How did he get *that* reading?? (Again) – Recensio 7

From one of the miracle stories of St Nicholas (BHL 6164), appended to John the Deacon.  The story so far.  St Nicholas has sneaked up on a gang of robbers who have looted a customs-house, which was left under the saint’s protection.

Tunc dixit ad eos Sanctus Nicolaus, “O infelices et miseri, quid agitis? Numquid ignoratis, quoniam ego ipse ibidem eram, quando hoc malum perpetrastis? Nam oculi mei conspexerunt, quando has et illas res abstulistis.” Quantitatem et numerum etiam cunctarum rerum, quae de theloneo abstulerunt, singillatim eis exponens, addidit dicens, …

Then Saint Nicholas said to them, “O unfortunate and wretched ones, what are you doing? Do you not know that I myself was there when you committed this evil? For my eyes saw when you took away these things and those things.” Then he gave the quantity, and also the number of all the things which they had taken from the toll-house, listing one by one to them, saying, , …

He tells them, “All is known!” and they panic and take the stuff back. The Latin text here is what I now think the author wrote.

But I was working from the Falconius edition (1751) here, and when I got to this bit, I was a bit puzzled.  Here it is:

Latin text of the passage in the edition of Falconius.

Which is a bit weird.  How is “quanti autem” the accusative for “exponens”?

Well I happened to have a manuscript open (BNF lat. 989, 10th c.), and I saw this:

Latin text of the passage in manuscript BNF lat. 989.

That made more sense.  “Quantitatem” rather than “Quanti”.  The etiam has moved up, so we end up with “etiam &”, a phrase not  uncommon in John the Deacon.  But the “etiam”‘s do move around in the manuscripts.  It’s probably just a copyist error in this particular manuscript.

Next I looked at the Mombritius, the first edition, published before 1480, and I got this:

Latin text of the passage in the edition of Mombritius.

This confirmed the “quantitatem”, but left the “etiam” alone.  Only “numerum” has now become “munerum”, “the quantity and value also of all the stuff…”.  Nobody else has “munerum”, so this suggests to me that the Mombritius edition was based on a manuscript in Gothic hand, where such slips can be rather easy…

Cartoon showing Gothic book hand and its unreadability

I do love that cartoon!

Next I opened another manuscript, Wien ONB 416 (12th c.), which belongs to a separate family from the other manuscripts:

Latin text of the passage in the manuscript Wien ONB 416.

Here again we have “Quantitatem & numerum etiam”, rather abbreviated.

Then I looked at the Lippoman edition (1515), and all became clear.

Latin text of the passage in the edition of Lippomanus.

Here is our “Quanti”, as Falconius gives it!  And here also is his “autem”, or rather “tatem”!!  The silly fool was copying Lippomanus, clearly in a great hurry, and didn’t notice the hyphen.  So he gave two words, “Quanti autem”, where the nice clear printed copy before him read “Quantitatem”.

It’s hard to believe that Falconius did this, so I would tend to think that his compositor/typesetter did it.  Which means that when Falconius sent his edition to the press, he sent a marked-up copy of Lippomanus to the press, rather than writing out his own copy first.

We get an awful lot of information here about these early editions.

  • The editio princeps, Mombritius, ca. 1480, was printed from a manuscript in Gothic hand, and misread.
  • The second edition, of Lippomanus, ca. 1515, may have used Mombritius but certainly did not copy it.  Instead it gives the manuscript reading.
  • The third edition, of Falconius, 1751, was done carelessly and quite possibly by writing changes into a copy of the Lippomanus edition.  There was no change at this point, but the typesetter misread the exemplar before him and got it wrong.

That’s rather nice, really.  I’ve learned a lot from a little.  Once again, I’ve learned not to rely on Falconius.


Learning by doing again – Recensio part 6

I’ve now collated my Latin text – all 6 sentences of it – with 2 early editions and 24 manuscripts.  I have at least another 6-10 manuscripts accessible to me to collate.

As I thought, this is a case where you have to learn by doing.  You have to attempt to collate the text and manuscripts, somehow, anyhow, just guessing how, in order to learn how to do it, and how not to do it.  Then you go back and do it properly.

I’d like to record a couple of things which are emerging from my first pass.

Point 1.  Don’t alter your “base text” in mid-collate.  If you do, the early parts of your collation become uncertain.

The “base text” does not matter at all.  It is NOT your final text.  Rather it is some “textus receptus” that you found in a crummy early edition.  It’s a random text.  But…. it’s the text against which you collate.  Unless you want to record every word of every witness, you have to have something where your silence says “the text at this point in the manuscript is the same as the ‘base text'”.

I have indeed made this mistake already.  What I should have done is have another line, which is my proposed text.  That I could alter as I choose.

Point 2.  Expect to collate more than once.

When you start, you don’t know how the text varies.  You don’t know if it varies only in individual words, or if whole chunks are involved. So take this part of the text:

favet**, credite mihi, favet nostrae devotioni.

Now you have to compile your collation using a) the word or phrase, followed by b) the manuscript shelfmark.  Doing it the other way doesn’t work for more than about three manuscripts.  So you gaily put something like this:

favet – Angers 802, Arras 462, Milan P113supp

and you mark which “favet” in the base text.

But as you proceed, you find that you’ve goofed.  You should have recorded both “favet”s separately.  Your notes start to change, until you reach something like this:

REDO THE WHOLE THING “favet, credite mihi, favet” (he favours) – Mom.; erased BNF 2627, Fal., Angers 802, Arras 462, Milan P113supp, BNF 1864; BNF 5287, BNF 5296C, BNF 5344  RECHECK, BNF 5573, BNF 11750, BNF 12600, Saint-Omer 715
“adjuvat credite michi et favet” – BNF 5284, BNF 5345
“adjuvat…adjuvat” – BNF 5360, BNF 5607, BNF 18303, BNF NAL 2335, Rouen 1383
“adjuvat” (=he helps) – Wien ONB 12831 (15th), Orleans 342, BNF 989 RECHECK

The unit of variation isn’t the individual “favet” but the pair favet…favet.  There are four possible combinations of “favet” and “adjuvat“.  You’ve got to record this part of the text differently.  You would have done it differently from the start, had you known.  You’re going to have to go back to the first few manuscripts and see how they look against what you now know about the variability of the text.

