The Lindisfarne and Rushworth Glosses
During the tenth century, complete interlinear Old English glosses were added to two of the most prestigious Latin gospel books produced in early medieval England. The Lindisfarne Gospels, penned by Eadfrith around 687–698, are among the most magnificent of all the illuminated codices that survive from early medieval Europe and one of the crowning achievements of the so-called ‘Golden Age’ of Northumbrian art. An Old English gloss was inserted above Eadfrith’s Latin text sometime between 950 and 970 by a scribe who identifies himself in a colophon as the priest Aldred. This colophon reveals that Aldred provided the gloss at least partly to establish his credentials on joining the community of St Cuthbert, which by the tenth century had relocated to Chester-le-Street in the wake of Viking raids.
1 The Lindisfarne Gospels are preserved in London, British Library, Cotton MS Nero D. IV; Ker §165. Images of this manuscript are available online at: https://iiif.bl.uk/uv/#?manifest=https://bl.digirati.io/manifests/ark:/81055/man_10000006.0x000001. On the production and reception of the codex, see Janet Backhouse, The Lindisfarne Gospels (Ithaca, NY: Cornell University Press, 1981). For discussion of Aldred’s gloss, see Stanton, Culture of Translation, pp. 49–53; Michelle P. Brown, The Lindisfarne Gospels: Society, Spirituality and the Scribe (Toronto: University of Toronto Press, 2003), pp. 90–102; Karen Louise Jolly, The Community of St. Cuthbert in the Late Tenth Century: The Chester-le-Street Additions to Durham Cathedral Library A.IV.19 (Columbus: Ohio State University Press, 2012). Another Old English gloss was added in the late tenth century to the Rushworth (or MacRegol) Gospels, an Irish codex originally made
c. 800. This gloss was written out by two scribes who made use of Aldred’s gloss on the Lindisfarne Gospels: Farman provided the gloss for all of Matthew, Mark 1–2.15 and John 18.1–3, while Owun was responsible for all the other sections.
2 The Rushworth Gospels are preserved in Oxford, Bodleian Library, MS Auct. D. 2. 19 (SC 3946); Ker §292. Manuscript images are available online at: https://digital.bodleian.ox.ac.uk/objects/b708f563-b804-42b5-bd0f-2826dfaeb5cc/surfaces/f0fd9c40-0e3d-45b4-9a1d-19d092474ad4/. For the texts of Lindisfarne and Rushworth glosses, in parallel with two MSS of the Wessex Gospels, see Walter W. Skeat, ed., The Holy Gospels in Anglo-Saxon, Northumbrian, and Old Mercian Versions (Cambridge: Cambridge University Press, 1871–87). See also Kenichi Tamoto, ed., The MacRegol Gospels or The Rushworth Gospels: Edition of the Latin Text with the Old English Interlinear Gloss Transcribed from Oxford Bodleian Library, MS Auctarium D. 2. 19 (Amsterdam: John Benjamins, 2013); Tadashi Kotake, ed., Rushworth One: An Edition of Farman’s Old English Interlinear Gloss to the Rushworth Gospels, Medium Ævum Monographs 44, new series (Oxford: The Society for the Study of Medieval Languages and Literature, 2023). The combined work of these three tenth-century glossators provides important evidence for the ongoing use of Anglian dialects at a time when late West Saxon had come to dominate written documents produced in the south of England:
3 On the development of late West Saxon as the standard form of written English, see below, pp. 92–3. both Aldred and Owun wrote in the Northumbrian dialect, whereas Farman uses Mercian spellings. The addition of Old English glosses to these highly prized and already ancient Latin Gospel books bears witness to the prestige of the vernacular in tenth-century England.
As with the psalter glosses discussed in Chapter One, both the Lindisfarne and Rushworth glosses were designed to serve as guides to the Latin text which they accompany rather than as a substitute. For the most part, the glosses therefore closely follow the syntax of the Latin source-text above which they sit. While interlinear glosses of this nature were often used as teaching aids, it seems highly unlikely that such valuable codices would be used in the schoolroom to teach novices Latin. Nevertheless, the importance of making the meaning of the Gospels as clear as possible evidently exceeded any considerations about the integrity of the manuscript page for these tenth-century glossators and their respective monastic communities. The function of these gospel glosses as linguistic cribs is made clear by the frequent provision of alternative translations indicated by the abbreviation ꝉ for Old English oððe/Latin vel (‘or’), which appears whenever the scribe was either unsure of the meaning of a Latin term or wished to provide a range of vernacular options.
