Welcome to the blog! Details about me and about the purpose of the blog are on the aptly-named About page.
So, I’m about to start work on a critical edition. I have been a fully-certified professional medievalist, licensed to commit scholarship, for a decade or more, but this is my first edition. I have located the notebook from the seminar in textual criticism I took in grad school, which has a page on which I copied down “What To Do When You Begin”. It’s reassuring, in that it reminds me of what I once knew about how collation is supposed to work, but it also more or less predates the invention of the internet and therefore its practical instructions are all about sheets of notebook paper. I’m not ruling out that approach a priori, but I would like to hear from tech-savvy readers: If you have done an edition from more than one manuscript – I have about twenty – how did you organize the work of collation?
9 responses so far ↓
Jon Myerov // May 28, 2009 at 10:31 am |
Good luck. I find critical editions fascinating. I’ll look forward to your updates here.
Carin // May 28, 2009 at 11:19 am |
Jon – Thanks for the inaugural comment! I aim to fascinate.
Larry Swain // May 28, 2009 at 3:20 pm |
Welcome to the blogosphere! To answer the question, though I only had four mss to work with, not 20, here’s what I did. Before that, though, let me say I’m assuming that your preparation of the edition is for a traditional, in print publication rather than an online critical edition, in which case some of the suggestions would be a little different.
First, I transcribed the manuscripts onto paper. Yes, good old fashioned paper. Then, I typed them in. I did this because though it takes more time, it a) allowed me to review and revise my transcription and make sure I had been correct on my readings in the first place and b) going through it a second time made me even more familiar with the text and the individual scribe’s work in the mss.
I typed them into Nota Bene. I did a bit of research on programs, and NB comes out way ahead of Word, Word Perfect etc for things like linguistics and textual editing. Another good program from Europe was called Classical Text Editor that was highly recommended. I chose NB since both its HQ and customer service and I are in North America, I figured it would be easy to get help (and they have users forums et al too).
Once I’d typed all my transcripts in, then it was time to choose a base manuscript, which in my case was pretty easy to be honest. But I copied that transcript into a working file, and then compared each transcription to the base, making notes of different readings. I worked in sections of anywhere from 10 to 50 lines depending on the subject matter and mss punctuation.
Once the comparison was done, I then would begin the process of actually editing, normalizing orthography, introduce modern punctuation where appropriate (and as lightly as possible), choosing between readings.
Essentially, I used a computer and word processor to do it exactly as I would if it was all done on paper with a pencil. There are things that a computer can do for you. It makes it a little faster for example to use the processor’s compare documents feature and track changes for some things, but you still have to go in and make the same kinds of decisions and look at the text…in the end, it didn’t really make it any faster for me to do it with the compare feature as it did with my own eyes.
Its on the back end that a computer really helps: the establishment of word frequency, indices, searching, etc, table of contents etc.
Also, NB and other processors have good export features where you can take the files and save them for the web or as pdfs and that really helps.
I don’t know that I’ve enlightened any, but that’s my .01 and hope it helps a little.
Carin // May 28, 2009 at 5:01 pm |
Thanks, Larry. This is exactly the kind of detailed, idiot-proof narrative of the process that I was hoping to elicit.
Some comments and further questions:
First, yes, for the sake of this discussion, let’s assume we’re talking about a traditional print edition (for CCCM). There will likely be a spinoff electronic project, but it will probably be a single-ms edition with commentary, and the choice of ms will depend on the results of the work on the critical edition, so let’s talk about the old-fashioned critical edition for the moment.
Second, I should note I am working on a Mac and will continue to do so; I have too much investment (emotional and time, more than money) to switch wholesale for the sake of this project, especially since I have projects that will dovetail with it that are already under way on the Mac. Now, if CTE is soooooo wonderful that I should change systems to be able to use it, I would consider using BootCamp to run Windows on the Mac, but for the moment I don’t think the tradeoffs would be worth it. I’m happy to hear arguments to the contrary, though.
On word processors: I will definitely want to work in a fully-Unicode-compliant program, not only for general interoperability, but because I want to capture glosses and punctuation for other purposes while I’m transcribing.
Now, on to the transcription process. I understand the value of writing and then typing up your transcriptions, but doesn’t that create a situation where you essentially have two texts (your transcription and your typescript) to collate against each of your manuscripts, in addition to the collation you will eventually do of your transcriptions against each other? I’m just wondering whether the payoff is worth the labor.
I am also wondering, generally, whether transcribing each manuscript in full is the sensible way to proceed. On the one hand, it would enable various forms of automated comparison, as you suggest. On the other hand, it’s an awful lot of labor up front. My notes from my long-ago editing course suggest that one choose a base manuscript for the purposes of collation (as opposed to the base ms for the edition, which obviously you can’t decide in advance) on the basis of convenience. My notes then suggest that the most efficient way to proceed is to read each ms successively against the transcription of the first manuscript, noting down only the variants. The advantage of this method would seem to be that one would, in principle, end up with a literal conspectus of variants: i.e. one could look at a page containing the base transcription and variants from all mss for a single chunk of text and begin to get an impression of which mss were varying together. Of course, one does not then end up with a complete transcription of every manuscript.
