Nothing is more synonymous with Anglo-Saxon archaeology and history than the fabulous treasures which were uncovered in 1939 at Sutton Hoo, yet there is nothing “typical” or “representative” about the vast majority of this truly remarkable assemblage.  The enormous highly decorated shield, for example, integrating hundreds of components, lacks any particularly appropriate English parallels, and its closest comparators come from similarly impressive burials from the roughly contemporaneous Vendel Culture in Eastern Sweden.  So remarkable and “exotic” is the Sutton Hoo Mound 1 (SHM1) shield, that the still foremost study of Early Anglo-Saxon Shields; Dickinson & Härke (1992) all but excluded it from consideration. Anglo-Saxon shields with even modest metallic decoration (such as appliques or boss apex discs hardly visible except at very close quarters) are exquisitely rare and occur mainly from the mid 6th century; it is unclear to what extent this reflects a decline in the fashion or is simply an already rare item among furnished burials becoming invisible with the decline of the rite itself. Variation in the lavishness of shield ornamentation could be a battlefield indicator of rank and status, though most fittings are too small to have been particularly visible in this context, and although it is possible to sort the limited sample of decorated shields into “ranks” of embellishment (Dickinson (2005) and Mortimer (2011)) there is a danger of implying false near-equivalence between the only “top rank” shield of SHM1, and “second rank” shields (the most elaborate from the more commonly regarded “English sample”) such as 6th century Bidford-182 (see our reconstruction and article here) and the Tranmer House Shield (Sutton Hoo Grave 868), which are, in terms of crude count of components, at least 20 times less elaborate. Given extremely limited organic preservation, painted designs, wooden or leather appliques might be feasible, but the picture from grave archaeology is that, despite being the largest, most conspicuous “display surface”, even a high-status Anglo-Saxon warrior’s shield would have looked quite plain.

​continental influence) territories like Mercia,  but, at least according to waning grave archaeology, these decades saw a gradual shift from the classic “insular” styles of Anglo-Saxon shield bosses with their low domes and often broad apical buttons, to the tall “sugarloaf” style already popular on the continent, which would (at least according to limited iconography) dominate well into the late Anglo-Saxon period. Our ability to thoroughly examine this phase of shield evolution is limited, for it coincides with the twilight of the furnished burial rite. The still most comprehensive (though very much out-of-date) study on these later shields, Evison (1963) suffers from a degree of conflation between insular and true sugarloaf (later designated Dickinson Type 7 / Hines SB5) bosses, and importantly, its chronology is skewed (relative to modern understanding) having been written prior to the pivotal re-dating of the SHM1 burial and its gear, which had knock-on impacts on comparative dating of all other Anglo-Saxon archaeology. 

​The true, fully developed sugarloaf shield boss form (Evison’s “tall curved cone”, Hines SB5-b/c) appears to have gradually emerged via, or in tandem with a short-lived phase of tall but relatively straight-coned bosses (still within Dickinson’s Type 7, or Hines SB5) with relatively tall, straight, sloping walls and limited overhanging carination (Evison’s “Tall straight cone”, or Hines SB5-a), in the second to third quarter of the 7th century. Notable examples of this form come from Wenden and Melbourn (Cambs), Croydon (Surrey), Faversham (Kent), West Knoyle Farm Stourton (Wilts), a particularly tall-walled example from Bury St Edmunds (Suffolk), and (at least according to Evison) somewhat lower (transitional?) examples from Salisbury Race-Course, Coombe Bisset (Wilts) and the famous 16th grave of the cemetery at Mount Pleasant, Alton (Hants).  With a relatively smooth continuity from “low” to “tall” bosses, sorting transitional examples (as noted by Evison) is an arbitrary business; a perhaps defining characteristic of tall sugarloaf bosses (as opposed to low insular bosses) given significant weight in the Hines (2013) analysis appears to be the number and shape of rivets, as sugarloaf bosses (including the tall-straight form) tend to be attached with small dome-headed rivets, often greater in number than the typical 4-5 larger disc-headed rivets of earlier boss types. Of course, especially given evidence for re-fitting of bosses, sorting and dating shield-bosses based mainly on the shape of their rivets could be unwise. 

​Importantly, the Coombe Bisset example was found with a garnet and shell cloisonné pyramid fitting (so-called “sword pyramid”) and the Alton case was found alongside a sword with cocked-hat pommel and hourglass grip collars, and the famous silver-gilt, garnet, and gold-filigree decorated buckle. The Bury St Edmunds and Melbourn examples were found with bronze buckles with folded back-plates secured by three small rivets (Bayliss and Hines BU7); a fashion of the 7th century found elsewhere in SHM17, and the Staffordshire Hoard.  Indeed, of any, this relatively rare type of shield boss shows the greatest co-occurrence with items similar to those in the Hoard. 
​We therefore chose the “tall straight cone” shield-boss for this shield, taking the boss from Melbourn, Cambridgeshire, as our exemplar. As with previous shield projects, we turned to Jason Green of Wieland Forge to produce the boss for us.

Another difficult aspect was the size of the board. The size of Anglo-Saxon shields is a contentious issue (discussed here), particularly due to misinterpretation of the otherwise invaluable data outlined in Dickinson and Harke (1992) concerning “maximum” and “minimum” shield sizes. The enormous SHM1 shield was around 92 cm in diameter (in line with ranges for the Vendel Culture analogues 0.84 to 1.1m), while the next largest – Ford 2 (with sugarloaf boss, and seax with silver fittings) was 90cm across; however the remainder of shields in Dickinson and Harke (1992) with probable board dimensions are considerably smaller than this, with the median of Anglo-Saxon grave shield max diameters (n=112) approximately 65cm.  That said it is noted that shield boards appear to increase in size towards the end of the early period (ie. Into the 7th century) and therefore a board on the upper end of the evidenced range would be reasonable.  From our own experience with replica shields, and our own size we have found 70-80 cm to be a respectable size for an average sized male, and (with one eye on the fittings we planned for it to comfortably frame) chose to make our new shield 77cm – not so enormous as the Sutton Hoo or Ford examples, but still well on the upper end of the evidenced range. 

