Freitag, 14.00-14.30
In this paper I will present some of the findings of a study on the development of direct and indirect speech in ancient Babylonian, in a period of 2000 years (ca. 2500 BC to 500 BC). On a synchronic level, a clear distinction between direct and indirect speech is sometimes hard to maintain. I will try to show how a synchronic fuzzy situation actually arises as the result of a historical development. In Babylonian, this historical development is largely the story of the quotative construction with the particle 'umma'. The development of the umma construction can serve as a model textbook example of a grammaticalisation process. We can follow it through a very long time span, and along a large section of the grammaticalisation cline. I will show how during a period of two millennia, what starts off as an independent whole clause (probably even with ceremonial meaning), is gradually reduced phonetically and syntactically to become a dependent quotative marker. But once having reached the status of quotative, the umma construction is not content with staying where it is, and continues its grammaticalisation journey. It is gradually extended to domains which were previously reserved for subordinate (indirect speech) complements. Thus the erstwhile purely quotative construction is extended to verbs such as hear and fear. At the same time as the extension into new domains, the quotative construction gradually looses some of the properties which distinguish it as direct speech (e.g. it no longer fully retains the deictic system of the original speaker). The result of this long development is that what started off as two constructions with clearly distinguishable roles (quotative on the one hand, and indirect complementiser on the other), later become two alternative strategies of complementation, used in more or less complementary distribution, but not along traditional lines of direct and indirect speech. In the last part of my paper, I will address the issue of the extension of quotative markers from a typological perspective. A general path for the extension of quotatives to complementisers has been suggested in the literature (Saxena 1995). The suggestion was that the development occurs along a line which follows Givons binding hierarchy of complements. (This implies e.g. that quotative constructions should first be extended to verbs like know and only later to verbs like fear). Based on the Babylonian data, and supported by evidence form Hebrew, Sanskrit and Japanese, I take issue with this conclusion, and suggest a different line of extension of quotative markers, based on the distance from a prototypical speech situation.
Saxena, A. (1995) Unidirectional grammaticalization, Sprachtypologie und Universalien-forschung 48,4:350-372
e-mail: gd116@cam.ac.uk
zum Programm der AG 11
zur alphabetischen Übersicht der Abstracts
zur zeitlichen Übersicht