The implications of the Staffordshire Hoard for the understanding of the origins and development of the Insular art style as it appears in manuscripts and sculpture

George and Isabel Henderson

As the full gamut of decorative forms on the various items of the Hoard come to be assessed, what light may we expect the Hoard to shed on the chronology of Insular illuminated manuscripts, or at any rate on their likely liaisons with one another and the drift of stylistic developments within them? It is generally agreed that in their use of motifs, and in their basic principles of design, metalwork and manuscripts in Britain in the early-medieval period mesh into one another, for example, to put it at its most simplistic, in the Ardagh chalice and the Lindisfarne Gospels. An interesting suggestion has recently been aired that very early decorative practice in manuscript art may underlie some of the most accomplished pieces from Sutton Hoo, rather than necessarily the other way round (Nees 2007, 15). Hitherto, of course, Sutton Hoo has been the principal focus for comparisons between metalwork and the manuscripts. The discovery of the Prittlewell burial chamber has been said to undermine supposed certainties about Sutton Hoo's material remains and context (Nees 2007, 16). However, the character of the finds are a very reasonable match for Bede's historical record of Essex in very early Christian times, in the person of Saba (HE II,5).

If Prittlewell did indeed disturb the status of Sutton Hoo, what must the effect be of the thousand and a half items in the Staffordshire Hoard? It is of course a forceful reminder of the difference between the critical situation before July 2009 and now - the change in the weight of the evidence which the emergence of the Hoard has caused. The manuscript situation is pitifully different. The Faddan More Psalter is only tentatively creeping into the light (Kelly et al, 2006) and the possibility of vellum manufacture at Portmahomack, Tarbat, Easter Ross, only hints at loss, does not make up for it (Carver 2008, 124-5). Scholars of manuscripts, particularly students of script, relish warning us about the likely loss of whole generations of manuscripts between the extant items, frowning on attempts to elucidate an apparent process. It is ironical that within the sword pommels in the Hoard generations of items are being postulated.

The Book of Durrow f. 192 vo

On the matter of the animal designs in the Hoard, scrutiny of the whole will show what range of types of Style II animals, and others, are employed. Do they move beyond the safety-pin jawed type to the type seen in Sutton Hoo with long upper and lower jaws used as space fillers and linear patterns, animals which undulate more freely, so that they really parallel the varied animals represented on such crucial parchment leaves as f.192vo of Durrow?

The Book of Durrow f. 125 vo

Equally, f.125vo is stamped over with rectangular panels containing decoration matched on the Sutton Hoo sword belt fittings (G. Henderson 1987, 32-3, pls 26 and 27). Are even closer parallels going to emerge when all the Staffordshire Hoard items are cleaned?

The chronological problem

The good reason for placing the Book of Durrow at the beginning of the manuscript sequence is the universally acknowledged relevance of the comparanda available in the decorated metalwork in the Sutton Hoo ship burial. The closeness of the connection brought with it the proposal of a date in the seventh century derived in a general way from the numismatic evidence present in the burial. The suggestion mentioned above, that very early manuscript design could have influenced metalwork, also supports a seventh-century date with a consequently likely origin in the south-east of England, the area to which, thanks to the Roman mission, we know that books containing sophisticated decoration were being introduced.

The new Staffordshire material shares design and technical features with the products of the Sutton Hoo goldsmith. The study of the Hoard will refine both that technical relationship and the nature of the designs and layout employed. To this great increase of gold and garnet cloisonné metalwork displaying abstract and animal ornament can also now be added the many single metal-detector finds scattered widely over the English landscape which use the same style and repertoire, and share with the Staffordshire material no fixed date and no certain location of the place of manufacture. As was evident at the Symposium, a great deal of relevant continental material has been organised into a typological chronology. Karen Høilund Nielsen has also recently looked at the whole phenomenon of the employment of Germanic Style II in England, from its introduction to its falling out of use - and it did fall out of use, after a relatively short period (Høilund Nielsen 1999). This research, together with a deepening knowledge of cloisonné work and technical advances in the approach to the materials employed by the jewellers, will have a fundamental bearing on the significance of the Staffordshire pieces, making possible finer distinctions leading to more precise chronologies for the Sutton Hoo comparanda. From what we learned at the Symposium it is obvious that this precision will have a bearing on the dating of the analogues in the Book of Durrow.

