Sunday, 10 June 2012

Læwatan: Staffs Hoard Seax 1

Læwatan; a Wælseax inspired by the Staffordshire Hoard
“þa gen sylf cyninggeweold his gewitte, wællseaxe gebrædbiter ond beaduscearp, þæt he on byrnan wæg; forwrat Wedra helm wyrm on middan.”  
‘The king himself nowtook command of his wits, reached for the slaughter-sax of keenest battle-sharpness, that he carried in his harness;the Geats’ Protector hewed the wyrm in twain’

One of the most striking objects in the Hoard is the solid gold hilt-plate k567. This piece, finely decorated with interwoven Salin Style-II biting beasts, is of a lozenge shape that would be expected to adorn the hilt of a sword, yet its triangular slot clearly indicates it once adorned a single-edged killing knife or "waelseax". Such seax fittings are rare, and it is not clear how such a piece would be integrated into a seax handle-design.

What would such a weapon have looked like, what length would the blade have been, and how would the handle have functioned?
To answer these questions, I embarked on a project to produce a feasible reconstruction of this seax.

K567 Staffordshire Hoard seax hilt-plate replica
The lozenge-shaped lower guard k567 is 7.35cm. long, 2.25cm. wide and 1.1mm. thick. It weighs in at 24.82 grams. (worth around £700 at today's’ prices).
The slot, which would have accommodated the blade, is 37mm long and 4.5mm wide. In the case of lower hilt-plates, the shape of slots corresponds with the cross-section of the blade, and the triangular shape of the slot in k567 indicates a single cutting-edge.
The size of the piece suggests it would have adorned a sizable blade, while the precise dimensions of the slot are consistent with the cross-section dimensions of a "Narrow seax" or (Schmalsaxe type-1 in Schmitt's typology). Within the range for this blade-type, and in proportion to the hilt-plate, a blade length of 35cm was decided upon.

A precise replica of the hilt-plate was produced by George Easton of danegeld.co.uk, while a seax blade to the specifications of the piece was commissioned from Paul Binns (paul-binns-swords.co.uk).

The k567 hilt-plate would certainly not have been fitted to the seax on its own, and precedent set by migration-period swords suggests metal-organic-metal sandwitches would have been favoured for weapon guards. For this reason, it was decided that the k567 replica would be fixed to a complimentary hilt-plate either side of a horn layer -work completed by George. While it is clear k567 would have been the lower-most hilt plate, the upper plate of this sandwich would've accommodated the tang of the blade -rectangular in cross section.

The blade used was a fine example of Paul Binns' magnificent metalwork, displaying the classic herringbone pattern, composed of two twisted bundles, each of 9 layers, forged flat with an iron back and a carbon-steel edge. Such blades are commonly found without any fittings, and the long tangs suggest that, most often, sword-like hilts were not favoured. How, then, could a sword-like lower guard (as implied by k567) be integrated into a seax-hilt design?

We experimented with various hilt-designs using scale drawings. The rather flattened lozenge shape of k567 implies a handle with flattened oval cross-section. Such handles are naturally preferable to more tubular handles, allowing the orientation of the blade to be "felt" in battle.
The long tangs of such seaxes suggest long handles, and while larger examples may have, in some cases, been used two-handedly, seaxes of intermediate size may have had such long handles in order to allow dual use; while a grip close to the blade allows precise cutting, a grip towards the back allows greater hacking force akin to a machete (Mortimer, 2011). It may also be the case that long-handles facilitated attacks involving a loose grip sliding from the guard to the back mid-swing, much like an axe is wielded, that could have maximised hacking power whilst reducing shock to the user.

Working with this theory, a handle should be shaped such that the user's hand can feel what grip-position is being used. In battle, this would allow a user to switch between grip styles depending on the desired attack without much conscious effort.
Having tried various approaches, a slim hourglass-shape was settled upon, expanding toward the lower-guard and upper end. Such a shape allows the use of all grip styles, while preventing the seax falling out of the hand when using a sliding grip. In addition to this, and almost more importantly, such a shape visually integrates the lower guard smoothly into the design.

The handle was formed by carving a block of English walnut (Juglans regia; a material traditionally used for the best gun-stocks) into a shortened version of the target shape. This was sanded, roughly polished, then cut into five sections. Between these sections, disks of flat cow-bone (from pre-prepared sheets available from the invaluable Reenactors Market) were added. Such bone is a wonderful material, resembling modern hard plastic in its working properties, and provides opportunity to highlight our ancestors' frugal use of virtually every part of a carcas. These disks of bone, when in place, would provide additional grip, and visual interest to the handle.
Drilling holes in these parts to accommodate the tang of the blade was a challenge, but was eventually achieved with a rather tight fit. As seax blades tend to be found with the ends un-peened, it is thought that handles would've been held on by friction, so a tight fit of these parts was essential.

To provide visual continuity with the lower guard, it was felt that some atrophied form of upper-guard should be included. Flush with the end of the handle, this piece caps and protects the organic handle from damage. A matching piece of horn to that in the lower-guard was sandwiched between two copper-alloy plates, and drilled with a hole sufficiently tight to clamp the tang and hold the entire assembled handle in place.

The end of an Anglo-Saxon blade's tang would have been much softer than the hard steel of Paul's seax blade, and it would have been possible to peen the end over the upper-guard piece to secure the whole handle in place. In this case, the snug fit of this part was sufficient, but the tang still protruded a few millimetres out of the guard. To hide this, as on their swords, our 6th-7th Century ancestors would have affixed a cnæpp or knob (sometimes referred to erroneously as "pommel" or, marginally better, as a "pommel-cap").

For this seax, a miniature replica of the fascinating 5th-6th Century hoard cnæpp k711 ("Woden's Pommelcap", or "Dave") was commissioned in gilded copper-alloy. Eventually, four gilded pins would be used to fix this onto the lower-guard piece.

After numerous dry runs, the final assembly was undertaken. Some gentle mallet-work was required to made each piece seat properly, and each was added in sequence, fixed to its neighbours using animal-glue. Finally, the "upper-guard" piece was installed, and the cnæpp fixed in place with the gilded pins penetrating into the guard.

Læwatan was now complete, but the importance of good maintenance practises with such pieces cannot be overstated. The blade was cleaned of any finger-marks and dust, and polished using a mixture of beeswax and linseed oil to protect it from corrosion. This step is absolutely crucial for a display item, as beautiful pattern-welded blades attract inquisitive fingers, and the oils and acids naturally secreted by human skin can rather rapidly lead to finger-print shaped areas of surface rust. It is always worth protecting vulnerable steel from such corrosion, and beeswax is particularly useful in this regard, forming a transparent and corrosion-resistant film on the surface. This is a technique our ancestors likely used.

To protect Læwatan, a suitable sheath would be essential. Fabricating a worthy sheath was the next task....


Mortimer, P. (2011) Woden's Warriors. Woden's Warriors: Warfare, Beliefs, Arms and Armour in Northern Europe during the 6th-7th Centuries. 1st edition. Cambridge, Anglo-Saxon Books. 

No comments: