The following is a reposting of excerpts from  a 2009 report by two undergraduate students of mine,  Emily Anderson and Sasha Puchalla.  As part of a course assignment, they checked the  TCP EEBO transcription of Marlowe’s Tamburlaine. They worked from a spreadsheet with a ‘verticalized’ representation of the text in which every word was a data row containing the spelling, the lemma , the part-of-speech tag, and five words of context preceding and following. This output was generated by Phil Burns’ MorphAdorner program. The students were asked to check these data against the EEBO digital page image of the source text. The technical environment was sub-optimal in many ways, but across several decades of undergraduate teaching I remember this charming piece as a wonderful example of what very bright undergraduates can do and how many fundamental philological insights and reflections are generated even by an encounter with a text in its original form even if the encounter is brief and  the “original” is only a digital surrogate.

Some of the most common corrections we made concerned either typographical errors or abbreviations. Many times, we marked ‘wd’ [the assigned symbol for faulty word division ] where an abbreviated character name was separated from the following period in the excel sheet, which might make a significant difference to anyone who is doing a search of how many times Marlowe uses a terminating period in Tamburlaine. Certainly ‘s’ and ‘f,’ as well as ‘u’ and ‘v’ were the most commonly confused letters in the transcription. Also commonly mixed up were the lowercase ‘i’ and ‘j,’ and ‘o,’ ‘c’ and ‘e.’ Most of the time the mistakes did not noticeably alter the meanings of the words, but sometimes, in the case of ‘checks’ versus ‘cheeks,’ for example, we were able to make a correction based on the context of the word.

Though the work was inherently tedious, we also found it somewhat inspiring, in that we were clarifying the work of a long-dead author. In poetry classes, we have both been taught to make everything a creative choice, even if it is as small as the choice between a comma or a period, or of capitalization. During editing, so many of the problems we ran into were discrepancies of this small nature, which made doing a careful job seem all the more important, in the effort of preserving Marlowe’s artistic choices. On the other hand, there were so many different spellings of words that it made us question whether or not such concern was even warranted. On one hand, Marlowe could have meant to vary the spellings of certain words, and if we looked more carefully, maybe we could have found some pattern in which, for instance, one could draw conclusions about different speakers depending on what spellings they utilized. On the other hand, maybe the variations were solely the careless work of the printer, or of Marlowe himself.

As we worked together, we were surprised at how frequently we could not quite reach consensus on a particular word or mark. Sometimes making a correct edit is much more difficult than recognizing the error, and personal thought-processes and understandings of words and their functions on both semantic and pragmatic levels become determining factors and points of irregularity throughout editing. Though we tend to enter this process believing we are completing an objective task in editing a text, we find that it is not uncommon for us to reach different conclusions about both seemingly menial details and clear errors. Even as we edit the digital copy, we realize that we are editing a transcribed and edited version of a work printed by an unknown someone who is presumably not the author, which means that our edits are already at least twice removed from the original text. This sparks, even further, questions about the value of digital copies of texts. In the increased convenience of a digital text, we inherently assume accuracy, but given the subjectivity of editing, the probability of error in digital texts seems to be unexpectedly high.

One thing the editing process made abundantly clear to both of us is that if all 17th Century texts look like this, it is no surprise that so many people were illiterate. The difficulty of seeing and processing the words on the page could have easily been daunting enough to deter people from reading even without the added burden of comprehending the text. We also wondered at the potential contribution of the many variable spellings of words to widespread illiteracy. Regardless of what causes these inconsistencies, one can only imagine how aggravating an irregular array of spellings for a single word could be for someone with a very basic and underdeveloped level of literacy.

For both of us, reading the text with the purpose of editing it word for word and reading the text with the purpose of taking it in on a larger scale produced extremely different understandings of the work. We recognized after having completed our edits that even having gone through the text word for word, had we not read the text before, we would not have been able to produce a coherent summary of the plot or characters. What we got from editing that we did not get from reading the text as a whole, however, was a very focused understanding and appreciation of each word and phrase on a semantic level. Instead of absorbing the general outline and broader picture of the text’s plot and characters, we found ourselves recognizing stylistic and structural nuances in the language of the text that distinguish Marlowe’s writing from that of other authors. We joke that we have “become fluent in Marlowe.”