Bilingualism

The role of bilingualism in the case of contact between Frisians and Middle Dutch speakers in Noord-Holland also deserves close consideration. Thomason and Kaufman (1988:67), as part of their discussion of intensity of contact as one of the factors affecting interference through borrowing, state their finding that "all cases of moderate to heavy structural borrowing... involve a group of active bilinguals who speak the source language fluently and use it regularly for at least some ordinary communicative purposes." In cases of interference through shift, however, they find that substratum interference in the target language is not a probability in cases where the substrate language users shift to the target language over a long period of constant contact leading to full bilingualism (Thomason and Kaufman 1988:119-120). They posit that the presence of many fully bilingual speaker in the shifting population leads to fewer learners' errors becoming incorporated into the language, as those speakers have a more 'correct' knowledge of the target language. Guy (1990:63) agrees; he states that in societies with large numbers of fluent bilinguals, "borrowing might be quite extensive but imposition should be low" while where there is a small number of very limited-proficiency bilinguals, "borrowing could be limited, and imposition would be extensive in the speech of a few." In his refinement of Guy's sociolinguistic types of language change, Ross (1991:126) claims that both types of change, 'borrowing' and 'imposition', "occur through the medium of bilinguals" and bases the distinction on their dominant language, which involves a question of terminology further discussed below. Van Coetsem later also recognized that bilinguals can play a unique role in language contact situations. In advancing his Neutrality Hypothesis nine years after the publication of his book, he sees bilingual speakers as neutralizing the distinction his theory draws between Receiver Language Agentivity and Target Language Agentivity as the two primary types of contact-induced language change. Since both the Receiver Language and Target Language function as the bilingual speaker's "first language", the distinction vanishes (Van Coetsem 1997). In a possibly similar situation to that of the Frankish inhabitants of Noord-Holland between the 12th and 15th centuries, Tsitsipis (1997: 385-391) discusses the political aspects of the process of language shift in his study of the Albanian community in southern Greece. Members of this group have lived in Greece for over five centuries and are all bilingual, so that even though they are descended from the immigrant rather than local population, parallels may exist with the situation of Frisian speakers in Noord-Holland. Tsitsipis introduces two processes under terms borrowed from the field of political science: that of 'subordination', whose results are subtle and slow to manifest, in contrast to 'antagonism' in which "an open questioning of the one language by the other would take place with more radical, even faster outcomes" (Tsitsipis 1997: 389).

The role played by bilingual speakers in cases of language contact and linguistic interference is explored by scholars in other areas of linguistics as well. (Discussions of multilinguals parallel those of bilinguals - once a speaker has gone beyond speaking only one language, there is apparently nothing that changes in essence with the addition of more languages. Since it is cumbersome to say "bi- and multilinguals" repeatedly, and the situation I am interested in really concerns the speakers of two languages, I shall use the term 'bilingual' with the understanding that it does not exclude multilinguals.) Weinreich (1953) mentions studies focusing on bilinguals and possible disadvantages due to speaking two or more languages. (footnote 14) It is interesting to note that while earlier studies mainly saw negative aspects to being a bilingual speaker, more recent ones have found this to be a state which leads to improved competence in solving even non-linguistic problems (Jeßner 1997: 19-24). One ongoing debate has centered on the definition of the term bilingual. Does this mean only the speaker who possesses native fluency equally in two languages, or does it also include one who has differing levels of competence? Can only formal or informal acquisition of a second language lead to "real" bilingualism, or can it be perhaps a combination of both? As Jeßner points out, most research on bilinguals suffers from being carried out in our primarily monolingual country. In many cases, the bilingual speaker is seen as "the sum of two monolinguals in one person" rather than following a "wholistic view" which sees a bilingual speaker as one who has "a specific linguistic configuration characterized by the constant interaction and coexistence of the two languages involved"(Jeßner 1997: 18). This means that only full, balanced bilinguals are considered in many studies. Different labels have been affixed to other people who use more than one language in daily life, such as 'semilinguals', 'dominant bilinguals', etc. Extremely complex gradients of bilingualism have been proposed to solve this question (Rot 1991:61-62). Furthermore, tests developed for many studies of bilingualism take no note of the communicative needs of sociocultural norms, thus trying to establish competence and its implications without considering that differing standards of competence may obtain in different social situations (Grosjean 1985:468-471, quoted in Jeßner 1997:18).