Style historians must resist compliance with the now common view that metalwork can have no place in the dating of manuscripts. One palaeographer observed 'Any particular jewel can have a long life and may be old before it is reflected in a manuscript' (O'Sullivan 1994, 84). Such a truism cannot be sustained as a general principle and, quite unjustifiably, tries to neutralise art-historical perceptions of development within a manuscript style attributable to metalwork. No art style can be fossilised. If copied slavishly, then after a period of time the essential rhythms of the style are stultified and there is misunderstanding of the model. Taste may dictate a desire for echoing an old style, but no masterwork will refrain from some personal interpretation which can easily be discerned. As has been remarked, 'there is always a creative element in zoomorphic ornament' (I. Henderson 1997, 151). The repertoire of the Book of Durrow, a masterwork, was not achieved by scissors and paste - the result of sticking together designs taken from old-fashioned mounts or even heirloom pieces. The art of Durrow displays rigorous control and engagement with a coherent living style. And of course an artist will subtly develop his material. Note, for example, the observation of Haseloff that the animals in procession on f.192vo of Durrow have a degree of naturalness added to the Style II animal. Haseloff attributed this to Mediterranean influence, which for some served to dilute the English element in the style. In other words, the artist was developing the style precisely as one would expect. In spite of this exotic element, Haseloff's personal view dated Durrow to about 650 (Haseloff 1987, 46). Such influence could have belonged to a much earlier period, or come from sources more committed to naturalistic animal art such as the art of the Picts, a source tapped for other purposes by the Durrow artist. What we then see in the later manuscripts is precisely the kind of creative development, described above, of the still discernible earlier Germanic style.

The lack of a definite context accounting for the origin of the Staffordshire Hoard must be regretted, but it has the advantage of opening minds to the possibility that the style flourished in places where the evidence for its use is sparse or not available at all. The awareness of this wider dispersal was already underway as a result of the finds at Dunadd, Argyll, the secular Dalriadic stronghold. These have been attributed to interaction with Northumbria, revealing that the Germanic style went beyond its nonetheless important 'Bamburgh beast'(Lane and Campbell 2000, 152-4; 243-7). The economic and social necessity for acquiring gold treasure was compellingly demonstrated at the Symposium, allowing for the movement of such treasure from any region to another. That Mercia might itself have manufactured some of the Staffordshire goldwork is supported by a recent careful assembly of old and new finds, many of the latter by detectorists, of a body of precious metalwork which could hold together as a Mercian style of the eighth century (Webster 2001). This style had a significant extension to the north, which could be explained by political affiliations, but if some of the Staffordshire Hoard was truly Mercian, then a more enduring northern association becomes explicable. More importantly, a convincing seedbed would be provided for the later Mercian metalwork and sculpture.

We are assuming that the incised naturalistic animal art of the Picts influenced the animal Evangelist symbols in Durrow and the Echternach Gospels. The later relief-carved monuments of the Picts favoured a cross-slab format, particularly suited to the reception of manuscript art. This means that Pictish sculpture in relief is a useful source for providing evidence for the repertoires of the (surely many) lost illuminated Gospel books. In early Pictish sculpture in relief, and in the fragmentary relief slabs at Monkwearmouth, we can observe the beginnings of the combination of early metalwork animal-art style with other features of the Insular art style, notably interlace. Such combinations might well already have occurred in manuscripts now lost to us.

The making of the Book of Durrow, which we personally regard as early, has much to offer in the understanding of the interaction between media. How early the individual components were available inevitably will influence our view of chronologies in manuscripts and sculpture alike. If the early eighth-century date for Durrow is accepted (see below), then chronologies for sculpture, largely based on dating of other media, would certainly require radical revision. However if a date as early as towards the beginning of the seventh century is to be accepted for both the Sutton Hoo material and the evidently related items in the Staffordshire Hoard, then the consequence is equally radical. That development within art styles exists and remains a visible framework for chronology appears to us indisputable. Its implications cannot be set aside, however revolutionary the result.

Some art-historical pointers already visible in the Staffordshire Hoard

The long poker-face of the gold helmet fitting StH 678, with squint-set pointed oval eyes and thin closed mouth, is reminiscent of the impassive economically delineated head of the calf symbol on f.124vo of Durrow.

