Diglossic switching and the phenomenon of ʻblendingʼ: evidence from Egypt

This paper argues on the basis of empirical data that the definition of diglossia given by Ferguson (1964) does not take account of the overlap between the Η (high) and L (low) varieties in the Arabic language. My data consists of spoken contemporary Egyptian monologues. These monologues are taken from different walks of life. I analyse four monologues. Two are political speeches, which contrary to Ferguson's expectations are a mixture of Η and L, and two are monologues by two cultured Egyptians on television. From this overlap between Η and L, springs a phenomenon I call 'blending'. Blending is different from diglossic switching and code-switching, in that it does not have a discourse function, and does not seem to be as rational an act as code-switching (cf. Myers- Scotton 1998b, Gumperz 1982). I will give five examples of blended features from H and L. This phenomenon resists interpretation within the framework of 'diglossia' and challenges ideas about code-switching. Thus, it is interesting on its own right, as a phenomenon that can explain more about both diglossia and code-switching. If one applies the idea of Myers-Scotton (1997, 1998a), of a Matrix language, then one would indeed find problems with the existence of blending, since a lot of blended forms are system morphemes. Yet, one can ap­ply Myers-Scotton's recent idea of a composite matrix language (1998a) to the Egyptian community. However, this kind of diglossic-switching may be stable and the appearance of blending does not necessarily imply language change or death. To try to explain what goes on in the Egyptian community, I will define those features which characterise blending. I propose that there are two kinds of blending, ʻidiosyncraticʼ blending and ʻintrinsicʼ blending. Selected papers on theoretical and applied linguistics, Vol 13 (2000): 13th International Symposium on Theoretical and Applied Linguisitcs, Thessaloniki 22-24 April 1999