Text annotation and Textual Criticism in the Manuscript

This month I had the pleasure to participate in a workshop on Glossing in Celtic Contexts to discuss some of the work I am doing on the editorial glosses in the Franciscan manuscript (MS UCD-OFM A7) of Félire Óengusso. As you can read here, this manuscript was written by Ruaidhrí Ó Luinín on behalf of Cathal Mac Magnusa Meg Uidhir (d. 1498). Ó Luinín was not a passive copyist of the text, however—something I have previously discussed in a lecture on the Preface to the Félire. In my talk, which I am expanding at present, I discussed the particular way in which this scribe marks text-critical annotations.

While this is still work in progress, the talk reminded me that for those of you who are interested in such things, I might share with you some of the more common text-critical graphs you would encounter when reading insular manuscripts.

Some of these you might have spotted already, at the top of the page, in the margin of the header image, which is from one of the Rawlinson manuscripts. There you can see the graph .i. (an i with enclosing dots either side) marking the insertion of a gloss. The enclosing dots (punctus) are often used to mark special characters and numerals. In Irish manuscripts the .i. graph stands for ed ón ‘that is’ and is the most common graph used for adding notes to a text. It is the medieval equivalent of modern ‘i.e.’, which stands for Latin id est ‘that is’, which is also what .i. was modelled on. In the example below, the scribe has added the note .i. do dia (to God) to the phrase slóg mar ba dixu (a host that was higher)*. 

Image illustrating the graph .i.

Bodleian Library, MS Rawlinson B 505, f. 211r Photo: © Bodleian Libraries, University of Oxford

Another very common graph which you might have recognised on the header image, is the graph for ‘or’, in Irish (or uel in Latin): ƚ. This graph is used in a range of different functions. Sometimes it represents the word ‘or’, at other times just the letters. In my ongoing research, I am looking at variation of the use of the ƚ graph by one particular scribe, in one specific text, to indicate variants to the main text. A good example is found among the set of notes to the first line of the quatrain on June 13th. While the gloss above the line begins with .i., mbresta is glossed with ƚ primda. In fact, this is not so much a gloss as a textual variant: this is is the word found in this place in the line (instead of mbresta) in manuscript Rawl. B 505. Of course, not all examples are as neat at this one (and the graph for ‘or’ is also still used simply to indicate the word)—but more about this another time!

UCD Archives, MS UCD-OFM A7, f. 22rb, June 13. Reproduced by kind permission of UCD-OFM Partnership.

Not all graphs on the page represent letters, however. Further to right on the header image, you can see a symbol resembling • • at the beginning of a marginal gloss. This symbol is a critical sign indicating that the gloss is a continuation of a gloss that was started elsewhere. It operates as a tag mark, tagging this line to the original gloss, and is therefore a reading aid. 

Image illustrating a tag mark

In this case, it refers back to this partner tag mark earlier in the same line:

Image illustrating a tag mark

Bodleian Library, MS Rawlinson B 505, f. 211r Photo: © Bodleian Libraries, University of Oxford

The full gloss comments on the name Luciani in the main text: .i. prespiteri, id est prespiteri Antiochie Maximiano imperatore passus est (i.e., a presbyter, i.e. a presbyter of Antioch who suffered under the emperor Maximian). [Here this refers to Maximinus Daza rather than the earlier emperor Maximian.]

Finally, another feature that is very common is something called the ceann fo eitte (head under wing). It is more of a punctuation feature, in that it usually marks the end of a paragraph, verse, or line. Sometimes it is repeated to fill up the remaining space. Other times writing continues after the ceann fo eitte, often continuing from the line below.

Here is a selection from MS B 505:


Interested in exploring more? I can recommend this helpful article by my colleague Evina Steinová, which deals with all sorts of technical annotations, as well as point you too the special section on Punctuation, critical signs, and numerals in the Tionscadal na Nod area of CODECS, published by the van Hamel Stichting. Happy reading!

*The discussion about whether or not this text uses the comparative for the superlative sense is one for another day!

