written by Frances Dowle (University of Oxford), recipient of a travel and fieldwork bursary from the Philological Society
With the generous support of the Philological Society’s Travel Bursary, I was able to attend the 28th International Lexical-Functional Grammar (LFG) Conference, which took place this year at the University of Rochester, NY from 21st to 24th July this year.
The funding from the Philological Society allowed me to travel in person to present my poster on complementizer-verb interactions in Welsh.
My poster sought to provide an explanation for why the modern day main clause complementizers /mi/ and /vɛ/ cannot occur with certain forms of the verb ‘be’ in Welsh (Borsley et al., 2007: 35). I argued that the behaviour of the forms in question could be explained with reference to their diachronic development (Jones, unpublished). In south Welsh dialects, the forms excluded from occurring with a main clause complementizer have clearly developed as a result of the phonological erosion of the old affirmative complementizer /ə(r)/. The erosion leads to the complementizer surviving only as the onset consonant of the vowel-initial finite ‘be’ verb forms:
(1) ər + ɔɪð > (*vɛ) rɔɪð
C.AFF + be.IMPERFECT.3SG > (*C.AFF) AFF.be.IMPERFECT.3SG
A similar process of erosion took place with the negative complementizer:
(2) nɪd + ɔɪð > dɔɪð
C.NEG + be.IMPERFECT.3SG > NEG.be.IMPERFECT.3SG
These phonological changes resulted in the development of specialised forms of the verb ‘be’: the affirmative ‘r-be’ forms and the negative ‘d-be’ forms. The old, vowel-initial forms of ‘be’ were retained only after other types of complementizer, such as os ‘if’. When new affirmative complementizers emerged in Welsh, speakers did not begin to use these complementizers with the ‘r-be’ form, even though the origin of these forms is opaque to modern speakers of the language. I argued that because forms like /rɔɪð/ occupy the same position in a clause as sequences of overt complementizers like os ‘if’ and a vowel-initial verb form, speakers can acquire the knowledge that forms like /rɔɪð/ function as both complementizer and verb. In other words, the ‘r-be’ forms like /rɔɪð/ are assigned two categories in the c-structure (tree structure in LFG): both C (complementizer) and I (inflected verb). In LFG, this dual category status of a single word can be modelled using Lexical Sharing (Wescoat 2002). Familiar principles of category assignment and complementary distribution then explain why this ‘r-be’ form cannot co-occur with either a complementizer or a finite verb (since words of the same category generally cannot cooccur outside of coordination structures). Further details can be found on the 28th International LFG Conference website (https://sas.rochester.edu/cls/lfg23/program/), where a .pdf version of my poster is available.
As well as having the opportunity to present my work, I also gained a lot from being able hear about others’ work, and from the many follow-up conversations that resulted from the huge range of interesting talks. I was particularly interested to hear about Elaine Ui Dhonnchadha’s work on implementing a computational grammar of Irish in XLE (a computational implementation of LFG). I was particularly interested to hear her proposals for modelling mutations in Irish, as this is an aspect of computational grammar development that presents unique challenges, and is also pertinent to Welsh. I left her talk feeling inspired and I hope to start exploring this aspect of LFG work more in the future. There was also a special session at the conference on the theme of lexical integrity, which is particularly relevant to my doctoral research and the poster that I presented, since Lexical Sharing is one of several recent challenges to the idea of lexical integrity that has surfaced within LFG.
I am very grateful to the ever welcoming and supportive LFG community for sharing their time, energy and expertise with me across this fantastic week. I had so many wonderful discussions, ranging from detailed theoretical topics to general advice on career development. This is truly a very special academic community.
I am incredibly grateful to the Philological Society for their generous travel bursary, without which I would not have been able to attend this wonderful conference.
References
Borsley, R., Tallerman, M. & Willis, D. (2007) The Syntax of Welsh. Cambridge: Cambridge University Press.
Jones B.M. (Unpublished) The Complementizer System in Informal Welsh. URL:https://users.aber.ac.uk/bmj/Ymchwil/complementizer_phrase3.pdf
Wescoat, M.T. (2002) On Lexical Sharing. PhD Thesis. Standford University.
The Philological Society Blog 