For the code accompanying my Ph.D. thesis, click here.

An electronic copy of the thesis is available here. There are character encoding problems on this portal—please find the abstract in pdf format here. Chapter 4 (pp. 55–104) in the thesis describes the FST implementation in great detail.

(including clarifications)

Section/page	Erratum
1.2, p. 5	consonant (clusters) consonantal digraphs
p. 62	I do not systematically include
p. 63	read read
4.3.4, p. 64	creating new mappings as part of a subsequent transducer by means of an operation known as composition.
4.3.4, p. 64	unlenited m and n as ˆM and ˆN, respectively.
footnote 12, p. 64	în ^ in
p. 65	by be by-passed
p. 65	LEXICON A) (LEXICON X)
4.4.3, p. 70 (clarification)	There is an apparent contradiction between the statement that my implementation choices have "pre-empted what would be covered in the tokenization/pre-processing stage", and the open question in 4.4.2 (p. 69) "whether morphological analysis should be preceded by tokenization, or whether both activities should be integrated as part of one model". It should be borne in mind that my morphological FST can currently only successfully deal with typographic variation (hyphens, mid-high dots or no spacing) relative to combinations of those proclitics and (monolithic) stems that have been encoded in the lexicon.
p. 74	One (or more) monolithic stems
p. 97	mut.script mutation.script
p. 97	an the initial consonants l, n and r,
p. 97	single l/n/r for these initial sounds
6.4.3, p. 131	can be linked to its lemma.
6.5, p. 134	'illumines').
6.5, p. 134	Old and Modern Irish
6.5, p. 135	the tools for modern language tools for the modern language
p. 136	dealing with non-standard language texts,
Bibliography, p. 141	Kyle P. Johnson, et al. 2014–2017 appears under 'K' but should appear under 'J', after Johnson, Charles Douglas. 1972.
Bibliography, p. 142	McCone, Kim. 2005. A first Old Irish grammar and reader: Including an introduction to mMiddle iIrish (Maynooth Medieval Irish Texts 3). Maynooth: Department of Old and Middle Irish.
Appendix A, A.1.1.6, p. 151	, (cf. section A.1.1.1)
Appendix C, C.3.1.3, p. 219	under code line 3: mut.script mutation.script

Fransen, Theodorus (2019). Past, present and future: Computational approaches to mapping historical Irish cognate verb forms (Doctoral dissertation, Trinity College Dublin, The University of Dublin).

@phdthesis{fransen2019,
Author = {Theodorus Fransen},
Title = {Past, present and future: Computational approaches to mapping historical {Irish} cognate verb forms},
Year = {2019},
School = {Trinity College Dublin, The University of Dublin},
}