Jun 28, 2013

An Encounter with Bede

While looking for early English examples of omission signs and symbols called signes-de-renvoi, I came across the "St. Petersburg Bede", or "Leningrad Bede," one of the oldest manuscripts of Bede’s Historia ecclesiastica gentis Anglorum.  In the facsimile copy at the University of Saskatchewan, there are only five visible annotations in the whole manuscript  (f.11v, 21v, 68r, 89r, 90v), despite many more inline marks that would suggest a corresponding marginal note or correction.  After checking bibliographic records, it appears the manuscript was trimmed at some point in its life!  

For most people studying Bede, I’m sure these annotations are not missed much in such a clean and exemplary text for its age, which contains the oldest known example of a historiated initial. But for someone looking for annotations, it is an unfortunate loss.  The decision to trim the pages was likely for usability, as well as an aesthetic choice.  Turning the pages is harder when they are not even, and the book’s opening is cleaner when it is uniform.  

Since I am working an article on the symbols used to guide the reader to annotations in the margin and swiftly back again, I will not overdramatize the fact that some of these are missing from one of the oldest examples of Old English text we have. But I cannot help thinking about the practice of trimming manuscripts within our own context in the world of digital humanities, and this serves as an example of the type of information the Medieval Codes project endeavors to find in early English manuscript and print.  The goal of digital texts is to transpose text from one medium (manuscript) to another (digital text and code).  In this process, we have many  usability and aesthetic choices to make.  It’s true, we don’t trim pages, and we even ensure that they are photographed in their entirety, but we make concessions in other ways.  Due to space constraints and load times, archival images are usually restricted to somewhere between 300 and 1000 dpi, depending on the size of the document.  This is fine for one type of reading of the text, for accessing some forms of information, recognizing the words.  On the other hand, looking at particular words, marks, and symbols or decoration, is not always easy.  This is not to disparage archival practice, and it is not a plea for the necessity and primacy of the manuscript page, but a comment on the difficulty of reconciling usability, aesthetic choices, and financial considerations, with the importance of conserving archival material.  

Many important aspects of a text are outside of the “linguistic dimension,” to use Jerome McGann’s terminology from The Textual Condition.  Some of the “bibliographic codes” are encoded in the text’s organization and get lost when marginal content is disregarded or physically removed.  There are tactile and extra-visual features of manuscript, social functions and psychological processes which are important to understanding manuscripts and printed books as well. While archival images and digital text encoded and displayed on a computer may not be able to capture all of the information in manuscripts, it is important that we study these extra-textual features. 
"we are mining a technology with a long history of human interaction for tools that our ancestors found effective for accessing information"
This may bring up questions of how to digitally encode signes-de-renvoi, which are both idiosyncratic and based on convention, or whether a page turned or made into a tab for easy access can adequately translate to a link or bookmark in a digital text.  More importantly, by paying attention to how medieval texts are organized, we are mining a technology with a long history of human interaction for tools that our ancestors found effective for accessing information, keeping in mind that speed and effectiveness are not necessarily related. Computers were built with calculation in mind, but books were created to facilitate human navigation of text. We have a lot to learn from books that can only enrich our digital experience with text. Part of the process is looking outside of the purely textual, and into the social, psychological, and bibliographical aspects of our navigation of books. 

Ben Neudorf

Jun 18, 2013

Defining the terms

In this inaugural post for the Medieval Codes project, it seems appropriate to lay out what I think we are trying to do and what we're not trying to do.

When I google 'medieval codes', the first results are gaming sites. So I'll start with this: we are not offering cheat codes for 'medieval' games. Sorry.

  • We are investigating ways in which medieval people, writing English texts, organised, stored, transmitted, and searched for information. Some of that information might conceivably be about combat games that actual medieval people played, but you may have to read a manuscript to get there.
The word 'codes' also suggests cryptography or secret writing, but that's not our primary concern either.
  • We are interested in medieval text forms as communication technologies. In societies where literacy was less common and had different social implications than it does in today's industrialised nations, how did medieval people learn, negotiate, experiment with, and develop the technologies of writing?
Although we have chosen the word codes deliberately to recall the most common context of that word today, digital environments, the project is not primarily about encoding medieval texts in digital forms.
  • We start with the observation that medieval texts are already encoded, by virtue of being written down. A better understanding of, and sensitivity to, the ways in which they were first encoded is important when we recode or remediate these texts into other forms -- print or digital.
When we suggest that medieval English texts were a form of code, we are creating a mental tool for analysis, a model. Since the early 1600s, model in English has meant ‘A three-dimensional representation, esp. on a small scale, of a person or thing or of a projected or existing structure; esp. one showing the component parts in accurate proportion and relative disposition’ (OED s.v. model, sense 4a).
"We are interested in medieval text forms as communication technologies."
In various disciplines that involve design, such as architecture or sculpture, such models are common tools. In the sciences, model came to mean something more abstract: ‘A simplified or idealized description or conception of a particular system, situation, or process, often in mathematical terms, that is put forward as a basis for theoretical or empirical understanding, or for calculations, predictions, etc.; a conceptual or mental representation of something’ (OED s.v. model, sense 8a). More succinctly, ‘to model is to create a useful fiction’ (McCarty 2008, p 259). Modelling is thus a central intellectual tool of the sciences, including computer science, and theories frequently involve or are expressed as models. For example, our understanding of atomic structure throughout the history of 20th-century physics proceeded as a series of models.

Digital humanities has often been described as modelling (Unsworth 2002, McCarty 2004), insofar as this activity involves creating digital representations of the subject matter of humanities research – texts, for example. This project is a modelling exercise. Analogies between medieval texts and modern (especially digitally mediated) texts will inevitably arise. The purpose of the model, however, is not to make facile comparisons, but to engage in a historically sensitive investigation of the ways in which the encoding of information has developed in English texts, from the Middle Ages to the present. In particular, we are looking for:
  • Aspects of medieval code that have been remarkably persistent in many different contexts. One example is the Roman alphabet, originally designed to encode Latin, adapted to English, and so basic to our experience of written English today that we think of it as natural, although it isn't.
  • Aspects of medieval code that tended to be lost in modern media, especially in print environments. One example is devices for qualitative visual indexing, such as visual art (decorated capitals and the like) that had a search function as well as an ornamental one.
  • Ways in which the analogy between medieval text and modern (digital) codes breaks down. The most obvious difference is that medieval text was intended to be read by humans, whereas modern computer code is meant to be read by machines. This is not a trivial difference, especially at a time when many humans are reading text that is mediated by machines. Ultimately, this project is an exploration of some of the very complex and far-reaching implications of technology.
At this stage of the project, besides messing around with this beautiful website (comments and suggestions welcome), we are pursuing two kinds of research simultaneously. I'm working on clarifying terminology such as model, text, information, and code for the purposes of the project. Ben is developing a case study that will apply the model to a specific feature of medieval textuality. We'll try to post here regularly, every week or so. Come back for more.

Yin Liu