SOURCE: The "Book of Kells": An Illustrated Introduction to the Manuscript in Trinity College Dublin, New York: Thames and Hudson, 1994, pp. 9-43, 89-92.

[In the following excerpt, written in 1993, Meehan discusses the Book of Kells's historical background, decorative plan, influences, symbolism, themes, and purpose.]

The Book and its background

It has always been difficult to write about the Book of Kells without resorting to hyperbole. Those who have tried to describe it betray almost a sense of disbelief, as though it had emerged from another world: 'the work, not of men, but of angels', as the thirteenth-century historian Giraldus Cambrensis put it; 'the product of a cold-blooded hallucination', in the words of Umberto Eco.¹

The Book of Kells, to express it more prosaically, is a large-format manuscript codex of the Latin text of the gospels. Preceding the gospels themselves are 'etymologies', mainly of Hebrew names (only one page survives); canon tables, or concordances of gospel passages common to two or more of the evangelists, compiled in the fourth century by Eusebius of Caeserea (so that a passage in any one gospel can, in theory, quickly be found in the others); summaries of each of the gospel narratives, known as Breves causae; and prefaces characterizing the evangelists called Argumenta. The first gospel text (Matthew) begins on folio 29r.

It is the most lavishly decorated of a series of gospel manuscripts produced between the seventh and ninth centuries, when Irish art and culture flourished at home and in centres of Irish missionary activity overseas. The artistic style of the period is commonly known as 'insular'. The Book is written in a bold and expert script of a type best described as 'insular majuscule'.² The text is based on the Vulgate (the version of the Bible completed by St Jerome in 384 AD) intermixed with strong elements of the version that preceded it, known as the Old Latin translation. This text is decorated and at the same time elucidated with images of great iconographic subtlety. Important words and phrases are emphasized and the text is enlivened by an endlessly inventive range of decorated initials and interlinear drawings. The great decorated pages, upon which the book's celebrity mainly rests, comprise symbols and portraits of the evangelists, introducing the gospels; portraits of Christ and of the Virgin and Child; and illustrations of the temptation and the arrest of Christ.

Where, when and why was the Book of Kells written? To answer these questions we must go back to the arrival of Christianity in Ireland probably in the fourth century, and its consolidation through the work of St Patrick and others in the fifth. The new religion was to exercise enormous patronage in the commissioning of a wide range of articles. It needed altar vessels, reliquaries, vestments, bells, staffs, flabella (liturgical fans) and above all books. 'Pocket' gospels like the Book of Dimma and Book of Mulling (Trinity College Dublin MSS 59, 60) were used for missionary work or private devotion, while larger volumes like the Lindisfarne Gospels (Cotton Nero D. IV), the Echternach Gospels (Paris, Bibliotheque Nationale lat. 9389) and the Book of Kells itself were designed for use on the altar. The Irish church developed particular characteristics, its orientation and control being primarily monastic rather than episcopal. As Roman practice changed, it came to differ from Rome in the date on which it observed Easter, and its clerical tonsure was distinctive. When the Irish entered the missionary field, problems arose from these differences. At the Synod of Whitby in 664, these were resolved, for the north of England, in favour of the Roman side. Irish missionary work in England had begun with the foundation of an abbey at Lindisfarne by St Aidan in 635.

Other Irish missionaries were prominent in continental Europe. Around 590, St Columbanus journeyed with eleven companions from Bangor, county Down, to Burgundy, founding monasteries at Fontaine, Annagray and Luxeuil before moving to Switzerland and from there to northern Italy, where he founded an influential house at Bobbio. One of his disciples, Gallus, stayed to live as a hermit in Switzerland at a place which came to be named after him. Another Irishman, St Kilian, was a missionary at Wurzburg in Germany, where he was martyred in 689. The Irish missions exercised a widespread influence through their foundations, not least on calligraphy and decorative techniques. This influence can still be observed in the many insular manuscripts and artefacts which remain today in European centres.³

The date and place of origin of the Book of Kells have long been subjects of controversy. Investigations have ranged far and wide, but have focused particularly on two locations: Iona, an island off Mull in western Scotland; and Kells, county Meath, in Ireland. The monastery of Iona had been founded in about 561 by St Colum Cille (to whom the Book of Kells was traditionally attributed) but in 807, after a series of devastating Viking raids, it established a house of refuge at Kells and for many years the two monasteries were governed as a single community. It was close to the time of this migration that the Book of Kells must have been written. The arguments concerning the origin and date of the manuscript are set out in Appendix I.

In the middle ages the manuscript was revered at Kells as the great gospel book of Colum Cille. This was the term used when the Annals of Ulster record the theft of such a book in 1007 from the western sacristy of the great stone church of Kells and its recovery 'after two months and twenty nights' covered by a sod. It is safe to assume—though it is of course an assumption rather than a fact—that the book in question is the Book of Kells. The phrase 'the most precious object in the western world' (primh-mind iarthair domain) occurs in the annalistic entry, but it is not clear whether this refers to the book itself or to the ornamental shrine, stolen in the episode, which protected and enriched it. The manuscript was inarguably at Kells from late in the eleventh century, when it was used to record property transactions.⁴ In 1090, it was reported by the Annals of Tigernach that relics of Colum Cille were 'brought' (by which we should probably understand 'returned') to Kells from Donegal. These relics included 'the two gospels', one of them surely the Book of Kells, the other perhaps the Book of Durrow, with which it has long been associated.⁵

In 1655, Samuel 0 Neale, reporting for the Down Survey, commented on the belief of the townspeople of Kells that the manuscript was 'written as they say by Columbkill's own hand, but [is] of such a character that none of this age can read it'. The book was lost to the town in the seventeenth century. The 'great stone church' of the annals had survived a transition from monastic to episcopal control in the twelfth-century reform of the Irish church, but it did not survive the disruption of the Cromwellian period, when it was in such a ruinous state that it was fit only for the stabling of horses. The governor of Kells, Charles Lambart, earl of Cavan, sent it to Dublin for reasons of safety around 1653. A few years later, it reached Trinity College, the single constituent college of the University of Dublin, through the agency of Henry Jones, a former scout-master general to Cromwell's army in Ireland and vice chancellor of the university, when he became bishop of Meath in 1661. The association with St Colum Cille persisted, at least in the popular mind, until as late as the nineteenth century, and it was as 'St Columba's book' that it was introduced to Queen Victoria in 1849. Among scholars, it was regarded in more secular terms. James Ussher, who collated its text and the text of the Book of Durrow with the Vulgate, seems to have been the first to use the term 'the book of Kelles', some time after 1621.

