Style II and all that: the potential of the hoard for statistical study of chronology and geographical distributions

Karen Høilund Nielsen (University of Southern Denmark)

Although the current English Heritage radiocarbon dating project includes a chronological seriation of selected grave-good types, the single object in the Staffordshire Hoard that would fit into this seriation is the small buckle with three rivets. Furthermore, the Staffordshire Hoard is unlikely to be a closed chronological unit.  Because of this, to consider chronology we must instead examine the objects from the hoard on the basis of their decoration, in styles related to Salin’s Style II and animal style.

The background for dating Style II comes from a major project in the 1980s and 1990s on Style II in Scandinavia, continental Europe and Anglo-Saxon England (for Anglo-Saxon England, see Høilund Nielsen 1999). For this project about 124 objects with Style II were recorded from Britain. The Staffordshire Hoard appears to add about 80 to 90 new objects with Style II to the total, not including fragments of embossed foils, and this number may rise when the dirtier items are cleaned. That represents an increase in material of about 50% to 75% - a good reason to start from square one again!

The method of analysis used in the 1990s was similar to that used by Salin in 1904, but with modern statistical methods and computer equipment; to analyse the stylistic elements, see their combinations, and analyse those combinations using correspondence analysis. Anglian areas and Kent did not work well when examined together, and so had to be treated separately. The result was good parabolas each for Kent and East Anglia, showing the chronological order of the Style II elements (Høilund Nielsen 1999, figs. 2, 5 and 7).

The objects from the Staffordshire Hoard that are decorated with animal interlace fit best within the Kentish material. The Kentish objects have a very strict and organised composition, which is exemplified in details of the objects like the brooch from Kingston Down grave 205, the buckle from Sarre grave 68, and the buckle set from Gilton grave 23.

Objects from the Staffordshire Hoard can show the same strict regularity, whether with or without animal elements. StH17 and StH21 are examples without animal elements, but StH 1148 has legs and perhaps a head.

On the other hand, some objects have interlace which is much more dismembered, more uneven, perhaps more relaxed. Here you can see StH 686 with the head elements highlighted in red, with a comparison from Market Rasen in Lincolnshire, both showing this more relaxed style.

On the basis of the Kentish material, there appears to be a chronological development from strict and regular animal interlace early on, then more uneven and perhaps sloppy interlace, then a loss of animal elements.

The original analysis of material from the Anglian area was dominated by material from Sutton Hoo. The early part of the seriation probably included Scandinavian imports; later on the style was developed locally, but building on this Scandinavian basis. Later still we have elements familiar from the Book of Durrow, the Crundale pommel and the Allington Hill mount. Finally come elements from very early painted Gospel manuscripts.

Placing the Staffordshire Hoard objects within the Style II seriation

I have not run a new Style II seriation including the Staffordshire hoard objects, but have simply estimated where they might fit into the old one. From this, the earliest object in the hoard would be pommel StH 711, which may be a Scandinavian import. On one face of this pommel (above) is a possible human mask, and on the other side (below) are boars’ heads which compare with those on the gold-and-garnet pommel from Hög-Edsten in Sweden. The details of StH 711 clearly fit in with the very early part of the Anglian seriation (Høilund Nielsen 1999, fig. 7).

A group of gold-and-garnet ornaments (e.g. the birds StH 16 and StH 1084, and the pair of pyramids StH 451 and StH 1166) have cloisonné animal decoration which fits into the later part of the Anglian seriation. This would make them contemporary with the later items from Sutton Hoo Mound 1.

   

The slightly rustic Style II ornament on the large cross StH 655 can be compared with the same motif in embossed foil on a maplewood bottle from Sutton Hoo Mound 1 (Bruce-Mitford 1983, fig. 261a). The wooden bottle from Sutton Hoo fits into the middle of the third quadrant of the seriation (ringed in the diagram below), and it is possible that the cross should fit in at more or less the same place.

The helmet cheekpiece StH 453 has a similar composition to the Allington Hill plate. If we focus in on the details of the animal style on StH 453, however, we find a body that fits into the later part of the seriation, but some details of the head that actually fit into the early part. Perhaps we are actually seeing for some reason a revival of older elements – a kind of ‘retro’ phase, in modern parlance.

    

Examples of this combination of elements of ostensibly different dates can be found on several objects in the hoard. The strange object StH 130 again has an ornament which combines elements from the later part of the seriation with elements that seem to be a revival of early Style B or very early Style II. Another example is the seax mount StH 449. The details look late, but the overall effect from a distance is that of a frieze of running dogs, characteristic of early Style II.

   

Perhaps another example is StH 271. The overall impression of these figure-eights is similar to late Vendel Style in Sweden, for example a mount from Vendel grave VII (Stolpe and Arne 1912).

There is a group of gold-and-garnet items including StH 72, StH 980, StH 660, StH 967, and pommels StH 284 and StH 1167, which have cloisonné animal motifs which are from the same stylistic family as the motifs on the Crundale pommel. Pommel StH 347 is extremely close to the Crundale pommel. These appear to fit nicely into the final section of the old seriation, which includes elements from the Book of Durrow and other very early manuscripts.

    

 

There are particularly close parallels between f192vo in the Book of Durrow and the seax fitting StH 567.

In summary, most of the items in the Staffordshire hoard which could be related to the Anglian seriation fit into its the later part, with a few elements belonging to the very earliest part. These may be due to a genuinely early date of some objects, or a degree of foreign influence, or perhaps my suggested revival of earlier motifs.