To illustrate the painstaking, scholarly approach of these glossators, I present below the Vulgate text of the opening of the Gospel of Matthew with the Douay-Rheims translation underneath, followed by a semi-diplomatic transcription of the Latin and Old English text as it appears in the Lindisfarne Gospels (Mt. 1.18–19), Cotton MS Nero D. IV, fols 29
r–29
v. Fig. 5 shows fol. 29
r which contains the opening of Mat. 1.18 up to the word
Iospeh:
4 Text cited from Skeat, ed., Holy Gospels, pp. 27–9.18 Christi autem generatio sic erat. Cum esset desponsata mater eius Maria Ioseph, antequam convenirent inventa est in utero habens de Spiritu Sancto. 19 Ioseph autem vir eius cum esset iustus, et nollet eam traducere, voluit occulte dimittere eam.
[18 Now the generation of Christ was in this manner. When as his mother Mary was espoused to Joseph, before they came together, she was found with child, of the Holy Ghost. 19 Whereupon Joseph her husband, being a just man, and not willing publicly to expose her, was minded to put her away privately.]
~
Figure 5. London, British Library, Cotton MS Nero D. IV, fol. 29r, ‘Lindisfarne Gospels: Mt. 1.18, with Old English gloss’.
Fol. 29r:
onginneð | godspell | æfter | matheus |
Incipit | euangelium | secundum | mattheu |
cristes |
18CHRISTI5 This word is written in the manuscript in the form of a magnificent, illuminated Chi-Rho-Iota, spelling the first three letters of the name Christi (fig. 5). |
cynnreccenisse ꝉ cneuresuu | suæ ꝉ ðus | wæs | mið ðy |
-RATIO | SIC | ERAT | CUM |
wæs | biwoedded ꝉ beboden ꝉ befæstnad ꝉ betaht |
ESSET | DESPONSATA | |
moder | his | | |
MATER | EIUS | MARIA | IOSEPH |
Fol. 29v:
ær ðon | hia gegeadradon ꝉ gecuomun |
ante quam | conuenirent |
bigetten ꝉ infunden | wæs ꝉ is | in | hrif |
inuenta | est | in | utero |
hæfde | of | halig | gaast |
habens | de | spiritu | sancto |
ioseph | cuðlice | uer | hire |
19IOseph | autem | uir | eius |
mið ðy | wæss | soðfæst |
cum | esset | iustus |
gebrenge ꝉ gelæda |
traducere |
ah he walde | deiglice |
uoluit | occulte |
forleitta | hea ꝉ ða ilca |
dimittere | eam |
The majority of Aldred’s gospel glosses are substitution glosses, with one or more Old English terms supplied above each equivalent Latin word. In verse 18 alone, for example, Aldred supplies two alternative readings for five individual Latin terms:
generatio >
cynnreccenisse ꝉ cneuresuu,
sic >
suæ ꝉ ðus,
conuenirent >
hia gegeadradon ꝉ gecuomun,
inuenta >
bigetten ꝉ infunden, and
est >
wæs ꝉ is, while he gives no less than four vernacular options for the Latin verb
desponsata (‘betrothed, engaged’) >
biwoedded ꝉ beboden ꝉ befæstnad ꝉ betaht. The provision of such a multiplicity of Old English equivalents may have arisen from Aldred’s consultation of a variety of sources. Alternatively, as Stanton suggests, this range of options might reflect a contemporary drive to expand Old English vocabulary and to extend the semantic range of existing terms.
6 Stanton, Culture of Translation, pp. 51–2. While the second of the two glosses given for Latin
generatio, Old English
cneuresuu, appears frequently in prose and verse as well as psalter glosses,
7 DOE s.v. cyn-recenes: ‘account of (someone’s) lineage, genealogy’. the first term,
cynnreccenisse, is confined to the Lindisfarne gloss to Matthew, where it also appears above the words
generationis (in Jerome’s Prologue to Matthew) and
genealogia (Mt. 1.0).