So… your first pass through the collation process is really just a means to learn what the “units of variation” are.

Point 3.  Note down any cases that explain a divergence of reading.

I’ve been wondering why I have variations between “expectet” and “expectat“.  But I have now come across the practice of writing “expectet” as “expect&“.  That abbreviation at the end, in some versions of book hand, looks almost identical to a lower-case “a” with a “t” right against it.  So that by itself explains the “at” variant.

Likewise, as we saw in a previous post, in Beneventan minuscule I was really uncertain what the word was.  It looked to me like “notata“.  But someone who knew the hand told me it was “nacta“.  The collation shows that these are the two main variants of the text at that point.  Conclusion?  An early ancestor of every manuscript of the “notata” family was written in a Beneventan book hand, and the scribe misread it – just as I did – when he made a copy in Carolingian book hand.  The text was composed in Naples, so the existence of such a copy in Beneventan is almost inevitable.

I’ve also noted that some manuscripts use the Tironian symbol for “et” (=”and”).  Usefully there is a unicode character for it: ““.  When the “” is faint and narrow – and it often looks just like the lower portion of an “i” -, then the scribe may simply not see it.  It can quite easily be mistaken for text on the other side of the leaf.  I found myself looking twice, in one manuscript.  So the presence or absence of an “et” is not necessarily significant.

Point 4.  The early editions are bad stuff.  The editors have introduced changes which are not found in any manuscript, and seem entirely unnecessary.

I have not, of course, examined every manuscript.  But I’ve looked at rather a lot now.  There is a consistent pattern of differences, between the manuscripts on the one hand, and both of the early editions (in different places) on the other.  I know that people tend to assume that the editors just printed what they had before them.  I am less sure of this now.

I’m going to carry on with the collate, and process the remaining half-dozen manuscripts into the working document.  But once I reach the end, I will do it all again; except that this time, I know what to look for.


Learning by Making a Collation – Recensio part 5

I’ve commented before on “learning by doing”, how you have to actually attempt something in order, not to do it, but to find out how to do it.  You never get it right first time, because when you make your first attempt, what you’re actually learning is how not to do it.  When you try again, you know what to avoid.  Here are some notes from my experiences.

A few days ago, I changed tack, more or less by accident.  Chapters 14 and 15 of John the Deacon’s Life of St Nicholas – in the early editions – are clearly spurious, and I’ve neglected them.  But I started work on John’s text with chapter 15, because it was only half-a-dozen sentences.  A year on, rather guiltily, I thought that I should return to it, and apply the lessons learned over the year to the translation.

Inevitably I used my machine-generated collation of the two early editions, and noted the variants; and then, equally inevitably, I started using the PDFs of the manuscripts that I have on disk.

Because this “chapter” is not actually authentic, the lessons learned about the manuscripts may not apply to the genuine portions of the text.  But never mind.

Because the text is so short, I felt able to collate the whole chapter.  This was rather a good thing, because you could do so much more, and you felt much more positive about the experience.  You could feel the wind in your hair.  You really could collate all the manuscripts; although your apparatus swiftly got very big!

Angers BM 802 - the start of the text

Here’s the first sentence.  I’m only 16 manuscripts and 2 editions into the task:

Laetemur ergo**, carissimi, laetemur in domino, et diem** festum sancti Nicolai** salubriter ** celebremus**, quoniam si nos ille concorditer festivos** inspexerit, favet, credite mihi, favet** nostrae devotioni.**

Let us, then, rejoice in the Lord, dearest friends, let us rejoice, and celebrate wholesomely the feast-day of Saint Nicholas, for, whether he examines** our festive selves amicably, he favours, believe me, he favours our devotion.**

** FUTP – hidden future.
** favet takes dative

Now here’s a look at my working notes for the apparatus so far.  Note that I quickly found that you absolutely must have each variant on a separate line, variant first, witnesses later.  This bulked up the stuff quite a bit.

** “in domino et” — Mom., Angers, Milan P113supp, Arras 462, BNF 989, BNF 2627, BNF 5287, BNF 5344, BNF 5360
“in domino” (omits “et”) — Orleans 342 (a space, tho)
“in domino ut” – BNF 1864, BNF 5296C
“Laetamur ergo in domine carissimi” – Fal.
“Laetamur ergo fratres karissimi laetamur in domino” – BNF 5573
“Laetamur itaque fratres karissimi laetamur omnes in domino et diem festum sanctissimum nicholai devotis mentibus celebremus: quia si nos concordes conspexerit sue intesse festivitati? adjuvat credite michi et favet nostrae devotioni” – BNF 5284,  BNF 5345 are an abbreviated version of the text.
** “diem”: Mom., default
“dum” – Fal.;
“venerabile” – Orleans 342; BNF 989;
“hunc venerabile” – BNF 5360
“diem venerabilem” – BNF 5607
** “festum sancti Nicolai” – Mom., default.
“festum sanctissimi patroni et pastoris nostri Nicolai… more text … celebremus” – BNF 5287.  Seems to be text added to create a pattern on page.
“festum venerabilem sanctissimi viri nicholai” – BNF 5344
** “salubriter” – Mom., default.
“solenniter” (solemnly) – Fal.
** “celebramus” – Mom, Fal;  OTHERS?
“celebremus” – check all mss viewed before BNF 2627
“celebremus” — BNF 2627, BNF 5284, BNF 5287, BNF 5296C, BNF 5344, BNF 5345, BNF 5360, BNF 5573, BNF 5607
** “festinos” – Mom. (checked against PDF);
“festivos” – Fal.,  default.
**  (REDO) “favet, credite mihi, favet” – Mom.; erased BNF 2627, Fal., Angers 802, Arras 462, Milan P113supp, BNF 1864; BNF 5287, BNF 5296C, BNF 5344  RECHECK, BNF 5573
“adjuvat credite michi et favet” – BNF 5284, BNF 5345
“adjuvat…adjuvat” – BNF 5360, BNF 5607
“adjuvat” (helps) – Wien ONB 12831 (15th), Orleans 342, BNF 989
** “devotioni” – Mom. Fal., etc; – treat as default, signal only others.
“devotionem” – Orleans 342, BNF 5607
“devotion3” – BNF 1864

You’ll note that I write “default” in places.  Originally I started out indicating every variation, other than minor word order stuff, for every manuscript.  I ended up with huge long lines of witnesses.