Another critical detail was the thickness and shape of the board. The SHM1 shield board was slightly dished at the edges, and was thinner at its edges than at the strap-holder (half way out from the boss), while another shield from Pewsey (grave 34) was at least 2 mm thinner at the edge than at the boss (from 7 to 5mm). The careful carving of these boards in this manner may be for weight management, with thinner edges both reducing overall weight, but also, if carved off the front edge, subtly shifting the centre of mass of the shield closer to the grip, and bringing the weight closer to the grip improving manoeuvrability.  With the SHM1 shield, its slightly dished shape would place the grip-point actually forward of much of the board, thereby helping the carrier manage its considerable weight. 
Our shield would not be so large, but following these precedents we carved down approximately the outer third of the board, on the front only, to an almost blade-thin edge which, after application of skin product, would reach approx. 5mm in total.

We also took care to slightly recess the shield-boss into the front of the board. If this was done, it would be almost impossible to detect archaeologically, but is one of the possible solutions for the problem identified by Dickinson and Härke (1992) of the sloping flanges of Anglo-Saxon bosses, which if mounted flat, leave a gap beneath. By carving a complimentary recess into the board prior to application of the skin-product layers, a tight fit is guaranteed, as well as a pleasingly smooth surface on the front of the board, from the iron flange to the board cover. This also brings the heaviest element of the shield – the iron boss (and therefore, along with it, the shield’s centre of mass) a couple of millimetres closer to the grip, making the shield less cumbersome. 

​To both sides we applied 1.5-2mm thick veg tan bovine leather.  Although Dickinson and Härke (1992) note some cases where only one side of the shield was covered, the majority of shields seem to have had skin product on both sides, which, presumably held on by a sturdy (yet archaeologically fugitive) glue, would form a 3-layer composite greatly increasing strength and durability.  It is not clear to what extent skin product on shields was tanned, though the Sutton Hoo shield is described as being covered with “vegetable tanned cattle hide”.  Modern veg-tan leather is a reasonable analogue, easily wet moulded around the now 3D shape of the board, and working well with our historic glue of choice; hide glue (this oldest of glues made from the collagen of animal connective tissues, would be indistinguishable from the collagen present in the skin product). Modern veg-tan is, however, typically pale and anaemic compared to more tannin-saturated historic leather of other periods and therefore not representative of the aesthetic likely seen on Anglo-Saxon shields. We use a modern dye blend on such items carefully chosen to mimic the warm hues of tannin-saturated historic leathers. Although this is an attempt to simulate the (probable) look of their tanning process, dyed leather was not unknown to the 7th-8th century Anglo-Saxons, as evidenced by the binding of the famous Stonyhurst / Cuthbert Gospel, of rich red-coloured tanned goatskin (probably imported).    We followed this with much polishing, using our own blend of pure beeswax and raw neatsfoot oil (rendered fat from cattle) – both substances available to the Anglo-Saxons, which serve to condition, waterproof and provide shine to the leather surface. 
​

Grips for Anglo-Saxon shields were formed in various ways; most commonly a piece of wood was lap-jointed in some manner (most commonly off-centre) spanning a pre-cut hole in the board with the joints reinforced with rivets; the iron element which typically survives, attached by those same rivets, is merely to reinforce the structure. Our own preference, however, is for an approach Dickinson termed “Type D”; forming the grip from the board wood, or rather, cutting two unequal lunate holes rather than one circular one, with the (off-centre) bridge between forming the basis of the grip. This approach is evidenced, accounting for approx 20% of cases, and is associated with slightly higher burial wealth among a sample which excludes Sutton Hoo (Dickinson and Harke, 1992). This grip was carved such as to be rounded on the front face to fill the hand, and wrapped tightly with a strip of leather; an approach seen on various finds including, importantly, preserved texture of binding on the grip reinforcer with the tall-straight sugarloaf boss from Melbourn. Shield grips were often off-centre, as this produces a more ergonomic and comfortable result; more space is needed for the back of the hand than for the fingertips (for archaeological examples see Dickinson & Harke, 1992). 

The boss was installed with eight small dome-headed rivets, shined up to contrast with the forge-blued surface of the boss. Although far plainer than the Sutton Hoo shield boss, or more typically “Anglo-Saxon” bosses from the 6th century, we were keen not to spoil Jason’s precise work by embellishing it beyond what is evidenced for its type. The Melbourn boss, according to one old report, had silver rivets, though this has not been confirmed, and the contrasting iron seen here is a small but conservative gesture towards that aesthetic. 
​

​Given the size of the board we decided on 12 decorated clips; four with rectangular plates (in this case made by pressblech) with Salin Style-2 interlace designs of a kind represented among the countless foil fragments from both Sutton Hoo and the Staffordshire Hoard, four with wyrm-heads with garnet cabochon eyes borrowed from one of the two designs seen on the Sutton Hoo shield (produced for us by George Easton / danegeld.co.uk) and four adapted from the so-called “horse head” terminals from the Staffordshire Hoard (STH 678 et al). ​ ​  These terminals, though commonly referred to as "horses", could in fact be a different style of dragon-representation or even boar - the latter typically only distinguishable from other beast terminals by the presence of tusks.  If they do indeed represent horses, they would be the first  representation of this animal on Anglo-Saxon war-gear outside of pressblech foils. We have discussed the significance of horses in this context previously, here (link).

