From my diary

I had forgotten how much I despise Microsoft software.  A couple of hours ago I decided to make a fix to my QuickLatin code.  More fool me.  Three hours later, I am no further forward and have spent the entire time struggling with their wretched development environment.  It was all working before I started.  I had to deinstall Visual Studio and reinstall, then reinstall an add-on, then work out why it wasn’t showing, then disable and re-enable to make the window pop-up…. bah!

I had to break off and retreat to my sofa for some intensive googling from my phone.  This did produce some results; but nothing that I could do would persuade Visual Studio to connect to its own marketplace.  The meaningless error message carefully conceals whatever the problem might be.

Part way through I wondered if I ought to purchase a more recent version.  Well, that led me down a rabbit hole as well.  All they want to sell you is incredibly expensive monthly subscriptions / licenses.  I never did find out what they would want to sell me what I wanted.  No wonder pirate keys circulate on the web.

Just so much pain, just to get started with anything.

In corporate IT departments they have pages on their intranets, describing just what “incantation” will make the software work as they need it to.  These can be very lengthy and detailed indeed.  Sixty sections of instructions is nothing.  But no ordinary person can spend that amount of time.

I’m still busy collating the text of John the Deacon against the manuscripts.  This was just something different to do.


Analysing the manuscripts of the Life of St Nicholas by John the Deacon – part 2 – the 12th c. manuscripts

In my last post, I analysed the 9-11th century manuscripts of John the Deacon, and found that they fell neatly into three families.  These I have colour-coded as green, blue and purple.  I’ve only really got three data points, so this is all a bit provisional.  The other three turned out not to vary much.

This evening I have completed the task of applying the same 6 passages to the 12th century manuscripts.  The same three families appear; but we also  get a brown family, with mixed readings.

This is perhaps to be expected.  But this determination is relying on a single data point in each case, which is certainly too few to be conclusive.

I had to download another four manuscripts last night.  One of these proved to have enormous page images, so that the whole download was 3.2Gb in size!  This proved too much for Adobe Acrobat Pro 2020, which combined all the images into a PDF, but then refused to save the PDF as “too large” (?!)

I’ve also found a second manuscript in Beneventan book-hand, where again the “Nacta” looks awfully like “Notata” if you don’t know the unusual shape of Beneventan “a” and “t” (which is well explained in this link).

I’m also finding more examples of abbreviated versions of the text, or a text which really belongs to a different version of the Life of St Nicholas.  These, of course, I have to ignore.

I shall have to ponder what all this tells me!


Apologies for spam

Spam is now starting to appear here in the comments.  For this I apologise.  It seems that Akismet have decided to force everyone to pay for their product.  To make us do this, this evening they have disabled my API key without warning.  Please bear with me while I deal with this.

Greed is horrible.

In the short term, all comments will require manual approval by me.  Sorry.


Analysing the manuscripts of the Life of St Nicholas by John the Deacon

I have made a full collation of all the 9-10th century manuscripts of John the Deacon’s Life of St Nicholas, as far as the beginning of chapter 6, where manuscript Q (BNF lat. 17625) breaks off.  I’ve recollated the first chapter, since I did that in a rather perfunctory way.

But how do I work out the relationships of the manuscripts?  I’m doing this blind – I can find no “how to” guide – so I’m just guessing, and trying things out.

This evening I decided to pick three places in chapter 1, where the collation already suggested differences in the manuscripts, and collate the 11th century manuscripts for these places in the text.  I’ve put a “heading 3” in my text, so that I can see the Latin around the area:

Screenshot of my Word text with H3 markers visible

I’m experimenting with a trial of Adobe Acrobat Pro 2020 (permanent license), which allows me to open the PDFs in tabs, unlike my elderly copy of Acrobat Pro 9.  I took the opportunity to add bookmarks and stickys to the PDF of each manuscript, as far as chapter 6, as I went.

After I had collated the 10 manuscripts for the three places in chapter 1, I felt the results were a bit thin.  So decided to collate another three places from chapter 5, where I knew that a line, or a phrase, was omitted.