8 DOE s.v. cnēoris: ‘1.a. progeny, line of descendants, tribe; […] 2.a. group of individuals born about the same period, generation’. As the editors of the
DOE note, the range of terms supplied for
desponsata may reflect Aldred’s hesitancy over the carnality of Joseph and Mary’s relationship: the first,
biwoedded, is a common word, while the second,
beboden, is rare in the sense used here;
9 DOE s.v. be-bēodan: ‘D.3.a.i. to commit, entrust (someone acc. to the care or protection of someone, or to the keeping of God dat.)’ the third,
befæstnad, is unique in this sense,
10 DOE s.v. be-fæstnian: ‘2. to pledge (someone) in marriage, betroth’. and the fourth,
betaht, is only recorded in this sense in one other instance.
11 DOE s.v. be-tǣcan: ‘2.d. to commit in marriage, betroth’, appears in the eleventh-century Old English translation of the Rule of Chrodegang. Eric G. Stanley notes that Aldred supplies a marginal gloss to this verse, further clarifying its meaning: to gemanne nalles to habban for wif (‘to take care of, not at all to possess as a woman’): ‘The Lindisfarne Gospels: Aldred’s Gloss For God and St Cuthbert and All the Saints Together Who are in the Island’, in The Lindisfarne Gospels: New Perspectives, ed. Richard Gameson, Library of the Written Word 57 (Leiden: Brill, 2017), pp. 206–17, at 210. In the same article, Stanley also discusses Aldred’s ‘encyclopaedic’ marginal gloss on Mt. 1.18, which confusingly states that the Old Testament priest Abiathar was forebiscop (‘high bishop’) in Jerusalem and bebeod (‘committed’) Mary into Joseph’s care (210). As we shall see below, Ælfric went to considerable lengths in his
Catholic Homilies to prevent any misunderstanding of the Virgin Birth and to promote its orthodox interpretation throughout the English Church.
12 See pp. 125–30.In addition to weighing up the various possible translations of each Latin word, Aldred pays careful attention to the grammar of his source-text: for example,
the genitive inflection of the opening Latin word
Christ-i is mirrored in the Old English gloss
christ-es.
Personal names are often left untranslated (e.g.
Maria,
Ioseph) when the meaning is obvious to Aldred’s tenth-century readers, though
Ioseph is supplied once in the gloss at the beginning of verse 19 to clarify the referent of the sentence. Inflected Latin verbs are sometimes expanded to form an Old English verb phrase, the process known in linguistics as ‘periphrasis’, as in the supplement gloss of
uoluit as
ah he walde. Despite these small concessions to vernacular idiom, for the most part the Lindisfarne gloss closely adheres to the syntax of the Latin source text that it sits above: for instance, Latin NOUN + GENITIVE DETERMINER constructions are replicated in the vernacular (e.g.
mater eius >
moder his), a feature which we also observed in the glossed psalters.
13 See p. 46. There is only one exception in this extract to this general practice of syntactical mirroring, where Latin
de spiritu sancto is rendered as
on halig gaast, with the placement of the adjective
halig before the noun reflecting standard Old English word order.
14 This reversal of word-order is noted by Ruta Nagucka, ‘Glossal Translation in the Lindisfarne Gospel according to Saint Matthew’, Studia Anglica Posnaniensia 31 (1997), 179–201, at 181. See further Giuseppe Pagliarulo, ‘Word Order in the Lindisfarne Glosses?’, Neophilologus 94 (2010), 625–35. These features of the Lindisfarne gloss point to its primary purpose as a guide to the Latin, facilitating close study of the meaning of each individual Latin word at the level of grammar and sense. Whenever more information is required to aid understanding, the reader’s eye moves upward from the Latin majuscule of the main text to the Old English minuscule of the gloss. The production of the gloss itself might be viewed as an act of prayerful study in which we witness Aldred himself meditating on the vocabulary, grammar and syntax of his scriptural source before weighing up its various possible meanings.
15 On monastic reading in this period, see Malcolm Parkes, ‘Rædan, areccan, smeagan: How the Anglo-Saxons read’, ASE 26 (1997), 1–22. Aldred’s provision of multiple glosses for certain Latin words in turn invites his fellow monastics to ruminate on the full meaning of the scriptural text. As well as assisting in the acquisition of Latin vocabulary essential for a novice, Aldred’s rich and variegated gloss could thus also prove useful in the practice of
lectio divina, the constant prayerful reading of sacred books that lay at the heart of the monastic
opus dei.