My notes also suggest that the ideal procedure to avoid prejudicing oneself about which ms is the “correct” one would be to collate MS B against MS A, then MS C against MS B, and so on, but that would require the full-transcription method, and then automated comparison might make that easier to do.
Thoughts?
Aidan // May 29, 2009 at 5:46 am |
Hi Carin, I’ve used CTE, but only for the layout capabilities, not for the automated apparatus functions and the like. There is a significant learning curve, but it is a powerful tool once mastered (and was used for some of the CSEL editions, I think). That said, I’m not sure that in iteself it would justify running dual operating systems and the like.
If you are use tradition collation methods (hand transcriptions or typed into a basic text editor), you can do that work essentially manually (or computer-assisted manually like Larry) and then use something like InDesign, QuarkExpress or any number of other programs for layout.
Much more would depend on the type of text you are dealing with. I assume that it’s relatively stable (unlike, say a model sermon for which the model doesn’t exist, but many mss copies of individual expansions do) and that it’s Latin.
I know John thought computers not so helpful for collating. The Institute for Textual Studies at Birmingham (http://www.itsee.bham.ac.uk/)issued a program called Collate at one point; this may be something that is being phased out in favor of more primarily electronic-edition tools (Anastasia and then its successor SDPublisher, http://www.sd-editions.com/SDPublisher/index.html).
If you are interested in practice in general (beyond LatTxtCrit), David Greetham’s Philology Redux (http://ecdotica.org/images/pdf/2006/ecdotica3_saggi_06.pdf) is an easy discoursive read (perhaps too chatty, but it’s interesting to see how periods and fields influence differing forms of textual criticism and foreground similiarities).
Also, a few years ago an English translation of Timpanaro’s book was published (2005, I think), which makes it a little more accessible than when it was assigned reading.
As for practical matters, I’ve been working on a colloborative edition of a translation/composite anonymous text and we’ve passed collations around as word docs, which has been helpful in inserting individual notes and comments (without getting as messy as sprawling handwritten comments and the like inevitably become). In doing this, I’ve found it helpful to have a full recording of all mss (rather than just the variants) because they often make sense when see as phrases and clauses rather than individual words. I followed the lead of a colleague who essentially laid things out phrase by phrase and then left number of tabs in between (also solving the problem of not leaving enough space in handwritten collation when ms C routinely contains more words and elaborated phrasings compared to ms B with which the collation started).
The time when I collated only variants (only for the sake of collation to establish relationships not an edition) recording agreement with ms 1 only with a check or something, I found myself doubting the reliability of the transcription more (possibly a good thing, I suppose). For example, ms 1 might have e-caudata for ae, but ms 5 usually e. In a case where I marked that the reading from ms 5 agreed with ms 1’s æstus (take æ for e-caudata), I’d ask myself if that was truly the case since I’d expect scribe 5 to write estus, go back and check, find that scribe 5 did in fact use e-caudata in this case and then do the same thing again a few days later.
Admittedly visually recording every word (and puncuation and gloss) can obscure an eye-catching conspectus of variants, so I use highlighters (and then if necessary key a note to the area which I have in another document, e.g. 1) example of A and B against C, which is in error; this allows simple ordering of similar and differing points once all the categories are sorted you can go back and hightlight with different colors, each for one set of criteria).
Without knowing too much about the text, it’s hard to say more than every situation is special. Even CCCM editions can take differing approaches (like Father Jeauneau’s Eriugena and well everything else). I’m sure you’ll get excellent (and better informed than me) opinions in the fall and spring. Best of luck!
Aidan // May 29, 2009 at 5:50 am |
Sorry, about the bad link-formatting to ITSEE in previous post. Space needed between ) and issued. Cheers.
Carin // May 29, 2009 at 6:37 am |
Hi Aidan! Thank you thank you for all the links and good ideas. I can definitely imagine the self-doubt that would arise from variant-only transcription. I know that I will want a record of spellings, beyond what may turn out to be editorially significant, because this is, you know, ALCUIN, and knowing how scribes were actually transmitting his lovely grammatical advice in the 9th c. is part of the point of the project.
(The text is Alcuin’s Grammar, so yes, it’s Latin and it’s fairly stable, even if the tradition turns out to be rather complicated.)
Now I am going to go follow all your lovely links.
Flavia // May 29, 2009 at 11:36 pm |
I’ve nothing to add at this point, but will be reading eagerly for advice, as I’m soon to be co-editing an edition from several MSS myself–and know, I think, much less than you about the process.
(Love the new digs, btw, including the wee little Carin favicon!)
Carin // May 30, 2009 at 6:02 am |
Flavia – I will look forward to talking about this, as well as the usual suspects, at one of the previously-discussed locations!