After a while I decided to erase the biggest list, where the others were just one or two witnesses, and replace it with “default”.  It doesn’t mean that I am going to accept that reading, just that I will only signal deviations from it.  This is just to reduce the sheer bulk.  I’m still not sure that I did right here.  It conceals just how many witnesses there are for each reading.

In some cases I misunderstood what the “unit of variation” is.  So initially I noted “favet” as the variant; only to find that the whole phrase varies, from “favet, credite mihi, favet” (“he favours, believe me, he favours…”) in all sorts of ways.  Either or both “favet” can be replaced with “adjuvat” (helps), and other words can appear.  When I found this, I realised that I would need to recollate for this phrase, rather than the note that I put originally just on “favet”.

Here we get the same old lesson.  The first time through is just a trial run.  You don’t know what you need to look out for until you’ve finished, at which point you will need to redo portions of it.  You can’t just collate the text once.

In this case the “variation unit” was larger than one word.  I’ve had three or four places, so far, in which I have discovered this only after a dozen manuscripts have been collated.  Of course I can’t be sure if I looked that closely at the other words in the newly enlarged phrase, so I’ll have to go back and redo.

In some cases, I only noticed later.  I didn’t notice that “celebramus” varied until I was well in.  So of course I wasn’t really looking for it in the early manuscripts.

In some cases I just could not read the manuscript.  So I made a note, and carried on.  Maybe it would become intelligible later.

This did indeed happen in one place yesterday.  The day before there was a mysterious word inserted in one place, that I couldn’t even read – 5 letters, following “NIcholaus”.  But later that day I came across a manuscript with a different variant – “Nicholaus mirreorum litie”.  Even that was baffling, until I realised it said “Nicholas of Myra in Lycia”!  The baffling word “litie” was “Lyciae” with a medieval twist.  Ho hum.

As you actually write the collation, you become uncomfortably aware of how subjective it is.  You always ignore clutter.  So… again, once you have your draft collation, you will need to go back and try again.

I’m doing all this to prepare a translation, so it is helpful to add the English meaning of the alternatives sometimes.

I have 53 manuscripts on disk, and the number is growing all the time, thanks to the wonderful, wonderful accessibility of the manuscripts held at the Bibliothèque Nationale Français in Paris.  You can find almost anything by shelfmark starting from here, although some are still microfilms.  The monochrome versions are harder to work with, when you are navigating around the text, looking for a “Laetamur”.  The medieval scribes used colour to indicate the start of paragraphs, which is a big help for the eyes.  Without it, you can miss your bit in the mass of apocryphal miracle stories.

What I do is go to the Bollandists site, the Bibliotheca Hagiographica Latina Online.  My text is BHL 6108, so I search for that, and display the list of manuscripts by “fond”, i.e. the library holding it.   And once you hit “Paris, BNF lat. 12345”, you’re in easy street.  The BHLO will tell you the folio numbers.  Well, I keep another tab open with the BNF entry point, go and find it, display the manuscript, download it, and mark up the folios I want in the PDF.  (I sort of wish that I could contribute my markers back to the BNF site, but of course you can’t.)  Once I’ve found the “Laetamur, karissimi, laetamur” bit, I can then collate it against my working file.

That’s what I’ve been doing, and, actually, it’s rather fun!

I’ve still got quite a few BNF manuscripts to go.  Not all of them actually contain my text – the BHLO is not 100% accurate.

French manuscripts are generally rather accessible.  The IRHT website helps you find non-Paris manuscripts.  Outside of France it is hit and miss.  The accessibility of English manuscripts is a disgrace.  American manuscripts are few, but not much better.  Italian manuscripts, other than those in the Vatican, are generally not online.  The Vatican ones would be much, much more helpful, if you could download them, but at least most of them are there.  Still, in five years I imagine this will improve out of all recognition.

As ever, I hope these notes will be useful to people coming to collate a text for the first time.  The first time, you’re just working out where everything is.  Your second pass is the real thing.


It’s starting to work! – Recensio part 4

This afternoon I went to my draft text and translation, and, as per my last post, starting from the top, looked for a place in the text where the editions differed in meaning.  I did not have to go far before I found this place, on “in vocem” or “in clamationem“.

Latin and English text, working notes

To those wondering how I got this, remember that I started my task by creating an electronic text of the Falconius edition, and then translating the whole thing, one sentence at a time.  But when I had finished, I decided that the Mombritius edition text was better.  So I created an electronic text of that, and then I compared the two texts electronically (using dwdiff – but it could have been several tools).  This got me a list of differences.  Then I revised my translation to follow Mombritius.  As I went through the difference list, in order to do so, I noted down the differences that seemed significant to me.  That is, I ignored typos, spelling differences, etc., and only took those where a difference of meaning was apparent.  I noted the meaning as well!  The result was this document, which I am using to work on the text.

Verum, quia scio me penes literatissimos magistros inefficacis esse sermonis, ideo deprecor omnes, qui hujus operis studiosi lectores accesserint, ut non facillimam prorumpant inclamationem,** et me indoctum meque** judicare inertem incipiant.

Until today, I had the Mombritius text, “in vocem” here.

So, just as I did earlier, I opened up my directory of manuscripts, and I started to work my way down the files.

Screen grab of directory in Windows Explorer

Note that I’ve found it endlessly useful to include the century in the file name.