StH 678 helmet fitting

The Book of Durrow f. 124 vo

The Book of Durrow f 2

The three-dimensional helmet fitting also reminds us of the sort of model evidently in the mind of the artist of f.2 of Durrow, where the Evangelist symbols are turned frontally and the lion, particularly, takes inspiration from a three-dimensional model such as the Sutton Hoo whetstone stag or the Carnwath bull (G. Henderson 1987, 47, pl.52). Here metalwork is undoubtedly making an impact on book painting. The cross which these frontal symbols accompany on f.2 is on a modest scale, kept within an inner compartment, not yet reaching out to the outer edge of the illuminated page like the crosses on f.26vo of Lindisfarne, p.220 of Lichfield, or f.126vo of the Augsburg Gospels (Alexander 1978, pls. 38, 77 and 119). It is interesting to note that all these crosses, like that of Durrow, employ a single motif throughout the length and breadth of the cross, so that the repetition of the entangled Style II animals on the folded-up Staffordshire Hoard altar cross StH 655 is in that very same stylistic tradition, arguably at a stage even more tentative and experimental than Durrow's use of an all-over design of woven cords.

The Durrow cross expands at the ends of the arms and stem, where large circular garnets were evidently set in the Staffordshire cross. The centre of the Durrow cross is given some emphasis by square insets, but does not rival the emphasis of the great garnet (or whatever) at the centre of the Staffordshire cross. This extra emphasis is like the introduction of the bust-length image of Christ at the centre of the cross on the Trier Gospels four-symbols page, f.1vo (Netzer 1994, 103-107 and pl. 1), while half-circles, like truncated settings, end the cross arms. The iconographic elaboration here in Trier suggests the possibility that, in the Staffordshire Hoard altar-cross, the idea of the five wounds of Christ and the blood-like colour of garnets were elements in the design scheme. It is indeed intriguing that the designer chose to decorate his cross arms and stem with the self-same design used on the neck of the maplewood bottle from Sutton Hoo (see Karen Høilund Nielsen's paper), bringing the idea of secular drinking close to the liturgical rite of the Christian altar, a primitive, uninhibited juxtaposition perhaps not too far from the outlook of Saba's sons when they demanded a share of 'the white bread' being distributed only to the faithful by Bishop Mellitus (HE II, 5).

Stylistically interesting is the fact that although long narrow beak-like forms occur on the decorated letters in Durrow, they are strictly part of trumpet spirals, with no zoomorphic touches, whereas the long slim motifs which frame the great central setting on the Staffordshire Hoard cross are bird heads with round eyes and the customary Anglo-Saxon cleft beak terminal.

Discussion of the script and epigraphy has not dwelt much on the beast-head line-ending which appears on both the obverse and reverse of the inscribed gold strip (StH 550) in the Staffordshire Hoard.

Though the beast-head gives the effect of profile, it is in fact flattened, seen from above, with two eyes, of the same pointed oval shape. The jaws curve up and are wide open, a design reminiscent of the wide-open mouths of the serpents incised on the Eccles buckle (Webster and Backhouse, 1991, no.7) and the filigree serpents between the legs of the boars on one only of the four parts of the Sutton Hoo shoulder clasps (Inv. 4a; Bruce-Mitford 1978, fig. 386, 434 and 436).

The front of the Eccles buckle Detail of shoulder-clasp Inv. 4a, Mound One, Sutton Hoo

The line-ending on the inscribed gold strip has a bold, gauche, casually confident, graffito-like quality. This style trait or tendency is present in the head of the lion symbol on f.191vo of Durrow (with incidentally the same framing band folded back at the nape of the neck) and it is the trait or tendency which accounts for the chunky simplified form of the lion symbol in comparison with its likely model, the Pictish wolf from Stittenham, Inverness-shire.