16 On lectio divina, see Jean Leclerq, The Love of Learning and the Desire for God, trans. Catherine Misrahi (New York: Fordham University Press, 1961).The subsidiary nature of Farman’s gloss on the opening section of Matthew in the Rushworth Gospels is evident in his placement of the first word,
kristes, to the right of the large, illuminated capitals CRI (CHI-RO-IOTA, an abbreviation for
CRISTI) and partially above the next word
autem (fig. 6). Again, the Old English gloss is written in Anglo-Saxon square minuscule above the insular majuscule of the Latin:
17 Text cited from Kotake, ed., Rushworth One, p. 90. The Latin text of the Rushworth Gospels is broadly similar to the Vulgate but contains some readings which are closer to Old Latin, and has therefore been classed as belonging to an ‘Insular Vulgate’ group: see Tamoto, The MacRegol Gospels or The Rushworth Gospels, pp. xlii–xlvii.~
Figure 6. Oxford, Bodleian Library, MS Auct. D. 2. 19, fol. 2v, ‘Farman’s gloss on Rushworth Gospels: Mt. 1.18–19’.
kristes | soþlice | kennisse | þus wæs |
18CR(IST)I | autem | generatio | sic |
| þa þe | hio wæs | bewedded ꝉ be- |
erat | cum | esset | dis- |
-fæst ꝉ in sceat alegd | his moder |
-ponsata | mater eius |
maria | iosefae | ærþon |
maria | ioseph | ante- |
hiae tosomne cwoman | hio wæs gemoeted | in hire innoþe |
-quam conuenirent | inuenta est | in utero |
hæbbende | of þæ(m) | halgan | gaste | ioseph | soþlice | hire wer |
habens | de | sp(irit)u | s(an)c(t)o | 19Ioseph | autem | uir ei(u)s |
swa | he wæs | monn | soþfæst | ⁊ | ne walde hie | |
cum | esset | homo | iustus | et | nolet eam | tradu- |
| wolde | degullice | forletten | hio |
-cere | uoluit | occulte | demittere | eam |
As Kenichi Kotake notes, the fact that the Old English words
þus wæs have been erased above the last part of
generatio at the end of the first line of folio 2
v yet placed correctly above the following word
sic suggests that Farman was copying from an existing gloss rather than translating word by word.
18 Kotake, ed., Rushworth One, p. 199. While the scribe responsible for the Latin text has indicated the beginning of each biblical verse by encircling the initial letter with a series of red dots, Farman inserts an asterisk (*) before each verse in his gloss to help readers match the Old English text to the Latin.
However, where Farman differs from Aldred is in his provision of far fewer alternative translations, with the result that his gloss comprises a steady flow of vernacular phrases which at times begins to approach the rhythm of Old English prose. Hence, for example, where Aldred provides two translations of
generatio (
cynnreccenisse ꝉ cneuresuu),
Farman offers only a single Old English term (
kennisse).
19 DOE s.v. cennes (1. birth […], 2. childbirth’) records 10 occurrences, mostly in prose. Moreover, while Aldred follows the syntax of Latin genitive phrases
mater eius (>
moder his) and
uir eius (>
uer hire), Farman opts for more idiomatic English word order (
his moder;
hire wer).
20 On the relative independence of Farman’s syntax compared to other interlinear glosses of the period, see Kotake, ed. Rushworth One, pp. 69–75. Farman’s concerted efforts to produce a gloss that is readable on its own can also be seen in his decision to translate even those personal names which Aldred omits (verse 18:
maria,
iosefae). This move towards a more natural prose style is also evident in Farman’s expansion of the Latin verb
conuenirent (translated by Aldred
as
hia gegeadradon ꝉ gecuomun, ‘they gathered
or came’) to the phrase
hiae tosomne cwoman (‘they came together’); the insertion of the adverb
tosomne helps clarify the two senses implied by Aldred’s gloss into a single, fluent clause. Similarly, where Aldred follows the Latin closely in translating
in utero as
in hrif (‘in womb’) and
esset as simply
wæss, Farman inserts personal pronouns to aid his readers:
in hire innoþe (‘in her womb’),
he wæs (‘he was’). Farman also adds the dative pronoun
þæ(m) in rendering the Latin
de, to further clarify the grammar. Indeed, only one Latin word, the infinitive verb
traducere (‘to expose’), is not glossed by Farman. Kotake suggests that Farman may have skipped this word because he took the personal pronoun
hie, which appears as part of the phrase written above the preceding words,
nolem eam (>
ne walde hie), as the accusative object of
ne walde (‘and he did not desire her’).