Of course each time I go looking for a passage in the PDF of the manuscripts, I add bookmarks and sticky notes to where I found it.  This does make navigation easier.  I have not attempted to mark up everything in one pass in advance.  Rather I am doing what I need to do as I need it.  After all, I can always come back!

Here is the current state of BNF lat. 2627:

Bookmarks and sticky notes in a manuscript PDF

Apologies for the size.

I found book marks by just picking up on red initials.  So in that picture, I didn’t bother to bookmark Mane itaque – because it’s not one of the main divisions in the text like Pontificalis or Praeterea.  But I could have done.

On my first pass, I added a sticky note for where I was looking at Nacta / Nactus / Notata.  Three lines down, there is “O novi iacob stropha”, from this morning.  I only add a sticky for that where there is an omission, because I always know that it’s just below the Nacta text.

Notice that the sentences in this 11th century manuscript all begin with a small capital.  The big red capitals allow you to find big places in the text.  Once you’re on the right page, the small capitals allow you to find the sentence you want.  When I was looking for these two places, I found myself looking for “Rumpe”!  Because that was a line or two above.

These little tricks all allow you to speed things up.

But back to what am doing right now.  Well, I clicked on every one of those manuscripts.  And I noted down the reading.

I started, of course, with:

** Mom. “in vocem”; Fal. “in clamationem”, crying out against; Corsi: “in cachinnationem”, in immoderate laughter.

Initially I added the manuscripts after the editions.  But actually it’s better to turn it around, and give the text, with the edition against it, and then add manuscripts on the end.

So I ended up with this:

  • “in vocem” – Mom., Lipp. Means nothing.
  • “in clamationem” or “inclamationem”, crying out against, criticism, abuse – Fal., Angers BM 802 (11th ), Balliol 216 (13th), Harley 3097 (1124), BNF lat. 196 (12th), BNF lat. 5284 (13th), BNF lat. 5308 (12th), BNF lat. 5346 (13th), BNF lat. 5624 (13th), BNF lat. 989 (10th), BNF lat. 1864 (14th), BNF lat. 2627 (11th), BNF lat. 18303 (before 968), Bruges BP 402 (13th), Cambridge CCC 9 (11th), Durham B.IV.14 (12th), Fribourg L 5 (13th), Milan P113supp (10th), Munich Clm 3711 (11th early), Orleans BM 342 (10th), Vat. Arch.Cap.S.Pietro A.5 (11th), Vat. lat. 1197 (11th), Vat. lat. 9668 (12th), Vat. reg. lat. 477 (12th), Vat. reg. lat. 496 (11th), Wien ONB 12831 (15th),

“in cachinnationem”, jeering, immoderate laughter – Corsi, Berlin theol. lat. qu.140 (11th), Linz 473 (13th), Munich Clm 12642 (14th),

Because I did this immediately after the last post, some of the manuscripts started to sound familiar!  That group at the bottom had an eccentric reading for the “O novi Iacob stropha” search too.  It’s a group, a family of manuscripts that share common errors.  This is precisely what we are looking for: a way to group manuscripts in order to get a stemma if we can.

The first collation I did took quite a while.  The one this afternoon was quicker.  This one took very little time.  Why?  Because I’m getting used to it, and developing my way of working.

Of course I am lucky to have four different early editions.  If I did not have this, if I only had one, then I would have to manually read through a manuscript PDF and manually compare it with my electronic text.  If I didn’t have an electronic text at all, I would have to transcribe one manuscript, and use that as my framework electronic text – not my final text – to translate, and on which to hang readings, in order to analyse the text.

I am rather enjoying this!  Maybe I’ll look for another passage next!


O novam Jacob stropham! – Recensio part 3

The earliest printed editions of a text are often merely a printed version of some manuscript that the editor had to hand; or are based on a prior edition, plus readings from such a manuscript.  In some cases all the manuscripts were destroyed afterwards, and we only have the printed edition.  This is the case with  Velleius Paterculus, and also with Tertullian’s De ieiunio.  So these editions are a “manuscript witness”.

I’ve scanned four such editions of John the Deacon to Microsoft Word, and carried out a machine comparison.  There are quite a few differences.  But in order to establish a “family tree” of manuscripts, which differences are significant?

At the moment I have two tentative guidelines.  They may be wrong, but it’s what I have.

  1.  The scribes do not care all that much whether they put down “at”, “et”, “ac” or “atque” – all of which mean “and” – regardless of which was actually in the text in front of them.  So “variants” which mean the same thing are not really useful to us.  What we need is a difference in the text which has a real difference in meaning.
  2.   Because the endings of so many words are abbreviated in medieval copies – “ū” for “um”, etc – these variants may not be significant either.  Let’s not spend a lot of time over “explicare” vs “explicarem”.

The next real variant is not much further down the text from the last one.  At the dead of night, St Nicholas has secretly visited the house of the poor man, tossed a bag of gold through the window, and secretly disappeared.  So now, time for a quick comparison with a biblical figure! The text continues:

O novi Jacob stropha!**  Ille commentatus est, qualiter Laban, mercedem non amitteret; hic autem, ut coelestibus non privaretur commodis.

O the trick of the new Jacob!**  The former devised it, with Laban, to avoid losing his wages; but the latter,  to avoid being deprived of heavenly rewards­.

The reference is to Genesis 30:32-3, where Laban agrees to pay Jacob for looking after his sheep by allowing him to keep any offspring that are striped; but, trickily, Laban gives him only monochrome sheep.  Jacob gets round this by putting branches of various colours in the drinking troughs, which cause the sheep to produce vari-coloured offspring.  By his trickery, Jacob gets the wages that he was promised.  St Nicholas, by his own strategem, gets the heavenly reward promised to those who do good in secret.  It’s not a great comparison, but there’s no doubt that this is what John is attempting to say.

