Sadler Benjamin. A linguistic analysis of dated poems from the book of Taliesin

University of Oxford, 2018. — 327 p.The lack of surviving Welsh vernacular manuscripts from before the thirteenth century, and the general scarcity of Welsh-language material elsewhere from that early period, has meant a great deal of controversy around those texts which may be early but are found in later manuscripts. The confidence with which many texts were given early dates has disappeared, and even those texts for which early dates are still proposed meet with varying levels of scepticism. Work has been done to develop our understanding of the language's historical developments, however Haycock has remarked that it is something of an embarrassment that there exists no agreed and dependable set of linguistic criteria for pre-c. 1100 verse. This thesis takes five poems from the fourteenth-century Book of Taliesin, for which dates have been offered between the ninth and eleventh centuries, on the basis of their historical context. The same scribal hand copied four other manuscripts, the contents of which date from the twelfth and thirteenth centuries. The language of the poems is analysed, using the language of the other manuscripts as a linguistic control, informed by other scholarship, to assess the survival of archaic forms through the process of subsequent redaction to arrive at the Book of Taliesin. The main finding of the thesis is that the language of all five surviving manuscripts by the Book of Taliesin scribe presents a largely uniform picture. There are a handful of examples that do diverge from the scribal norm. If the dates given to the poems are accepted, the evidence points to extensive redaction, removing those forms that had become unfamiliar or out-of-date. In this scenario, there should be no embarrassment that early texts cannot be dated on the basis of their linguistic features: the language of early texts may have been so altered as to appear contemporary to texts many centuries their junior.