Thijs Porck

Home » Posts tagged 'Latin'

Tag Archives: Latin

Missing line of 1400-year old poem found! Columbanus’s ‘De mundi transitu’ once more complete!

In the 9th century, an anonymous scribe copied a poem attributed to the Irish missionary Columbanus (540-615) and made a crucial mistake. He accidentally omitted one line from the poem. Since this 9th-century manuscript is the only extant copy of the 120-line poem, modern editors of the poem have been forced to leave this line in the poem blank. This blog post reports on the fact that this missing line has been found(!), as I reveal in a new Open Access publication (see the link available at the end of this blog post).

Left: Fresco of Columbanus in the Cathedral of Brugnato (Italy) [source]; Right: Opening lines of De mundi transitu, with marginal attribution to Columbanus (Zürich, Zentralbibliothek, Ms. C 78, fol. 159r)

A missing line in Columbanus’s De mundi transitu

The Irish missionary Columbanus (540-615) founded monasteries in France and Italy and was also known to have written pieces in verse. His De mundi transitu is a short but highly wrought poem, written in a style that is typical for Hiberno-Latin poets of the early Middle Ages. It consists of 120 heptasyllabic lines, divided into thirty stanzas of four lines each. In addition to rhyme, which occasionally extends to three syllables, the poem also features alliteration. The topic of the poem is the fleeting nature of earthly delights; everything that is joyful in this life (beauty; wealth; youth) passes:

Cottidie decrescit
Vita praesens quam amant;
Indeficiens manebit
Sibi poena quam parant.

[The present life that they love declines daily; the penalty they prepare for themselves remains unfailingly.]

Columbanus, “De mundi transitu”, ll. 21-24.

Therefore, Columbanus (540-615) holds, a true Christian should focus on the permanence found in Heaven.

This beautiful and powerful poem survives in only one manuscript, which was copied in the 9th century in the Abbey of Saint Gall, Switzerland. When the 9th-century scribe copied the poem, he made a crucial error. Rather than copying line 106 of the poem, his eyes skipped across the page of his exemplar and he accidentally copied out the last part of the wrong line: line 102, which now occurs twice in the manuscript, but does not make any sense in the second instance (a description of Heaven):

Zürich, Zentralbibliothek, Ms. C 78, fol. 160r (underlined, in read, the repeated part of line 102)

Ubi cibo superno
Plebs caelestis pascitur,
Ubi nemo moritur
Quia nemo nascitur,
Ubi aula regia
caelestis pascitur
In qua male resonans
Nulla vox audita est

[Where on celestial food the heavenly folk are fed; where no one dies because no one is born; where the royal hall the heavenly … are fed, in which no voice is heard resounding evil]

Columbanus, “De mundi transitu”, ll. 101-108.

The erroneous repetition of part of line 102 on the place of 106 may have been triggered by the fact that both lines 101 and 105 start with the same word: ‘Ubi’. The scribe did not notice their mistake, but later editors of the poem, who could only rely on this faulty manuscript, have been forced to leave open line 106 or come up with their own lines. However, during my research, I chanced upon a partial copy of Columbanus’s poem in another manuscript, which contains the missing line 106! But before I can talk about that part of the story, I need to talk about the fact that old people do not groan in Heaven.

Heaven is a place where the old man does not groan

When I read the poem De mundi transitu for the first time, I was struck by one of its descriptions of Heaven. According to Columbanus, Heaven is a place “Ubi senex non gemat, / Neque infans vagiat” [where the old man does not groan and the infant does not cry]. In my research into early medieval ideas about old age, I found that Heaven was typically described as a place where old age was absent (see Heaven is a place without old age: Age and the afterlife in early medieval England) and I knew of only one other text that described Heaven in the same terms as Columbanus had done: an anonymous Old English homily, found in a twelfth-century manuscript. This Old English text described Heaven as a place “þær eald ne graneð, ne child ne scræmeð” [where the old one does not groan and the child does not scream]. The phrasing was eerily similar to Columbanus’s De mundi transitu and, upon closer investigation, I found that the Old English homily featured a near-translation of some fifty lines of the poem – something that no one else had noticed!