​Previous interpretations of this item (such as a replia shield produced early-on for exhibitions at Tamworth Castle) have placed it with the fish upwards, and birds divided by a vertical line of symmetry, yet when placed horizontally it appears not as two birds, but a single bird catching a fish at the water’s surface, with the reflection of both below the waterline. An additional pair of tiny stylised birds facing the same direction at the rear of the fish could be conceptualised as reflections of the bird in the turbulence created by the fish's fins. It is interesting to note that this one piece, if indeed shield ornamentation, spans and brings together the two best represented classes of zoomorphic shield appliques; those (like Mill Hill, Shelford, or SHM1) which depict birds of prey, and those (like Sheffields Hill, Barnes, or Eriswell) which represent fish. These motifs meet, as separate pieces, on the Tranmer House cemetery shield (Sutton Hoo Grave 868). The juxtaposition of birds and fish on shields, it has been suggested, represent binary concepts; sky and water, above and below; life and death. This is arguably brought into its sharpest relief with this unique piece from the Hoard with its vivid depiction of the waterline as veil between worlds.
​
In this context the bird; probably a white-tailed sea eagle (Earn) one of the three Wodanic beasts of battle, pulls its catch through the boundary between worlds, from its living world to its death; perhaps symbolising the harvesting of heroes by Woden. The placement of this object on the original shield board was, perhaps, a symbolic reminder that in battle, the veil between the world of the living and the dead was the shield itself. 

​While the meaning of such items would normally be a matter of pure speculation, the Hoard analysts (publication pending) have discovered that the head of the fish was deliberately and precisely broken off in antiquity, perhaps to “dis-empower” the motif. If so, this would support Tania Dickinson’s thesis that animal art on shields represented symbols of superstitious / apotropaic protection for their wielders (Dickinson, 2005).
​
George produced a wonderful replica of this piece for us, and with it still falling into the category of “bird fitting” we, following the precedent set by the SHM1 shield, placed it on the right side, with the birds facing inward. 

This shield, at least in terms of physical size, represents our largest "treasure reconstruction" to date, and is the culmination of months of work, and two years of collaboration with our associate craftspeople.

Integrating a total of 10 individual zoomorphic elements and a bronze rim, but a conservative mid 7th century boss and grip, this build is, we hope, representative of the kind of shield that may have been carried into battle by the warrior elites who wore the Staffordshire Hoard.
​

Bayliss, A., 2017. Anglo-Saxon graves and grave goods of the 6th and 7th centuries AD: a chronological framework. Routledge.
Bruce, S.G., 2015. The St. Cuthbert Gospel: Studies on the Insular Manuscript of the Gospel of John (BL, Additional MS 89000).
Dickinson, T.M. and Härke, H., 1992. Early Anglo-Saxon Shields (Vol. 110). London: Society of Antiquaries of London.
Dickinson, T.M., 2005. Symbols of protection: The significance of animal-ornamented shields in early Anglo-Saxon England. Medieval Archaeology, 49(1), pp.109-163.
Ellis Davidson, H.R. Gods and Myths of Northern Europe
Evison, V.I., 1963. Sugar-loaf shield bosses. The Antiquaries Journal, 43(1), pp.38-96. 
Fern, C. 2014. "Dramatic golden fish and eagle: A link to Anglo Saxon kings?", [Online: Video]. History West Midlands, IDM Media production. Youtube.  [URL: https://www.youtube.com/watch?v=QsYzrmO3jn8 ]​
John Hines., 2013. Anglo-Saxon Graves and Grave Goods of the 6th and 7th Centuries AD: A Chronological Framework [data-set]. York: Archaeology Data Service [distributor] https://doi.org/10.5284/1018290
Mortimer, P., 2011. Woden’s Warriors. Warriors and Warfare in 6th-7th Century Northern.
Stephenson, I.P., 2002. The Anglo-Saxon Shield. Tempus.

A wide range of shades, from white through creams to greys and browns, and even black were achievable simply by means of sorting sheep fleece before spinning, or by different retting techniques and sun-bleaching of plant fibres like flax, hemp and nettle-fibre.   
In the Anglo-Saxon period there seems to have been an increase in access to purer white wools, and with it came expansion of potential for bright colours from dyeing using techniques which first began to be practised in North Europe in the late Iron Age.

Many native plants can be used for dyeing yet relatively few specific plant dyes appear in the archaeology; most famously, indigo blue from woad (Isatis tinctoria), luteolin yellow from weld (Reseda luteola) or dyers greenweed (Genista tinctoria), and reddish colours (alazarin) from the roots of various bedstraws including most famously, dyers madder (Rubia tinctorum).   The latter, though popular among the Romans, is peculiarly all but absent from early Anglo-Saxon archaeology, returning tentatively in the 7th century, usually in expensive trim. Other dyes (tannins, lilacs and purples from lichen varieties, and others) were in use, but many secondary colours are possible simply by overdyeing these three principle plant dyes, but also through modification using alternative mordants and additives.  

Plant-dyed cloth has a particular quality in terms of look and lustre that is hard to simulate or even adequately describe, and there is therefore more than just an intangible payoff for living historians pursuing truly plant-dyed costumes. It is further, false to assume that "plant dyed" means "dull", "soft" or "pastel" tones, as the results of successful plant-dyeing can be blindingly bright, though often fading with time. ​ The affinity of light, natural de-greased wool for dyes is quite staggering, and only enhanced by mordants. Plant fibres, on the other hand, do not have a strong affinity for dyes, and at best, plant dyes produce softer hues on them. To what extent the Anglo-Saxons used plant dyes on the likes of linen is a matter of debate; there is next to no evidence of dyed linen in the archaeology, though it could be argued this is an artefact of preservation, with pigments more likely to be retained (to be detected by scientific analysis) in wool, over 1000 years in the earth, compared to linen. 
​
​Over a number of years, our team (members Lindsey and Æd in particular) have been exploring plant-dyeing, experimenting on various scales, with various dyes and recipes, drawing on the advise of a number of more experienced experts. We are particularly indebted to the Mulberry Dyer, Deb Bamford, for her guidance and expertise.    Plant dyeing has a satisfying "alchemical" quality, often involving mysterious colour changes, odd ingredients, and not uncomplicated recipes. 