Quire structure added to KBR 5100-4

I have been experimenting a little with visualising quire structures recently and have now added a link to such a visualisation in the description of manuscript KBR 5100-4. You will be able to navigate a schematic illustration of the quires that contain Félire Óengusso based on my analysis of the manuscript last February. Images are courtesy of the Koninklijke Bibliotheek in Brussels and the schematics tool used is the Schoenberg Institute for Manuscript Studies Collation Modeler (VisColl), the wonderful work of Dot Porter.  The visualisation can’t currently display large quires (such as we see often in Irish manuscripts) very well, but Dot tells me there is a new version on the way that can, at which point I will update the link 🙂


If you are not sure what I mean by quire and quire structures, this next section is for you!

When discussing how a manuscript is put together, scholars often have to refer to its structure. This is not unlike modern book. In fact, if you have a hardback to hand while reading this, I suggest taking one out and looking at it from the top. When you look closely at the point where the pages meet the spine, you will very likely see that the pages were not pasted together as a loose stack, but as small groups or booklets of pages. (Depending on the publisher.) Like in this book on the right for example.

This is true also for most manuscripts, except that we do not always call the leaves of a manuscript pages (and where they are called pages, this is because they have been numbered as pages).

Most manuscripts consists of similar ‘booklets’ bound together, which we call quires (or gatherings). Each quire consists of a number of sheets or folded together. A simple way to create a small quire or booklet yourself is to take a number of A4 or A3 sheets, say four, place them on top of each other, and then fold the bundle in half across the short side. You will now have a booklet with four folded sheets or bifolia and eight folia: each folded sheet is called a bifolium and forms two leaves or folia in your booklet.

If you were to paginate these like in a modern book, you would number them 1 to 16, but with manuscript folios we normally refer to the front and back side, the recto and the verso, of the folio. This is called foliation. Instead of having  a page number 1 on the first leaf of your booklet, you would refer to it as folio 1r, and to the backside as 1v.

Quarto. Source: www.lib.umich.edu

To create such neatly folded quires, a scribe would have taken a sheet of paper or vellum (animal skin) and folded it to right size.  This might mean folding a sheet more than once. For instance, if you take one sheet of paper, fold it in half, and then fold it in half again, you have a quarto. Your quire now has four leaves or folia.

If you look again at the image of the quire structure of KBR 5100-4 above, you will see, on the left, a series of curved lines with numbers. This schematic illustrates how the part of the manuscript that we call quire 7 fits together. Each of these lines represents a folded sheet or bifolium and each folio is referred to by recto and a verso. In the visualiser loose, individual images have been stitched together, but it shows you each side of the whole bifolium. This way we can visualise the structure of the manuscript without having to take it apart.


If you want to know more about manuscript terminology and making manuscripts, you could start here:

https://blogs.bl.uk/digitisedmanuscripts/2017/09/a-rough-guide-to-making-a-medieval-manuscript.html

https://www.bl.uk/catalogues/illuminatedmanuscripts/glossary.asp

(this page was edited in 2020 to add an image)

The little things (KBR 5100-4)

Having recently visited the Royal Library in Brussels for the second time to study MS 5100-4, I wanted to share with you some small discoveries that put a smile on my face.

For those of you new to the manuscript (see here), this is a composite manuscript consisting of individual booklets of texts copied by Michél Ó Cléirigh as part of his work for the Franciscans in Louvain, who at that time began to collect religious materials from Ireland for posterity. The separate ‘copy books’ can be identified, among other things, by the scribal pagination. There are three numbering systems in the manuscript: foliation in dark pencil, scribal pagination, and from quire 2 (second copy book) onwards, light pencil foliation that continues onward from the scribal pagination of quire 4. The main sections in the manuscript, as indicated by the scribal pagination, consist of a section containing poetry, the prologue and commentary to the Félire, the Félire itself, the Martyrology of Gorman, and the Martyrology of Tallaght.

Quire Contents Dark pencil  ff Light pencil ff Scibal pagination
2—4 Poetry 6—39 1—67 (39r)
5—6 Preface and Commentary to Félire 40—65 68—93 1—45 (ends 65r)
7—8 Félire Óengusso 66—93 94—121 1—51 (ends 91v)
9—19 Martyrology of Gorman (9, part of 18, and 19 are autographs and notes) 94—180 122—207 1 (=100r)—141 (ends 170v)
20—21 Martyrology of Tallaght (foll. by Comainmniughud) 180—201 208—228 1—12 (ends 192r, text ends 197r)
KBR MS 5100-4, sections with scribal pagination © Nicole Volmering, March 2018 (research carried out in February-March 2016 and February 2018)

Index to the commentary, KBR MS 5100-4 f. 42r (own image) Low resolution images posted courtesy of the Koninklijke Bibliotheek van België.