By the mid nineteenth century, the manuscript was on display in the Long Room of the College library, and gradually assumed a new role as an art object. As such, it has attracted an enormous degree of scholarly examination.⁶ Much recent research has tilled the fertile ground of biblical exegesis, concentrating on the spirituality of the book and on its relevance to liturgical practices. Attention has been drawn to the role of eucharistic ceremonies in the life of the Irish church,⁷ and to the possible significance of Easter rites and festivals in the specific decoration of the Book of Kells. Increased emphasis has been placed on the importance of pericopes (short passages selected for public reading) such as 1.18, the first pericope of Christmas eve, the opening word of which is the Chi Rho monogram (Chi and Rho forming the abbreviated Greek form of the name of Christ).⁸ This was elaborated to an increasing size in insular manuscripts until in Kells it took up almost the whole page.⁹

It is emphasised at present that certain pages and motifs are capable of carrying different layers of meaning, or at least of interpretation; that the images on the page should be read in a number of different ways simultaneously; that the complexity of the decoration mirrors the 'verbal virtuosity' of the Irish exegetes of the period, as Benedict of Aniane put it,¹⁰ and is meditative and allusive in intent. Free association of ideas is taken to be part of the intellectual baggage of the monks. Research employing this technique, some of it carrying in itself resonances of the exegetical, has unquestionably advanced our understanding of the manuscript, though any re-creation of the spiritual world of the Columban monk must depend to a certain extent on conjecture.

Particular difficulties lie in the way of a fuller understanding of the circumstances which occasioned the production of the book, and the uses to which it was put when first it was made. Its many textual errors and the virtual illegibility of many of the major text pages and the bands of display lettering,¹¹ added to its sheer sumptuousness, mean that it was unlikely to have been produced for daily readings. Perhaps it was made as a piece of altar furniture intended for use only on special occasions. It may have been made for a particular event. Henry suggested that it might be possible to tie the making of the book to the bicentenary of the death of Colum Cille in 797, but this is not easily ascertained, since it is uncertain if such anniversaries were celebrated at the time. A connection has been postulated between the production of the book and the translation of St Colum Cille's relics to a new shrine in the middle of the eighth century,¹² but this date seems around a generation too early for the Book of Kells. If we are able to accept that manuscripts were produced for events such as this, it might be supposed that a different gospel book, though one now lost, was produced for the occasion. Perhaps we can reach no closer than to conclude that the book came about through the happy conjunction of a perceived need and the opportunity afforded by the presence of several artists of conspicuous talent.

During the last century, the Book of Kells, both in Ireland and abroad, has become an inspiring symbol of Irish nationality and creativity. Remote as the modern mind is from its arcane imagery, there is a lasting recognition of the artists' skills and devotion and an enduring fascination with the enigma surrounding its production.

Decorative influences and parallels

The repertoire of ornament employed in the Book of Kells had long been in development in manuscript art and applied crafts. Red dots used around the shape of a letter for the purposes of highlighting it appear first in the earliest surviving Irish book, the gospel manuscript known as 'Usserianus primus' (Trinity College Dubin MS 55), which dates from the late sixth or early seventh century. In the device known as 'diminuendo', the letters of words introducing a new section are formed in decreasing sizes. The earliest appearance of this technique in an Irish context is in the Cathach (Dublin, Royal Irish Academy, 12. R. 33), a psalter traditionally, and perhaps correctly, attributed to the pen of St Colum Cille. The two devices are commonly used in tandem in the Book of Kells to pick out the opening letters of verses. Interlace, which in the Book of Kells reached a high order of ingenuity, developed from the art of antiquity. The first surviving insular manuscript to use interlace—in a static, ribbon pattern—was a fragmentary late seventh-century gospel book now at Durham (Dean and Chapter Library, MS A.1 1.10). In the later seventh-century Book of Durrow (Trinity College Dublin MS 57) interlace took on zoomorphic as well as abstract elements in designs which, combined with the complexity of Celtic spiral and trumpet devices, point the way towards the sophistication of Kells. In Durrow, parallels with the motifs of jewellery and metalwork are particularly striking.

It is not possible within the scope of this study to provide more than a sketch of the varied artistic models and predecessors of the Book of Kells. The extent to which the manuscript was paralleled by the art of the Mediterranean countries, and by Byzantine, Coptic, Assyrian, Armenian and other sources is a matter of debate. Devices like the snake ornament of folio 33r or the joint articulation of animals, seen, for example, in folios 19v, 76v or 302r, are found in the Pictish stone carvings of Scotland. There is no agreement on the direction in which this influence was running; on whether, in other words, the manuscript artists were influenced by sight of the carved stones, or, as is likely, the stone carvings were copied from the more portable medium of books. The style probably originated from the common source of metalwork, perhaps from the technique of cloisonn6 enamel.¹³ What seems indisputable is that once a motif, whether figurative or symbolic or purely ornamental, is employed in any medium, it enters a common currency of decoration, and can be borrowed, re-used and adapted.

Parallels with metalwork as well as stonework are frequently to be observed. On pages like 34r and 124r, the use of tight red dotting on a blank ground is reminiscent of stippling on metalwork, such as on the bowl of the Ardagh Chalice. Many parallels can be observed in subject matter. One example is a small circular mount found at Togherstown, County Westmeath (National Museum of Ireland 1929: 1400) which contains three figures grasping each other's legs. This is similar to three roundels at the top of folio 130r, with the difference that in the manuscript the figures grasp each other's arms. Several parallels together are found in two dismembered finials from what was once an Irish shrine of the highest quality (now Musee des Antiquites Nationales, St Germain-en-Laye, France: 52 748 [a, b]). Its bosses are similar to those on 331; its snake ornament is of a type used extensively in Kells; and the human head, here within a snake's jaws, and long curved neck resemble the initial on folio 58v. Other heads in the Book of Kells, such as the portrait of Christ at his arrest in folio 1141, with its prominent, bulging eyes and precisely formed beard, resemble a small eighth-century Irish head once part of another metalwork shrine.¹⁴ A remarkable late seventh- or early eighth-century hoard discovered in 1985 at Donore, only a few miles from Kells, displays several, unsurprising parallels with the art of the manuscript, not least in the similarity between the profile lion of the cast-bronze door handle and the biting lion of folio 124r. The trumpet and spiral motifs of one of the discs from the hoard is close in style to those of folios 33r and 34r, the designs in both media constructed through the use of compasses. In a different medium again, a small bone pin of a man squatting, his arms wrapped around his knees in a naturalistic pose, was found last century at Newbridge, County Kildare (NMI W 13). This resembles a figure in folio 86r line 2, forming the T of Tunc, and the tiny detail of a man, upside down, clasping his knees at the top of the shaft of the first I of the Initium page (folio 130r).