The first three words of the text, however, vary in some interesting ways.  I only have 46 manuscripts at the moment, but here are the readings:

  • O novam Jacob stropham. — Mombritius (1477), Lippomano (1553)
  • O pueri Jacob stropham.  (what?!) — Falconius (1751)
  • O nova Jacob stropha. — Corsi, based on Berlin theol. lat. qu. 140 (11th c.), BNF lat. 5284 (13th c.), BNF lat. 5308 (12th), BNF lat. 5345 (13th), Vat. lat. 1271 (12th c.), Bruges BP 402.
  • O novi iacob stropha. — BNF lat. 2627 (11th c.), BNF lat. 18303 (=early 10th c), Angers BM 802 (11th c.)  Balliol 216 (13th c.), BNF lat 196 (12th c.), BNF lat. 1864 (14th c.), BNF lat. 3791 (12th c.), BNF lat. 3809A (15th c.), BNF lat. 5572 (11thc), BNF lat. 5573 (12th c.), Durham B.IV.14 (12th), Fribourg L 5 (13th c.),  Milan P113 supp. (10th c.), Munich Clm 3711 (11th), Orleans BM 342 (10th c.), Vatican Arch.Cap.S.Pietro A.5 (11th c.), Vat. lat. 9668 (12th), Vat. reg. lat. 477 (12th), Vat. reg. lat. 496 (11th c.), Vienna ONB 12831 (15th c.)
  • O novi iacob tropha — BNF lat. 1765 (13th c.)
  • Omitted: clamque discessit is followed directly by Mane itaque, omitting the whole digression. Vat. lat. 5696 (11th c.), Vat. Arch.Cap.S.Pietro A.3 (12th c.)
  • Omitted: clamque discessit is followed directly by Hic est magister bone, omitting two sentences, but retaining some of the digression, then Mane itaque.  Vienna ONB 416 (12th c.), Klosterneuburg 701 (14th), Linz 473  (13th c.) – The Linz manuscript is a contaminated text, however, containing material from BHL 6118.  Munich Clm 12642 (14th).
  • . Vat. lat. 5696 (11th c.), Vat. Arch.Cap.S.Pietro A.3 (12th c.)

A couple of oddities:

  • BNF lat. 989 (10th c.) is impossible to read, but the last word is stropha.
  • Cambridge, Corpus Christi College 9: O novi iacob stropha, but, above the “i” in novi there also appears an “a”.

Now by chance I got some help from a google search.  I wasn’t familiar with the word “stropha”, a strategem or trick.  Googling the words above produced a passage about this in Jerome’s “Hebrew Questions on Genesis”, (Quaest.Heb Ad Gen.30.32-3):

Itaque Iacob novam stropham commentus est, et contra naturam albi et nigri pecoris, naturali arte pugnavit.

Jacob therefore invented a new trick, and by natural art fought against the nature of the white and black cattle.

There’s an awful lot of the same words in there, isn’t there?  Although they’re doing different things.   This perhaps explains why we find all those accusatives like stropham in our text.  Quite possibly they are the result of the copyist being more familiar with Jerome than with John the Deacon. On seeing the unfamiliar text, the copyist “normalised” it.  Jerome has “Jacob” as a nominative, the subject of the verb in his sentence.  But it can’t be the same in John the Deacon.

“Iacob” is indeclinable, so we could read the genitive, sometimes as “Iacobi”, “of Jacob”.  The sense is that Nicholas is the new Jacob.  So “novi” and “Jacobi” would agree.  We end up with (in English word order) stropha novi Jacobi, “the strategem of the new Jacob”.

Of course I only have a selection of manuscripts.  But all the same, it’s clearly necessary to look at them.


How to lose the first letter of a word in transmission

In my last post I looked at how to decide what the genuine reading was of a single word in John the Deacon’s Latin text.  Among the variants was “Nacta” and “Acta”.

Purely by chance this evening I have come across a perfect illustration of how Nacta became Acta.  It is to be found in Ms. Vatican Barb. lat. 586, on fol. 3v, where the text appears like this:

Nacta written but the initial never inserted, leaving "acta".
Nacta written but the initial never inserted, leaving “acta”.

There it is.  The word is “Nacta”.  The scribe has left a space for the “N” to be illuminated, for a decorated initial to be inserted.  To help the artist, he’s put a written “N” in the space, and the text reads “acta”.

In this case the N is big, and bold, and clear.  But what if it wasn’t?  What if it was small, tiny, faint?

Clearly this has happened, sometime in the past, in some other manuscript.  The copyist did not notice the “N” and wrote “Acta”.  How do we know?  Because “Acta” is one of the variants that I found in some of the manuscripts, listed in my last post.

This, folks, is how you lose letters from the front of a word in transmission.

Update: Stephen Carlson points out that it actually looks as if it was originally an A, which was erased and the N written in.  The first “a” of “acta” is different to the other, and the surface looks erased!  And the other initials have been marked up in red.  So maybe… it means the opposite?!  Acta, corrected to Nacta!  Here’s the other A:

Folio 6v – A


Inventa ergo… Or maybe not – Recensio, part 2.

Time to plunge into the text and see if I can find any errors in the manuscripts that might help me divide them up into families.

When I was collating the text of John the Deacon’s Life of St Nicholas, I came across a passage, which is interesting for the sheer number of textual variants, for the first word of the sentence.  St Nicholas has learned that a starving man, unable to afford a dowry for his three daughters, has decided to prostitute them.  He decides to do something about this.

Inventa ergo** cuiusdam noctis hora, sumens non modicum aurum, ligansque in panno, perrexit ad domum viri, quam undique circumspiciens, per fenestram quae competens videbatur, clam intro projecit, clamque discessit.

Therefore, when the hour of a certain night arrived,** he took not a little gold, and tying it in a cloth, he went to the man’s house, which he surveyed from all sides, and then, through a window which seemed appropriate to him, he secretly threw it inside and secretly departed.

I noticed this place when I was machine-comparing the editions.