Even if your main interest is only in part of the manuscript, it is always helpful to get a sense of what the scribe was doing with the compilation as whole. In this particular manuscript I came across some nice personal touches. For instance, the commentary to the Félire is preceded by an index (fol. 42r). This paints a lively picture of the scribe trying to organise and navigate through his material (which a little bit of a jumble of short stories). The index includes the page numbers as well as occasionally the year of the saint’s death.

Tags indicating the start of each month of the Martyrology of Gorman. KBR MS 5100-4, f. 100r (own image) Low resolution images posted courtesy of the Koninklijke Bibliotheek van België.

There are further signs of active organising too, like this lovely set of little tags, neatly pasted in, marking the start of each month of the Martyrology of Gorman. A few have broken off, but on some the name of the month is still readable. The text of the martyrology, with numbering and dominical letters, begins on f. 100r starting on f. 100 (f. 128 in light pencil), at one time numbered 1 by the scribe, to mark the start of the text (the new quire beginning with the previous folio).

Also visible on this photo is the ruling, which shows that the main text was written one line of text per line, while the glosses were written roughly half-size, filling the ruled space with two lines of text. The number 1, marking the start of this section, sits perfectly between the double-ruled bounding lines.

Fingerprint, KBR MS 5100-4, f. 193v (own image) Low resolution images posted courtesy of the Koninklijke Bibliotheek van België.

My favourite discovery however is on fol. 193v. At the bottom right just above the start of a catchword (referring to Lasar on the next page), a fingerprint was left on the page, presumably due to a small ink spill at the time of writing. Might this be Michél Ó Cléirigh’s own fingerprint?

Update: As of September 2019 you can now finally inspect the full contents of this manuscript on: isos.dias.ie!

(Edited in 2020 to add reference to digital catalogue and move page)

Stokes and the Félire: Four Decades of Editing

Earlier in the summer I gave a paper at the Second European Symposium in Celtic Studies (1 August) about the background to Stokes’ editions of the Félire Óengusso. I am now sharing some of it on this blog for those of you interested in such things :).  

As many will know, Stokes produced two editions of the text, one published in 1880 as the first volume of the Transactions of the Royal Irish Academy Irish Manuscript Series, and one published on 1905, with the Henry Bradshaw Society, which had been founded in 1890. While Stokes’ 1905 edition, now widely distributed in the reprint format issued by the Dublin Institute for Advanced Studies (1984), is still the standard edition of the text, it has not escaped scholars that there are some inherent problems with the manner in which the edition was laid out and printed. Anyone who has used the edition to consult the copious glosses and scholia will have come away from it somewhat baffled with the arbitrariness of their presentation: Stokes indicated from which manuscript he took the individual lines or paragraphs, but he did not include (and it is therefore impossible to determine based on the edition) how much of these are also in the other manuscripts. The apparatus to the main text is likewise eclectic, as well as erroneous in various places, and then there is the odd omission of certain copies of the Preface. By contrast, the 1880 edition, while more limited in scope, is laid out much clearer. It would in many ways of course be rather unfair to hold Stokes’ editions up to the standard of modern editorial scholarship, which has made significant advances and is still continuously changing, given that his studies of the text were carried out over many decades, and at a time when manuscript facsimiles, digital or otherwise, were not nearly as readily available as they are today. However, Stokes’ was not known to be a particularly careless scholar and so, before labouring to remedy the problems, I am pausing to ask the question: can we identify anything in Stokes’s approach to the manuscripts or his editing process that might shed light on the format and shortcomings of the edition?

