By Elizabeth Polcha, WWP Encoder and Ph.D. Candidate in English
Since late last fall, I’ve been encoding a text that poses some interesting markup challenges because of its use of Orientalist language: Scottish author Eliza Hamilton’s 1796 epistolary novel, Translation of the Letters of a Hindoo Rajah. While I was excited to encode Translation because my own research considers eighteenth-century colonial literature, I focus on Caribbean and American literature. So, as an encoder, I approached Translation with an interest in how Hamilton is using distinct language to construct colonial notions of race and gender, but with only a limited familiarity with Orientalist print culture and history.
Before I lay out the details of how I’ve been encoding linguistically distinct language in Translation, it is necessary to explain just how Orientalist (and orientalist, to use Edward Said’s version of the term) this novel is. And no, Translation is not actually a collection of letters that Hamilton translated from Hindi.1 The “translated” letters of Hamilton’s text are fictional, mostly authored by the titular character and protagonist, Zāārmilla, the Rajah of Almora. Hamilton supplements the letters with a “preliminary dissertation,” lengthy footnotes, and a glossary of terms. She strategically includes these textual addendums as a way of demonstrating her expertise in the Orientalist scholarship of her time. Also, as you can see from the macrons included on “Zāārmilla” and on another major character’s name, “Māāndāāra,” Hamilton is a fan of using diacritical marks as a kind of typographic flourish. In writing Translation, Hamilton participated in a scholarly discourse rooted in a Western imperialist fascination with Eastern Asia, citing British colonial scholarship like Nathaniel Halhed’s A Code of Gentoo Laws Or, Ordinations of the Pundits and Orientalist groups like The Asiatic Society.2
Part of our encoding process at the Women Writers Project is to begin with a preliminary document analysis. This means that once we’ve acquired a text to encode, we look through the text carefully to take note of its structure and textual features before opening up an XML file and marking up our text in TEI. During my preliminary document analysis of Translation, aside from noticing the epistolary structure and Hamilton’s unusual diacritical marks I’ve described above, I also noticed quite a few Hindi and Sanskrit terms and phrases that seemed to be roughly transliterated into English (such as “Poojah” or Pūjā, पूजा, a Sanskrit-derived word for Hindu ritual prayer). From my document analysis I knew that it would be important to look up the etymology and meaning of Hamilton’s transliterated terms in order to decide how to most accurately describe them using the TEI. My encoding practice for Translation so far has involved occasionally switching between my XML file, the Oxford English Dictionary (OED), and Google Books in determining the best way to tag specific terms and phrases.
The WWP follows the TEI Guidelines for capturing specialized language with the element <distinct>, which means that we use <distinct> to tag language that is “archaic, technical, dialectal, non-preferred.” In addition to <distinct>, <foreign> and <term> were also particularly important in my encoding of Translation. The WWP also uses the @xml:lang attribute with a value from the IANA language registry to provide standardized identifications for non-English words and phrases.3 This means that my encoding process involves paying attention to the etymology of distinct words and phrases in order to assign each <distinct> or <foreign> element an IANA language code.
For example, in the first letter in Translation, Zāārmilla refers to a character’s “Ayammi Shadee,” which Hamilton defines in a footnote as “the present made to a young woman by her relations during the period of her betrothment” (58). In determining how to encode this term, I first searched for it in the OED—which returned no results. I then searched in Google Books, which brought me to Halhed’s A Code of Gentoo Laws, Hamilton’s original source. Eventually, I determined that “Shadee” must be Hamilton’s (and Halhed’s) version of the Hindi word, shadi, or, marriage.
This term stood out to me in the text not only because it was capitalized and footnoted, but also because I did not recognize it. If Hamilton had simply used the word “Marriage” there would be no need to tag it with a more descriptive element, but because the WWP is interested in tagging non-English and linguistically distinct language, I needed to figure out the best way to encode the term. I ended up encoding “Ayammi Shadee” using the element <foreign>, which is used to tag non-English words in cases where there is not another more appropriate element, such as <name>, <persName>, or <placeName>. I also used the @xml:lang attribute with a value of “hi” for Hindi.
As in the example above, one of the challenges of marking up non-English and linguistically distinct terminology in texts like Hamilton’s Translation is that it is sometimes difficult to know when a word is being referenced in the text as a foreign language term, or when the text is using a term that has been adapted into English as a loanword. For example, the English word “pundit” is a loanword from the Sanskrit term “pandit” meaning knowledge owner, or, according to the OED, “a person with knowledge of Sanskrit and Indian philosophy, religion, and law.” So, when Halhed includes “Ordinations of the Pundits” in the title of his text, he is referring to a “pundit” as an intermediary who could clarify Indian law for colonial authorities.
It is also difficult to distinguish when a term can accurately be tagged “foreign” or “distinct” (<distinct> is the element we use for linguistically or dialectically distinct terms that are not distinct enough to constitute a ‘foreign” language), since what is considered foreign or distinct to me may not have been foreign or distinct to an eighteenth-century reader. The WWP aims to best represent the documents we encode within the context in which they were written and published, which is part of the reason why the OED is so often a valuable resource for encoders—we wouldn’t want to mark an early modern spelling of a particular word as a typographical error using the elements <sic> and <corr>, for example. But it is also important to recognize that each encoder approaches the encoding process with her own understanding of the text. My choices in marking up the term “Ayammi Shadee” are based on my understanding of the WWP’s encoding practices and my analysis of the text—and these choices will be reviewed by other encoders and may change as Translation moves through our proofing process and into final publication on Women Writers Online.
What I love about working for the WWP is the endlessly evolving way we think about markup, and the collaborative nature of the encoding process. From the many discussions I’ve had in encoding meetings with my WWP colleagues about Hamilton’s Translation, we’ve shifted slightly in our thinking about elements like <distinct>. Ultimately, the complicated way Hamilton uses Hindi- and Sanskrit-derived terms has helped me to think more critically about the linguistic complexity of eighteenth-century colonial writing.