This paper introduces the ongoing ERC-funded project Chronologicon Hibernicum, which studies the diachronic developments of the Irish language between c. 550-950, and aims at refining the absolute chronology of these developments. It presents firstly the project organization, its subject matter and objective, then gives an overview of the potentials and challenges in studying the Early Irish language. The project combines historical linguistic analysis, corpus linguistic methods and Bayesian statistic tools. Finally the paper explains the impact of this project in preserving the Irish cultural heritage and the lessons learned in the first three years.

This article introduces Corpus PalaeoHibernicum (CorPH), a corpus currently consisting of 78 texts in Early Irish (c. 7th–10th cent.) created by the ERC-funded Chronologicon Hibernicum (ChronHib) project by bringing together pre-existing lexical and syntactic databases and adding further crucial texts from the period. In addition to being annotated for POS, morphological and syntactic information, another layer of annotation has been developed for CorPH – ‘Variation Tagging’, i.e. a tagset that numerically encodes synchronic language variation during the Early Irish period, thus allowing for much improved research on the chronological variation among the material. Another new pillar of studying linguistic variation is Bayesian Language Variation Analysis (BLaVA), in order to address the challenge that “not-so-big data” poses to statistical corpus methods. Instead of reflecting feature frequencies, BLaVA models language variation as probabilities of variation.

This paper intends to study the history of the Old Irish word aue ‘descendant, grandchild’ in both qualitative and quantitative approaches. The former approach tries to demonstrate what forms this word evolved into from the early Old Irish period up to the end of the Middle Irish period, and to establish the phonological changes it underwent in accordance with our present understanding of the history of the Irish language. The latter approach is based on a linguistically annotated corpus of the Annals of Ulster, and shows the distribution of variant forms of aue in relation to the period they are attested in. The discrepancy between the two observations is discussed and various hypotheses are raised to explain it.

No abstract

Background to the volume This volume is a collection of eleven chapters that showcase the state of the art in corpus-based linguistic analysis of the old, middle and early modern stages of Celtic languages (specifically, Old and Middle Irish, Middle Welsh, and Cornish). The contributors offer both new analyses of linguistic variation and change as well as descriptions of computational tools necessary to process historical language data in order to create and use electronic corpora. On the whole, the volume represents a platform for the exploration of corpus approaches to morphosyntactic variation and change in the Celtic languages and, for the first time, situates Celtic linguistics in the broader field of computational and corpus linguistics. These chapters were originally prepared for lectures hosted by the Chronologicon Hibernicum project (ChronHib), an ERC-funded project at Maynooth University, Ireland (ERC Consolidator Grant 2015, H2020 #647351). The lectures occurred at three separate workshops (December 15, 2016, April 4, 2017, October 13-14, 2017), which brought together an international group of researchers with various backgrounds to help the ChronHib team gain insight into preparing linguistically marked-up text for statistical research on language variation in Old Irish. At the first event, all aspects of corpus building and use, such as morphological tagging, syntactic parsing and maintenance and sustainability of online databases, were discussed. In subsequent events, two main themes emerged: first, the necessity of developing computational tools such as morphological taggers/analysers and lemmatisers, and second, that careful use of corpora with a focus on new search queries yields progress on previously intractable problems of Celtic morphosyntax. 2 ChronHib and CorPH The overall goal for ChronHib is to develop a statistical methodology of linguistic dating in order to more precisely date the diachronic development of the Early Irish language (Old Irish: seventh to ninth century, Middle Irish: tenth to twelfth century) and thereby to predict the age of the large number Open Access.