Oxford, Bodleian Library, MS Bodley 343 – a manuscript with Old English homilies

Finding a Latin source for an Old English text is rare and such a substantial one offers all sorts of opportunities for research (how did the Old English homilist adapt his source; what did he leave out; what did he add, etc.) – I was able to write a 23-page chapter about this find (see the link to the full article below). Unfortunately, the Old English homilist did not include a literal translation of the last lines of Columbanus’s poem (including line 106). He did give a near-literal translation of lines 103 and 104 (“Ubi nemo moritur / Quia nemo nascitur”) as “ne þær nan ne swæltæð, for þam ðe þær ne byð nan acenned” [where no one dies, because no one is born there], but what follows seems more of a paraphrase. However, whilst researching this homily and the poem, I found additional and crucial traces of the missing line 106 of Columbanus’s De mundi transitu.

Since I wanted to make sure that the Old English homilist had not used another Latin text as its source, I did some detective work to see if any other texts mentioned old people not groaning in Heaven (and various other phrases found in both Columbanus’s poem and the Old English homily). When I searched for variants of the line “ubi senex non gemat” [where the old man does not groan] in an online database with medieval Latin texts (Brepols’s Library of Latin Texts), I found a text that contained a version of that line, along with some fourteen other lines of Columbanus’s De mundi transitu, including (YES!) line 106 – it turned out to be the work of another person from Ireland: Sedulius Scottus.

An Irish teacher’s notebook: Sedulius Scottus’s Collectaneum Miscellaneum

Left: Fresco of Columbanus in the Cathedral of Brugnato (Italy) [source]; Right: Sedulius Scottus [source]

Sedulius Scottus (fl. 840-860) was a 9th-century Irish teacher, scribe and poet, who fled Ireland after the Viking raids and ended up in continental Europe, where he certainly stayed in Liège (present-day Belgium) and later possibly moved to Milan (Italy). Scottus wrote various works, including poems and biblical commentaries. One of his works, commonly referred to as the Collectaneum Miscellaneum, is a collection of quotes from mostly biblical and patristic sources (a florilegium). For some passages, names of source authors have been added in the margins of the manuscript, but the origin of some strings of text in Scottus’s notebook are still unknown. On one of the pages of his notebook, Sedulius included some lines of Columbanus’ poem De mundi transitu. The passage from Columbanus’s poem was included between two quotes from Church Father Augustine (354-430) and the entire passage is attributed to ‘AVGVSTINVS’ in the manuscript of the Collectaneum:

Passage from Columbanus in Sedulius Scottus’s Collectaneum Miscellaneum (edition by Dean Simpson, 1998), underlined in red ‘ubi senex non gemit’ [where the old man does not groan]

The misattribution to Augustine of the entire passage (the lines from Columbanus are located in between two Augustinian quotes) has probably resulted in this passage having gone unnoticed in the scholarship on Columbanus’s De mundi transitu. This is unfortunate, since Sedulius Scottus seems to have had access to a copy of the poem that was complete, or, at the very least, did contain line 106 that is missing from the only extant 9th-century manuscript of the full poem!

Here are the lines from the poem in the 9th-century manuscript (without line 106) and the equivalent lines in Sedulius Scottus’ notebook (with line 106!):

Columbanus, De mundi transitu, lines 101-112 and the corresponding lines in Sedulius Scottus’s Collectaneum

It looks as if Sedulius Scottus decided to copy out the lines from Columbanus’s poem that describe Heaven and, crucially, Sedulius’s version of the poem does not appear to have had the scribal error of the 9th-century manuscript from Saint Gall. His version of lines 105-108 read “Ubi aula regia / regis summi sita est, / in qua male resonans / nulla uox audita est”, the meaning of which makes perfect sense: “Where the royal hall of
the highest king is located, in which no voice is heard resounding ill”. The line “regis summi sita est” is also a perfect fit with the rest of Columbanus’s poem – it has seven syllables (like all the other lines), there is alliteration with the last word of line 105 (“regia / regis”) as well as alliteration within the line it self (“summi sita”); and it rhymes with line 108 (“sita est […] audita est”). In other words, it is very likely that the line “regis summi sita est” indeed belonged to the original poem by Columbanus! Thus, by sheer chance and some detective work, the line that was lost by the carelessness of a 9th-century scribe in Saint Gall can now be restored and Columbanus’s poem is once more complete!