One of the most ancient, and most famous dye plants, with evocative "magical" associations, woad - a relative of cabbage - works totally differently to other dyes, as the dye chemical from the plant is wholly insoluble. Crushing and simmering of the leaves will simply release the blue powder from the leaves, which will collect in the pan (and can, patiently, be dried, if desired, to be mixed with binding agents to form a paint! Such a paint was used, among other places, in the detailing on the cover of the Stonyhurst / Cuthbert Gospel). The challenge with woad, therefore, is to find a way to dissolve it so that it can soak into the fabric, and here, to the rescue, comes some surprisingly advanced chemistry.
​

What is required is a reducing agent; a chemical which will strip the oxygen from the dye molecules. Through this, the indigotin (insoluble) is converted to indoxyl, which is soluble and can soak into the fabric, although this will rapidly turn back to insoluble indigotin on contact with air. Through this process, temporarily soluble dye can enter the network of fibres which comprise the cloth or yarn, but afterwards transform to insoluble powder within its matrix and, for the most part, cannot wash back out. But what reducing agent? As with so many ancient and medieval crafts and industries, the answer was urine. Specifically the ammonia which is produced by its fermentation. This foul business is the reason dye industries tended, along with tanning, to be banished to the outskirts of medieval and later settlements. 

Though not knowing the actual chemistry underpinning the process​, generations of experimentation and the sharing of expertise would have led to an intuitive understanding of how to achieve good results, and it is interesting to speculate how the process by which these particular leafy greens and stale urine, when associated, can be used for dyeing. Traditional herbalism holds that woad leaves can be used as wound-dressings, and as one of the most ancient "cleaning agents" stale urine may have been used by ancient people to clean wounds. If both were applied together, ancient people may have found that their skin magically turned blue as the dressing was taken off, and through this, would have become aware of this mechanism of dyeing. 

​In practice, at least on a small (bucket with lid) scale, our experience of woad-dyeing with urine was frustrating as well as foul. The fabric would come out of the green vat vaguely pale green, and turn pale blue on contact with air, but weak activity of the reducing agent often meant that it would be wholly used-up reacting with air introduced along with the cloth or yarn, and take a day or two to regenerate.

A bolt of fine diamond-twill wool, of a kind frequently preserved on the back of early Anglo-Saxon brooches, was the starting point for a 6th century Anglo-Saxon dress, which was then dipped multiple times into an active woad vat; entering white, absorbing the yellow to green dye, and then spectacularly converting to brilliant blue on contact with air.  Additional, shorter dips (long enough to allow more dye to soak in, but not so long that the dye already locked in would be attacked by the reducing agent) allowed the colour to be made deeper and stronger, and to even out some patchiness. Pale and dark patches occur as the vat typically will have "blooms" of concentrated dye near the surface, while, additionally, in any one dip, it may be difficult to ensure the whole bolt of fabric is fully submerged without trapping bubbles.  Dyeing "in the yarn", however, would mean any variation in shade would even out once it was woven, and it may be that this approach predominated. 
​
We finished the dress with brightly coloured 6th century Snartemo-V-style tabletweave in madder red, weld yellow, and woad-weld green. ​

We were keen to produce another garment that displayed the full potential of madder. The dye in madder (alazarin), like most other plant dyes (but not like woad!) is soluble; steeping the chopped madder roots will release the dye into solution, rather like making tea, and this "tea" will easily soak in (and wash out!) of the fabric. What is required, then, is another chemical to help trap the dye; a mordant. Much is made of the etymology of mordants (related to the French mordere- to bite) but this is unhelpful, for, unless something has gone very wrong, no chemical should "bite" into the fibres. Rather, mordants are soluble (and polar) chemicals which will easily soak into fabric, but have a tendency to "clump" together with dye molecules when they are mixed. The resulting clumps, chains and blobs ("dye complexes") of linked dye and mordant molecules, if formed while inside the network of fibres which make up the yarn or cloth, will be entangled, too big to easily wash out. Thus the dye will stay in place. Because they sit linked up with the dyes, some mordants can modify the resulting colour. 

As previously noted madder-dye is rare in the early Anglo-Saxon archaeological record, and is gradually re-introduced during the 7th century. Its scarcity may at least partly be accounted for by the fact that madder dyeing (more-so than the extremely strong yellow dyes) is particularly reliant on expensive mordants and as the pigment's colour changes with pH, acidic mordants such as natural tannins are not a good choice. 

Wary of piercing this fine fabric, tiny loops in the tablet-weave were integrated as it was stitched on, which provide attachment points for a brooch without pushing a pin through the weave.

Both garments demonstrate the vivid colours available to our ancestors. Having been dyed "in the cloth" the colour is uniform rather than alternating, with the weave pattern only visible as texture. Though it is likely the majority of dyeing was done "in the wool/yarn", colour-patterning so often seen on Anglo-Saxon living historians' costumes is actually rare in the Anglo-Saxon archaeological record, and hardly ever occurs, even in better preserved late Anglo-Saxon textiles.  The look of this wool, with its subtle diamond pattern but single colour, is therefore arguably more representative of these finds. 

References:

Owen-Crocker, G.R., 2004. Dress in Anglo-Saxon England. Boydell Press.

Rogers, P.W., 2007. Cloth and clothing in early Anglo-Saxon England, AD 450-700 (No. 145). Council for British Archeology.

The Mulberry Dyer. [Online] https://www.mulberrydyer.co.uk/
 ​

The inspiration for this project was the magnificent lyre from Grave 58, Trossingen, Baden-Wuttenberg, South-West Germany, uncovered as part of emergency excavations in winter 2001/2.   During these excavations 12 graves of this late Migration-Age presumed "Alemannic" cemetery were uncovered, of which Grave 58 was the most spectacular; a princely chamber-grave with organic grave-goods remarkably well preserved due to the waterlogging and uniquely heavy clay subsoil. 