One phenomenon usually associated with bilinguals who are not entirely competent in both languages is that of code switching. This is the practice of switching from one language to another during the course of a conversation. Scholars who focus their research on code switching distinguish between different types, such as switches made in the middle of a sentence, the alternation of entire utterances, or merely inserting a few words of another language (Malmkjær 1991:62). Along with other aspects of bilingualism, code switching has until recently been seen as a rare and negative behavior, a manifestation of the speaker's lower level of competence in the language being spoken. More recent studies, however, have shown that the occurrence of code switching may occur frequently among full bilinguals and be due to other factors than the speaker's lacking the correct word in one language and resorting to another language's equivalent as a stop-gap measure. Speakers may code switch because the other language - in which they are thoroughly competent - does not actually have a word that exactly fits what they wish to say. They may code switch to indicate a changing social situation, to exclude other listeners, or even as a psychological defense mechanism (footnote 15) (Titone 1991: 440-441). The phenomenon of code switching has implications for contact-induced language change because it may be one of the specific bridges bilinguals build between their languages that have long-term effects. Two studies of core lexical borrowings, of English conjunctions borrowed into the Swedish spoken in the U.S. and into French in Ontario, suggest that code switching is the mechanism responsible for these borrowings. The conclusion drawn in these studies is that such sentence connectors may "start out as code switches... which by dint of repetition become loanwords" (Mougeon and Beniak 1991:211, quoted in Pfaff 1991: 348 - 349). Pfaff further points out that similar code switching usage may also be at the root of other borrowed items, to include even function words, plural morphemes, and structural borrowing in general (Pfaff 1991:349). This then fits the Thomason and Kaufman model of interference through borrowing mentioned above.

During the process of language shift, a great deal is going on in the concerned community. Adult speakers who grew up speaking only the local language are learning the new language with varying degrees of interest and proficiency. Children born after the language contact and the beginning of the process of shift have begun to grow up learning both languages, either with the original local language at home and the other outside of the home, or both simultaneously, depending on their parents' proficiency. These bilingual children grow into adults who interact with monolingual speakers of both languages, as well as with other bilinguals who may have differing levels of proficiency in the two languages. In such situations it is clear that, as Nemser (1991: 361) points out, "common manifestations of contact unite the processes of language convergence and foreign language acquisition." This means the convergence also of several branches of linguistic study, to include both child and adult language acquisition theories. For although children must necessarily have a higher than usual place in the process of language change in such a volatile situation, it must be clear that the adult speakers play the main role, as they always do. Nemser also echoes Bloomfield's assertion that the native speaker continues adopting and borrowing features from his fellows - "the very things which make up the infantile language learning" - to the end of his life (Bloomfield 1933: 46, 444). Studies of bilingual speakers and aspects of bilingualism must also have a place here, and perhaps ought even to take a fairly central place. For, as Jeßner ( 1997: 27) points out, the majority of the world's population is bilingual, so that research centered on the monolingual individual can be even misleading. In cases of contact-induced language change this is especially important, since the bilingual speakers can be the very locus of those changes. If, as it seems, the shift in language from Frisian to Dutch occurred over the course of three centuries in Noord-Holland, then the presence and role of bilinguals in this process must be considered. The length of time involved suggests some generations of partially or fully bilingual speakers, most likely in the cities. The countryside is more likely to have fewer bilinguals, because of a lower level of contact with Dutch speakers and less of a need or desire to communicate with them. Thomason and Kaufman's framework ties together the numbers of bilingual speakers in the shifting or borrowing populations and their level of competency with certain observed changes due to such language contact. I propose to apply this framework "backwards": using the observed changes resulting from medieval Hollands-Frisian contact to draw conclusions about the level of bilingualism in the province of Noord-Holland in the Middle Ages.