The scheme of decoration

The Book of Kells integrates and develops its decorative inheritance with breathtaking assurance, embellishing the text on every page. Two pages (folios 29v and 301v) lack colour, but it was not planned that this should be so. The text on folio 29v is framed with an elaborate, unfinished design in ink. In the case of folio 301v, the bowl of the letter a on line 11 was lightly decorated by the scribe, but no pigment was added.

The major decorated pages are complex in composition and iconography. They comprise canon tables (folios lv-6r); pages grouping symbols of the evangelists (folios Ir, 27v, 129v, 187v and 290v); a depiction of the Virgin and Child surrounded by angels (folio 7v); portraits of St Matthew (folio 28v), Christ (folio 32v) and St John (folio 291v); narrative scenes, the earliest to survive in gospel manuscripts, representing the arrest and the temptation of Christ (folios 114r, 202v); a page wholly of decoration depicting a double-armed cross with eight roundels embedded in a frame (folio 33r). The famous Chi Rho page (folio 34r) introduces St Matthew's account of the nativity. The opening words of each of the gospels are elaborated extensively: Liber generationis (Mr: 1.1) on folio 29r; Initium euangelii iesu christi (Mk. 1.1) on 130r; Quoniam (Lk. 1.1) on 188r; and In principio erat uerbum Iet) uerbum (Jn. 1.1) on 292r. Other passages given additional decorative emphasis are on folios 8r (the opening of the Breves causae of Matthew); 13r (the beginning of the Breves causae of Mark); 12r, 15v, 16v, and 18r (the opening words of the Argumenta of the four gospels); 19v (the words ZACHA[riae] at the opening of the Breves causae of Luke; 114v (the opening of Mt. 26.31, Tunc dicit illis ihs omnes uos scan[dalum]; 124r (Mt. 27.38, Tunc crucifixerant xpi cum eo duos latrones); 183r (Mk. 15.25, Erat autem hora tercia); 200r-202r) (Lk 3.22-38); 203r (Lk. 4.1, Ihs autem plenus spiritus sancto); 285r (Lk. 23.56-24.1, … Una autem sabbati …).

The decorative plan appears to have been that each of the gospels should be introduced by a page of evangelist symbols, a portrait page and an elaboration of the opening words or letters. It was common for pages of major decoration to be painted on single leaves, so that the transcription of the rest of the text might continue without interruption on conjoint leaves while the more time-consuming task of decoration was in progress. Single leaves were liable to become detached and lost or misplaced in subsequent bindings. This may explain why no portrait now accompanies St Mark's or St Luke's gospel. Some uncertainty attaches to folio 187v, at the end of St Mark's gospel, which may have served, though incomplete, as a symbols page for the following gospel, that of St Luke. The present placement of several pages is doubtful. It seems likely that the cross-carpet page at folio 33r originally faced folio 34r.¹⁵

It is evident that the Book of Kells was not completed. A crucifixion scene was probably intended for the blank page 123v, where it would have been observed by onlookers gathered in groups on folio 124r. It was normal practice in the manuscript for the reverse of pages bearing elaborate decoration to be pricked and ruled for text with more than normal care. This was the case for folio 123r, though its verso remained blank. The intended image may have been similar to that in the Durham Gospels (Durham, Dean and Chapter Library, A.II.17 folio 38³v). It is also possible, as Françoise Henry suggested, that an image of the Last Judgment was planned for folio 289v. Such an image—or it may be an ascension—occurs in the eighth-century Irish manuscript St Gall 51 p. 267, a manuscript which also has an image of the crucifixion on p. 266. The plan of folio 29r was changed in several places. Compass circles are left unused, and the original scheme for the roundel in which the ER of LIBER was placed underwent a simplification, intricate snake decoration being left less than half completed, and the roundel being filled with a single wash of yellow. The face of the angel at the top left of 29r was given no features. On 95r, the head emerging from a decorated Et was provided with eyes but neither nose nor mouth. The decoration of folios 30r-31v was begun, but came to a halt at an early stage. On other text pages, like 301v-302r, the colouring process was not concluded. On 188r, it was surely intended that a head should be added to the cloak at the top of the page, and placed, like its companions, in the mouth of a lion. On 288r, a lion was left unfinished at the right of the page. There is something strange and uncharacteristic about the decoration of folio 203r in the context of the other decorated pages which may be explained if it is work done later or not completely finished.

The purpose of the decoration

The decoration of the text pages was mostly of the initial letters of verses, forming part of the text itself rather than serving as an adjunct to it. Hence, decoration punctuated the text and aided legibility, at the same time repeating the thematic preoccupations of the fully decorated pages. These functioned in a number of different ways. They continued the text. They illustrated it. They had a liturgical use. They also had an evangelical purpose. The importance of the visual in instructing the uninitiated was recognised from the outset by missionaries. Pope Gregory the Great explained that images provided 'a living reading of the Lord's story for those who cannot read'. This was why images were used in a missionary context. Bede explains in the Historia Ecclesiastica that when St Augustine came to England to preach the gospel in 597, he came bearing 'a silver cross and an image of our Lord and Saviour painted on a panel',¹⁶ which he raised high for king Ethelberht to see. It was important that the icons of Christianity should impress by their splendour. In 735 St Boniface, the English missionary to the Germans, commissioned from the abbess Eadburga a copy of the epistles of St Peter written in gold, so that, as he said, he might impress honour and reverence for holy scripture before the eyes of the carnal in his preaching.¹⁷ In larger manuscripts, the overall sweep of designs was constructed in such a way as to make them visible at a distance to a congregation sitting in the body of the church. The Echternach gospels contains such an image in which St Matthew displays his gospel in an open position (Paris, Bibliothèque Nationale MS lat. 9389 folio 18v). The opening words can be seen (Liber generatione ih u xpi), but no decoration is visible. A manuscript could be observed in this way on the altar, and recent research has made it clear that gospel books were displayed to the faithful in processions.¹⁸ Intricate and detailed decoration, like the roundels of folio 33r, was for contemplation only by those who had access to the manuscript. Not all of the elaborated pages of key phrases are immediately legible, nor, it might be argued, were they necessarily intended to be. They functioned as an aide memoire for those already familiar with the text, and familiar with patristic writings and exegetical texts which alluded to the themes it is possible to discern in the Book of Kells. The artists did not necessarily comprehend the full significance of all that they were copying. Some details may have been half remembered, imperfectly understood, or inaccurately copied.

