Written by Beth Beattie (University of Glasgow), recipient of a travel and fieldwork bursary from the Philological Society.

The Forum for the Regional Languages of Scotland and Ulster (FRLSU) is an organisation dedicated to researching and promoting the languages used across Scotland and Ulster. It has a broad scope, covering historical and contemporary language, as well as spoken, written, and signed forms of communication. The primary languages which FRLSU works with are Irish, Scottish Gaelic, and Scots, and these languages receive a generous portion of FRLSU’s advocacy efforts. The FRLSU conference held at Ulster University from 29^th to 30^th November 2024 was the first conference held since the pandemic, and it was dedicated to the memory of the Belfast poet and critic John Hewitt with the aim of highlighting work that shines a spotlight on Scotland and Ulster’s minority languages. This is one of the aims of my own research on Older Scots, so I applied, and was accepted, to present my research at this conference and contribute towards improving the standing of Scots in academic research.

My presentation was based on a subset of my PhD research, titled ‘Establishing a Scottish Reformation Discourse: A corpus-based approach.’ There has been next to no research using corpus methods on Scottish Reformation discourse; existing research either explores the different uses of Scots and English in texts (Gribben, 2006), or focuses on the language of specific individuals (Mullen, 2021). Corpus methods are beginning to be applied to English religious discourse, but after having first been touched on over forty years ago, this methodology is only now beginning to be more thoroughly explored (Hudson, 1981; Smith, 2020; 2021; forthcoming). This paper aimed to build on this developing area of research to establish the extent to which the lexical choices of Scottish Reformation writers were different to those in England. 

To answer this question, I created a 260,000-word corpus of sixteenth-century Scottish religious polemic texts and compared it with an equivalent corpus of English texts from the same period. The creation of the Scots corpus was of particular relevance to the FRLSU conference attendees because I had to contend with technical and ideological challenges presented by the multilingual nature of sixteenth-century Scotland. Aside from those written in Latin, the surviving texts from this period are written in Older Scots and Early Modern English. These two languages have many linguistic similarities, due to both deriving from different dialects of Old English, but they are different enough to present challenges when features in the same corpus. 

There are no corpus tools that are compatible with both Scots and English. One option was to translate the Scots texts into English, which would make the entire corpus compatible with existing English-trained software. However, as is a problem with all translation, meanings of English words do not map perfectly onto Scots words. This is a significant challenge when exploring word choice. Furthermore, you have the ideological issue of effectively ‘erasing’ Scots to make way for English. The growth in Scots language awareness and activism meant that this option is not suitable within the current landscape of Scots scholarship, so I instead chose to keep the Scots texts in Scots and customise or develop software trained on Scots for pre-processing. I created a custom spelling normalisation training for VARD2 trained on Older Scots (Baron and Rayson, 2008), and I also presented preliminary information on the Older Scots part-of-speech tagger I am currently building. After this, I then grouped Scots and English together manually, which worked well because this was a short study, but I am exploring ways to do this automatically for my PhD thesis.

Scottish Keyword	Scottish rel. freq.	English rel. freq.	Log ratio	English Keyword	Scottish rel. freq.	English rel. freq.	Log ratio
Melchizedek	315.02	24.03	3.71	signify	40.39	431.01	-3.42
vocation	226.17	17.62	3.68	English	56.54	551.18	-3.29
Reformation	201.94	17.62	3.52	sense	48.46	424.6	-3.13
Calvin	452.34	43.26	3.39	believed	64.62	480.68	-2.9
debate	193.86	19.23	3.33	proved	44.43	323.66	-2.86
justly	274.63	28.84	3.25	justified	44.43	323.66	-2.86
ministry	197.9	24.03	3.04	another	76.74	546.37	-2.83
verity	553.3	78.51	2.82	works	133.28	938.92	-2.82
latter	218.09	32.05	2.77	among	96.93	578.41	-2.58
woman	791.59	121.77	2.7	Jews	72.7	434.21	-2.58
command	379.64	59.28	2.68	comes	44.43	259.57	-2.55
expressly	214.05	33.65	2.67	often	44.43	257.96	-2.54
universal	335.21	52.87	2.66	image	68.66	395.76	-2.53
punishment	234.25	40.06	2.55	needs	72.7	413.38	-2.51
Satan	266.55	46.47	2.52	saints	84.81	469.46	-2.47

Table 1: Comparison of top fifteen keywords in Scottish and English subcorpora

For exploring the lexical choices of the corpora, I identified the most frequent and statistically significant words and phrases in each corpus. The results of the lexical keyness analysis are shown in Table 1, which shows the relative frequencies and log ratio (binary log of the ratio of relative frequencies) of the keywords of each subcorpus. The Scottish texts feature higher frequencies of explicitly religious words than the English, such as religious figures like Melchizedek and Satan. Contemporary issues are also present in the Scottish corpus, demonstrated by the overuse of ‘Reformation’ and ‘Calvin.’ There are explicitly religious terms in the English corpus as well, like ‘image’ and ‘saint.’ These words play key roles in discussions surrounding the veneration of saints and their images in the Church of England. Despite both corpora having the same proportion of Catholic- and Protestant-aligned texts, Protestant-coded words like ‘Calvin’ and ‘godly’ appear more frequently in the Scottish corpus, and Catholic-coded words like ‘sacrament’ and ‘saints’ in the English. These patterns are also found in the two- and three-word phrases, indicating that there are fundamental differences in religious discourse in Scotland and England at the national level.

I received encouraging feedback on my presentation at the FRLSU conference and I was able to discuss potential avenues for further research. These include adding a qualitative angle to my research, as a mixed-method approach to this topic would help to bring together existing research and corroborate my findings so far. I was also able to discuss my research more broadly with experts in the field, and through the FRLSU network I have found a mentor for a postdoctoral research project following my PhD. Thanks to the PhilSoc travel grant, I was able to gain so much from attending this conference, and I am incredibly grateful for PhilSoc’s assistance.

References

Baron, A. and Rayson, P. (2008) VARD 2: A tool for dealing with spelling variation in historical corpora. In: Proceedings of the Postgraduate Conference in Corpus Linguistics. Aston University, Birmingham.

Collinson, P. (1983) Godly People: Essays On English Protestantism and Puritanism. History Series. London, The Hambledon Press.

Gribben, C. (2006) John Knox, Reformation History and National Self-fashioning. Reformation & Renaissance Review, 8 (1), pp. 48–66.

Hudson, A. (1981) A Lollard sect vocabulary? In: Benskin, M. and Samuels, M.L. eds. So Meny People Longages and Tonges: Philological Essays in Scots and Mediaeval English Presented to Angus McIntosh. Edinburgh, Middle English Dialect Project, pp. 15–30.

Mullan, D.G. (2021) Scottish Catholic Responses to Reformation Teachings after 1558. In: A Companion to the Reformation in Scotland, c. 1525-1638. Leiden, Brill, pp. 177–203.

Smith, J.J. (2020) Godly vocabulary in Early Modern English religious debate. In: Jonsson, E. and Larsson, T. eds. Voices past and present – Studies of involved, speech-related and spoken texts: In honor of Merja Kytö, Studies in Corpus Linguistics. Amsterdam, John Benjamins Publishing, pp. 95–112.

Smith, J.J. (2021) Lexical choices in Early Modern English devotional prose. Journal of Historical Pragmatics, 22 (2), pp. 263–281.

Yadomi, H. (2019) Language, identity and community: A sociolinguistic analysis of language practice of early modern English preachers. PhD Thesis, University of Glasgow. Available from https://theses.gla.ac.uk/id/eprint/75164.