  • Mombritius, Lippomano: inventa ergo … hora – the hour having been found/reached, therefore.
  • Falconius: nactus ergo … hora – (he) having reached, therefore … the hour.
  • Corsi: acta ergo … hora – the hour having come, therefore.
  • Mai: infamiis notata igitur – their disgrace having been noticed, therefore.

That’s a lot of differences, and that’s what, from a text criticism point of view, we need to find!  So… good news!  Now here are some thoughts, based on what I generally know about these editions.

  • Mombritius printed some unknown (probably late) manuscript.  Lippomano may have just reprinted Mombritius at this point.
  • Corsi used Falconius, but also a Berlin manuscript.  At this point in the manuscripts, there is an initial.  Is acta really Nacta, copied from a manuscript where the initial “N” had never been painted in?  So we could ignore it?
  • Mai’s edition is a printed version of an abbreviated form of the text, which turns into a paraphrase.  Maybe the scribe of the abbreviation found something odd here – maybe just something he read as atta? and improvised?

Maybe we have manuscripts missing the initial letter.  Let’s go and look, and see what we have.  Maybe we have a point at which the manuscript tradition diverges?  (This will also help me get more of the manuscript material in order on my disk.)

The first PDF, alphabetically, in my folder of manuscripts is Balliol 216.  This I made from a zip file of images, downloaded from the website, and pulled into a PDF using Finereader 15.  I’m opening it for the first time (in a very old copy of Acrobat 9 Pro).  I wince a bit as I see images on their sides and upside down.  I read the folio numbers as I page down, and get Nicholas at folio 33r as expected.  I bookmark it, and save the PDF properties so that the bookmarks will open whenever I open the PDF.

The text isn’t that great to read – a Gothic hand, drat it – but I know what I’m looking for.  It’s an initial.

Ooo.  On folio 34 there’s an erasure.  I note that in the bookmarks.

I page down.  Some of the photos are lying on their sides.  I rotate them.  I look out for familiar initials and bookmark them.  Acrobat is amazing.  Pity you can’t actually buy a copy any more.

I page down, looking for the end of the text.  I must have passed it, because I have a red initial “Igitur postquem beatissimi nicholaus ex hoc mundo migravit” – “After blessed Nicholas snuffed it”; but I know this isn’t part of my text, but some of the tedious miracle stories often added on the bottom.  So fol. 42r is past the end.  Bookmark that.

Back up.  Aha!  Bottom of f41v is what I’m looking for – “remearunt ad propria” – “they went home”, plus some standard stuff “magnificentes doninum jesum christum”.  That’s the end.  Bookmark it.

So I’m not going to find a handy initial.  Rats.  Hmm… I can make out “Tunc om” and then an abbreviation.  I got to my working file: it’s tunc omnes, and I’m in the middle of chapter 7.  Too far.  Mark it up anyway.  His ita transactis, the start of chapter 7, can’t be far – oh yes, there it is.  Sticky note, and bookmark.  Back up I go… ah, there’s Laban!  Good old Laban, I’m not far now.  And … there it is!

Balliol MS 216 - position of our passage
Inventa ergo? Not in Balliol 216! It’s Notata igitur.

Immediately we find… “Notata igitur!”  (Words before it are patrem tuum qui in caelis est, your father who is in heaven.  Unlike me.)  Different again from any of the manuscripts, although clearly the Mai abbreviated text is working from something of this type.

I won’t drag you through this process for each manuscript.  But I’m doing the same thing in each case.  What do I get?

  • Balliol 216 (13th) = Notata igitur
  • Berlin theol. lat. qu 140 (11th) = Acta igitur, which is Corsi’s reading from just this manuscript
  • BNF lat 196 (12th) = Acta igitur, with the capital.
  • BNF lat. 989 (10th c) = v faded.  I think it’s a Notata igitur, with the capital, after some image manipulation.  The N and the ata are clear.
  • BNF lat 1765 (13th) = Nacta igitur.  But something is odd about this ms – the text is a lot shorter and ends with “accepit insulam”, part way through chapter 7, then another text, which seems to be called the “Relatio Simplicii” in another ms (below) and then an odd ending from BHL 6108a.  Then the Passio of St Lucy.
  • BNF lat. 1864 (14th) = Notata ergo.  This text ends with the usual remearunt, but then follows with material printed by Falconius as chapters 14 and 15 – the first manuscript copy I have seen of this.
  • BNF lat. 2627 (11th) = Notata ergo.  This too ends with chapters 14 and 15.
  • BNF lat. 3791 (12th) = Nacta ergo.  The front of the ms is missing.  This copy ends with remearunt and then follows the Life of St Lucy.
  • BNF lat. 3809A (15th) = ???  There’s definitely an ergo but what’s the first word, with the initial, following the “a – li – ud. -“?  It looks like “Clam“? “without knowledge of the hour”?  The thing ends with the ch.14, and a bunch of miracles, then the life of St Ambrose.

  • BNF lat. 5308 (12th) – Transacta ergo.
  • BNF lat. 5573 (12th c.) – Nacta ergo, but marginal correction to facta.
  • Fribourg L 5 (13th) – Nacta igitur.  This does not seem to have the usual remearunt, but does have chapters 14, 15 and then ending from BHL 6108a, and then the “Relatio Simplicii” about the transitus of St Nicholas.
  • Milan P113 supp – Nacta ergo.  This ends with “chapters 14 and 15” and then the Life of St Waleric (who?)
  • Munich BSB Clm 12642 (14th) – Nactus ergo, but the Nactus appears to be in a different hand, so an erasure and correction.
  • Vatican Barb. lat. 583, f.44v – blessed if I know!  It’s something in Beneventan, which I can’t read.  I’ve posted to Twitter.

  • Vatican – Nacta ergo.  But with an unilluminated N.  Easy copyist error to write “Acta”.
  • UPDATE: Vat. lat.1271 (12th c.) – Inventa ergo.  Finally!

I’m beginning to wear out here, so I will stop for now. I’ve learned quite a bit. Clearly I need to catalogue exactly how each copy ends.

But notice what is not found in any of these?  The “inventa” that we started with!