The sacred text itself was copied in the Book of Kells with a remarkable degree of inaccuracy. For today's observer many of the images in the book can be understood only imperfectly. The current idea that the images could carry composite or 'multivalent' meanings even for their artists is an extraordinarily useful but potentially misleading device. If we bring to texts so rich in interpretive possibilities an original readership which knew scripture well, and artists as skilful and apparently as allusive as those of the Book of Kells, the possibility does occur of several equally plausible interpretations of particular pages or motifs. Since however there is so little chance that the artists' intentions can be proved, we are also in danger of coming eventually to the position that the pages mean whatever we want them to mean. Yet around the time of the writing of the Book of Kells, it was widely acknowledged that scripture presented difficulties of interpretation and needed careful exegesis. Adomnán exemplifies this idea in his account of visions experienced by St Colum Cille, particularly in his comment that 'everything that in the sacred scriptures is dark and most difficult became more plain' as a result of an ecstatic vision.¹⁹ Adomnán reinforces too the individual nature of Colum Cille's enlightenment. His visions are of an intensity which would blind those less worthy. The interpretation of spiritual matters was only for the elite. As Colum Cille announced to his disciple Lugbe, 'There are some, although few indeed, on whom divine favour has bestowed the gift of contemplating, clearly and very distinctly, with scope of mind miraculously enlarged, in one and the same moment, as though under one ray of the sun, even the whole circle of the whole earth, with the ocean and sky about it'.²⁰

Decorative themes: the book and the cross

Giraldus Cambrensis maintained that close study would reveal 'ever fresh wonders' in a manuscript like the Book of Kells. This cannot be disputed, but the wonders of the manuscript are those of technique and variety rather than theme, which in essence is singular. The decoration of the entire manuscript glorifies aspects of Christ's life and message and reflects the principal moments of that life. There are recurring images of Christ's birth, his sacrifice, commemorated in the institution of the eucharist, and his resurrection. His name and face, his properties, attributes and symbols are constantly before the reader, in a form of counterpoint with the text of his life.

The book was the medium which conveyed the fundamental message of Christianity. The opening words of the gospel of St John stressed this by opening with the statement, 'In the beginning was the word'. The book itself is a constant motif in Kells, depicted over thirty times. It is in the hands of Christ on folio 32v, of the evangelists Matthew and John (28v, 291v) and the evangelist symbols (1r/v, 2v, 3r/v, 4r, 5r, 27v, 290v and 292r). It is brandished by angels on 3r, 34r, 183r, 202v and 285r. Books are also seen in the hands of more enigmatic figures on 1v, 4r, 8r, 29r and 187v. Dr Farr has suggested that the figure on the left of 29r, as well as other figures in the manuscript, is the deacon. The deacon was sixth of the seven clerical orders, just below the priest himself. Appropriately, he was responsible for the altar and its furnishings and read the gospel at mass.²¹ Books were regarded as having talismanic properties, and are known to have played a central role in life on Iona. In Adomnán's time, for example, the monks there placed books of St Colum Cille on the altar along with his garments in an attempt to gain favourable winds for the passage by sea of timbers to be used in construction work.

The cross was an explicit reminder of Christ's passion and death. A full page was devoted to the cross in the Book of Durrow, and a similar image is on folio 33r of the Book of Kells. The cross occurred as a prominent feature of manuscripts at least as early as the Coptic Glazier codex (Pierpont Morgan Library, New York, Glazier 67) written close to the year 500, and in Usserianus primus (Trinity College Dublin MS 55) folio 149v around a century later. In this position and in its use on book covers, there is a sense in which the cross served an apotropaic or protective function. For early Christian writers, the cross also had a cosmological value, in signifying the celestial axes from east to west and from north to south.²² Countless crosses, taking every imaginable form, occur throughout the Book of Kells. Display script forms a cross on 124r. On 114r, Christ's limbs are arranged in a cross shape. A saltire cross is on 187v and 290v. Further crosses are inside the Virgin's halo on 7v. On the same folio, the upper left corner extension has ribbon interlace worked in such a way that a cross is formed. Elaborate crosses surround St John in his portrait on 291v, with crosses worked into the interlace designs at the top and foot of the figure. Crosses occur inside the bodies of individual letters, and as interlinear devices, in every form which the ingenuity of the artist could muster. Striking examples are on folio 174r (between lines 15 and 16) and on 314r (last line). In the Passion sequence in St Luke's gospel, most notably on 283r, cross forms dominate the page.

Angels

Taking their name from the Greek word for 'messenger', angels were an unseen but constant spiritual presence, intermediaries between God and man. They had sight of God in heaven and it was known that they would be with him at the Last Judgment. Christ was surrounded by angels at the most important moments of his life on earth. They were present at his birth (Lk. 2.9-15), and on 7v, where he is on the knee of his virgin mother. This page faces the reference to his birth in the Breves causae of Matthew on 8r. Three angels appear on the Chi Rho page (folio 34r) at the moment of the nativity (Mt. 1.20, 24). Angels were with Christ in the desert (Mt. 4.11) and in his agony (Lk. 22.43). They were prepared to defend him when he was arrested (Mt. 26.53) though no angels appear on folio 114r, where the arrest is shown pictorially, or on folio 116v-117r, where it is described in the text. Angels also witness Christ's resurrection (Mt. 28.2-7 and Jn. 20.12), and appear on folio 285r, at the resurrection narrative in Luke. On this page, on 7v, 32v and 202v, angels appear in groups of four. These are probably to be identified as the principal archangels, Michael, Gabriel, Raphael, and Uriel,²³ bearing flabella, flowering boughs or books, where the image becomes a precise one of the messenger bearing the message.

The sight of angels was afforded only to the worthy. St Colum Cille was 'held worthy to receive in shining light the sweet and most pleasant visitations of holy angels'.²⁴ The frequent references to angels in Adomnán's life, occupying the whole of the third book, are paralleled in the decoration of the Book of Kells. Angels, like the sacred book, helped to make clear the unseen mysteries of faith, but only for those who were prepared or were in a position to see. Several of the angels in Kells are obscurely placed, especially so on the last line of folio 48r, where a winged angel in an 'orans' pose, with hands outstretched in prayer, is placed within the initial 0 of Omnia. The drawing of this angel is similar to those in an eighth-century manuscript probably copied in Würzburg from an Irish model (now Wurzburg, University Library M.p.th.f.69 folio 7v).²⁵ Angels assisted Colum Cille, as the gospel book did, in the work of evangelism, by striking the pagan magician Broichan²⁶ and by engaging in fearsome battles with demons. In one case the demon carries a rod, as the Devil does in folio 202v. Colum Cille was often observed by his contemporaries in the company of angels, taking on, for hagiographical purposes, the attributes of Christ himself. One angel whom he met in a vision carried a 'glass book' (vitreum … librum). Colum Cille was accustomed, Adomnán tells us, to see 'the souls of just men borne by angels to the height of heaven'.²⁷ In his turn, the saint's own soul was accompanied to heaven by great hosts of angels, the island lit bright by their presence.

