Building a Tool for the Analysis of Translations: The Case of Epistemic Modality in Edgar Allan Poe's Stories

Building a Tool for the Analysis of Translations: The Case of Epistemic Modality in Edgar Allan Poe’s Stories
Zupan, Simon, University of Maribor, Slovenia, simon.zupan@uni-mb.si
Juuso, Ilkka, University of Oulu, Finland, ilkka.juuso@ee.oulu.fi
Opas-Hänninen, Lisa Lena, University of Oulu, Finland, lisa.lena.opas-hanninen@oulu.fi
This paper investigates epistemic modality in Edgar Allan Poe’s Gothic stories and their translations into Slovenian and Finnish. To facilitate this we have built a tool that allows the researcher to align the translations, search for the instances of epistemic stance and compare the original text with the translations. From the point of view of translation studies, epistemic modality is interesting in the context of Gothic stories, because it helps to create the mood of the story and any major shifts in the translation may cause a loss of the essence of the Gothic. On the other hand, the tool we have built is applicable to any investigation into a text and its translations and will, we hope, significantly aid those working in the field.

Epistemic modality has traditionally been regarded as the “manner in which the meaning of a clause is qualified so as to reflect the speaker’s judgement of the likelihood of the proposition it expresses being true” (Quirk et. al. 1992: 219). In other words, it shows the position that speakers adopt with respect to the truth value of what they are saying (hence it is also referred to as epistemic stance). As Halliday (Halliday and Matthiesen 2004) has pointed out, speakers can adopt two extreme positions; they either qualify their proposition as true (e.g. “Translation scholars are in need of efficient electronic text-analysis tools.”) or, alternatively, as not true (“Translation scholars are not in need of efficient electronic text-analysis tools.”). These two positions are referred to as positive and negative polarity, respectively. More importantly, speakers can also adopt various positions in between the two poles. They can thus claim something to be “more” true (“Translation scholars are definitely in need of efficient text-analysis tools.”) or “less” true (“Translation scholars are probably not in need of efficient text-analysis tools.”). It is precisely these intermediate positions that are strictly referred to as epistemic modality because they indicate the speaker’s (apparent) inability to ascertain the truth of the proposition they are making and consequently qualify it as polar.

This has important implications for the Gothic stories, in particular from the point of view of their translation into other languages. As Simpson (2004) has pointed out, authors of Gothic stories often employ epistemic modality to add uncertainty to them. These narratives thus abound in expressions such as “possibly”, “perhaps”, “undoubtedly”, “it might have been”, “I believe” and the like, which all indicate the protagonists’ (apparent) inability to be able to tell what was behind a particular event or experience. In turn, this makes that same event or experience appear mysterious. It also lays ground for evoking in the reader the prototypical Gothic effects such as that of discomfort, uncanniness or eeriness. Consequently, these linguistic features of the original text need to be preserved in the translation. As our preliminary research has shown, however, this is not always the case. Comparison of one of the Slovenian translations with the original has shown that the translator failed to preserve epistemic modality in some of the examples by turning them into polar sentences. As a result, those passages in the target text lost some of their potential to evoke the same Gothic effects as their corresponding passages in the original.

In order to further investigate the extent of this phenomenon, we analyse several of Edgar Allan Poe’s stories. These include The Fall of the House of Usher, Berenice, The Masque of the Red Death, Metzengerstein, and Ligeia, all of which can be seen as Gothic stories or stories with Gothic elements.

To facilitate the comparison of the source text and the target text(s), we have built ParallelTexts, a tool that allows the researcher to view two or three texts simultaneously in a synchronized manner. It makes use of XML markup, using pages and paragraphs, or other user-defined elements, for the purposes of synchronization of texts. There is support for marking up additions, deletions and other shifts in the text. The user interface shows a table-like view of the texts and allows for simultaneous scrolling of the texts. Figure 1 below shows the basic principle of the use interface, where the text in red on the left-hand side is highlighted to indicate that it is not found in the text on the right-hand side, ie the comparison version or the target text. Equally, the text in green on the right-hand side indicates that this text has been added into the target text and has no equivalent in the original text.

Fig. 1 A first version of ParallelTexts, with some basic instructions for using it

Full Size Image

With appropriate input data, ParallelTexts allows us to examine various aspects of both the original text and its translation(s). First, it helps us identify most instances of epistemic modality in the original text and allows easy access to their corresponding sentences in the translation(s). Second, the tool allows us to statistically examine the texts and determine various quantitative features, among them the number of occurrences of epistemic modality and their distribution in both the original and the translation. Finally, the quantitative data collected with ParallelTexts will also allow us to qualitatively examine individual instances of epistemic modality in both texts and determine the cumulative effect of epistemic modality translation shifts on the translation as a whole. We also expect that the tool will prove to be useful for other similar translation studies text analyses. We believe that with some modification and appropriate input data, ParallelTexts could be used to compare any type of linguistic features in the original text and its translation. We wish to demonstrate the tool at DH2011 and all feedback and suggestions for further improvement of the tool will be most welcome.

References:
Halliday, M.A.K. and Matthiessen 2004 An Introduction to Functional Grammar, Arnold London

Poe, E. A. 1982 The Complete Tales and Poems of Edgar Allan Poe, Penguin Books London

Quirk, R., Greenbaum, S., Leech, G. N., and Svartvik, J. 1992 A Comprehensive Grammar of the English Language, Longman London and New York

Simpson, P. 2004 Stylistics: a resource book for students, Routledge London and New York