This I did in a separate Word document.  I had a list of manuscripts; and I indicated the 6 places, comma-separated, against each.  In retrospect a spreadsheet might serve better.  They all started out as black text.

But the results were rather interesting, and here they are:

List of manuscripts and variants

Once I was done, I colour-coded manuscripts that were basically the same.  I have three groups!

Not all of my “places” were significant, at least in the 11th century.  Thus I chose “inclammationem” because I had a bunch of witnesses on both sides:

  • “inclamationem”, “crying out against” – Fal., M, P, Q, O, B, C; “in cachinnationem”, “in immoderate laughter” – Corsi, A, Linz 473 (13th), Munich Clm 12642 (14th); “in vocem” – Mom., Lipp.;

But in actual fact there was no variation on this, at least not in the 11th century.  It looks as if it must originate later.  Likewise the sentence beginning “hactenus” and the clause starting “trade” are unimportant.

But “et laudem / ex laude” and “aede / sede” form a clear group.  Likewise the weird Nacta / Notata / etc lines up with them, and splits the “ex laude” group further.

That’s a useful result.  I have learned a bunch about ten manuscripts from this exercise, which took me less than an hour.

So far so good.  Onward.


From my diary

For the last week I have been steadily collating a group of manuscripts against my text of John the Deacon’s Life of St Nicholas.  I have a list of the earliest manuscripts accessible to me, in century order.  I have a 9th century manuscript (M), and three 10th century manuscripts (P, Q and O), which form the core of what I am doing.  These are the “four horsemen” of my edition.  I also have the four early editions.  Finally there are two manuscripts of the 10-11th century (V and B) and a couple of randoms that I am consulting as well.

It’s actually rather rewarding.  It brings you into very close contact with the text.  You start to get a feel fairly quickly for the relationships.  Quite often the four horsemen disagree with the early editions, but the readings of the latter sometimes appear in V.  P, Q and O are closely related; so much so that when P is unreadable – and it is getting very unreadable – you can often work out what it is saying from Q or O.

I’ve steadily made a complete collation as far as the third sentence of chapter 6.  But I have now run into a problem.

P was OK when I started but is largely unreadable except on the spine-side of the page.  But I have just turned over the page in Q, to folio 261v, and found, to my horror, that there is no more.  The text breaks off after the third sentence of chapter 6.  There’s some irrelevant material written later; but there are no more folios in the manuscript.  Indeed the suddenly worn state of the page indicates that this was indeed the last leaf.

The last page of Q – BNF lat. 17625 (10th c.), fol. 261v

That’s a nuisance.  I was quite happy to continue collating away.  But I shall need to change my approach.

I already have three or four places which I have labelled “major variants” – where a sentence is missing in some witnesses, or else the variation is great enough to separate the manuscripts into families.  It’s not as many as I would like, but it ought to be enough to start to classify later manuscripts.  I have thirteen manuscripts of the 11th century to hand.  Maybe it is time to mark these up, for these variants.  Is there, perhaps, a copy of P / Q, with which I can carry on the full collation?

The other problem, that today I have started to feel, is that it is a pain to have fifteen Adobe Acrobat windows open, containing the editions and the manuscripts.  I’m managing because I have three screens on my PC, and I make sure they all open in the right-hand screen.  But what I want, I suspect, is tabs.  I want them all to open in a single window, and allow me to arrange them.

Foxit Reader does allow me to open them in tabs; but I don’t seem to be able to open the set in one go.  Pity.  Much worse is that I can’t add sticky notes.  Bookmarking chapters, and adding stickies is something that I am doing as I go through collating each manuscript.  It really helps.

So I shall have to divert from the task for a bit in order to solve these side-problems.


A new BHL-type database of Latin hagiographical texts and manuscripts at the IRHT?