21 Kotake, ed., Rushworth One, p. 200. Overall, Farman supplies supplement glosses more frequently than his predecessor Aldred had done, expanding on the source text whenever it was deemed necessary for the sake of clarity.
Printed in isolation from the Latin source text, Farman’s gloss certainly appears a step closer to the smooth, idiomatic vernacular prose style of the Wessex Gospels discussed below:
18 kristes soþlice kennisse þus wæs þa þe hio wæs bewedded (oððe befæst oððe in sceat alegd) his moder Maria Iosefae ær þon hiae tosomne cwoman hio wæs gemoeted in hire innoþe hæbbende of þæ(m) halgan gaste.
19 Ioseph soþlice hire wer swa he wæs monn soþfæst and ne walde hie […] wolde degullice forletten hio.
Indeed, Farman might have repurposed an earlier, free-standing translation of the Gospels, adapting it into his gloss.
22 Kotake suggests that Farman’s exemplar ‘could have contained a freer translation than a word-for-word interlinear version and accordingly that his faithful copying from such an exemplar made his gloss more idiomatic than one would expect to find in an interlinear gloss text’ (Rushworth One, p. 70). Alternatively, he may have intentionally crafted these more idiomatic translations, anticipating that his readers – perhaps still developing their Latin proficiency – would benefit from glosses that aligned more closely with their native language than the Latin source. Farman’s gloss thus transforms the Rushworth Gospels into a bilingual gospel book, which like the Paris Psalter could be read either in Latin or English.
23 For Toswell’s argument that some of the glossed psalters are similarly bilingual, see above, pp. 35–6.The Lindisfarne and Rushworth (MacRegol) glosses provide a window onto the painstaking and intensive study of the Gospels undertaken by tenth-century English monks. In Steiner’s terminology, Aldred and Farman (and his co-glossator Owun) affirmed the value of the Gospels themselves by making the initial decision to translate them from Latin into English.
24 On Steiner’s translation theory, see above, p. xiii. As we have observed, in both cases, the priority of understanding the Latin source outweighed concerns about preserving the integrity of the magnificent codices that contained the Vulgate text. In the case of the Lindisfarne Gospels in particular, practical considerations may have also been a factor in the production of the gloss, notably Aldred’s desire to establish himself in the monastic community at Chester-le-Street. The addition of these glosses might even be said to enhance the pages of these Latin gospel books with words from the by-now prestigious written Old English of the Kingdom of England. Once the decision had been made to provide an interlinear translation to these Latin gospel books, the glossators ‘plundered’ each individual lexical item of the Vulgate text, reflecting on various possible meanings before supplying readers with either single or multiple equivalents in the target language. With a few small exceptions, such as personal names where the meaning is obvious in the Latin, the entire text of the Gospels was incorporated into the target language, allowing monastic readers to match up the Latin words with their vernacular equivalents when studying the scriptural text. Finally, while Aldred made little effort to provide compensatory measures, cleaving closely to the word order of the Latin source and supplying mostly substitution glosses, Farman often provided supplement glosses which when read together begin to approximate Old English prose. Both these splendid Latin gospel books were furnished with Old English glosses for the same principal purpose: to facilitate the constant meditation on the words of Scripture (
lectio divina) that lay at the heart of monastic life. Although it seems highly unlikely that such prestigious manuscripts were ever used in the classroom, these glosses would nevertheless provide any monk or nun looking at these beautiful codices with the opportunity to gain a deeper understanding of the Latin gospel text’s literal and spiritual meaning. The Old English gospel glosses were thus produced within an enclosed space for highly specialised use by members of monastic communities. For a tenth-century translation of the Gospels into Old English prose intended for readers outside the cloister, we must turn now to the
Wessex Gospels.