Later: By chance I’ve found a perfect example of why the text cannot be “Acta”.  It’s in my next blog post, here.

Later still: Or maybe it was originally Acta, “corrected” to Nacta?

The Munich copy of vol. 2 of Mombritius, “Sanctuarium”, p.163, showing “inventa”


How to Compare Manuscripts – Recensio part 1

The Latin text that I am working on has never had a critical edition.  I am actually not sure what the author wrote at points, because the editions differ so much.  What to do?

These days we have lots of manuscripts online.  But … how do we go about comparing them?  Where do we start?

Googling has not produced anything very useful.  So I thought that I would record my own thoughts, as far as I have got.

The only practical guide that I have seen is in Martin West’s Textual Criticism and Editorial Technique (1973).  On page 66 he tells us:

The manuscript is compared with a printed edition word by word, and the differences written down. Some people write them in the margins of the edition, but even if the copy is interleaved this does not give one room for more than a few manuscripts’ variants, and I usually use a separate notebook. It is essential in this case to record in writing which edition has been used for the collation, for if that is not known a collation loses much of its value. (One must bear in mind the possibility that one’s collations will one day be used by someone else, and one must therefore make sure that it is clear in this and in all other respects how they are to be interpreted.)

It is best to choose an edition which is light to travel with, will always be easily available, and keeps close to the paradosis (to minimize the amount of writing necessary); and to use the same one for each collation.

Every effort should be made to prevent confusion between the collations of different manuscripts. If they are done into the printed copy, the best thing is to use different coloured inks[2]); in a notebook, the manuscript should be identified at the top of every page.

Care must also be taken to avoid ambiguity about the location of the variant. In prose texts the lines should be numbered down each printed page and the numbers used for reference. If the variant is for a word that comes twice in the same line, or might be read as being for either of two similar words, it must be made clear which one is in question.

[2] Collations should always be in ink. If washable ink is used, beware of rain.

He goes on to add that, if you have decided to ignore some trivial points, like iota subscript (in Greek), make a note that you have done so.  Always record corrections and marginal notes.  It’s a good idea to note where the page turns; an omission in another manuscript at precisely that point is evidence of copying.

These are all good, practical points.

Now I’d like to add a couple of my own, as far as I have taken things, which is not very far.  What I write relates to Latin, but no doubt applies more or less equally to Greek.

  1.  In order to compare manuscripts effectively you do need to be familiar with the Latin text.  Otherwise you simply won’t be able to find your place.  I started downloading manuscripts at an early stage, but could do nothing with them.

2.  The best way to get familiar with a Latin text, and its peculiarities, is to prepare a translation of it.  This forces you to grapple with every word, and to work out what the author is saying and how he says it.  I know a Swedish philologist who intended to edit one of the works of Tertullian.  In preparation for the task, he prepared a translation of it, into English (!)  It may seem burdensome, but it is really a huge aid.

In my own case I have a Word file with the text broken down into a sentence or two, with my draft English translation interleaved.  This also gives me a place to write notes and… to start noting manuscript variations.  Here’s a bit from my current opus:

Example of interleaved Latin and English
Excerpt from the working file on the text and translation

Of course a single file can be overwhelming.  I work a chapter at a time, and only combine them once I’ve done three passes on each.

Windows Explorer, showing directories
Windows Explorer, showing directories
Directory showing files in progress
Work in progress!

3.  You need an edition to use as a base text.  It doesn’t matter what it is, or whether it is any good or not.  In fact a late pre-critical text, “vulgate” text, with interpolations, but punctuated, can be ideal for this purpose.  It’s likely to be based on some, dead common manuscript.  In that case, it will save you a lot of typing.  It’s just the rail on which you will hang your notes, and it is far easier to mark it with “this bit not in XYZ”.

4.  Get it into an electronic form, so you can use it as I did above.  I use Abby Finereader Pro 15, which isn’t that expensive and does Latin very nicely.  Once you have it in electronic form, you can do searches on it, when you’re staring at some manuscript and can’t remember “where does it say ‘Armata'”?  You can do comparisons automatically with other editions too, as I remarked in my last post.  You can copy and paste bits of the manuscript images into a Word document if need be.

5.  Get some manuscripts downloaded in PDF form.  The Gallica website Bibliotheque Nationale Français is great for this.  Finding manuscripts can be a pain, but this will certainly get easier as time goes by.  You should have a PDF editor on your machine, which will allow you to extract just the pages that you need.  Then you can mark it up.  I’ve used bookmarks to indicate key points in the text, and sticky notes to indicate, within a page, just where something is.  For instance in the image below, “laetamur” is the start of “chapter 15” of the printed text.

Manuscript with markup

I have a folder full of manuscripts, indicating whether they are microfilm or real, and with the date in the file name (which I recommend):

So far I have been looking at these, and marking up divisions in the text, and indicating where the chapters of the edition are.  This makes it faster to access the pages that you want, and helps you to start getting to grips with the text.

That’s as far as I have got for now.  My working file already indicates places where the editions disagree, so I will start noting the readings of the manuscripts for the same.  I’ll write another post when I’ve done more!


Getting Started With Collatex Standalone

Collatex seems to be the standard collation tool.  Unfortunately I don’t much care for it.  Also interestingly, the web site does not actually tell you how to run it locally!  So here’s a quick note.

Collatext is a Java program, so you must have a Java Runtime Environment (JRE) installed, for version 8 or higher.  I think Windows 10 comes with a JRE anyway, but I can’t tell because long ago I set up a Java development environment which overrides such things.

You download the .jar file for Collatex from here.  Download it somewhere convenient, such as your home directory c:\users\Yourname.

Then hit the Start key, type cmd.exe, and open a command window.  By default this will start in that same directory.

Then run the following command in the command window.

java -jar collatex/collatex-tools-1.8-SNAPSHOT.jar -S

This starts a web server, on port 7369, with error messages to that command window.  (If you just want to start the server and close the window, do “start java …”).