It seems that something is going on at the IRHT (= L’Institut de recherche et d’histoire des textes).  For those who do not know, the IRHT is the French manuscripts people.  They do all sorts of very useful things.  But there is no announcement, nor much online.  It’s a new database, designed to allow you to look up a particular Latin hagiographical text, and find what manuscripts it appears in.

We already have something of the sort.  A couple of decades ago, the Bollandist “BHL” index (= Bibliographica Hagiographica Latina) was turned into the BHLms database.  I use it intensely.  You look up a text by the BHL reference and it will tell you the libraries that have copies in manuscript, the manuscript reference number (= shelfmark), and the pages or folio numbers.  It’s great. But it is obviously old.

Last night I saw a job advert.  It’s for 6 months, starting in September.  It’s actually in various places, but I saw it here.  Excerpts (Google translate) –

Job offer – Study engineer (M/F). Feeding a database devoted to Latin hagiographic manuscripts

As part of the “Latin Legends” project – the result of a partnership between the IRHT, the Société des Bollandistes (Brussels) and the University of Namur – the IRHT offers a 6-month fixed-term contract to contribute to the data load of a new database dedicated to Latin hagiographic manuscripts.

This database, still in the process of being run in, is intended to list all the manuscript witnesses (medieval and modern) of Latin hagiographic texts identified by the “Bibliotheca Hagiographica Latina” (“BHL”). Once published, it will work in conjunction with the other databases hosted by the IRHT (Medium, Pinakes, Biblissima).

Sounds interesting!

– he will contribute to integrating the data from the Bollandist files (approximately 9000 handwritten files, now digitized on the IRHT servers). If necessary, he will complete the information, and, in case of doubt, will verify its accuracy.  …

The recruited research engineer will work under the scientific direction of Cécile Lanéry-Ouvrard (IRHT Researcher), within the Latin section of the IRHT (Aubervilliers site). He will also be in contact with the other researchers and technicians involved in the “Latin Legendaries” project (in particular Cyril Masset, from the IT department of the IRHT, and Fernand Peloux, researcher at the CNRS in Toulouse). On occasion, he may be required to correspond with the Société des Bollandistes de Bruxelles, or with other researchers specializing in hagiography. 

Part of the work may be carried out remotely, with this restriction, however, that he has permanent access to the tools necessary for the proper performance of his activities (access to IRHT servers, access to bibliographic tools, etc.)

The list of skills is formidable, but this isn’t really an IT role, as one bit amusingly makes plain:

some skills or previous experience in the field of databases and their operation would be welcome, to facilitate exchanges with the IT department of the IRHT, responsible for maintaining the database.

So it basically requires manuscripts skills, rather than hard-core SQL.  I wonder what the database actually is.

A bit of googling reveals another worker on the project, Antoine Charrié-Benoist, who is listed as doing data load for the BHLms database, and gave a paper about the project.

It would be good to know more.  But clearly whatever database is planned will be immensely useful.  Excellent news.


From my diary

I have returned to working on the Latin text of the Life of St Nicholas by John the Deacon, collating it against a bunch of manuscripts.

Working on the text is a question of repeated passes, as I learn more and work out what I need to do.  Last time I combined the 13 chapters into a single file, but the Latin text and English are still interleaved.  I don’t have a stemma.  But I do have quite a lot of manuscripts in PDF form on my disk.  There are too many to collate the lot.

Last time through I collated the early editions – Mombritius and Falconius, with a certain amount of the modern but non-critical Corsi edition.  I also looked at whatever manuscripts I then had.  So I have notes under the Latin text, and indeed the English, which look like this:

Quid plura**? Ingruente** inedia, tres virgines, quas habebat filias, quarum nuptias etiam ignobiles spernebant viri, fornicari constituit, ut earum saltem infami commercio, infelicem ageret vitam.

** “quid plura (dicam)”, what more should I say? An idiom.
** Mom., Fal.; Corsi: “ingrediente”

What more can I say?  With his hunger** increasing, he decided to prostitute** his three virgin daughters**, whose hands in marriage even humble men spurned, so that by their infamous trade he might at least carry on his unhappy life.

