Migration Period Swords and Fancy Hilts & Pommels

What It's All About

Not much seems to be known about Elis Behmer. But his one and only book "Das zweischneidige Schwert der Völkerwanderungszeit" (The double-edged sword of the migration period) is still the standard for our topic here. It was published in 1939, a time when "Germanic" stuff was prominent in Germany (and Scandinavia); the book should be read with this in mind. Behmer's book is apparently his PhD thesis work¹⁾ and contains a huge amount of hard information in the form of (black-and-white) pictures of a large number of objects plus detailed information about their origin and whereabouts.

Behmer established three main groups (A, B and C) for migration period swords, with 4, 3 and 2 subgroups, respectively. In contrast to the classification given here, Behmer does not consider the blades but looks at everything else, in particular the hilt (with pommel, grip, cross guard) and the various metal parts associated with a scabbard (chape, locket, belt attachments etc.).
His classification of the sword from the earlier periods (say 350 AD - 450 AD) is based to a large extent on the Danish bog finds and thus overlaps with the modern systems. However, for the time after 450 AD, his system is still the one everybody seems to use.

Here are a few excerpts. First the group A, subgroup 1 hilts:


Typical group A, subgroup 1 hilts

Behmer sees swords with this kind of hilts (and proper other parts) as the oldest "Germanic" types. They are typical for finds in Danish bogs. The one the left is indeed from Nydam and on display in the Schleswig museum. There are also finds from graves, and according to Behmer this kind of sword was prevalent all over Northern Europe (including England) around 300 AD - 400 AD. Since we know that a least the blades found in Danish bogs were all of Roman make, this would indicate that the Germanic tribes used (pattern welded) Roman blades but attached their own hilts. This is quite possible; there are indications from all eras that one and the same blade may have seen different hilts. It is, however, hard to prove.
Behmer, like many archaeologists, is reluctant to make definite statements. All the numbers given are therefore mostly my interpretation of what he conceals in lengthy prose.

Here is group A, subgroup 2:


Typical group A, subgroup 2 hilts. On the right a new photography of the real thing.

This type is more or less restricted to Denmark, typically found in bogs like Kragehul. In this case Behmer also looks at the blade in some detail. It is typically narrow (4,5 - 5,5 cm) and pattern welded. The blades have fullers - up to six - and go back to the Roman spatha for mounted warriors, says Behmer. The time horizon for this type is 350+ AD

Group A, subgroup 3 is of some interest because it is the first one to sport major amounts of gold (or silver). The hilt is covered with the noble metal and gold cells inlaid with almandine (a red garnets variety) in the cloisonné technique. Here are pictures from Behmer:


Typical group A, subgroup 3 hilts.

The blade is broad and bears no resemblance anymore to Roman types (says Behmer). The blades tend to have one (broad) fuller but "rich damascening is only found on exceptions". I'm not sure if that is true since many of these blades are badly corroded and "damascening" is not obvious and only shows up in X-rays (that Behmer didn't have).
I would tend to put the following sword hilts in the same category, and so does Behmer:

Sword details;
1 Spatha from Gültlingen; 460 AD - 480 AD. Only the showy side of the hilt is gold; the gold is "Rhinegold". Landesmuseum Stuttgart.
2 Spatha from Sindelfingen; rather similar to 1. Table 8 in Dorothee's thesis.
3 Spatha from am 2006 auction (Hermann Historica). Supposedly from 375 AD - 450 AD and Eastern Europe. Estimated at € 30.000 - €45.000.
4 Recently found spatha (plus sax etc.) from Pleidelsheim; ca. 400±. Rather similar to 1 again.
More pictures here

The group A3 swords are the swords of the Alemmani (or Alamanni). These swords spread from their homeland in South Germany to the West, East and North, states Behmer. Well, maybe, but not in the way Behmer describes it. In his reasoning the "Heruli" play a major role, an East Germanic tribe who migrated from Scandinavia to the Black Sea in the third century AD. The Heruli, however, according to more modern insights, were not as important as earlier historians believed.
"Goldgriffspathen " (gold grip spathae) appear around 450 AD – 490 AD and are nowadays seen (by some) as derivatives of a Romano-Byzantine design. The culmination of this kind of sword was found in the grave of Childeric I, who died on 481 / 2 AD. We also have an early lavish use of red almandine gemstones for decoration.


