the situation in the case of Dutch and Frisian. The speakers of Old Low Franconian/ Middle Dutch in the province of Noord-Holland borrowed some Frisian words from the language of the earlier inhabitants. Those originally Frisian-speaking locals, at the same time, were shifting to Middle Dutch, and taking aspects of their own language along. Together, these two processes resulted in the development of Hollands dialect, as discussed in detail below.
|