For a more thorough, sourced and nuanced discussion of this find, as well as the relationship between Columbanus’ poem De mundi transitu and the anonymous Old English homily you can now read my Open Access (i.e. freely available) publication “Columbanus’s De mundi transitu in Early Medieval England: A New Source for an Old English Homily (Irvine VII) in Oxford, Bodleian Library, Bodley 343” (click the link). This chapter is part of the book The Anonymous Old English Homily: Sources, Composition, and Variation, ed. Winfried Rudolf and Susan Irvine (Brill, 2020). I am currently working on a new critical edition and translation of Columbanus’s De mundi transitu, which will have the missing line filled in!

Word processing in early medieval England: Browsing British Library, Royal MS 8 C III

Many manuscripts were produced in early medieval England and quite a few have gained great renown for their beautiful illumination (such as the Lindisfarne Gospels), their famous texts (e.g., the Beowulf manuscript) or their interesting history (like the Codex Aureus, once kidnapped by Vikings). By comparison, British Library, Royal MS 8 C III, a late tenth-century manuscript, is relatively obscure. With hardly any illumination, some fairly standard texts in Latin and no exciting ‘back-story’, this Anglo-Saxon manuscript does not seem to have invited much scholarly (let alone popular) interest. This lack of attention is undeserved. As this blog post will demonstrate, this manuscript is full of interesting examples of ‘word processing’ in early medieval England.

Initials: Planned, faced and bitten

In most manuscript containing multiple texts, like Royal MS 8 C III, the start of each text is signalled by an initial letter that is larger than the rest of the text. These letters could be executed fairly simple or lavishly decorated. In case of the latter, the initials could be made by a different individual from the scribe responsible for the text; the scribe would then leave space on the page for the initials to be added at a later stage.

The first two texts in British Library, Royal MS 8 C III demonstrate this practice. For instance, the first word of the text starting on fol. 6v, a Latin exposition on the Mass, reads “rimum” but should probably have read “primum” [first]. The very first text in the manuscript, pseudo-Jerome’s De diversis generibus musicorum, even misses the first few words. In other manuscripts, this text starts with “Cogor a te ut tibi dardane de aliis generibus musicorum”, but here, on fol. 2r, the words “Cogor a te ut” are left out. They had probably been intended to be added as a full line of decorated letters, since a lot of space was left open at the top of the page:

Blog.Royal8ciii - initialisplanned

Initials planned, but not executed © British Library, Royal MS 8 C III, fols. 2r, 6v

These two instances of unexecuted initials notwithstanding, Royal MS 8 C III does feature several, simple initials. In two of them, the scribe (or a later reader) added a face; a third was rather beautifully decorated with a dragon biting an O so as to form a Q:

Tweet.RoyalMS8CIIIJustification: Space out your words or stretch out your N’s

If we want our text to be spread out evenly across the page, with straight left- and righthand margins, all we need to do is tell our word processor to “justify” the text. The word processor will then increase or decreates the letter- and word-spacing, creating our desired layout of the text. The Anglo-Saxon scribe of Royal MS 8 C III also appears to have liked justification; on fol. 81r, for instance, he ends his text with a heavily spaced line that reads “deo gratias” or rather “deo      gra      ti      as”:

Blog.Royal8ciii - justifycation with spaces gra ti as

© British Library, Royal MS 8 C III, fol. 81r

Elsewhere, our scribe experiments with justification of his lines through the extension of the letters N and V, seen here in the words “domino” and “unitur” (last line) and “invisibli” (penultimate line).

Blog.Royal8ciii - justification

© British Library, Royal MS 8 C III, fol. 41v

Dangling word-ends

What if you can’t fit the end of the last word on the last line of the page? Do you hyphenate and force your reader to turn the page in order to finish the word, or do you add a lovely flourish and add your word’s end in the bottom margin? The Anglo-Saxon scribe of Royal MS 8 C III opted for the latter:

Blog.Royal8ciii - Endwords small

Consummatis, habitation, formidine with end words hanging in the lower margins. @ British Library, Royal MS 8 C III, fols. 45r, 46r, 48r

Avoid the hole!

