This paper applies Matras’ (1998) pragmatic-detachability hierarchy to Texas German (TxG) in order to determine what factors have triggered the borrowing of English discourse markers (DMs) such as well, you know, and anyhow, among others, into TxG. Part one summarizes the findings of Fuller’s (2001) analysis of English DMs in Pennsylvania German (PG) to establish a mode of comparison. She argues that DMs that are non-lexical and operational in nature are high in the pragmatic-detachability hierarchy (Matras 1998). This explains why they are easily borrowed in language contact situations. Fuller shows that DMs of donor languages such as English well are the most pragmatically detachable DMs, because they are not lexical or deictic and are related to operational procedures. She concludes (2001: 367) that Pennsylvania German is “in the process of adopting an English-origin discourse-marking system.”

Part two applies Matras’ (1998) and Fuller’s (2001) findings to TxG. Based on an electronic corpus of 85 hours of sociolinguistic interviews with 70 fluent speakers of Texas German (287,000 words, recorded 2001-2005), we show that TxG speakers use ja and mal in similar pragmatic situations and similar syntactic positions as PG speakers, but to a much higher degree. Fuller’s thirty hours of interviews with eighteen PG speakers yields only a single instance of ja and fifteen instances of mal. In contrast, our TxG corpus contains more than 200 instances of each DM. Our comparison shows that although both PG and TxG have similar DMs, the two dialects differ when it comes to the frequency of German-origin DMs. Even more interesting is a comparison of Salmon’s (1990) analysis of TxG with our data. Salmons claims that German DMs doch, aber, and noch “appear to have been lost or radically reduced in function and frequency for even the most fluent speakers in central Texas.” (1990: 462) This observation, among others, leads him to conclude that his informants “have only one system, basically the American English system, having lost all but traces of the German system.” (1990: 473) However, we find hundreds of instances of each of the three DMs in our TxG corpus. In addition, other German DMs such as aber (‘but’) and weisst (‘you know’), which are higher up on Matras’ (1998) hierarchy, occur much more frequently in our corpus than in Salmons’ (1990) or Fuller’s (2001) data.

The final part of our paper addresses a number of possible explanations for these differences: (1) Degree of leveling between donor dialects; and (2) functional and distributional differences between DMs in TxG and PG (see also Salmons 1990). (3) Differences in corpus size and sampling locations. Finally, we propose a modified form of Matras’ (1998) hierarchy for TxG.


References:

• Fuller, Janet (2001): “The principle of pragmatic detachability in borrowing: English-origin
•  discourse markers in Pennsylvania German,” in: Linguistics 39: 351-369.
• Matras, Yaron (1998): “Utterance modifiers and universals of grammatical borrowing,” in:
• Linguistics 36: 281-331.
• Salmons, Joe (1990): “Bilingual discourse marking: code switching, borrowing, and convergence
• in some German-American dialects,” in: Linguistics 28: 453 – 480.