** deponent verb
** Lit. “three virgins, whom he had as daughters”

My Word document is under version control (using Git), so I can safely remove stuff.  The English notes will get deleted.  The Latin needs revision.

But the early editions are hardly a reliable source for the text. What I should be using is the earliest and best manuscripts.  Unfortunately I don’t know what the “best” manuscripts are.  But it was fairly obvious that, if I collated against the earliest few – whichever they were – then I ought to improve the standard of the text.

So I went through my collection of manuscripts and established what the earliest ones are:

9th century

  • Milan P.113sup (last half 9th) = M


  • BNF lat 989 = P
  • BNF lat 17625 = Q
  • Orleans 342 = O


  • Vat. lat.1271 = V
  • Vat. lat.5696
  • Munich CLM 3711 (early 11th) = B

11th 3rd quarter? or “post 950”?

  • BNF lat 18303 = C

Plus a bunch of 11th century manuscripts.  I have this list open in a Word document.  I assigned sigla to the first four manuscripts, which I knew I wanted to collate against my text.  BNF lat. 18303 is a funny one; my information on the date of the text varies wildly.  But it’s clear, little abbreviated, and I just plain like it.  So I’m using it as a second-string source.  Others in the list, as I start to use them, get sigla.

Why am I using the later mss at all?  Because my text derives from the early editions.  If all the early manuscripts disagree, it’s nice to know if there is a manuscript recording the edition reading or not.  I’m not spending much time on that, but a glance at a few later ones can sometimes tell me.

Because I don’t have a stemma, I have no idea how independent the first 4 manuscripts are.  The only way to find out is to try collating them, to learn by doing.  If they are all identical, but different from the early editions, then plainly there is another family of manuscripts around.  It’s a guess, basically; the manuscripts are early, so they ought to have less corruption.  But it’s practical for me to collate 4 manuscripts.  It’s not practical to collate 60.  Even if I know that “recentiores non deteriores”, that “later may not be worse”.  But I won’t know until I’ve done a lot more collating.

It seems that creating a critical edition is just like everything else.  It has to be done iteratively, repeatedly working again and again through the text, learning all the while.  It’s hugely wasteful of time; but there isn’t any other way.  You learn as you do it.  As you search, and research, you find resources and have to go back and use them.

For instance last night I discovered the “History of St Nicholas” in the Golden Legend, in Latin, and in Caxton’s English.  I was googling for a particular phrase, and up it came.  Of course the Golden Legend derives from John the Deacon, so some of the Latin is the same, so the English is a control on my own translation.  Except that Caxton is very loose!  (Is there a modern translation?)  So… that’s another resource.  I ought to go back through my text and translation and check against it.  That would be another pass, once more through the text.

I’m probably not as far along as I think I am.  I feel that I am close to completion; yet there is all this text critical work to be done.

As I collate, I am finding that M, the Milan manuscript – the only one of the Milanese manuscripts that I could get – is indeed somewhat different from P, the earliest Paris manuscript.  But this becomes unreadable through wear by chapter 4.  Q seems to be much the same as P; the Orleans manuscript is mostly the same, but has at least once gone completely off-piste.

I’ve begun chapter 4, and I have found that the Mombritius text is, as I thought last time, more reliable than the Falconius text.  But I am finding the Falconius reading sometimes, and sometimes only in V or  By the time I reach the end of chapter 13, I will have a collation of the lot, and a much better idea of the text and these manuscripts and their character.  No doubt I shall find that I have to go back yet again to apply whatever I learn this time.

Oh well.  Onward.


Plutei manuscripts online at the Biblioteca Medicea Laurenziana, but … not useful

Quite by accident, while googling for a Latin incipit, my attention was drawn to MS. Firenze, Biblioteca Medicea Laurenziana, Plut. 20.2.  Not, of course, to the online manuscript itself, which remained resolutely hidden to my Google search.  But rather to the MirabileWeb page, here.