The evangelists and their symbols

Two of the evangelists are portrayed on 28v and 291v, fixing the viewer with a direct stare. St Matthew holds his gospel in his left hand. Although his pose is upright, he is placed against a chair or throne, which features the symbols of his fellow evangelists, and is set in a frame composed of snake ornament. St John, extravagantly haloed and seated on a throne, his scribal accoutrements to hand, is one of the most powerful images in the entire book. The feet, hands and haloed head of another figure are at the centre of the outer edges of the frame. The figure was partially dismembered by a binder's knife in the nineteenth century, and the short, bifurcated flowering rod held in his right hand has been badly damaged by abrasion. Since part of the left hand and half of the head are now missing, identification is problematic. Against suggestions that it represents God embracing the cosmos, Dr Werner favours an indirect representation of Christ crucified.²⁸

Full pages (27v, 129v and 290v) are devoted to the evangelist symbols. All have wings and are framed in panels around crosses. The origin of the symbolism lies in the prophecies of Ezekiel (1.4) and the Apocalypse (4.2), each of which described winged figures with the appearance of a man, a lion, a calf and an eagle. These became identified with the evangelists, St Gregory's homilies on Ezekiel explaining the symbols as the four stages of Christ's life: birth, death, resurrection and ascension. Christ was a man in his birth, a calf in his death, a lion in his resurrection, and an eagle in ascending to heaven, so that, in representing the evengelists, the symbols were also representing Christ.²⁹ On 27v, all four symbols have haloes. A cross is placed over the calf's head, while three crosses are placed around the head of the eagle. The man for Matthew carries a rod topped with a flowering cross, while the eagle clutches his gospel. The symbols on 129v lack the explicit addition of haloes, but are enclosed within prominent yellow circles which may perform the same function. The man shares his frame with another like image, or perhaps the representation is that of an angel; the lion is accompanied by the calf and the eagle; the calf is with the eagle and, perhaps through error, another calf, where a lion might have been expected; and the eagle is with the calf and the lion. The intention is to stress the unity of the gospels despite their diversity of authorship and approach. Flabella are prominent inside each of the panels.

In the preliminary texts the symbols take diverse forms. On Ir they occupy the second column. Facing left to right, they are placed in the following order: man, lion, eagle, calf. The conventional forms appear in the canons from lv-3r, but on 3v, the lion has the hooves of a calf, and the calf is given the mane and paws of a lion. The man is in an unusual squatting position, accommodating to the shape of the arch. On 4r, the calf, in an odd, spreadeagled representation, has the tapering torso, ears and paws of a lion, and three spikes suggestive of a mane. The eagle on folio Sr holds his gospel with a human hand.

Stray allusions to the evangelist symbols may occur elsewhere, such as the calf of Luke on 201 v. The eagle, or at least a generalised raptor, more common then than now, is recognisable in several places, such as 174v, or 273v, where it is in flight. The eagle on 212v resembles strongly the symbol of St John in the Book of Armagh folio 90r. It may be possible to explain as symbols of St Matthew some of the small human figures which occur throughout the book, such as the man in the first column of numbers on lv, or the figures on 8r, one of whom holds an open book or wax writing tablets, or those who occupy the supporting columns of lv, 2r and 3v. Why they pull their own and, others' beards is not clear here, or on 329v. The ubiquity of the lion is discussed below.…

Conclusion: 'The secrets of the artistry'

This necessarily brief survey has attempted to convey something of what can be discovered about the Book of Kells. Much still remains uncertain. The precise relationship of its gospel text and decoration to that of other early gospel manuscripts is not established.⁷⁰ The minor decoration of the text pages has not yet been classified fully and evaluated. The relationship between scribe and artist is not known in detail. A precise place still has to be found for the Book of Kells in the development of zoomorphic initials in insular and continental manuscripts.⁷¹ Some of the pigments, particularly the organic ones, have not been identified with certainty, and in the present state of knowledge they cannot be. Colour symbolism in the manuscript awaits further research. Meanwhile, the long debate among commentators in different disciplines concerning the origin of the manuscript and the date it was written remains unresolved, and from a historical perspective these are issues of the first importance.

Yet these uncertainties are part of the fascination of the Book of Kells. For over a thousand years it has remained an enigma, a work to be wondered at rather than understood. In many ways, Giraldus Cambrensis' description of a book he saw at Kildare in 1185 can stand as a description of the Book of Kells:

'It contains the conc6rdance of the four gospels according to Saint Jerome, with almost as many drawings as pages, and all of them in marvellous colours. Here you can look upon the face of the divine majesty drawn in a miraculous way; here too upon the mystical representations of the Evangelists, now having six, now four, and now two, wings. Here you will see the eagle; there the calf. Here the face of a man; there that of a lion. And there are almost innumerable other drawings. If you look at them carelessly and casually and not too closely, you may judge them to be mere daubs rather than careful compositions. You will see nothing subtle where everything is subtle. But if you take the trouble to look very closely, and penetrate with your eyes to the secrets of the artistry, you will notice such intricacies, so delicate and subtle, so close together, and well-knitted, so involved and bound together, and so fresh still in their colourings that you will not hesitate to declare that all these things must have been the result of the work, not of men, but of angels."⁷²

Appendix I: Historical Background

No account of the Book of Kells can ignore St Colum Cille, abbot of Iona, with whose name it has been associated since the Middle Ages. Christianity was well established in Ireland, and supported by the aristocratic families, by the time Colum Cille was born in Donegal into the ruling dynasty of Uí Neíll in 521 or 522, a great-great-grandson of Niall of the Nine Hostages, the founder of the dynasty. Around 561 Colum Cille travelled to Scottish Dál Riata with twelve companions from among his kin, including his uncle and his cousin and successor as abbot, Baithéne. It is not certain whether this was with the aim, as Bede claimed, of converting the northern Picts, or whether, as may be more likely, that mission grew out of his foundation of two monasteries. Colum Cille settled first on the island of 'Hinba', perhaps to be identified as Canna,⁷³ then on Iona, an island off Mull. Iona grew to be the prosperous head of a confederation (or paruchia) of monastic houses exercising wide influence over ecclesiastical affairs in Ireland and in the north of England, where Lindisfarne became its most prominent foundation.