Hilt and parts of scabbard mounting of the sword of Childeric I
More migrations period swords with garnet hilts
Source. All over the Internet

Childeric I (ca. 440 – 481/482) was a king of the Salian Franks, those Franks in the North-East (present day Belgium, North France). He fathered Clovis I, who would unite the Franks and founded the Merovingian dynasty that eventually subdued and wiped out the Alemannis . His sword, however, was evidently imported from the Alemanni. Since it must have been made some time before his death we can date it to 450±.
His tomb, around Tournai, Belgium, was discovered in 1653. The many precious objects it contained were kept in a library in Paris. In 1831 they were stolen and melted down. Only a few pieces survived including the pieces shown above. From the#rest we have only drawings.

I'll skip the peculiar but not so interesting group A, subgroup 4 swords (Eastern type) and continue with group B, subgroup 5. You know these hilts from Nydam and especially Esbjol.


Typical group B, subgroup 5 hilts.
Source: Photography from Schleswig-Holstein Landesmuseum, otherwise Behmer's book.

Those hilts are almost exclusively form Danish bogs; they date to the end of the 4th century (380+) up to about 500 AD. Behmer includes some rather different looking hilts from Norway in the same group; see below. I'm not sure about that.

I could go on like this for quite a while but remember: This is about iron and steel and not about gold and jewels! So for the remaining subgroups I only give you an overview:


The 9 migration sword subgroups of Behmer Large picture
Source: Some Polish sword site; can't give details, sorry.

Congratulations to whoever compiled this picture! Scaling and combining the important pictures with Behmer's often long-winded prose in a meaningful way is a lot of work. As far as I'm concerned, I will only look at one topic in what follows:

I'll skip the subgroups 7 - 9. They are more or less a transition from the "fancy hilt" group 6 to the "Viking sword" that has its own backbone chapter. For the remainder of this module we look into the fascinating if somewhat decadent subgroup 6.

The Pyramid Pommel

The general topic that triggered this module is: "Evolution of Pattern Welding"; we may also call that: "How Pattern Welding Matured and Died". That development might well be symbolized with the well-known "pyramid pommel" often associated with the Merovingians and the Vendel culture. This statement needs to be qualified in what follows, but first let's look at the object thus described


Hilts with a pyramid pommel Large pictures of other ones
Source: Left: Photographed in the Copenhagen museum Right: Found in Tezzo d'Adda; Italy, 625±; from Olga Stroganova's web site

These pommels are typically made from precious metals, often inlaid with (then very precious) almandine (a garnet variety, mostly red but other colors do also exist). They are flimsy and neither well suited for fixing rivets nor for bashing heads in. They are prominent in Wilfried Menghin book "Das Schwert im frühen Mittelalter" (Swords during the Early Middle Age) from 1982.

We have a peculiar pommel shape, more or less reminiscent of a pyramid or actually more of a bicorn hat or cocked hat (the kind Napoleon and Lord Nelson liked to wear). Pyramid pommels also come with a ring attached to one side as shown above. In Behmer's systematic they belong to the large group B, subgroup 5. In less well preserved specimen we see a lot of rivets, one rivet is quite prominent in the right-hand side hilt above.
The first question to consider now is: what are pommels for? Here is the list:

It serves as a solid endpiece on which the end of the tang can be securely fastened by riveting, i.e. by hammering it flat. Riveting the tang in this way fixes all the pieces forming the hilt - at the minimum cross-guard, grip and pommel - securely to the blade.
It provides the necessary weight needed to balance the sword.
It serves as decorative element, signaling the wealth / importance of the sword owner.
It bears insignia (e.g. a cross) that signals membership to certain groups (e.g. a " T" for Templer).
It supplies a bit of magic , e.g. by enclosing relics, being engraved with symbols or runes, or by other means. Suffice it to mention "sword pearls" in this context.
It keeps the hand from slipping. That is quite important if you thrust down, e.g. from horseback.

Quite a list! If you now look at the sword hilts in the picture above, you see:

Hilts belonging to Behmer's subgroups 1, 2 and 5 essentially incorporate points 1 and 6 of the list above, with a bit of point 3. Points 2, 4 and 5 are absent. As far as decorative functions are concerned, the pommel is not much different from the rest.
Eastern swords (Behmer's subgroup 4) are not contained in this system. Their tang is short and does not extend all the way to the pommel. Only the grip is fixed to the tang (I'm not sure how) and the pommel is kind of "nailed" to the wood forming the grip.
Subgroup 5 is different. Points 1 and 2 are pretty much missing but points 3 - 6 are very pronounced. Of course, the tang is still riveted to a part of the many pieces forming the pommel now but that is done in an invisible way "inside" the whole construction.

Let's look at few pictures to illustrate this.