Parchment (made of animal skin) was expensive and, so, it would generally be used, even if the parchment was slightly damaged. Upon finding a little hole in one of his pages, the scribe of Royal MS 8 CIII decided not to rip out the page (and risk jeopardizing the construction of the book!), but simply wrote around it:

Blog.Royal8ciii - Avoid the hole

© British Library, Royal MS 8 C III, fol. 53r

Here we can clearly see the scribe increased the space between “in” and “baptismo” so as to avoid the hole.

Triangular text!

In addition to experimenting with justification, juggling the ends of his words, and writing around holes in the parchment, the scribe of Royal MS 8 C III has one more spectacular word processing trick up his sleeve. Halfway through a rather standard theological text about the Mass, and for no apparent reason, he starts laying out the text of six consecutive folio sides (fols. 70v-72v) in a triangular form:


Given the value of parchment, why waste so much of it to form textual triangles? It is rather a mystery. Triangular-shaped texts are extremely rare in medieval manuscripts and I may devote a separate blog to their appearance in the future.

For now, I hope to have shown you that British Royal MS 8 C III is worth our attention. If you’re convinced, why not browse the manuscript yourself? It has been digitized and is available here.

If you liked this post, you may also appreciate the following blog posts about manuscripts:




Anglo-Saxon Cryptography: Secret Writing in Early Medieval England

In this day and age of cyber espionage, encryption of information is becoming increasingly more important. But even in the early Middle Ages, scribes developed techniques to encode their messages, as this blog post reveals.

Codified colophons

SecretWriting.TrinityCollegeB325 fol 99v

© Cambridge, Trinity College, B.3.25, fol. 99v (source)

At the very end of an eleventh-century manuscript copy of St Augustine’s Confessions, an Anglo-Saxon scribe wrote “Fknktp Lkbrp χρp prfcpnkB rfddp”. Rather than garbled gobbledegook, these words were written in a simple but popular code: the vowels have been replaced by their neighbouring consonants in the alphabet: a=b; e=f; i=k; o=p; x=u. The scribe’s words actually read: “Finito libero Christo [the Greek letters χρ is a well-known abbreviation for Christ] preconio reddo”, which is Latin for something along the lines of: “The book is finished, I give a laudation to Christ in return”. Apparently, this scribe was happy that his job was done and rendered thanks to Christ in an encoded message.

The same motivation seems to underlie another encrypted colophon at the end of an eleventh-century Gospel-book made in England: “DFPGRBTKBS AMΗN”:

SecretWriting.ReimsBibliothequeMunicipale9 154r

© Reims, Bibliotheque Municipale 9, fol. 154r (source)

The first two words of this colophon read “DEO GRATIAS” [thanks be to God]; the last word is “AMEN”, with a Greek capital Eta instead of the E (and a weird M and N, which I haven’t been able to identify).

One of the most ambitious encoded messages of this kind is found in Ælfwine’s Prayerbook, made in the 1020s in Winchester. The encoded message reads as follows:


© British Library, Cotton MS Titus D XXVII, f. 13v

Frbtfr hxmkllimus ft mpnbchxs afslknxs mf sckpskt skt kllk lpngb sblxs. B m .. n ;

[ve]l us     [ve]l us                 [ve]l us

AFlfwknp mpnbchp aeqxf dfcbnp cpmpptxm kstxm ppsskdfp [ve]l mf ppsskdft. Bmfn.

The first line is easy to decipher: “Frater humillimus et monachus Ælsinus me scripsit, sit illi longa salus” [Ælsige, the most humble brother and monk, wrote me, may a long health be to him]. The code “B m .. n” means “Amen”, the “e” is replaced by two dots (for which, see below).

The next two lines take some more effort. The first part of the third lines reads: “Ælfwino monacho aeque decano compotum istum possideo” [I posses the computus for Ælfwine, the monk and dean]. The second line (vel us, vel us, vel us), makes clear that the words “Ælfwino monacho aeque decano” can also be read as “Ælfwinus monachus aeque decanus”, thus changing the dative forms into the nominative forms. Combined with the last part of the third line which starts with “vel”, this reads: “vel Ælfwinus monachus aeque decanus me possidet. Bfmn.” [or Ælfwine, the monk and dean, possesses me. Amen].