From this I learned much about this hitherto unknown (to me) manuscript.  I did not learn the date of the manuscript – who needs to know that, eh?! – but I did learn that the 6th item in the book – a legendary for the whole year -, on folios “9-16”, is none other than the Life of St Nicholas by John the Deacon.  It’s not even listed in the Bollandist BHLMs, but it’s a manuscript of our text alright, and a jolly nice one.

But is it online?  Well, who knows?  The digital library there is at, which is as user-friendly as a cornered rat.  I selected manuscript, Plutei, asked for more shelfmarks, and was able to find Plut.20.2, displayed with an obviously temporary address, as a mass of unreadable thumbnails.  The site also told me that the manuscript is 11th century.  So it’s a (capital letters on) Manuscript Of Significant Interest to us.

What I actually needed was way simpler:  I needed  Now that’s a great URL address!  It’s simple, and it’s obvious.  Someone at the BML site is on the ball!  But I didn’t find any indication that this was the url on the site itself.  I only discovered this trick accidentally while messing around with Google.

At least I now know how to find BML manuscripts online.

Now into the images.  There’s no IIIF interface that I could see, so we are reliant on whatever browser the library staff (who won’t be using it themselves) care to give us.

At first sight it’s not too bad.

That’s a very nice, clearly written manuscript – all good – and all we need now is to download the part of it that I want.

Which you can’t do.  No PDF download.

My next thought was whether I could get individual images – I only need a dozen pages – but no luck here either.  There is a “download” of individual pages, if you right-click on them, except that it doesn’t do anything useful.  All it gives you is a screen grab of whatever is on the screen – either a tiny image, or part of an image.  No dice.

So I can’t actually work effectively with this manuscript, or consult it unless I want RSI from all the dragging and squinting.  The BML ought to talk to the Austrians at, if they want to force researchers to use their site.

In fairness this is clearly version 1.0 of the site.  It’s hardly usable, but it’s still better than nothing.  I can’t seriously work with the manuscript through that dreadful interface, nor can anybody else.  But no doubt things will change.


Forty-Seven Latin Miracle Stories of St Nicholas – Now Online in English

I’ve just uploaded a file containing the Latin text, with a translation, of 47 of the miracle stories of St Nicholas which are found in medieval manuscripts.  These are BHL 6130-6147 inclusive.  A couple of the texts I have transcribed from manuscripts online.  Most are from the Bollandist catalogues of the Brussels and Paris libraries, or from early editions.  The translations are basically from Google Translate, but I have at least read over them and fixed some obvious errors.  As usual I put this file and its contents in the public domain – do whatever you like with it, personal, educational or commercial.  Just don’t put your own copyright notice on my work, thank you.

Here are the files:

I’ve also uploaded them to here.

I made this file while working on the Life of St Nicholas by John the Deacon.  In the manuscripts this gets tangled up with all these texts, and it gets fairly confusing.  With this file, all I have to do is a Ctrl-F Find, and I can at once see just what the page of text in the manuscript image in front of me belongs to.

There are still more miracle stories to do, but I ran out of puff at this point.  Maybe one day I will return to it and add more!  Or maybe not.


Another drawing of the Septizonium

Uploaded on Easter Sunday to was an interesting volume, with the description: “Salzburg, Universitätsbibliothek, M III 40, MONUMENTA ANTIQUA ROMANA — Lombardei, 4. Viertel 15. Jh.”  I.e. 4th quarter of the 15th century, 1475-1500.  It contains sketches of a number of the monuments of ancient Rome then standing, and notes about these, which seem to be in Italian.  I wish I could read these.

Via this site, I learn that this book is Giovan Maria Falconetto, Monumenta Antiqua Romana.  HMML say here that it is possibly by Giovanni Maria Falconetto.  There’s a Wikipedia article about him, but I was unable to find any information about the manuscript.

This page (f.21r) was particularly interesting:

For this is surely the ruins of the Septizonium, the monumental facade erected by Septimius Severus at the end of the Via Appia as a main entrance to the Palatine complex.  The notes on the right of the drawing look like measurements!

Wonderful to see!