A great deal is known about life on Iona from the writings of the seventh-century abbot, A domnan (c628-704), the author of several works including a life of St Colum Cille. Iona is a small island (only about 5.5 km from north to south and 2.5 km at its greatest width) but with enough land to support the arable and grazing needs of a sizeable monastic community.⁷⁴ Its position, however, left it exposed to the threat of Viking attacks which began towards the end of the eighth century. It was pillaged in 795, and again in 802. In 806, sixty-eight of the community were killed in another raid. The following year, the survivors migrated to Ireland and began to erect conventual buildings at Kells, probably in wood, on a site granted to them in 804. Kells had earlier been prominent as a prehistoric burial ground, an ecclesiastical foundation, and a royal hill-fort associated with the southern Uí Neill.⁷⁵ In the words of the Annals of Ulster, this was 'the new monastery of Colum Cille' (noue ciuitatis Columbae Cille). The church at Kells, probably a stone structure, was not completed until 814. At this point abbot Cellach resigned from Iona and moved to Kells, leaving Diarmait installed as the new abbot of Iona. Cellach died the following year, still described in the Annals of Ulster as abbot of Iona. The annals are difficult to interpret on this point, but it seems that an overlap of abbots had taken place, perhaps because Cellach was elderly. From 814 until the death of Mael Brigte mac Tornain in 927, it appears that the abbacy of Kells and Iona was held jointly, its abbot regarded as the successor of Colum Cille, located in the house of refuge at Kells. Iona continued to be occupied after the settlement of Kells in 814, and another Norse attack on it is recorded in 825, when Blathmac was murdered for refusing to reveal the hiding place of St Colum Cille's shrine ('scrin'). This was probably the chief relic of the founder, the chest containing his corporal remains. The historical record is not sufficiently precise to allow for certainty concerning which relics were on Iona at this date, in particular whether a great gospel book associated with the founder was there, or whether it could be counted among the relics of St Colum Cille which went from Kells to Iona in 829, and came back two years later. It appears, though from separate historical sources, one composed in Ireland and one in Scotland, that a division of relics took place in 849, when some relics were taken to Kells and others to Scotland, probably to Dunkeld. The Annals of Ulster record that Colum Cille's shrine and other relics arrived in Ireland in 878, having been brought in haste following another Norse attack, though it is not stated whether the attack was on Iona or, as Bannerman deduces, on Dunkeld. Given his reputation as a scribe, it would be surprising if there were no book by Colum Cille in this consignment, though again it is not known whether a gospel manuscript of late eight- or early ninth-century date was included. It may be suggested that a gospel book of such recent manufacture would not necessarily have counted as a relic at all. Bannerman makes a case for concluding that such a manuscript, even one which we regard as highly as the Book of Kells, was placed beneath objects like the flabellum or the shrine itself in the hierarchy of Columban relics.⁷⁶

The community which produced the Book of Kells must have been rich and stable, with a large number of scribes and artists and an established library. This description would fit Iona before the raid of 795, or Kells in the period of calm it enjoyed after 814. It has not been possible however for scholars to agree on the date when the book was made and thereby fit it securely into either location. The date 'c800' can be cited as a suitable compromise, and the manuscript attributed to the Iona scriptorium, whether working at Iona or at Kells or partially at both locations.⁷⁷ Features of the decoration, principally the fact that the book is uncompleted, support the dramatic scenario of monks fleeing Iona in a currach, clutching their great half-finished gospel book, while Norsemen wreak havoc behind them. A pronounced change in the form of the canon tables at canon IX from the arches of lv-5r to the simpler grid style of 5v-6r was put forward by A. M. Friend in 1939 as evidence that the writing of the manuscript was interrupted by the Vikings and the subsequent move to Kells, but the occurrence of the same break in an earlier, fragmentary gospel book, British Library Royal 7.C.XII, makes such an explanation unnecessary.⁷⁸ Francoise Henry regarded Connachtach as the first scribe, working on Iona before the manuscript was completed at Kells. Connachtach was celebrated in a seventeenth-century compilation, the Annals of the Four Masters, as an eminent scribe and abbot of Iona, who died in 802. Although he was not known to the earlier and generally more informed Annals of Ulster, the record of his death may have made its way into the Annals of the Four Masters from a lost section of the earlier Annals of Tigemach. The status and succession of the abbots of Iona has not yet been established with certainty, but it may be taken that Connachtach was a historical figure, and one with a scribal expertise which was worth recording around the time when the Book of Kells was being written.⁷⁹ The texts of the canon tables and other prefaces in the Book of Kells derive from an exemplar which was very closely related to that used by the Book of Durrow.⁸⁰ Where the Book of Durrow itself originated is open to debate, though Dr Henderson has indicated a preference for Iona.⁸¹ A detail of the decoration of folio 201r of the Book of Kells has been seen as another indication of its place of origin. The right hand of a figure (this may be a man, as usually supposed, or a woman, since breasts are shown) with fins and a double tail of a fish grasps the name of one of Christ's ancestors, 'Iona'. This does not allude to the island, which was known as Í in the Middle Ages, but may echo St Colum Cille's name in religion, 'Columba', the Latin word for 'dove', which was 'iona' in hebrew, as Adomnán had pointed out.⁸² If it is a reference to the founder of the monastery—and such allusions are extremely rare—it is one which would apply equally to Iona or to Kells. Iona was well placed to act as an entrepôt for ideas and influences. The island received many visitors, such as the Gaulish bishop Arculf, who was blown off course on a journey home from the Holy Land and made his way to Iona during Adomnán's time. Adomnán took the opportunity to learn from him about the Holy Places, compiling a book on the subject which he presented in 686 or 688 to king Aldfrith of Northumbria, whose mother, incidentally, was Irish. It is not difficult to see a diffusion of information and artistic styles resulting from episodes such as this.

While Iona's claims are dominant at present, they are not conclusive, and it is difficult to escape the impression that no great quantity of evidence would be required for the pendulum of scholarly opinion to swing back, however inconclusively, towards Kells itself. In favour of an origin at Kells, there are pointers towards a date after 800 in similarities which Henry noticed with manuscripts produced in northern France around that date, principally the Corbie psalter (Amiens, Bibliotheque Municipale 18) and the Gellone sacramentary (Paris, Bibliothèque Nationale lat. 12048).⁸³ Manuscripts such as these are however no more susceptible than the Book of Kells is to precise dating. It may be that the production of a great gospel book was undertaken along with the erection of the cross of Saints Patrick and Columba as part of a programme of construction and decoration designed to enhance the prestige of Kells in its early years.⁸⁴