Stokes interaction with the Félire Óengusso over the course of his career, is marked by repeated – and characteristic – (self-)correction and revision. Stokes himself regarded his editions as works in progress and candidly claimed only a “temporary and provisional value” for his first edition, professing himself unable to “render with certainty several portions of the text” (1880: 1, 26). This edition was originally prepared for the Royal Irish Academy in 1871, but since it was not published until 1880, the final version includes corrigenda collected over the nine intervening years in the introduction. This first edition is a synoptic edition based on diplomatic editions of four primary manuscripts, Rawl. B 505, Rawl. B. 512, Laud Misc. 610 and the Leabhar Breac (RIA 23 P 16), supplemented in part by readings from the Franciscan manuscript, UCD-OFM A7, and has clear referencing throughout. For the 1905 edition on the other hand, he attempted a critical edition. This, unfortunately, is opaque and confusing in its approach. His editorial method is summarised in the phrases “in forming the text” and “with various readings from the ten MSS. in which it is wholly or partially preserved” (1905: vii). Note that the phrase says various readings, not variant readings, which ought to serve as a warning for scholars not to presume that the apparatus is complete. Indeed, concerning NLI G10, he specifically says: “All the various readings of this fragment, which seem of the slightest importance, are given in the footnotes marked C” (1905: xi, my emphasis). Likewise, of the Brussels manuscript (KBR 5100-4) he states that “all its important readings are, in the present work, marked B, and are given in the footnotes” (1905: viii, my emphasis). When this edition was published in 1905, more than 35 years after he first started work on the text, Stokes again felt that it was not what he had hoped it to be, writing:

“I am far from regarding the present edition as definitive. I know too well that I have not realised my ideal” (1905: vii)

It would have been wonderful of course, to know exactly what Stokes’ ideal edition entailed. At most, we can deduce from the printed text and from his comments on his previous editions and the manuscripts, that he envisioned a critically reconstructed text. In his 1905 Editor’s Preface he wrote:

“The first edition of the metrical Martyrology called in Irish Félire Óengusso Céli Dé was published in 1880, by the Royal Irish Academy, in parallel column, from four corrupt and uncorrected texts, hereinafter denoted by R1, R2, L and LB.” (1905: 17)

In his manuscript descriptions, ‘corrupt’ generally refers to Middle-Irish readings and interference from later scribes which deviate from the archetype, by which he seems to mean the original Old Irish version of the text (1905: xvii, xxiv). Now that I am myself working towards a modern critical edition of the Félire Óengusso, it will come as no surprise to anyone that I have repeatedly wished that he had set out his editorial policy and methodology more clearly, especially in light of some significant omissions in the edition, such as the calendrical material transmitted with the text, which I discuss is a forthcoming article, and, perhaps most glaringly, copies of the prose Preface. Such errors and omissions may at least in part have been influenced by his level of access to the primary sources, as it is not clear from the 1905 edition exactly what materials he had available to him and in what format. Much of this may be reconstructed from his publications and letters, and I hope to be able to add more details as research continues.

Exactly when Stokes began work on the Félire is unclear to me at present (there may be more information in his letters than I have unearthed to date, but I have not yet had the chance to peruse them all). One solid anchor point is naturally the first edition, likely completed in the summer of 1871. In a letter he wrote to John Rhŷs in Dublin, 18 July 1871, roughly a month after presenting his first edition to the Royal Irish Academy, he writes that he spent the previous week working on the Bodleian Irish manuscripts and also confirmed that he had received a favourable answer from the publisher regarding publication of the first edition:

Stokes-Rhys Correspondence (Copyright: National Library of Wales, 1v)

Stokes-Rhys Correspondence, 1v. Copyright: National Library of Wales.

“You will be glad to hear that Williams of Norgate have agreed to publish my three-text edition of the Félire of Oengus céle-dé (Culdee) with a copious vocabulary” (Letters and Cards from Whitley Stokes, National Library of Wales, 1v)

In his revised preface to the edition, written in 1882 and published in Revue Celtique 5, he recalls carefully collating the copy in Laud Misc. 610 in 1871 with a copy made for him at an earlier date by Mr Hennessy (1881-3: 340). At this stage he thus certainly had access to all three of the Oxford manuscripts as well as the Leabhar Breac, which formed the basis for his 1880 editionHe nevertheless refers to it as a three-text edition rather than a four-text edition, presumably because Laud Misc. 610 and the Leabhar Breac are edited beside Rawl. B 512 (which lacks the main text) for the Prologue and Epilogue and beside Rawl. B 505 (which lacks the former) for the main text. His edition of the glosses and scholia was mostly based on the Leabhar Breac, but he also used scholia from Laud Misc. 610 and Rawl. B 512  “with a few from the Franciscan copy” to fill lacunae in the Leabhar Breac text (1880: clxxviii). He does not, however, seem to have had access to this last manuscript yet, writing that

