Methodological Translation: _Lexicons of Early Modern English_ and TEI
Three eighteenth-century dictionaries by John Kersey (1702), Nathan Bailey (1738), and Samuel Johnson (1755), are to be released for free distribution online, the first of many texts from Lexicons of Early Modern English (LEME). We believe that the linguistic and intellectual expertise needed to make all LEME dictionaries maximally useful to researchers is multidisciplinary. Thus the Creative Commons provides a suitable license, and TEI an ideal encoding system, for this scholarly exchange. Here we discuss how the LEME-encoded documents of an XML database were transformed to TEI encoding, using an automated tripartite re-encoding system designed to preserve all document content while remaining consistent with the TEI schema.
We designed a flexible and permissive XML schema in RelaxNG from a representative sample of LEME documents, and developed a preprocessor to ensure the documents were valid per XML 1.0. Heterodox elements of LEME’s schema and common typos or errors were automatically corrected in the documents using regular expressions; unusual errors were corrected by hand based on schema errors. Taking advantage of XML’s tree-like document structure, we then translated elements from LEME tags to TEI ones using a top-down approach through the “tree.” While actual content was unchanged, the translator converted XML metadata such as attributes and tags to reflect the TEI schema’s expectations. Finally, a ‘relinking’ step identified and rearranged elements within the tree to produce a valid TEI-encoded document. Using the extensive lxml library of Python, a modern high-level programming language, we are able to generate complete and valid documents in a matter of seconds.
While our needs were particular to the design of the LEME schema and the dictionaries used, the resulting software demonstrates both the reliability of modern XML tools in automatically correcting and updating documents for wider scholarly use, and the growing importance of open licensing and encoding practices in disseminating texts among researchers.
New Format, Old Tricks? Texts’ Graphic Travels
What becomes of humanities content when we think of it as both data and document? This paper brings new database technology to an old debate, with a particular focus on the way humanities content travels when it is translated between digital formats on the journey from the page to the screen. Ed Folsom and N. Katherine Hayles brought this question to the fore in as special issue of PMLA in 2007. Their articulation of what becomes of narrative when it is encapsulated in a relational database became a touchstone for the hypermedia archive-based projects that followed. While Folsom suggests that databases are anti-narrative, Hayles disagrees, positioning databases and narrative as symbionts, separate species with particular ordering logic and habits that can support one another.
The proliferation of scholarship in the following decade about other textual forms including the memo (Guillory), job printing, and Xerography (Gitelman) has expanded our understanding of the way textual works travel between formats, but what of the text as represented in the newest database formats and published online? The Web's breadth expanded though the early 2000s with the inclusion of relational-database driven texts. There is room now, however, for further analysis: in the last decade there has been a quiet revolution in the technologies that order texts and in the scale at which those texts are mobilized. Graph, as opposed to relational, databases have enjoyed a boom in the last decade, one that Hayles and Folsom could not have predicted. This paper will bring Johanna Drucker’s recent call for a reconceptualization of what text is to the examination of the life of text in graph databases, particularly digital lives of texts that are meant to connect beyond the database, with case studies of those texts' lives within both academic and lay online projects, to trace the translation (and publication) of humanities content across formats, including plain text, XML, relational databases, and graph databases on its way out into the online public sphere.
Folsom, Ed. “Database as Genre: The Epic Transformation of Archives.” PMLA, vol. 122, no. 5, Oct. 2007, pp. 1571–79.
Gitelman, Lisa. Paper Knowledge: Toward a Media History of Documents. Duke UP, 2014.
Guillory, John. “The Memo and Modernity.” Critical Inquiry, vol. 31, no. 1, pp. 108–32.
Hayles, N. Katherine. “Narrative and Database: Natural Symbionts.” PMLA, vol. 122, no. 5, 20017-10, pp. 1603-08.
Drucker, Johanna. “From A to Screen.” Comparative Textual Media: Transforming the Humanities in the Postprint Era, edited by N. Katherine Hayles and Jessica Pressman, U Minnesota P, 2013.
Patterning Chaos: Periodical Publications and the Small Worlds of Literary Translators
It is generally convened that translators have had little influence in the production process of literary translations (Buzelin 2006, Jones 2009). However, translators working in local languages have a high degree of freedom in selecting the texts to work on, which translates into a more manifest individual agency. This is especially true about poetry translation, since this literary form does not carry significant economic power in the publishing world and is generally associated with poet-translators’ personal initiatives (Bradford 2009; Galvin 2014; Jacobs 2014; etc.).
Our paper suggests that the new regimes of attention (Cronin 2016) in literary translation and a paradigm shift towards a politics of microspection (Cronin 2012) reveal a poetics of fecundity based on an economy of literary barters, literary kinship, and cosmopolitanism. This poetics of fecundity is essentially grounded in the networks established, maintained, and grown further by literary translators themselves, rather than by institutions (be they writers’ union, literary journals, or translator associations). Since in such cases translatorial action is ostensibly impossible to predict, contemporary poetry translation appears as chaos and translation projects depend on the set of unique conditions that characterize the activity of each literary translator.
In this context, our paper explores the activity of Romanian poetry translator Olimpia Iacob, largely unknown to mainstream literary media, but who has translated over fifty American and Canadian contemporary poets in various local literary journals for the past fifteen years. Since her activity is only briefly and selectively presented in her own bios and no interviews with her exist, we shall explore by means of several digital tools the translator notes that accompany the corpus of her translations published in three literary outlets between 2002 and 2017. By running the corpus through a series of concordancers and text analysis tools, our study aims to reveal her author selection pattern for each literary journal, alongside the interpersonal complex networks she established as a result of her work.
The computational analysis will involve both in-house and general-use tools for text and network analysis and visualization, which will occasion a critical comparison between the former ones—developed at University of Ottawa—and the latter, alongside a critique of the underlying algorithmic approaches. This critique will focus simultaneously on the computationality of the methods and the data they process—in terms of the structure and tractability they require of those data and the embedded politics they manage or not to uncover and/or they inform the data with—and thus reach useful conclusions regarding the prospects of text-analysis and chaos- and network-theory-based reframing of translation studies.