When we compare the Staffordshire hoard to the Kentish seriation(Høilund Nielsen 1999, fig. 5), it seems that the animal interlace from the hoard begins in the middle and may extend later than the latest objects in the 1990s study. The result of a new analysis would probably form an extension of the seriation beyond the current end of the Kentish seriation.

 Approximate calendar dates, based on the conclusions from the 1990s, would most probably be post-600 and pre-700.

Moving on to the embossed foils, we can see at least two different sets of animal motifs. StH 75 is certainly different to StH 1383, StH 1412 and StH 1406, which as Leslie Webster has commented look as much like rim mounts for a wooden object as part of a helmet.

      

There is a clear similarity between the repeating animal on the longer strip StH 1383/1406/1412 and a square Danish horse-harness mount from the National Museum, Copenhagen; the head, body and two legs are arranged in the same way.

Foils more certainly from a helmet come in a variety of patterns. One set (e.g. StH 1423 and StH 1382) shows marching soldiers with unusual clothes. Although the eagle-crested helmets are familiar from the Vendel helmet foils (Stolpe and Arne 1912), at Vendel the warriors are wearing chain mail, not the strange short balloony tunics of the Staffordshire Hoard. There are no boar-crested helmets shown on the foils, although we know from the Benty Grange helmet that these did exist in England too.

 

These tunics and helmets are not Continental imitations, and although the Scandinavian links seem quite clear at this stage, the Staffordshire Hoard gives us a lot of new pieces of iconography whose Europe-wide links will have to be assessed. The Staffordshire Hoard foils are not telling exactly the same story as their Scandinavian counterparts; they do not come from exactly the same society.

Another set, including StH 1400 and StH 1624, has a motif similar to that on the embossed gold disc from Pliezhausen in Germany; we also see it on helmets from Vendel and Valsgärde.
       

The Staffordshire Hoard foils are closest to the Pliezhausen design, but differ in the way that the dagger enters the horse and the way in which the figure is holding the horse leg. The basic motif derives from the ‘Fallen Warrior’ motif found in Roman art.

A third set of helmet foils (e.g. StH 1432, StH 1529 and StH 1556) show kneeling warriors each with a spear and a shield. They also differ from the usual depiction of warriors in that they are looking upwards, so one might expect to find something or someone standing above them. No direct parallels to these have so far been found.

Finally we come to StH 1497, which has been nicknamed the ‘sea-horse’. This appears to be contemporary with painted gospel books and the very latest part of the Anglian Style II seriation. This late date may also be appropriate for the animal head at the end of the inscribed strip StH 550.

To sum up, here are a few bullet points:

• the 50-75% increase in Style II-decorated objects is a huge rise
• the animal interlace develops later than our hitherto-known Kentish sequence
• the Hoard may allow improvement of the later stages of the Anglian sequence
• with one possible exception (the late 6th-century silver pommel) everything seems later than 600 AD,   but not later than 700 AD (in stylistic terms, nothing is as late as the Lindisfarne Gospels)
• there is a possible re-introduction of old stylistic motifs – visible in the Style II and in the embossed foils
• the motifs have close links to various items from Sutton Hoo Mound 1
• there also seems to be continuous contact with the Scandinavian world
• the embossed foils add a number of new or different details to this iconographically important group

The number of animal interlace pommels and mounts has increased so much that the view that one group of them is essentially Kentish may now be questionable. I suggested in 1999 that the dismembering of the animal character of this group was a chronological phenomenon, but it may alternatively (or to some degree) be a regional variant.

On the other hand, if the Staffordshire Hoard objects really do belong to the Kentish (or southern) tradition of Animal Style II, the increased number of objects clearly suggests that a new analysis is necessary to understand the chronological as well as the regional development of this version of Anglo-Saxon Style II. In addition, the Anglian (or northern) objects need to be re-analysed as the new finds may add to - or even alter - our perspective of stylistic development in this area.

It would also be possible to seriate the Staffordshire Hoard finds as an assemblage by themselves, using correspondence analysis. On the one hand little chronological development may be revealed, as the period covered by the artefacts is rather short; but on the other hand, it would be easily done during the analysis of the finds.

The Staffordshire Hoard provides us with an invaluable source of new information. The embossed foils are an important contribution to the iconography of the sixth and seventh centuries, and allow a comparative analysis between Anglo-Saxon, Scandinavian and Continental motifs. As for Style II, there are now so many artefacts from Anglo-Saxon England that comparative analysis of Style II within Britain, as well as between Britain, Scandinavia and the Continent, will be much more useful. The metalwork will now be able to be studied without turning to illuminated Gospels (or even putative, now-vanished manuscripts); we can now discuss chronology, regional variation, the diversity or uniformity of the iconography, and the meaning of the images on a much better basis.

References

• Åberg, A., 1943-47. The Occident and the Orient in the Art of the Seventh Century (Wahlström & Widstrand, Stockholm)
• Bruce-Mitford, R., 1983. The Sutton Hoo Ship Burial vol. 3 (British Museum Press, London)
• Høilund Nielsen, K., 1999. ‘Style II and the Anglo-Saxon élite’, Anglo-Saxon Studies in Archaeology and History 10, 185-202)
• Stolpe, H., and Arne, T. J., 1912. Graffältet vid Vendel (K.L. Beckmans Boktryckeri, Stockholm)
• Stolpe, H., and Arne, T. J., 1927. La Nécropole de Vendel (Kungl. Vitterhets historie och antikvitets akademien, Stockholm)