You can then access the GUI interface in your browser on localhost:7369.  This is the same interface as the “Demo” link on the Collatex website.  You can load witnesses, and see the graphical results.

I think it’s best for collating a few sentences.  It’s not very friendly for large quantities of text.

UPDATE: 20 Dec 2022.  Apparently this is just a standalone thing, and is NOT how you use Collatex for real.  It’s actually done by writing little scripts in python.  A couple of links:




A way to compare two early-modern editions of a Latin text

There are three early modern editions of John the Deacon’s Life of St Nicholas.  These are the Mombritius (1498), Falconius (1751) and Mai (1830-ish) editions.  I have already used Abbyy Finereader 15 to create a word document for each containing the electronic text.

But how to compare these?  I took a look at Juxta but did not like it, and this anyway is ceasing to be available.  For Collatex I have only been able to use the online version, and I find the output tiring.  But Collatex does allow you to compare more than two witnesses.

The basic problem is that most comparison tools operate on a line-by-line basis.  But in a printed edition the line-breaks are arbitrary.  We just don’t care about them.  I have not found a way to get the Unix diff utility to ignore line breaks.

Today I discovered the existence of dwdiff, available here.  This can do this quite effectively, as this article makes clear.  The downside is that dwdiff is not available for Windows; only for MacOS X, and for Ubuntu Linux.

Fortunately I installed the Windows Subsystem for Linux (WSL) on my Windows 10 PC some time back, with Ubuntu as the Linux variant.    So all I had to do was hit the Start key, and type Ubuntu, then click the App that appeared.  Lo and behold, a Linux Bash-shell command line box appeared.

First, I needed to update Ubuntu; and then install dwdiff.  Finally I ran the man command for dwdiff, to check the installation had worked:

sudo apt-get update –y
sudo apt-get install -y dwdiff
man dwdiff

I then tested it out.  I created the text files in the article linked earlier.  Then I needed to copy them into the WSL area.  Because I have never really used the WSL, I was a bit unsure how to find the “home” directory.  But at the Bash shell, you just type this to get Windows Explorer, and then you can copy files using Windows drag and drop:

explorer.exe .

The space and dot are essential.  This opened an explorer window on “\\wsl$\Ubuntu-20.04\home\roger” (??), and I could get on.  I ran the command:

dwdiff draft1.txt draft2.txt

And got the output, which was a bit of tech gobbledegook:

[-To start with, you-]{+You+} may need to install Tomboy, since it's not yet part of the
stable GNOME release. Most recent distros should have Tomboy packages
available, though they may not be installed by default. On Ubuntu,
run apt-get install tomboy, which should pull down all the necessary [-dependencies ---]
{+dependencies,+} including Mono, if you don't have it installed already.

The [-…] stuff is the value in the first file; the {+…} is the different text in the second file.  Other text is common.

There were also some useful options:

  • dwdiff -c draft1.txt draft2.txt added colours to the output.
  • dwdiff –ignore-case file1 file2 made it treat both files as lower case.
  • dwdiff –no-common file1 file2 caused it to omit the common text.

So I thought I’d have a go.

First I went into word and saved each file as a .txt file.  I didn’t fiddle with any options.  This gave me a mombritius.txt, a falconius.txt and a mai.txt.

I copied these to the WSL “home”, and I ran dwdiff on the two of them like this:

dwdiff falconius.txt mombritius.txt --no-common -i > op.txt

The files are fairly big, so the output was piped to a new file, op.txt.  This I opened, in Windows, using the free programmer tool, Notepad++.

The results were interesting, but I found that there were too many useless matches.  A lot of these were punctuation.  In other cases it was as simple as “cujus” versus “cuius”.

So I opened my falconius.txt in Notepad++ and using Ctrl-H globally replaced the punctuation by a space: the full-stop (.), the colon (:), semi-colon(;), question-mark (?), and two different sorts of brackets – () and [].  Then I saved.

I also changed all the text to lower case (Edit | Convert Case to| lower).

I then changed all the “v” to a “u” and all the “j” to an “i”.

And then, most importantly, I saved the file!  I did the same with the Mombritius.txt file.

Then I ran the command again, and piped the results to a text file.  (I found that if I included the common text, it was far easier to work with.)

dwdiff falconius.txt mombritius.txt > myop2.txt

Then I opened myop2.txt in Notepad++.

This produced excellent results.  The only problem was that the result, in myop2.txt, was on very long lines.  But this could easily be fixed in Notepad++ with View | Word Wrap.

The result looked as follows:

Output from dwdiff
Falconius edition vs Mombritius edition

The “-[]” stuff was Falconius only, the “+{}” was Mombritius.  (I have no idea why chapter 2 is indented).

That, I think, is rather useful.  It’s not desperately easy to read – it really needs a GUI interface, that colours the two kinds of text.  But that would be fairly easy to knock up in Visual Basic, I think.  I might try doing that.

Something not visible in the screen shot was in chapter 13, where the text really gets different.  Also not visible in the screen grab – but very visible in the file – is the end, where there is a long chunk of additional (but spurious) text at the end of the Mombritius.

Here by the way is the “no-common” output from the same exercise (with my note on lines 1-2)

dwdiff no-common output

This is quite useful as far as it goes.  There are some things about this which are less than ideal:

  • Using Linux.  Nobody but geeks has Linux.
  • Using an oddball command like dwdiff, instead of a standard utility.  What happens if this ceases to be supported?
  • The output does not display the input.  Rather it displays the text, all lower case, no “j” and “v”, no punctuation.  This makes it harder to relate to the original text.
  • It’s all very techy stuff.  No normal person uses command-line tools and Notepad++.
  • The output is still hard to read – a GUI is needed.
  • Because it relies on both Linux and Windows tools, it’s rather ugly.

Surely a windows tool with a GUI that does it all could be produced?

The source code for dwdiff is available, but my urge to attempt to port a linux C++ command line utility to windows is zero.  If there was a Windows version, that would help a lot.

Maybe this afternoon I will have a play with Visual Basic and see if I can get that output file to display in colour?