by Edwin D. Floyd
It is pretty generally held that (1) poetic formulas (combinations of words, such as noun-adjective phrases) can be preserved for a very long time orally - but (2) the intervention of writing fairly soon imposes an irreversible break in the legitimacy of their use. Without much oversimplification, that is the "Parry-Lord hypothesis" concerning oral poetry, and it seems to be assumed in Watkins 1995 (How to Kill a Dragon), as he regularly takes Indo-European traditions, in India, Greece, Italy, Ireland, etc., down through a period of 200 to 300 years - but no further - after writing was first used in one area or another.
The foregoing summary may seem to correlate with the fact that, fundamentally, we learn language orally, not through any written medium. There is, however, a good deal of counter-evidence, in several different linguistic traditions, indicating a much longer stretch of vitality for traditional poetic patterns in a demonstrably literate tradition. Perhaps the most striking example is provided by Byzantine Greek. More than 1700 years after the introduction of alphabetic writing into Greece (around 750 BCE), ancient patterns such as kléos áphthiton "fame imperishable" were still being used in a highly traditional fashion in the tenth to twelfth centuries CE. The question is: How did this archaic usage maintain itself?
Part of the answer is provided by Nicolaus Callicles, poem 19 (composed around 1100 CE). At lines 41 and 44, Nicolaus uses the words des-pótēs "master" and én-don "within" in ways which correlate strongly with their etymological background ("house-lord" and "in-house" respectively). Also, in line 12, techthéntá me "me, having been born" has an extra accent. (The usual form of the participle is techthénta.) The extra accent represents a phonologically reasonable development; however, although it is discussed as a Homeric feature by ancient grammarians, this particular accentual pattern is only marginally utilized in written Greek. The extra accent, then, is more a feature of recitation than of writing; accordingly, Nicolaus' use of it adumbrates an oral component in Greek poetic recitation, from Homer (around 725 BCE) to the early 12th century CE. In his discussion of Pindar, Watkins 1995: 369 observes something comparable - but only from Homer to Pindar (about 250 years). Now, on the basis of Callicles 19 (along with other aspects of Byzantine poetry), we can extend Greek oral poetry, in some form, through another 1500 or so years.