Text Alignment

SunoikisisDC Digital Classics and Byzantine Studies: Session 2

Date: Monday April 15, 2024. 16:00-17:30 BST = 17:00-18:30 CEST.

Convenors: Megan Bushnell (University of London), Chiara Palladino (Furman University)

Youtube link: youtu.be/rX4aHy1v1Es

Slides: Google Slides

Outline

This session will introduce the topic of Text Alignment, particularly focusing on the alignment of texts in different languages or Translation Alignment.

In the first part, we will cover the general principles of Text Alignment as a manual and automatic task, and we will describe the related challenges in the establishment of cross-linguistic equivalences. Then, we will illustrate how alignment can be performed at two different levels of granularity for different case studies. First, we will show an example of "chunk" alignment, consisting in the creation and visualization of a dataset of aligned poetical lines from the Aeneid and its oldest Scottish translation. Second, we will show an example of "word-by-word" alignment used to compare competing English translations of the Hippolytus by Euripides.

In both cases, we will provide a short overview of the tools and techniques used to create the alignments and manipulate the resulting datasets, the different types of observations that different alignments allow, and the respective limitations. In the last part of the session, we will illustrate the Exercise required to complete this module.

Required readings

• Bushnell, M. (2021). "Reconstructing Gavin Douglas’s Translation Practice in the Eneados Using a Corpus Linguistic-Based Method", DHBenelux 3, pp. 1-25. Available: https://journal.dhbenelux.org/journal/issues/003/article-16-Bushnell.pdf.
• Panou, D. (2013). “Equivalence in Translation Theories: A Critical Evaluation, Theory and Practice.” Language Studies 3.1, pp. 1-6. Available: http://www.academypublication.com/issues/past/tpls/vol03/01/01.pdf
• Palladino, C., Shamsian, F., Yousef, T., Wright, D.J., d’Orange Ferreira, A. and dos Reis, M.F. (2023). Translation Alignment for Ancient Greek: Annotation Guidelines and Gold Standards. Journal of Open Humanities Data, 9(1), p.22. Available: https://doi.org/10.5334/johd.131.
• Palladino, C., Shamsian, F. & Yousef, T., (2022) “Using Parallel Corpora to Evaluate Translations of Ancient Greek Literary Texts. An Application of Text Alignment for Digital Philology Research”, Journal of Computational Literary Studies 1(1). Available: https://doi.org/10.48694/jcls.100

Further readings

• de Pedro, R. (1999). The Translatability of Texts: A Historical Overview. Meta, XLIV, 4, 1999. Available: http://www3.uji.es/~aferna/EA0921/4a-Translatability.pdf
• T. Yousef, C. Palladino, F. Shamsian, A. d’Orange Ferreira, M. Ferreira dos Reis (2022). "An automatic model and Gold Standard for translation alignment of Ancient Greek". Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), Marseille, 20-25 June, 5894–5905. URL: http://www.lrec-conf.org/proceedings/lrec2022/pdf/2022.lrec-1.634.pdf
• C. Palladino, M. Foradi, T. Yousef (2021). "Translation alignment for Historical Language Learning. A case study." DHQ 15.3. Available: http://www.digitalhumanities.org/dhq/vol/15/3/000563/000563.html
• T. Yousef, Translation Alignment Applied to Classical Languages (2023). PhD Thesis, University of Leipzig. Available: http://dx.doi.org/10.13140/RG.2.2.15623.57764.
• G. Crane, A. Babeu, L.M. Cerrato, et al. Beyond translation: engaging with foreign languages in a digital library. Int J Digit Libr 24, 163–176 (2023). https://doi.org/10.1007/s00799-023-00349-2.
• Copeland, R. (2009). Rhetoric, Hermeneutics, and Translation in the Middle Ages (Cambridge: CUP). Available to pu: https://www.cambridge.org/core/books/rhetoric-hermeneutics-and-translation-in-the-middle-ages/07E67BCB1CC7602BC1236AD414C42EF7.

Resources

• Digital Fragmenta Historicorum Graecorum (DFHG): https://www.dfhg-project.org/
• DFHG Automatic Translation Alignment: https://ugarit.ialigner.com/dfhg/
• Automatic Translation Alignment of Ancient Greek Texts: https://huggingface.co/UGARIT/grc-alignment
• Ugarit Translation Alignment Editor: https://ugarit.ialigner.com/
• Ugarit Automatic Alignment Tool (Beta): http://ugarit-aligner.com/
• Alignment Gold Standards for Ancient Greek (including guidelines): https://github.com/UgaritAlignment/Alignment-Gold-Standards

Sample Parallel Corpora

• DHFG: https://www.dfhg-project.org/
• Scaife Viewer: Iliad and its various translations in English: https://scaife.perseus.org/library/urn:cts:greekLit:tlg0012.tlg001/
• Bible Corpus: https://github.com/christos-c/bible-corpus/tree/master/bibles
• Sefaria (Hebrew): https://www.sefaria.org/texts

Exercise

Option 1) Select a portion of a text of your choice, in two different languages. You can use a text from the libraries provided for this session, or one of your own. Then, use AntConc to align the text at chunk level, choosing the finest unit possible (ideally, individual sentence or line).
After you have completed the chunk alignment, use Ugarit to align the same corpus at word level. You can test the Automatic Alignment Tool or simply work manually to be more accurate. Once you have finished both alignments, observe the results you obtained with the aligned datasets and consider the following questions:

• What can alignment tell you about the relationship between the two texts?
• What aspects about the two texts made alignment more or less difficult?
• What kinds of decisions did you have to make to overcome "imperfect" equivalences between sentences or words? How did you establish that there was sufficient overlap between the two texts to align them?
• What kinds of observations does each type of alignment facilitate or enable? Conversely, what aspects about the texts are obscured by each alignment?

Option 2) Select a short portion of a text of your choice, and two different translations (they can be in two different languages or in the same language). You can use a text from the libraries provided for this session, or one of your own. Then, align the two translations to the original at line level and/or at word level using the tools provided. Once you have finished your alignments, evaluate the two translations and how they mapped onto the original text:

• What can alignment tell you about the performance of an individual translation? How does it help you capture areas of expansion or omission?
• In what ways were the two translations different from each other?
• How did each translation map onto the original? How did they differ morphologically and syntactically, and what lexical choices were particularly interesting, in your opinion?
• What constitutes a "good" translation, for you? Craft a short answer based on this exercise.