Loanwords in Welsh: Frequency Analysis on the Basis of Cronfa Electronaeg o Gymraeg

Abstract

The Welsh language adopted words from several languages, the most important being Latin, Norman French and English. As noted by Prof. Hildegard Tristram, the issues of English influence on the Insular Celtic languages did not receive due attention because of political undercurrents of the British Isles [Tristram 2002: 258]. The research of T.H. Parry-Williams [Parry-Williams 1924] still remains the main work on the subject. The prevailing view of the Welsh-speaking community on this topic can be seen in the name of a series of articles in the Mabon journal during the 1970s: Sut i beidio ag ysgrifennu Saesneg yn Gymraeg (How not to write English in Welsh) (e.g. [Roberts 1973]). This prescriptivism is avoided mainly in dialectal and code-switching studies, which cannot be prescriptive by definition, but still there are many issues awaiting description. In our paper we would like to present the result of our research, in which we analyse loanwords in two Welsh corpora. The first should be more precisely called a text massive, as only a part of it is available electronically yet. It consists of the 11 texts of the Mabinogi in the broader, Lady Charlotte Guest’s, sense and represents a classical sample of the Middle Welsh prose language. The second is the Bangor corpus of the Modern Welsh language. Selecting the loanwords in the top 1000 of the most frequent words in both corpora and comparing those two lists provides ideas on the English/Latin loanwords ratio in the language, their place in the whole vocabulary, and the correlation between Middle and Modern Welsh. Taking into account the less frequent loanwords allows refining the results.