This paper will consider the application of some NLP (Natural
Language Processing) techniques to Medieval Irish texts to
provide an alternative perspective on linguistic analyses of such
texts. Using Táin Bó Fraích as a case study, I present the
outcome of some preliminary experiments. The pilot study starts
with the creation of an annotated lexicon as a basis of automated
text analysis. Linguistic features such as part of speech
information are recorded in a machine-readable representation to
assist with subsequent linguistic analysis of this well-studied
text. Using CELT’s electronic version of Meid’s 1974 edition, I
conduct both statistical and linguistic analyses of textual features
such as sentence structure, lexical frequency and grammatical
types. I use the results of this analysis to raise some tentative
suggestions regarding Táin Bó Fraích, and in particular the
frequently noted relationship between the two distinct sections.
On this basis I hope to make some suggestions about the
potential usefulness of applying some NLP techniques to
Medieval Irish.