Abstract
āMouthingsā are unvoiced pronunciations of German words, produced along with manual signs by deaf signers of DSGS. At first glance, it is reasonable to suppose that these mouthings primarily represent an intersentential code-switching to, or borrowing at the lexical level from the deaf signerās other language, German. Although lexical code-switching as well as borrowing do explain the function of many of the mouthings in the spontaneous signing of three early and three late learners of DSGS in this study, further analyses of the data have indicated that the situation is often much more complex. Because these deaf bilingualsā two languages are communicated in different modalities, the spoken word, which has a primarily denotating function in the oral language, can in the silent mouthings of sign language, where the hands are the primary channels of denotation, take on nonlexical functions as well. This is particularly the case for the early learners, for whom the uses of mouthings seem to be more thoroughly grammaticized, with a larger range of systematically used subtypes: lexically and grammatically, stylistically to mark a particular kind of signing, āconstructed speaking,ā and prosodically, to mark boundaries of grammatical phrases and prosodic groups. For lexical items involving mouthings, criteria for distinguishing between codeswitches, nonce borrowings, and established borrowings are proposed.
Introduction
Swiss German Sign Language (DSGS)
Swiss German Sign Language (Deutschschweizerische GebƤrdensprache, henceforth referred to as DSGS) is used by an estimated 4000-5000 deaf persons living in the German-speaking cantons of Switzerland. (Deaf persons living in the French and Italian-speaking cantons have their own, separate sign languages.) DSGS is composed of five related dialects; there is no standardized version of the language. The tradition of schooling deaf children only in the oral language is still quite strong in German Switzerland and has influenced not only the spoken and sign language skills of Swiss deaf persons, but also their attitudes toward these languages. Deaf adults in German Switzerland live in a bilingual situation in which one of their languages-German-is actively furthered and positively viewed both by hearing and by deaf persons, whereas their other languageāDSGSāhas only recently been exposed to much public attention, is insufficiently researched, and is still viewed as a dispensable evil by a large number of hearing authorities responsible for deaf education as well as for medical and family counseling. (cf. the Appendix for a more detailed description of the sociolinguistic situation of sign language and its deaf users in German Switzerland).
Mouthing and DSGS
The āmouthingsā discussed in this report refer to the voiceless pronunciation of German words or word parts that accompany the production of manual signs. For example, in DSGS the manual sign MUTTER1 (āmotherā) is always accompanied by the mouthed word Mutter. In the sign language sentence, the mouthed element sometimes has the same or similar meaning as a manual sign; in other cases it modifies or is completely different from the simultaneously produced manual sign.
The component called āmouthingā that is analyzed here does not include all the forms that the mouth can make in sign language, but only includes those forms that clearly stem from the spoken language. Other uses of the mouth that are not based on the spoken language, and are not part of this study, are referred to here as āmouth gesturesā and include what have been termed in the literature ānonmanual adjectives and adverbs,ā or nonmanual components of āmultichannel signs.ā2
The mouthing component derived from spoken language has attracted more attention from researchers of some sign languages than of others, leading to the impression that mouthing is indeed a more important linguistic factor in European sign languages. Studies of mouthings in European sign languages include those by Schrƶder (1985) for Norwegian Sign Language, Schermer (1990) for the Sign Language of the Netherlands, Ebbinghaus & Hessmann (1994, 1995, 1996) for German Sign Language, and PimiƤ (1990) for Finnish Sign Language.
There are several possible reasons why mouthing seems more significant for DSGS than has been reported, for example, for ASL. One reason is that most Swiss German deaf persons do not customarily use finger-spelling, the principal technique for incorporating lexical items from the oral language into many other sign languages, including ASL. The lack of a standardized form of DSGS also has the consequence that, in this small country, oneās conversation partner quite likely is using a different sign dialect from oneās own. In this sociolinguistic situation, mouthing can serve as a redundant marker of a lexical meaning that might have different forms in the manual signs of other dialects. Finally, given the bilingual living situation of the Swiss German deaf population, heavily weighted as it is by educational policy and by general positive language attitudes of deaf as well as hearing persons toward the oral language, it is not surprising that mouthing of German words has become an important element in the signing of deaf Swiss Germans.