Riveted pommels in Schleswig (Nydam swords)
Source: Photographed at Schleswig-Holstein Landesmuseum, Schleswig, Germany

Those pommels from Behmer's subgroup 1 or from even older times do not supply any substantial weight (they are made from wood, bone or ivory). They certainly prevent your hand from sliding off and they do provide enough material for securely riveting the tang, always with some metal piece (a rivet plate or cap) right on the bone / wood. Here are some rivet plates and caps:


End pieces or caps for riveting the tang; schematic of riveting
Source: Rivet caps photographed at Schleswig-Holstein Landesmuseum, otherwise Behmer's book; my drawing

The two on the lower right-hand side are special: they have a pyramidal shape and belong to early "subgroup 6" swords.
As long as you use these pommel endpieces for riveting the tang by banging it (cold) into a mushroom shape, these pieces needed to be made from solid steel and thus could not be particularly decorative.

The next thing that happens (if we dare to put this into a time-line) was that somebody had a brain-wave and figured out that you can actually rivet your tang to a simple piece of metal like the elliptical ones above, and then put a fancy kind of endpiece on the whole thing, fastened with rivets to the metal plate. The tang wouldn't show any more and you could make (the end piece of ) your pommel as flimsy and decorative as you liked. For reasons of symmetry you used a similar construction for the cross-guard part. Now you could produce a hilt that was more precious than the blade and really announced the status of its bearer:


Vendel period hilt from Välsgarde Another one
Source: from Olga Stroganova's on "Pin It"

You either used gold inlaid with almandine or at least silver or bronze parts, more or less fancily decorated .

"Fancy pommel" science got a mighty kick in 2009 when the so-called "Staffordshire hoard" was discovered in England. It consists of about 3,500 precious items that are mostly related to weapon decoration (nothing for females in there). Most everything dates to 600 AD - 700 AD; a few early pieces go back to about 530 AD, the youngest ones are from 700+. The items must have been intentionally damaged, if only by prying them off by force from a hilt.
Why should one have damaged and buried extremely precious objects around 600 AD in England? We know that the English are notoriously a bit on the weird side but this is wasting money and that's usually where normal weirdness stops. A few lines from "Beowulf" might point in the right directions:

Then the one warrior plundered the other,
Stripping Ongentheow of his iron mail-coat,
His hard-edged hilted sword, and his helm,
Carrying the old man’s armour to Hygelac,
Who accepted the spoils, and pledged fairly,
To share the rewards, and promptly did so:

Later it continues:

They gave to earth the heroes’ treasure,
Gold under gravel, where it lies still,
And as useless, now, to mankind as ever.

So we had looting and burial of treasure. The latter, however, in connection with the burial of a hero and not "just so".

Among the items were 86 "pyramid" pommels; 64 made from gold; the rest silver (17) and bronze (5). 17 of the gold pommels were inlaid with garnets in the cloisonné technique (look it up yourself), the rest were heavily ornamented. But only one sported a ring! (see below). Compare that to the 29 pommels collected from a huge area around Uppåkra, Sweden, and the 5 pommels from the Snösbäck / Sveden ritual deposit. There are many single stray finds, of course, and we have about 400 pyramid pommels altogether. But with 86 pommels from one place (plus all the other objects) there is now plenty of material to study in a known context.


Some pommel end pieces from the Staffordshire hoard Large picture
Source: Medievalists.net – March 19, 2014


Details of Staffordshire pommels Large picture
Source: Internet at large (British museum, Birmingham museum, National Geographic, Current Archaeology, newspapers, ...)

Occasionally, such a pommel comes up in the antiquity trade so you can buy and actually own one. The link shows an example

There is more to these expensive masterpieces of early medieval gold-smithing than meets the eye. The surface of the gold below the garnet is often textured into a regular pattern of little pyramids that reflect the light back - just like the reflectors of your bicycle. This can be seen in the picture below or in the lower left pommel of the picture here.


The textured gold beneath the garnets
Source: National Geographic

Then, following point 5 from above, the fancy pommel may well have some magical / spiritual function, in more than one way. The shape of the gold cells inlaid with garnets, or the decorations otherwise employed, are not always random. They may have meanings:


Front and back of the Hög Edsten Pommelcap
Source: The Thegns of Mercia; UK Midlands-based group dedicated to exploring Anglo-Saxon and 'Viking-Age' history. Sth711: Woden's Pommelcap. © Birmingham museum That's what the source given had to say about this pommel. The actual pommel, however, is on display in the Stockholm archaeological museum.