“Of the part of this (with the notes and glosses thereon) dealing with the month of December, I possess a transcript made by the late Dr. Todd.” (1880: 3)

In the revised introduction he positively identified the manuscript he previously only referred to as that from St Isidore’s in Rome as that now known to us as UCD-OFM A7, by giving the title on the cover: “It is covered with parchment endorsed ‘Martyrologium Cathaldi Maguir sive ængussius Auctus No. 7′” as well as (in a footnote) Todd’s transcription of the scribe’s colophon (1881-3: 342). He further added that he had “lately, myself, transcribed the prologue, which is copiously glossed and affords some good readings … ” (1881-3: 342). Likewise, he adds that he had “lately transcribed the parts relating to January, February, March and April” from RIA 23 P 3 (1881-3: 342). A footnote in the same volume confirms that he had not yet seen the Cheltenham manuscript, now NLI G10 (1881-3: 302). What is unclear from these statements is in what format he was able to consult the manuscripts. It appears, for instance, that he may never have had the opportunity to inspect A7 in person, as he writes in a letter to Rhŷs in 1903

“I am slaving (and straining[?] my eyes) at the edition of Félire Oengusso, which I have promised the H.B. Society.  …. I copied from my photograph of the Franciscan ms. a note ….” (Letter dated 14/12/1903, NLW, Letters and Cards from Whitley Stokes 121r)

If this is the case, the quality of the photographs might have something to do with the fact that he missed the prose Preface in this manuscript.        

Stokes was actively collecting feedback on his edition in the second half of 1880, specifically asking Rhŷs for more notes, some of which are available in the Stokes-Rhŷs correspondence now in the National Library of Wales. He appears to have continued work on the Félire for at least another few years subsequently. Following publication of his revised introduction to his 1880 edition, with a further set of corrigenda, under the same name, in Revue Celtique 5, he published two related studies in Revue Celtique 6: one on the metre rinnard, and one on metrics in general, which largely deals with assonance, accentuation and alliteration in Irish metre and was intended as a feisty correction of Prof. Atkinson’s metrical analyses. (As an aside, he also candidly berated himself, admitting that: “When I edited the Calendar of Oengus I was ignorant of the true meaning of ard, and stupidly rendered the word by “alliteration” (1883-5: 273)).

Sometime in the early nineties he gained access to the Brussels manuscript, which he catalogued, and from which he then edited the Martyrology of Gorman (published in 1895). He explicitly thanks the staff of the library for their courtesy in the introduction to that edition. His work on the Félire then seems to have suffered somewhat of a pause, no doubt at least in part due to the fact that he was stationed in India for long periods of time. However, work on the Félire had resumed by 1901, when he was revising his text in preparation for a critical edition at the request of the Henry Bradshaw Society. In the corrigenda published in Revue Celtique 23 in 1902 he explicitly refers to the Cheltenham manuscript, now NLI G10, to which he thus must have gained access in the meantime. I have not yet been able to determine, however, whether he saw this manuscript in person, or, like in the case of A7, consulted a photograph. This possibility seems likely as Stokes appears to have missed the copy of the Preface in this manuscript as well.

Could the eclectic nature of the 1905 edition in part be due to issues with access to the source material (rather than primarily due to a laconic attitude to the apparatus)? This would certainly go some way towards explaining the most obvious omissions, briefly raised earlier, that is, the missing copies of the prose Preface. In the 1880/1905 editions Stokes edited the copies from the Leabhar Breac, Laud Misc. 610 and Rawl. B. 512, yet, for some reason currently unclear to me, the prose Preface appears to have escaped his renewed scrutiny in preparation for the 1905 edition. Stokes neglected to edit four of the extant seven copies. In the case of manuscripts G10 and 23 P 3, he appears to have overlooked them completely as he categorically states in both cases that the prose Preface is missing.
I give the locations of the remaining four here:

NLI G10
Concerning G10 Stokes says: “p.19,which probably contained a prose preface, is now illegible”, whereas, in fact, the Preface is on pp. 22-23.

