Friday, December 1, 2017

Digitizing Hervor: Part 3 - Creating the Facsimile Layer

In my previous post I introduced TEI encoding and the Menota project. As I mentioned, Menota supports three different levels of text representation: facsimile (as close to the text as possible), diplomatic (abbreviations expanded and some special characters removed) and normalized (word forms match those which would be found in a grammar or dictionary of the language in question). In today's post, I will talk about the process I used for creating the facsimile level of the transcription for the Digital Hervararkviða.

Abbreviations and Special Characters

Abbreviations and what we would call "special characters" (that is, they do not correspond to the characters typically available on an English-language QWERTY keyboard) are extremely common in Old Icelandic manuscripts. Actually, they are extremely common in most medieval manuscripts, but medieval Icelandic scribes seem to have increased the number of abbreviations they were using just at the time that everyone else was moving away from the practice. There may be any number of reasons for this--Iceland's geographic remoteness still lends itself to an extreme linguistic conservatism--but one important factor may have been economic.

Parchment was always expensive in the medieval world, and suitable sheets could not be made from older animals. That meant that in addition to the time-consuming work of turning a mere skin into parchment, that skin came with an opportunity cost of a calf who would never grow into a cow/bull, or a lamb which would never produce wool. In Iceland, perhaps more than in some warmer and less remote places, the cost may have been felt heavily.

Of course Iceland also has its own robust literary tradition throughout the Middle Ages, complete with its own orthography and stylistic tendencies (it has been noted, for instance, that the practice of Skaldic poetry was primarily an Icelandic art form). So it may be as simple as saying that this is the way they liked to do things. The abbreviations found in Icelandic manuscripts can be traced back, via the Anglo-Saxon scribes from whom the Icelanders seem to have chiefly learned their craft, to Roman and even Biblical times. There are three major sources: the Tironian notae, a system of shorthand invented by Cicero's personal secretary; the old Roman system of legal shorthand; and the nomina sacra, a system of abbreviating the names of God which early Christians ultimately borrowed from the Hebrew Scriptures, and which had become commonplace in Latin texts by the early Middle Ages.

Matthew James Driscoll identifies four basic types of abbreviations found in old Icelandic texts. I have listed them below, along with examples from the Hauksbok manuscript. But these are only examples which I think typical of these kinds of abbreviations and their use in the text; cataloging every variety of abbreviation in the MS would be too exhaustive for the purpose of this post.

1. Suspension, where only the first letter or letters of a word are written, generally followed, and frequently also preceded, by a point or, occasionally, with a superscript stroke.

e with superscript i for eigi.

h with horizontal bar for hann

h with superscript o for hon

q with points on either side for quað

long s with a point for segir

þ with horizontal bar for þat

minuscule t with superscript i for til
2. Contraction, where the first and/or last letters are written, normally with a superscript stroke, or, less commonly, a point or points.
the name Angantyr, contracted by a superscript stroke over the g

two long s with points on either side for synir

3. Superscript letters, a superscript vowel normally representing that vowel preceded by r or v, a superscript consonant that consonant preceded by a.

hug with a superscript r for hugar
4. Special signs (‘tittles’) or brevigraphs, sometimes originating in a letter or combination of letters.

meg with superscript "er" character for megir. This character is far and away the most common abbreviation used in the Hauksbok manuscript, with the possible exception of the Tironian et.
Learning these abbreviations was by far the most difficult (not to say maddening) part of this process, since none of the training I had in Old Norse up to this time had prepared me to encounter them. Still, I knew they were abbreviations because I had worked with the poem before (in the form of a critical text) and so I knew more or less what it should say. Working backwards in this way, I was able to figure most of them out on my own--but it was a halting process, and my first attempt at a transcription took about three weeks of hard work as I would get halfway through and then suddenly realize I'd been reading a word or a character wrong. I learned a lot, though, and so when I made my final pass (the XML encoding pass) I was able to get it done in about a week. All right, so it was a long week, but the final result is something of which I'm proud--a facsimile which faithfully preserves each and every abbreviation in the original text (or will when the proofreading is finished!).

For the most part, I stuck to the standard encoding for abbreviations as defined by the Menota handbook. When I encountered an abbreviation the handbook didn't cover, I turned to Driscoll's Marking Up Abbreviations in Old Norse Manuscripts. This is a very useful resource which I wish I'd found before I started this project. Where Driscoll also failed me, I used my best judgement, combining the characters at my disposal. Since most of these non-standard cases involved the application of a horizontal bar to a pre-existing letter, this wasn't too hard once I got the hang of it.

In addition to the abbreviations, Menota standards also allow for certain symbols not on the US QWERTY keyboard. Anyone who has spent any amount of time at all with Old English or Old Norse will already be familiar with ð (eth) and þ (thorn). Additionally, Menota recognizes the use of certain rotunda ("rounded") variants of standard letters. For instance, Old Norse often uses an r rotunda instead of the usual r following any letter with a rounded stroke on the right side (such as a b or an o). There are also "small capitals" which are often used to indicate a geminated consonant. The Menota handbook contains instructions for handling each of these characters; however viewing them in a web browser requires the user have a special font installed.

Paleography is, as Prof. Peterson says, something you can really only learn by doing. Tackling this manuscript and creating the most accurate facsimile possible turned into a six-week crash-course in the subject. I hasten to add, lest the reader be daunted, that it was also a whole lot of fun. In his plenary session at Mythmoot IV this year, Michael Drout talked about how much we owe to the philologists of yesteryear, and just how great a gap lies between them and us. Although we are capable of summarizing and sometimes even explaining their work, we're rarely capable of going back and "showing our work" -- explaining how we know the things we know, and how our forerunners came to the conclusions that they did. But until we can, we will not be equipped to question (let alone challenge) those conclusions. The discipline of philology becomes static, and a vicious cycle results.

That is one of the primary reasons I undertook this project. The first job of the philologist is, as Fulk says, the editing and preparation of texts for readers. I wanted to do the work--not just talk about the work that others before me have done. But in the process I have discovered a marvelous thing--the joy of reading Old Norse as it was written down on the page over 600 years ago; the joy of deliberating over each stroke of the pen as much as over each half-line or case-ending. As Svanhildur Oskarsdottir writes, philology "means love of words. So let’s do it – let’s fall in love."

Attached below you'll find a PDF of the Facsimile layer of this project. There are no frills, nor has the proofreading process been completed yet, so take it for what it is. But perhaps it'll give you a better sense of what it might have been like to pick up this poem and read it in the 1300's.

The next layer of this project, the Diplomatic layer, is still in progress. In the meantime, my next post will be a (hopefully shorter) dig into the meter of Eddic poetry, and of the Hervararkviða in particular.

The Digital Hervararkviða - Facsimile Layer

The Riddles of Gestumblindi: Riddle 10

As usual, here's the answer to the previous riddle : "Góð er gáta þín, Gestumblindi, getit er þessar. Þar fara svanbrúðir til hre...