Detail from the Book of Durrow f. 191 vo (detail) The Sittenham Wolf

Transformation of the fluent naturalistic Pictish animal sculpture into a robust piece of Anglo-Saxon metalwork is not the tendency in manuscripts such as the Echternach Gospels, the date of which, in the early eighth century, much scholarly literature is currently arguing is also the date of the Book of Durrow, focussing on a shared scribal convention of demarking text spaces by a 'flourish' (Ó Cróinín 1989). On f.115vo of Echternach the animal symbol is a somewhat spiritless but essentially accurate copy of the naturalistic Pictish animal type (Henderson and Henderson 2004, 32-33 and pl. 27). There is a definite change or difference in attitude here which should perhaps have - even probably have - chronological significance. The cultural shift is palpable.

Detail from the Echternach Gospels f. 115 vo

An unusual and interesting feature of the line-ending beast-head is its triple-pronged tongue. It gives a fierce defensive look to the head, perhaps associated with the pugilistic phrases of the inscription. Psalm 67, verse 23, mentions dogs licking the blood of God's enemies, and the Carolingian Stuttgart Psalter's illustration of that verse shows a lively dog's head down among streaming blood. But the Staffordshire beast is more likely to be a dragon. In the sixth of the Nordic mythology books in Saxo Grammaticus's Gesta Danorum, the dragon guarding a heap of treasure, as in Beowulf and the Sigurd legend, has a three-forked tongue which leaps out of its gaping mouth. For a long narrow strip of metal, inscribed with lettering, and ending in a ferocious dragon head, there is a parallel in the possible fitting from a knife scabbard (Webster and Backhouse (eds) 1991, no. 179). An outstanding Pictish example of such a format, a band of lettering ending in a fierce beast's head, is the silver-gilt chape from St Ninian's Isle (Wilson 1973).

Fitting found at Westminster Bridge Chape from St Ninian's Isle

Rather than derived from pagan mythology, however, the tongue and its form might perhaps have been intended to convey and emphasise the sacred nature of the text inscribed, which is after all an invocation of the God of Battles at a critical moment in the history of the Chosen People. The triple tongue might illustrate and represent the tongues of fire, symbolising the Holy Spirit, which are referred to in The Acts of the Apostles, 2, verse 3, as dispertitae linguae, tongues separated into parts.

Another line of thought is opened up by Jacqueline Borsje's study of Adomnán's Vita Columbae (VC). Borsje points out that the text referring to the vipers which St Columba renders powerless on Iona (VC III, 23) mentions the poison of their 'three-forked tongues, trisulcarum linguarum.' She traces this attribute back to classical texts, at least one of which seems likely to have been known to Adomnán (Borsje 1996, 145-151). In Virgil's Aeneid II, 475, the snake, coluber, having refreshed its strength, comes forth into the light 'darting from its mouth a three-forked tongue, linguis micat ore trisulcis.' Virgil uses exactly the same words about an amphibious snake in Georgics III, 439. These words neatly describe the line ending on the inscribed gold strip in the Staffordshire Hoard.

In VC the saint counteracts the poison of the evil vipers by his faith in Christ and the power of his blessing. In Georgics the snake is too dangerous to approach, and earlier the poet urges a shepherd to kill a similar venomous snake with a stone. Neither of these negative accounts of inimical creatures would seem an apt model for the form and function of the Staffordshire Hoard's line-ending, unless the sense was truly defensive of the text, as we suggest above. But a highly appropriate belligerent sense can be perceived if the source of the idea is Virgil's Aeneid II, 475. There the snake is introduced as a simile for Pyrrhus, 'primoque in limine, on the very threshold, proudly gleaming in the sheen of brazen arms', as he begins to storm the palace of Priam - so a quintessentially warlike context for the image.

It is an interesting coincidence that two of the quotations from Virgil in Bede's Ecclesiastical History come from the same point in the history of the fall of Troy, Sinon's fraudulent talk accounting for the wooden horse, in which Pyrrhus among others is meanwhile concealed, and the bloody death of Priam at Pyrrhus's hands - this last only twenty-six lines after the simile of the snake with the three-forked tongue, so a text certainly known in Northumbria, if nowhere else in Anglo-Saxon England! It would be culturally curious and suggestive if the designer of the Staffordshire Hoard's line-ending did indeed finish off the inscribed battle cry from sacred scripture with a visual allusion to a high moment in classical literature.