Grave 58 contained a man aged around 40 years old who, thanks to dendro-dating of the chamber timbers, we know was buried in the late summer of the year 580 CE.  Strontium isotope analysis revealed he was local to the Trossingen area. He was 1.78m tall and his skeleton showed little signs of physical labour. He lay in a frame bed constructed from beautifully turned beech posts, furnished with red and yellow woollen blankets, and was surrounded by many other sophisticated grave goods, including candle-holders, a turned round table, an elaborate throne, wooden canteen, various bowls, a 3.6m long spear, a bow, a shield of alder, a riding crop and remains of a saddle. A comb lay by his head, a sword (with silver inlaid hilt) lay at his right arm, and at his left was the most perfectly preserved warrior lyre so far discovered.  The overall picture is of a high status princely burial that must inevitably be compared to the likes of Sutton Hoo, Prittlewell and Taplow, though with fewer examples of jewellery, and, instead, a great array of wonderful organic items. Trossingen 58 creates the tantalising possibility that other contemporaneous princely graves may have been as, of not more lavishly furnished with beautifully made organic items (in addition to their preserved metallic finds) which have simply not survived.
​

​
​Prior to the discovery of the Trossingen lyre, it is fair to say that our understanding of Germanic warrior lyres was largely based on ill-preserved fragmentary remains from various contexts, preserved but isolated components, manuscript depictions, conservatism and sensible conjecture.

​Being so wonderfully preserved, Trossingen provides the first evidence for previously unknown features including sound-holes, and elaborate surface decoration beyond anything we could have dared imagine prior to its discovery. Its entire surface was covered with decoration, including friezes of animal interlace, and processions of warriors carrying shields and spears. 
​

​Our reconstruction, produced over approximately 6 weeks, is overall faithful to the original, though differing in some aspects to account for material limitations.  The first lyre Æd had ever made, it was produced following detailed instructions and advice from personal communication with experienced lyre-maker Michael J King (www.michaelking.com).

Given difficulty sourcing a large enough piece of Acer campestre wood for the body, we instead used a piece of English lime, having excellent properties for carving and lacking the characteristic "flame" seen on sycamore maple which could obscure the decoration. For the soundboard we sourced a minimally flamed sycamore tonewood (Acer pseudoplatanus - again an appropriate substitute for the original's field maple) though it wasn't quite long enough to allow the sound board to extend all the way up the pillars on the original. We therefore terminated the sound board a few inches lower on the pillars, in a fashion consistent with the design of the remarkably similar contemporaneous lyre from Oberflacht. 

The body of the lyre was cut, the sound-box carved using chisels, scraped out, and then, importantly, the profile of the body was gradually tapered down to be considerably thinner at the at the string arch than at the base, following closely the dimensions of the original.   The sound-board was carefully shaped and sanded, and positioned using ash (Fraxinus excelsior) positioning-pegs as on the original, and finally installed using hot animal-glue.   After a week of drying the outside of the lyre was shaped and sanded. 
The original lyre showed various signs of repair in antiquity, including non-original iron pins used to re-fix the sound board to the body. We chose not to include these so as to give an impression of the lyre when it was first made.  Similarly the original lyre had an unusual mixture of ash and hazel pegs, of different design, and it is suggested that some of these were replacements.  We chose to incorporate a matching set of ash pegs, corresponding to the spoon-shaped design of the original's ash set, whittled by hand and knife.  
The bridge was hand-carved, as the original, out of a piece of willow (Salix spp.).

​Recreating the elaborate surface decoration took a substantial part of the time spent on this build, and was undertaken closely following the instructions of lyre-maker Michael J King. This unusual technique - "Kolrosing" - is essentially a means of 'tattooing' the wood, leaving decoration which is totally smooth to the touch and does not remove material or alter the surfaces of the instrument, thereby not affecting its sound.
The technique involves finely sanding the surface of the lyre, then incising the design into the surface using a scalpel, in the form of many shallow vertical cuts, removing no material in the process. Charcoal is brushed into these cuts, and the surface is then sealed with an oil or wax and sanded over. 

As this is the only example of such a warrior lyre where the surface of the body was preserved, it opens up the possibility that other lyres were similarly decorated with this technique, although we will never know for certain. The evocative processions of shield-carrying warriors, all with unique faces, reaching for a huge spear, may be a reference to some story or myth in the owner's oral tradition, and may have served as a visual aid for storytelling in the mead-halls of the 6th century.   The decoration on the reverse of the lyre, however - particularly the hugely complex knot work of snakes, must surely be simply an aesthetic indulgence. 

​"Dreamgifu" - meaning "gift of music" in Old English, we hope, helps to illustrate the wonderful craftsmanship that those in what we call the "Early Anglo-Saxon period" applied not just in their now much-celebrated metalwork, but also in organic materials. We hope it also highlights the importance of music and performance in Migration-Age warrior culture. It could not have been a more perfect find to recreate for Connor - combining three of his greatest loves; historic music, tattoos, and snakes. 

The earliest and most abundant examples are narrow bands with simple patterns, from female graves in Kent and the Isle of Wight. These are almost always found near the head, and may have been used as glistening headbands for securing veils. More spectacular are the remains from the famous Taplow princely burial - the second most lavish burial found from the Anglo-Saxon period, which contained, alongside shields, feasting gear and sword, over 2.5m of exquisite, often very wide, finely woven metallic gold brocade.

Due to the shortcomings of the 19th century excavators of the burial, the position of the fine strands of gold found in the burial cannot be stated with any confidence, and as such, we cannot say for certain if the huge amount of gold brocade formed the decoration of a cloak or other garment, or perhaps a sword harness or belt.  The famous gold Taplow buckle had adhesions of both leather and tablet-weave-structure textile on its back, so it may have been associated with the brocade.

To illustrate what such a belt may have looked like, and as part of ongoing efforts to recreate and "bring to life" princely treasures from the 7th century, in 2016 we embarked on a project to build a gold brocaded belt inspired by the Taplow burial. The project began with team member Aed learning the gold brocade technique, first experimenting with a narrow band based on one from Chessel Down, Isle of Wight (in blue, at the top) before moving on to the more complex band from Taplow. The pattern we chose to reproduce is known as the "narrow band" for it is the narrowest element identified from the burial, although it is still wider and more complex than any gold brocade from any other early Anglo-Saxon burial.  The replica band took over 150 hours of high-concentration work.