The rather intricate code is simply an inscription to indicate the maker of the manuscript (Ælfsige) and its owner (Ælfwine). Given the rather complicated encoding, one might wonder whether Ælfsige’s modesty (he calls himself humillimus ‘most humble’) is feigned modesty.

Hygeburg, a cryptographic Anglo-Saxon nun

Another case of feigned modesty is found in the prologue of the Anglo-Saxon nun Hygeburg (fl. 780). As part of the Anglo-Saxon mission, she ended up in Heidenheim, Germany. She was an abbess and wrote a work called the Hodoeporicon, a saint’s life of the Anglo-Saxon missionary saint Willebald. In her introudction, Hygeburg confesses that she considered her womanhood a hindrance for writing hagiography, noting in her preface:

And yet I especially, corruptible through the womanly frail foolishness of my sex, not supported by any prerogative of wisdom or exalted by the energy of great strength, but impelled spontaneously by the ardour of my will, as a little ignorant creature culling a few thoughts from the sagacity of the heart, from the many leafy, fruit-bearing trees laden with a variety of flowers, it pleases me to pluck, assemble and display some few, gathered – with whatever feeble art, at least from the lowest branches-for you to hold in memory. (trans. Dronke 1984)

Hygeburg’s declaration of ignorance is undermined by the flowery rhetoric of her Latin prose, which suggests a high level of education. The encoded message with which she closes her message has a similar, subversive effect:


© München. Bayerische Staatsbibliothek, Clm 1086, fol. 71v (source)

In this message, Hygeburg has replaced all the vowels with abbreviations for ordinal numbers, e.g., “Secd” for “secundum” [second] meaning the vowel e. The code can cracked as follows:


© München. Bayerische Staatsbibliothek, Clm 1086, fol. 71v

With her encoded message, Hygeburg not only shows off her encryption skills, she also claims the text (and possibly the manuscript?) to be her own: “Ego una Saxonica nomine Hugeburc ordinando hec scribebam” [I, a Saxon nun named Hygeburg, have written this].

Dot codes in Anglo-Saxon manuscripts


My reproduction of dot-coded writing in Cambridge, Corpus Christi College, MS 326, p. 105 (Corpus Christi College does not allow the use of images of their manuscript, you can see a low-res image of this page here)

Another encryption method, similar to Hygeburg’s, is the replacement of vowels by dots. One dot equals the first vowel (a), two dots equal the second vowel (e), three dots mean the third vowel (i), and so on. A line in a tenth-century manuscript of Aldhelm’s De Virginitate, probably made in Canterbury, is reproduced above and reads: “V⋮V:V·L:F:L⋮C⁞MCR⋮ST:: ·M:N” (four dots in a line representing U; four dots in a square representing O). In other words: “Vive vale feli cum Cristo. Amen” – here, the word “feli” [with/for the cat]  is usually emended to “felix” [happy], so that it translates to “Live, be well, happy with Christ. Amen.” (Live, be well, for the cat, with Christ would make little sense, especially given the rather haphazard relationships between cats and medieval manuscripts, for which see: Paws, Pee and Pests: Cats among Medieval Manuscripts).

My last example is found in a tenth- or eleventh-century manuscript of Bede’s Vita Sancti Cuthberti, made in southern England. Here, the scribe has once more replaced the vowels with dots: ·=a – :=e – ⋮=i – ::=o – :·:=u.


© Copenhagen, Royal Library, G.K.S. 2034, fol. 13v (source)

Q:·:|⋮ SCR⋮PS⋮T :·:|⋮|:·:|·T :T Q:·: L:G·T L:T:T:·:R


Which, rather charmingly, translates to “May he who wrote live and may he who may read be happy”. This encrypted message suggests that encoding messages was an enjoyable pastime for scribes and that decoding these messages was considered a fun mental exercise for readers.


If you have enjoyed this blog post, why not follow this blog (see button on the right-hand side) and/or read the following posts:

Works referred to:

  • Peter Dronke, Women Writers of the Middle Ages (Cambridge, 1984)