The line-ending on the inscribed gold strip from the Staffordshire Hoard, carefully considered, seems likely to have a part to play in the discussion of chronology of the Insular manuscripts. Its simple directness and its scale and potency are difficult to parallel in manuscript art. It is more like an inverted initial, from which the lines to text expand backwards. Again the interesting question arises as to the priority of manuscript practice to this particular example of luxury goldsmith work.

The news that the Staffordshire Hoard contained a large quantity of sheet silver, embossed with military iconography, was particularly exciting, in anticipation, to students of Pictish art, where silver is the staple material and figurative art is outstanding and diagnostic in the period. It appears, however, that the ethnic content of the Hoard is not mixed. So far as we have seen them, the figures on the metal sheets, with their well-defined legs and ankles, are tightly grouped warriors recognisably similar to the military bands on the Sutton Hoo helmet. The bent legs of some of the figures, as we remarked at the Symposium, are reminiscent of the running shepherds in the Romulus/Remus scene on the Franks Casket. In the great dearth of narrative figurative art in this early period, when in many media, especially textiles, in a festive secular society there must originally have been ample examples, the full exploration of the silver repoussé figures is awaited with the keenest interest.


  • Alexander, J. J. G., 1978. Insular Manuscripts: 6th to the 9th century (Miller, London)
  • Borsje, J., 1996. From Chaos to Enemy: encounters with monsters in early Irish texts (Instrumenta Patristica 29, Turnhout)
  • Carver, M., 2008. Portmahomack: monastery of the Picts (Edinburgh University Press, Edinburgh)
  • Kelly, E., Meehan, B., Read, R., Ó Flóinn, R. and Wallace, P., 2006. 'T he Faddan More Psalter: a medieval manuscript discovered in County Tipperary, Ireland, 20 July 2006' Supplement to Archaeology Ireland 77 (Autumn 2006)
  • Haseloff, G., 1987. 'Insular animal styles with special reference to Irish art of the early medieval period' in M. Ryan (ed) Ireland and Insular Art AD 500-1200 (Royal Irish Academy, Dublin) 44-55
  • Henderson, G., 1987. From Durrow to Kells (Thames and Hudson, London)
  • Henderson, I., 1997. 'Paralleled zoomorphic ornament on Pictish sculpture at Nigg, Easter Ross, and St Andrews, Fife, and in the Book of Kells' in C. Karkov, R. Farrell and M. Ryan (eds), The Insular Tradition (State University of New York Press, New York), 143-166
  • Henderson, G. and Henderson, I., 2004. The art of the Picts : sculpture and metalwork in early medieval Scotland (Thames and Hudson, London)
  • Høilund Nielsen, K., 1999. 'Style II and the Anglo-Saxon élite', Anglo-Saxon Studies in Archaeology and History 10, 185-202)
  • Lane, A. and Campbell, E., 2000. Dunadd: an early Dalriadic capital (Oxbow Books, Oxford)
  • Nees, L., 2007. 'Weaving garnets: thoughts about two 'excessively rare' belt mounts from Sutton Hoo' in R. Moss (ed) Making and Meaning in Insular Art (Four Courts Press, Dublin), 1-17.
  • Netzer, N., 1994. Cultural Interplay in the Eighth Century: the Trier Gospels and the making of a scriptorium at Echternach (Cambridge University Press, Cambridge)
  • Ó Cróinín, D., 1989. 'Is the Augsburg Gospel Codex a Northumbrian manuscript?' in G. Bonner, C. Stancliffe and D. Rollason (eds) St Cuthbert, his Cult and his Community to A.D. 1200 (Boydell Press, Woodbridge), 189-201 and pl. 26
  • O'Sullivan, W., 1994. 'The Lindisfarne scriptorium: for and against' Peritia 8: 80-94 (Journal of the Medieval Academy of Ireland, Cork)
  • Webster, L., 2001. 'The Anglo-Saxon hinterland: animal style in Southumbrian eighth-century England, with particular reference to metalwork', Veröff. Joachim Jungius-Gesellschaft der Wissenschaften, Hamburg, 90: 39-62.
  • Webster, L. and Backhouse, J. (eds), 1991. The Making of England: Anglo-Saxon art and culture AD 600-900 (British Museum Press, London)
  • Wilson, D., 1973. 'The Treasure' in A. Small, C. Thomas and D. Wilson, St Ninian's Isle and its treasure (Oxford University Press, Oxford)