RIA 23 P 3
The prose Preface in 23 P 3 is on f. 12, a folio that probably once stood before f. 1. It precedes scholia to the text, which may explain in part why it was overlooked. In addition, it is fragmentary, lacking the beginning, and is badly effaced at the end. The text now begins with the story of Óengus’ humility and his revelation to Mael Ruain.

KBR 5100-4
Stokes does not mention the Preface of the Brussels manuscript in the 1905 edition, which is particularly surprising because he lists it in his catalogue of the manuscript in the edition of the Martyrology of Gorman and thus was clearly aware of its existence. In the manuscript itself, the Preface is separated from the main text, beginning at folio 40r of quire 5, in the section with scholia preceding the Félire proper, which starts at folio. 66r (aka 94r) in quire 7.

UCD-OFM A7
It will not come as a surprise to those of you who have worked with A7 that Stokes appears to have made no attempt to transcribe the prose Preface from this manuscript. The first recto is heavily stained and defaced in parts, rendering the text nearly illegible. The catalogue entry for the Preface merely states that folio 1v is:

“Defaced and largely illegible. Ends Is he nó chanad amhlaid. Verso continues the second preface, printed by Stokes, op. cit., from Laud 610, with the words so cen boi ic disirt Oenghuso. The text varies from that of Laud 610, and ends like Rawl. B 512, with quatrain beg. Taidet remain slighidh saethraigh.”

To the best of my knowledge no further attempt has been made to restore or decipher the text of this or any of the other missing copies, which is unfortunate, because the Preface has the ability to provide further clues on the transmission history of the text. In advance of a forthcoming article in which a more detailed study will be provided, I outline the structure of the Preface here for reference.

As most of you probably know, the Preface addressed various topics, which, for easy comparison, I have subdivided into four logical units based on content, narrative coherence and spacing the manuscripts. In general, the Preface has a clear internal sequence. With the exception of the Leabhar Breac, all copies more or less adhere to this sequence, which suggest that the logical order of the four units in the manuscripts is that presented here:

  1. Outlines the place, person, time and cause of composition of text, culminating in the story of Óengus inspiration for writing (partly omitted in Laud, G10 and 23 P 3)
  2. Describes how Óengus first reveals the Félire to Fothud na Canóine
  3. Describes Óengus’ interactions with Mael Ruain, including the reference to the epistle which fell from heaven and marked the location of the monastery (in L and P only); the narrative concerning Óengus’ humility, hiding as a slave in Mael Ruain’s kiln; and the story of the boy who learned his lesson while asleep, ultimately revealing Óengus to Mael Ruain.
  4. Discussion of the metre.

Thanks for reading!

If you enjoyed this, you can sign up to follow this blog by email!

Editions and Studies on the Félire by Stokes 
Stokes, Whitley, ed. On the Calendar of Oengus. Vol. 1.1. Transactions of the Royal Irish Academy Irish Manuscript Series 1. Dublin: RIA, 1880.
—–.  ‘On the Calendar of Oengus’, Revue Celtique 5 (1881-3) 339-80. (Signed Oxford, 6 June 1882)
—–. ‘On the Metre Rinnard and the Calendar of Oengus as illustrating the Irish verbal accent’, Revue Celtique 6 (1883-5): 273-297.
—–. ‘On Irish Metric’, Revue Celtique 6 (1883-5): 298-308.
—–. ‘Notes  on the Martyrology of Oengus’, Revue Celtique 23 (1902): 83-116. (Signed 13 January 1902)
—–, ed. The Martyrology of Oengus the Culdee. Henry Bradshaw Society 29. London: Harrison, 1905. (Repr. Dublin: DIAS, 1984)
—–. “Miscellen 2. Notes on the Second Edition of the Martyrology of Oengus, London 1905.” Zeitschrift für Celtische Philologie 6 (1908): 235–42.

Letters and cards from Whitley Stokes, 1871-1909
(c) Llyfrgell Genedlaethol Cymru – National Library of Wales

For the manuscripts – see links elsewhere on this page.