Research Questions
For this study, the principal research questions concern the status of mouthing in sign language: In what situations are mouthings instances of codeswitching between the oral and sign languages and when so used, in what environments and for what purposes does this codeswitching occur? To what extent are mouthings permanently integrated loan elements, or borrowings, in DSGS, and at what levels of the language do these borrowings function? To approach these basic questions, this study began with an analysis of data from spontaneous signing of three early and three late learners, all deaf, of DSGS. The first part of this report is therefore a description of the mouthings found in the data. In the last section, an attempt has been made to interpret these data by means of a theoretical model for code-switching and borrowing.
Subjects, Data Collection, and Coding
To investigate these questions, data was analyzed from a study of three early and three late deaf learners of DSGS.3 The comparison of the data of these two subject groups is helpful, because it clearly shows that the functions of mouthings are very dependent on who is producing them. The data used for this mouthing analysis are from the first 4 minutes of each subjectās unrehearsed short narrative about a personally experienced accident. The audience for these narratives consisted of the other five deaf subjects, most of whom already knew each other. Filming was done in the Deaf Center in Zurich, with no hearing persons present during the filming. The subjects were aware that their signed stories would be used for research purposes.
At the time of data collection, all early and late learners had been using sign language in their daily lives for more than 10 years, and all are fully integrated into the local deaf community. The early learners were all born profoundly deaf, had deaf relatives, and had attended schools for the deaf. They learned sign language from their relatives and from other children at the school for the deaf that they attended. The late learners have more varied backgrounds: Two had lost their hearing between the ages of 6 and 7; the third has been severely hard-of-hearing from birth and attended a strictly oral school for the deaf where signing was not allowed, even in the dormitories. All three late learners mastered sign language as adults, after they had completed their basic schooling. German is thus the first language and DSGS a later second language for all three.4
Two deaf coworkers (both early learners of DSGS) transcribed the German words that they thought were being mouthed by the videotaped subjects, paying attention to how the mouthed element was distributed over the flow of manual signs. When not otherwise noted, the mouthing was transcribed as a complete German word. If the mouthing was clearly a reduced form of the word, the deaf assistants noted the part of the word which they judged to have been actually produced.
Data Analyses
The small number of informants (six) and the fact that the data is composed of short anecdotes means that all the observations based on the analyses of this data are necessarily tentative. However, both informant groups, but especially the early learners, performed quite uniformly on the aspects of mouthing that were studied, which suggests that the major findings of this study could be investigated further using only early learners.
Initially, several different aspects of the mouthing data were analyzed. The four following kinds of analyses are briefly described in this section: frequency of mouthing, forms of mouthing, coordination of mouthing and manual components, and coordination of mouthing and manual meanings.
Frequency of Mouthings
Because the oral language was their first, and for many years their only, language one original hypothesis of this study was that the late learners would accompany more of their manual signs with mouthings than would the early learners and that these mouthings would primarily represent code-switching to German.
This hypothesis was disproved by a first analysis of the data that showed that both the early and the late learners accompany approximately the same, quite high, percentage of their manual signs with mouthings. Figure 1.1 shows the percentage of all manual signs that are accompanied by mouthing over the first 4-minute stretch of the spontaneous narrative signing. The three early learner subjects accompany an average of 80% of their signs with mouthings; the late learners, an average of 76%. Thus the first quantitative finding was that both groups of learners accompanied nearly four fifths of their signs with mouthings, although the late learners showed more individual variation than did the early learners.
The high percentage of signs accompanied by a mouthing component found in this study is partly due to the coding method used. Signs were counted as being accompanied by mouthing even if the mouthed element was only part of a German word accompanying two or more successive manual signs. Thus, in the signing of the early learner in Example (1) that follows, each of the six manual signs was coded as being accompanied by a mouthing or mouthing segment. In this case, however, only three mou...