 Upon completion it was sewn, using the same threads, to a piece of prepared, stained and burnished veg-tan leather, and the buckle and retaining loops were installed, such that the belt could be clasped without the buckle being covered, or the delicate brocading needing to pass through the buckle or be strained. 

Rather than the Taplow buckle (a replica of which we had used on a previous build) we chose to incorporate an exquisite replica of the contemporaneous and similarly styled buckle from Alton, Hampshire, produced by Ganderwick Creations. 

The final result was more subtle than we expected; the gold foil woven through the belt is quite subtle until light hits it at just the right angle and flares, revealing the pattern vividly against the ground weave. It looks particularly good glistening in firelight, and it may have been this that attracted our ancestors to the technique. After all, while coloured tabletweave will appear dull in the dim light of the meadhall, gold brocade glistens and sparkles in light cast by dancing flames.  What better assemblage to, along with replicas of finds such as the Staffordshire Hoard, illustrate that the 7th century really was the "Age of Gold"?

Beginning in 2014, we embarked on a project to envision what a complete, kingly-status, large broadseax from the late 7th century would have looked like, drawing on elements from the corpus of English broadseax burials, and taking inspiration from the later phase material from the Staffordshire Hoard.

Our associate specialist weapon-smith Paul Binns (of Paul Binns Swords) produced a blade of extraordinary beauty and breathtaking complexity. This has a slim band of ‘pattern-welding’ sitting below two deep parallel grooves, themselves bearing subtle ribbons of patternation, and a graceful wave-like pattern interface between the blade-core and the carbon-steel cutting edge.
The blade is 48cm long, 4cm wide and 8mm thick at the spine edge. Longer than is typical for a light broad-seax, and a little more slender than is typical for a heavy-broad-seax, this blade arguably falls into the category of “atypical broad-seaxes”; a more elegant form which arguably foreshadowed the advent of the more familiar and fearsome longseaxes of the 8th-10th centuries.
Paul, an artist of iron, relishes interesting projects and the opportunity to push his skills to new limits. I am sure he would not mind me saying that the blade he produced for us was, to a great extent, far more elaborate and impressive than we had really anticipated, and necessitated us re-thinking our original plans in order that the other elements; hilt and sheath, would live up to the magnificence of the blade. It is a nervous and sometimes agonising business, attempting to add to another craftsman's masterpiece, and we only hope we have done it justice.

​Gilt bronze hilt collars, decorated with filigree work, were custom-made for the blade by our associate historic jeweller George Easton, taking inspiration from a number of items from the Staffordshire Hoard. The hilt was capped with a gilt bronze replica of the pommel-cap STH-K347 made for us specially by Mr Easton for this project.
Although some broad-seaxes, notably those from Northolt Manor and Purton have very long grips of at least 24cm in length, most vary between 10 and 20cm. Taking into account the proportions and balance of the blade we settled on a hilt length of 19cm, with the actual grip (excluding the jewelled elements) measuring 9.5cm, producing a grip comfortably wide enough for one hand. Because of the size of the grip collars, the hilt was made wasp-waisted to some degree with an offset central widening to promote a comfortable grip. The haft was carved in sections from walnut using spacers of black cow-horn.
The result is, expectedly for such a weapon, blade-heavy, but nevertheless comfortable in the hand. Such weapons, and their even more substantial brothers – heavy broadseaxes – seem to have been optimised for heavy chopping blows, and this is no exception.

The thick veg-tanned leather was tooled to a conservative and well evidenced design, featuring lines delineating the shape and fullers of the blade contained within, with other tooled elements having been designed in a manner informed by, and sympathetic to the metallic elements of the sheath. It was lined with fine animal hide to further protect the blade (-an approach more commonly associated with wood-cored sheaths, but nevertheless evidenced having augmented the core-free sheath of the seax from Buttermarket G2297). The sheath was closed using a myriad of tiny copper-alloy pins on the suspension-flap following the style of Buttermarket G1306, and likewise, an L-shaped copper-alloy fitting of U-cross-section was added to the suspension-flap at the mouth of the sheath. Further decoration came from 7 rivets interspersed with 5 larger decorative copper-ally buttons 16mm in diameter showing what at first sight looks to be an abstract chip-carved design but is actually a devolved Salin Style-II picture of what I take to be a raven. Such large decorative rivets are seen on the sheaths of continental large seaxes but are not unknown in England, being found in Buttermarket and the smith’s grave at Tattershall-Thorpe, which was dated to the late 7th century. The buttons and rivets used here, though not precise replicas of either find, mirror their styles and arrangement closely and together, we hope, are highly representative of these overtly Frankish-influenced, highly decorative and wealth-displaying sheaths.

The suspension system relies on two iron three-way strap-dividers and utilises two of the large brass studs to secure the vertical straps. Adjustment is via a small iron buckle on the anterior strap and a larger iron buckle on the breast strap of the baldric. This system produces very comfortable carriage of this big seax with no damaging torsion being placed on the straps.

This is a speculative weapon reconstruction but the sheath and suspension have been kept very disciplined and within what is well-evidenced.

The name we have given to this ‘atypischer Breitsax’ is ‘Átorþorn’, which means ‘poison-thorn’ in Old English. The Anglo-Saxon term ‘þorn’ actually means ‘a sharp point on a stem’ and has always had the associated idea of its causing pain. In Norse poetry the term was sometimes used to describe a jötun. Our blade has the classic seax twist-welded pattern now described as ‘herringbone’ or ‘Anna’ but which was known in Anglo-Saxon times as ‘átertánum fáh’. This meant ‘poison-twig pattern’ and was clearly a kenning for the venomous adder whose markings it resembles. We particularly like the wave-like appearance of the area of the blade which marks where the hard steel edge meets blade core. It is now known that good quality seax-blades did sometimes show this interface which some have likened to the hamon on a Japanese sword.