A certain agnosticism over the question of provenance has followed the controversy associated with the name of the Belgian scholar François Masai, and his publication in 1947 of Essai sur les origines de la miniature dite irlandaise, a work which attempted to overturn the traditional attribution of the Book of Kells and other landmarks of Irish art to Ireland. For Masai, the Book of Kells was a product of the scriptorium of Lindisfarne, its inspiration Anglo-Saxon in origin rather than Irish. His follower, the palaeographer Julian Brown, put forward a case in 1972 for the Book of Kells having originated at an unknown scriptorium subject to Northumbrian influence in Pictland in eastern Scotland.⁸⁵ These views have become less fashionable, and weight is now given to the inevitable intermingling of artistic influences which occurs with the movement of people and the objects they bring with them. While the Irish exercised a strong cultural influence on Northumbria through the monastic missions, there was also some English presence in Ireland in the same period, at, for example, the monastery founded by Colman for Englishmen in Mayo following the Synod of Whitby. Bede knew this as a distinguished and devout foundation.⁸⁶ There were Englishmen on Iona even in Colum Cille's time. Adomnán mentions two of them—Pilu, and an English lay brother, a baker by name of Genereus—though they were no doubt singled out because they were in a minority.⁸⁷

Appendix 11: Losses, Additions and Marginalia

The Book of Kells has not come through the centuries intact or unchanged. In the late eleventh and twelfth centuries, blank spaces in folios 5v-7r and 27r were used to record the details of property transactions of concern to the monastery of Kells.⁸⁸ The leaves now measure around 330 x 255 mm, but were severely cropped in the nineteenth century, and the edges were gilded. At present there are 340 folios, but around thirty folios, including some major decorated pages, have been lost. According to notes on folio 334v, the folios numbered 343 in the year 1588, one fewer than in the year 1621, when they were counted by James Ussher, then bishop-elect of Meath. At the beginning of the volume, an estimated ten leaves are missing. These may have contained Jerome's letter to pope Damasus, usually known by its opening words, Nouum Opus. Around twelve leaves have been lost from the end (John 17.13-21.25). Other textual gaps suggest the loss of roughly another six leaves in all, or twelve pages of text, at the end of the Hebrew names (after folio 26); at Mark 14.32-42 (after folio 177); at Luke 12.6-18 (after folio 239); and at John 12.28-13.20 (after folio 330).

A note by Gerald Plunket on folio 337r records the last of these gaps. As another note on the same page indicates, the bifolium in question was found in 1741. These pages had suffered severe water damage. Their pigments are largely washed out, and conspicuous iron staining is evident. The bifolium was folded the wrong way when it was rebound, and was numbered accordingly by J. H. Todd, Trinity College Librarian from 1852-69, when he foliated the manuscript in the lower left margin of each leaf. The mistake was not corrected until the Book of Kells received its present binding from Roger Powell in 1953, so that the leaves are now numbered in the order 336, 335.

Gerald Plunket has left many other annotations on the manuscript in his characteristic thin brown ink. He signed his name and initials on a number of pages; provided transcriptions of the text for folios 8r, 29r, 203r and 292r, usually in the tail margin; and imitated the artists in such efforts as his fish in the margin of 98v. Plunket interpreted folio 32v as a portrait of Christ, an identification with which most scholars would now concur, by writing 'JESUS CHRISTUS' in the spandrels of the arch on either side of the subject's head. In the nineteenth century, this phrase was concealed under obtrusive white paint, which now gives a misleading visual focus to the page.

Queen Victoria's contribution to the Book of Kells is well known. Visiting Ireland in August 1849, she and prince Albert were permitted to sign it, or so they believed. What they signed were modern flyleaves which were removed from the manuscript in 1953. Victoria was neither the first nor the last to succumb to the urge to fill its blank spaces. In the fifteenth century, a poem complaining bitterly about taxation on church land was written on the blank folio 289v. At the foot of the same page, Richard White, the rector of a small parish close to Kells, added some anodyne historical notes in the seventeenth century. Other contributions include the signature of Sir Thomas Ridgeway, Treasurer of Ireland in the early seventeenth century, on folio 31v. The monogram of John Obadiah Westwood, the author of the earliest modern account of the Book of Kells, appears with the date 1853 on folio 339r.

Notes

Abbreviations Used in the Notes

Adomnán (1991):

Adomnán's Life of Columba, ed A.O. and M.O. Anderson, revised edn (Oxford 1991)

Age of Migrating Ideas:

The Age of Migrating Ideas. Early Medieval Art in Northern Britain and Ireland, ed R. Michael Spearman and John Higgitt (Edinburgh 1993)

Alexander (1978):

J. J. G. Alexander, Insular manuscripts, 6th to the 9th century, A survey of manuscripts illuminated in the British Isles 1 (London 1978)

Bede, HE:

Bede's Ecclesiastical history of the English people, ed B. Colgrave and R. A. B. Mynors (Oxford 1969)

CLA:

Codices latini antiquiores: a palaeographical guide to Latin manuscripts prior to the ninth century, ed E. A. Lowe I-XI, Suppl (1934-72)

Farr (1989):

Carol Farr, 'Lection and interpretation: the liturgical and exegetical background of the illustrations in the Book of Kells', unpublished PhD dissertation, University of Texas at Austin (1989)

George Henderson (1987):

George Henderson, From Durrow to Kells. The Insular Gospel-books 650-800 (London 1987)

Henry (1974):

Francoise Henry, The Book of Kells: reproductions from the manuscript in Trinity College Dublin, with a study of the manuscript by Françoise Henry (London 1974)

Ireland and Insular Art:

Ireland and Insular Art AD 500-1200, ed Michael Ryan (Dublin 1987)

Kells commentary (1990):

The Book of Kells, MS 58, Trinity College Library Dublin: commentary, ed Peter Fox (Faksimile Verlag Luzern 1990)

Kells conference proceedings (1994):

The Book of Kells. Proceedings of a conference at Trinity College Dublin, 6-9 September 1992, ed Felicity O'Mahony (Scolar Press, for Trinity College Library Dublin, 1994)

Lewis (1980):

Suzanne Lewis, 'Sacred calligraphy: the Chi Rho page in the Book of Kells', Traditio 36 (1980) pp 139-59, at p 158.

Mayvaert (1989):

Paul Mayvaert, 'The Book of Kells and Iona', The Art Bulletin 71 (1989) pp 6-19

Ó Carragáin (1994):

Éamonn Ó Carragáin,' "Traditio Evangeliorum" and "Sustentatio" ': the relevance of liturgical ceremonies to the Book of Kells', Kells conference proceedings (1994)

O'Reilly (1993):

Jennifer O'Reilly, 'The Book of Kells, Folio 114r: a Mystery Revealed yet Concealed', in Age of Migrating Ideas pp 106-114.

O'Reilly (1994):

Jennifer O'Reilly, 'Exegesis and the Book of Kells: the Lucan genealogies', in Kells conference proceedings (1994).