Having taken well over two years of planning and research, this project has, to paraphrase Winston Churchill, taken copious quantities of ‘blood, toil, tears and sweat’. We hope the end result has been a replica seax and sheath worthy of a late c7th king of Mercia, and brought to life what such a weapon assemblage might have looked like at the time of its making.

Drawing on research for the previous reconstruction (see discussion here), work began with fashioning the board.  Once again, 10-year seasoned ash planks sourced from trees felled less than 14 miles from Bidford were used. While ash sits among less commonly used timbers for Anglo-Saxon shield construction, evidence  from the Tranmer House cemetery suggests that this timber was used preferentially (and even imported from neighboring regions) for high status shield boards, perhaps due to its outstanding toughness, split-resistance and durability compared to wood such as willow, poplar or maple more commonly used for shields.  The toughness of ash surely posed a challenge for shield-makers, being highly resistant to carving, and it is this fact which most likely made other, weaker but more easily worked timbers more widely used.

Planks for the board were laboriously thinned to within a few millimeters of the target thickness, glued together with hot hide-glue, then re-thinned and painstakingly sanded by hand, before being cut to shape and sanded further, with board thickness declining gradually towards the edge which was itself carved thin on the front-face, producing a subtly very slightly convex outer face (yet with flat reverse) corresponding to inferences from cases such as the shield from Sutton-Hoo mound 1.    Unlike the previous reconstruction (Bidford-182) this shield was made larger but thinner, measuring approx 5mm thick at the centre - corresponding to the lower end of estimated shield thicknesses inferred from early Anglo-Saxon finds. Given the greater durability of ash compared to other potential shield woods, but also it's increased density, it is argued that shields made of this material may well have been carved thinner than otherwise to normalize weight without strength being compromised due to the excellent properties of the timber. 
 
Leather was added to the front face with hide glue, with a leather edge stitched in place and hand-burnished. The leather face was treated with natural oils and finally, polished with a protective coating of beeswax.  The grip - corresponding to the extant remains of the type 1b grip from the grave, was formed around an ash core and fixed in place by means of small clench-nails.    The boss - produced by our associate smith closely corresponding to that of the original, was rivetted in place along with two disc fittings and a rare lozenge fitting, in an arrangement based upon the positions of these finds relative to the boss in a photograph of the grave in-situ during excavation in 1923 (see below).  Larger, more elaborate lozenge fittings feature heavily on impressive shields from high status burials of the roughly contemporaneous related east-Swedish Vendel-culture, but all extant Anglo-Saxon examples are smaller, plain, and usually single. The fitting from Bidford grave 33, precisely reproduced here, was the largest found in England at the time of publication of Tania Dickinson's influential synthesis of early Anglo-Saxon shield finds in the 1990s, and is the 'type specimen' for such fittings in England.    Although the purpose of such lozenge fittings is not known, and they may have been purely decorative, it is possible they were added to shields to reinforce observed weaknesses or points of damage. Here, the lozenge has been positioned bracing two of the planks. 

The result - the final in this series, we hope provides a representative impression of what a more typical early Anglo-Saxon shield would have looked and felt like based where possible on evidence from the particular grave in question, and otherwise from understanding of early Anglo-Saxon shields more broadly. It is interesting to note that, despite being of considerably greater diameter than our previous reconstruction of the shield from grave 182 and constructed of precisely the same materials, the modest reduction in board thickness (through reduced thickness of both wood and leather) has led this larger shield to weigh slightly less than its brother.   Members of the public are often surprised by the weight of combat shields, but this reconstruction might demonstrate that Anglo-Saxon shields were perhaps not quite so heavy after all !

The modern town of Cannock sits in an interesting position in the heart of the Anglo-Saxon territory of Mercia, on the important Roman road of Watling Street, and only a few miles from the find-site of the Staffordshire Hoard.  Although the oldest part of the oldest building in the modern town (St Luke's Church) dates to 1100 AD, the parish and its surroundings are steeped in earlier history. 
The Thegns have been proud to be involved, this year, in helping St Luke's Church to communicate and promote Cannock's rich history as part of their 900-year anniversary celebrations. 
As part of this effort, the Thegns were commissioned to produce a special reconstruction to mark the 900 year anniversary. 

It is hoped that this reproduction will raise awareness of the detailed local history of the area dating back to the latter parts of the Anglo-Saxon period.  While the introduction to the charter mentions nearby town names including Bilston, Willenhall, and, significantly, Ogley (Hay) where the famous Staffordshire Hoard was found, the more detailed parish-boundary description for Hatherton contains a wealth of detail about the landscape which local people may still be able to recognise.   Finally, the detailed witness-list (including Æthelred, king of the English, and each archbishop and senior bishop in Anglo-Saxon England at the time) highlights that in the 10th century, this land transfer was sufficiently significant for the most important folk in the land to take an interest in the Cannock area.  

It is fortuitous, to say the least, that our ancestors chose to write their records on a material which if properly prepared and treated well can last over a thousand years, and therefore that these important pieces of local history have not been lost.     There is no reason why this reconstruction might not last equally long, and we hope that, in the keeping of Cannock's central church and oldest historic building, it might serve to help future Cannock residents feel connected with their area's rich heritage. 

Following the success of the previous project to reconstruct the famous 6th century shield from Bidford-on-Avon grave-182 with the kind help of the Shakespeare Birthplace Trust, we were honoured to be asked by the Trust to produce for them a reconstruction of a 6th shield to be shown to visiting school groups, alongside the original finds in their collection vaults, as part of their educational programme in support of teaching for the new national curriculum.   

Although simplifying some aspects of the construction to maximize durability and keep costs for the Trust low, we were pleased to be able to produce a  satisfying result; a highly durable but convincing reconstruction with details linking directly to items in the collection and drawing particular inspiration from Bidford-on-Avon graves 33 and 182. 

A size of approximately 60cm in diameter was chosen; a reasonable size to be handled by pupils in the target agegroup, close to the modal size for 6th century Anglo-Saxon shields, and also fitting comfortably in the upper half of the "medium shield" category (Dickinson and Härke, 1992). Key details of interest which were integrated into the reconstruction included a boss hand forged by our associate smith corresponding to the form of the well preserved boss from grave 33, in the SBT collection, with rivets capped with pure tin achieving a similar look to those of the boss from grave 182, tinned board disc fittings like those of grave 33, an iron lozenge fitting corresponding to the type specimen from grave 33, and a type 1a(i) grip. 

On completion of the project, three members of the Thegns of Mercia were pleased to be warmly welcomed to the Shakespeare Birthplace Trust in Stratford-upon-Avon to present the reconstructed shield to staff in person. It is hoped that this shield, displayed alongside matching finds in the collection, will help to bring the Anglo-Saxon age to life in the minds of visiting school pupils, inspire learning about the period, and encourage pupils to consider various basic aspects of interpretation of archaeology. 

With the kind help of the Shakespeare Birthplace Trust, over 2013-14 we researched the find and attempted a convincing reconstruction with the help of our associate craftspeople, hoping to recreate what the shield may have looked like when new, and with the intention to raise awareness of this important Midlands find. 

The effort began with substantial research; first to track the find down, and then to explore its context. With Bidford-182 sitting in the "borderlands" between the Wessex-influenced upper Thames valley region and the tribal regions that would later form Mercia, it was particularly important to investigate the affinities of the Bidford finds, and the items from grave 182 in particular. Before recreating the artistic elements, it was also worthwhile exploring their meanings. (Our discussion on the cemetery, its finds, affinities, and the meaning of the 182 shield art can be seen here). 

A shield boss closely matching the dimensions of the original was commissioned from our associate smith, while replicas of the decorative gilded copper-alloy fittings were commissioned from our associate historic jeweller. It was necessary to be extremely precise with these items so as to ensure that the gently curving fittings would, in shape, to the sweep of the boss flange. Our jeweler was also able to produce gilded disc fittings corresponding to the reported measurements of the originals. 

The board was constructed by Thegns members Dr Andrew Thompson and Æd Thompson, using 10-year seasoned ash which had been cut from trees felled less than 14 miles from Bidford-on-Avon itself. The board planks were thinned using authentic adzes and chisels to an appropriate thickness, then glued together using hot animal-glue, producing strong bonds between the plank edges. (More detail on the reconstruction here).  With no evidence to infer the size of the board of the Bidford-182 shield, a diameter of 68cm was chosen, taking into account inferences from neighboring finds. 

The board was then covered with oak-tanned leather, fixed in place using hot animal glue, while an edge, again of oak-tanned leather was sewn in place through pre-drilled holes with strong linen thread. The leather was then treated with a beeswax and oil mixture, applied warm and with considerable polishing, to achieve a rich, water-resistant and scratch-resistant finish.  Meanwhile, a grip had been prepared (Dickinson and Harke type 1a(I) ) corresponding closely to the degraded remains of the original. This was fixed in place using soft iron clench-nails (again corresponding to extant remains) atop a wooden component continuous with the boards themselves, then bound with thin calf-skin for added comfort - a technique well evidenced from other 6th century finds.   Although previously skeptical about the security of this rather simple type of shield grip (and preferring the common and more sophisticated type 1b) we found the result to be both surprisingly sturdy and comfortable. 

Following installation of the decorative pieces, the boss was installed using soft iron rivets peened onto roves on the reverse, while, for security, the decorative disc fittings on the board were attached in a similar manner. The shield boss rivets themselves were capped, prior to installation, with tin, as the originals had been, resulting in an impressive juxtaposition of black iron, gold mounts and silver-coloured rivet caps. 

The result, we hope, reflects well what the original shield may have looked like before deposition at some time in the 6th century, with the elements available for study effectively recreated, and conservative choices made with respect to aspects which could not be inferred from the archaeology. 

In April 2014 we conducted a scaled-down simulated Anglo-Saxon cremation, in order to investigate the process, and produce a replica of a fairly typical Anglian cinerary burial as it might be found during excavation. 

To achieve this, we built a small pyre of seasoned oak branches cut to length (simulating round logs), with animal bones placed on top. The pyre took the "criss-cross" form broadly accepted to have been the method for such cremations, and was lit using authentic techniques.      The resulting transformitive roaring flame lasted for approximately two hours, achieving temperatures far in excess of a typical campfire. It was interesting to note that this arrangement encouraged inward, rather than outward collapse, such that un-burned timbers, and the bodily remains themselves were deposited further into the heart of the fire as it burned.   Various aspects of this simulation had been carefully balanced; much of the energy in an open cremation is used up boiling off the huge quantity of water in tissues, but, on the other hand, various constituents of the body (particularly high-energy fats) are thought to promote high temperatures in the pyre.  The ratio of fuel to cinerary matter was carefully chosen to go some way to accounting for other issues resulting from the scaling down of the pyre. 

The following morning, ash, coals, and the degraded bone fragments remained. It was interesting to note that,for the most part the bones remained fairly intact, although they had lost most of their mass, becoming light and brittle.  In a full cremation it is likely that falling timbers would often smash these fragile pieces, but the relatively intact state in our simulation goes some way to explain why bone fragments found in real cremation urns survive in identifiable pieces. 

The remains, including ash, coals, and manually crushed degraded bone, were added to a small replica Anglian cinerary urn fashioned after examples from Loveden Hill, Lincolnshire.  Added to the remains were modest grave goods; a small token bone comb and a copper-alloy toilet set.  

The result was a fairly convincing cinerary urn and contents, ready to be buried at educational events and exhumed by plucky budding archaeologists.    Although simulated inhumations are becoming a familiar feature at educational events, it is very much worth emphasizing the importance of cinerary burials too. Such burials dominate in early cemeteries in Anglian regions, and although the goods associated with this rite are usually less impressive than are found in inhumations, the spectacular nature of such send-offs, and the materials necessary for them show that this rite was not the poorer cousin of inhumation, but rather an elaborate expression of wealth, culture and belief among the pre-Christian communities of England in the 5th-7th centuries.