Werner (1994):

Martin Werner, 'Crucifixi, Sepulti, Suscitati: remarks on the decoration of the Book of Kells', Kells conference proceedings (1994)

Zimmermann:

E. H. Zimmermann, Vorkarolingische Miniaturen I-IV (Berlin 1916-18)

¹Kells commentary (1990) p 14.

² The word 'insular', originally coined by the German palaeographer Ludwig Traube in 1901 to describe the script of the period in Britain and Ireland, has become commonly used as a broad and neutral term to describe the characteristics of the style in art and artefact as well as script. 'Majuscule' distinguishes the script as being of a higher grade than 'minuscule'.

³ Dáibhi Ó Cróinin, 'Ireland and the Celtic kingdoms of Britain', in The Work of Angels. Masterpieces of Celtic Metalwork, 6th-9th centuries AD (London 1989); idem, 'The Irish missions', in The Celts (Milan 1991) pp 659-662.

⁴Kells commentary (1990) pp 153-65.

⁵The Annals of Tigernach. The Continuation, A.D. 1088A.D. 1178, ed. Whitley Stokes, Revue Celtique 18 (1897) p 12. The Book of Kells resembles the Book of Durrow closely in its prefatory material, and may have been termed the 'great' gospel book of Colum Cille in the annals in order to distinguish it from the lesser and earlier gospel book of the saint: 'libellum', as the Book of Durrow is termed in its colophon. See Evangeliorum Quattuor Codex Durmachensis II (1960) pp 17-24. For the view that one of these books was not a gospel but rather the psalter known as the 'Cathach' of Colum Cille, see Máire Herbert, Iona, Kells, and Derry (Oxford 1988) p 93.

⁶ See the bibliographical references in Alexander (1978) pp 71-6.

⁷ Lewis (1980) p 158.

⁸ The Chi Rho is depicted decoratively on other pages, such as 62v, where it is made up of the body and paw of one lion and the leg of another. The same shape is found on 258r, formed by the tongue of one lion and the tail of another.

⁹ See 0. K. Werckmeister, 'Die Bedeutung der "Chi" Initialseite im Book of Kells', Das erste Jahrtausend. Text band 11, ed V. H. Elbern (1964) pp 687-710; Lewis (1980); George Henderson (1987); Farr (1989); Carol Farr, 'Textual structure, decoration and interpretive images in the Book of Kells', Kells conference proceedings (1994); Ó Carragáin (1994); Werner (1994); O'Reilly (1993); O'Reilly (1994).

¹⁰ Lewis (1980) p 140.

¹¹ John Higgitt, 'The display script of the Book of Kells and the tradition of insular decorative capitals', in Kells conference proceedings (1994).

¹² See note 74.

¹³ Robert Stevenson, 'Further thoughts on some well known problems', in Age of Migrating Ideas pp 16-26.

¹⁴ Moyse's Hall Museum, Bury St Edmunds 1987-27.1, reproduced in The Work of Angels. Masterpieces of Celtic Metalwork, 6th-9th centuries AD (London 1989) p 143, 136.

¹⁵ Henry (1974) p 182. The belief that 33r is now turned the wrong way, based on the artistic balance and colouristic values of the conjectural opening 33r/34r, is given some confirmation by codicological evidence. Most of the leaves of the manuscript have been trimmed more rigorously at the head than at the tail in the course of an unsympathetic binding in the nineteenth century. The head and tail margins of folio 33r conform to this pattern only if the page is turned upside down and faces folio 34r.

¹⁶ Bede, HE i.25.

¹⁷ Monumenta Germaniae Historia, Epistolae III (1892) p 60.

¹⁸ Ó Carragáin (1994).

¹⁹ Adomnán (1991) pp 208-9.

²⁰ Adomnán (1991) pp 78-9.

²¹ Farr (1989) pp 344-5.

²² Werner (1994).

²³ In the eighth-century Trier gospels (Trier, Domschatz 61 folio 9r), the archangels Michael and Gabriel are identified as such in introducing St Matthew's gospel.

²⁴ Adomnán (1991) pp 14-15.

²⁵ Alexander (1978) plate 265.

²⁶ Adomnán (1991) pp 142-3.

²⁷ Adomnán (1991) pp 14-15.

²⁸Kells commentary (1990) p 280; Werner (1994).

²⁹ Henry (1974) pp 194-6.…

⁷⁰ See the comments of Patrick McGurk, Kells commentary (1990) pp 61-9.

⁷¹ Much research remains to be done on parallels between the Book of Kells and particular continental manuscripts such as Paris, Bibliothèque Nationale, lat. 9427 or lat. 12168.

⁷²Kells commentary (1990) pp 320-21.

⁷³ J. L. Campbell, Canna. The story of a Hebridean island (Oxford 1984) pp 5-6.

⁷⁴ Ian Fisher, 'The monastery of Iona in the eighth century', in Kells conference proceedings (1994).

⁷⁵ D. L. Swan, 'Kells and its Book', in Kells conference proceedings (1994).

⁷⁶ For discussion of the problems of St Colum Cille's relics, see Meyvaert (1989) p 11; George Henderson (1987) pp 179-98; John Bannerman, 'Comarba Coluim Chille and the relics of Columba', The Innes Review 44 (1993) pp 14-47.

⁷⁷ Henry (1974) pp 216-220. For a review of the problem, see J. J. G. Alexander, Kells commentary (1990) pp 288-9; Peter Harbison, 'Three miniatures in the Book of Kells', Proceedings of the Royal Irish Academy 85 C 7 (1985) pp 181-94, at pp 190-3.

⁷⁸ Alexander (1978) p 74.

⁷⁹ Henry (1974) pp 220-1. Henry regarded Connachtach as scribe 'A': see above p 91. I am grateful to Dr John Bannerman for advice on Connachtach.

⁸⁰ Patrick McGurk, 'The texts at the opening of the book', in Kells commentary (1990) pp 57-8.

⁸¹ George Henderson (1987) pp 54-5.

⁸² Meyvaert (1989) pp 6-10.

⁸³ Henry (1974) pp 215-9.

⁸⁴ Françoise Henry, Irish art during the Viking invasions (800-1020 AD) (London 1967) p 138.

⁸⁵ See Carl Nordenfalk, 'One hundred and fifty years of varying views on the early insular gospel books', in Ireland and insular art pp 1-6.

⁸⁶Bede's Ecclesiastical history of the English people, ed B. Colgrave and R. A. B. Mynors (Oxford 1969) IV.4.

⁸⁸ See G. Mac Niocaill, Kells commentary (1990) pp 153-65; M. Herbert, 'Charter-material from Kells', Kells conference proceedings (1994).