The design on front and back is subtly different. The marked parts designate a boars head. Yes; I don't see it either but people familiar with early medieval iconography are sure about things like that. Maybe this picture helps:


A real boars head on a Staffordshire hoard pommel juxtaposed to its stylized counterpart
Source left: The Thegns of Mercia; UK Midlands-based group dedicated to exploring Anglo-Saxon and 'Viking-Age' history. © Birmingham museum

Boars are associated with the prominent Norse God Freyr who gets around by riding the shining dwarf-made boar Gullinbursti. Depictions of boars heads on the pommel might have given protection by magic.
Here is another pommel with garnets on one side and a kind of Celtic braid in gold on the other side.

Wilfried Menghin has a lot to say about pommels in general and pyramid pommels in particular. His book is where you look for details. Since it is from 1982, it does not contain the Stafforshire hoard and is thus already a bit outdated. It does still contain a lot of detailed information, though, that is of interest here. For example many maps of where specific things have been found; here is an example for pyramid pommels:


Map of pyramid pommel distribution
Source Slighlty modified from Menghin's book

Menghin also gives timelines for specific shapes: Pyramid pommel shapes and time of occurrence


Pyramid pommel shapes and time of occurrence
Source Modified from Menghin's book

Fancy pyramid pommels thus appeared around 550 AD - just about the time when the Merovigian and Vendel periods started. That is why these fancy hilts are often associated with these cultures.

The Pommel Ring Puzzle

Now you are prepared to appreciate the "pommel ring puzzle ". Some of these hilts come with a ring attached to one side; see above. The puzzle is simple: Why? The answer is simple, too: We don't really know!

We have some idea how this "pommel ring" developed. Some people call it "ring pommel" but that will lead to a mix-up with the those swords where the whole pommel is a ring.
In the beginning, it seems, somebody just used one of the rivets to make a little loop, enclosing a freely moveable ring. Also note the magical sign on one pommel.


Simple pyramid shape pommels with rings
Source: Behmer

Later an extra rivet was used for the ring holder and the rings got more massive. They were often made from gold or silver or at least from bronze, possibly gilded. The rings are more massive and do not move easily anymore


Merovingian spatha sporting a pommel with ring; found in Dizier, France Two more pommel ring hilts
Source: From Olga Stroganova's web site on "Pin It"


Pyramid pommel end with two kinds of garnets and unmovable gold ring
Source: Sword Forum; Kirk Spencer; unclear origin

In the end - around 700 AD - the "rings" aren't even separate pieces anymore but made from one piece of gold or gilded silver / bronze:


Vendel hilt; Valstenarum; Valstenarum find; Historiska museet Stockholm
Source: From Olga Stroganova's web site on "Pin It". Photograph courtesy of M.Bunker

This example gives us a weak clue to what rings were good for. The massive ring on this hilt was added some time after the hilt had been made. Pieces of the wonderful gold - almandine work had to be cut off and damaged to make place for the ring. There is evidence that this has happened with other hilts, too. There is also evidence that a ring was sometimes also removed from a hilt.
So it appears that some sword owner on occasion felt compelled to add a ring to his pommel, which might have been removed again somewhat later. Either by the owner or by whoever took over the sword. Now consider that there were far more fancy-hilt swords without a ring than with one, and you can start to develop your own answer to the "why are there rings" question. Here are some answers from the literature:

Behmer already discussed these suggestions:

The ring held a strap used to tie the sword to the scabbard. This is obviously wrong considering that many "rings" weren't hollow anymore, see above. And the riveted loop or eyelet would have been sufficient for doing this anyway, no additional rings would have been needed.
The ring assembly served as counterweight to balance the blade. That could have been done in simpler ways, not to mention that it wasn't efficient.
The ring assembly just was an ornament. The opposite is true. The effect of the expensive and elaborate ornamentation already there was destroyed. The hilt above would look better without the ring.
The ring supplied some magic. Well - maybe. But why was it then removed on occasion? And why did not every hilt sport a magic ring? If you had the means to acquire a gold - garnet assembly, you certainly could have procured a magic ring, too.

That does not get us very far. So let's look at newer propositions:

The ring is connected to the "peace band". Peace bands are mentioned in Viking sagas and supposedly were tied around the handle of the sword, securing it in place in the scabbard. Untying the peace band then would have been akin to a challenge to a duel or declaring war. That hypothesis is just a variant of No 1 above and very unlikely for the reasons given there. It is far more likely that the peace band thing would have looked like this:


Sword secured with a peace band and adorned with (reconstructed) "sword pyramids"
Source: The Thegns of Mercia; UK Midlands-based group dedicated to exploring Anglo-Saxon and 'Viking-Age' history. The Anglo-Saxon Sword riddles


Sword pyramid from the Staffordshire hoard
Source left: The Thegns of Mercia; UK Midlands-based group dedicated to exploring Anglo-Saxon and 'Viking-Age' history. The Anglo-Saxon Sword riddles. © Birmingham museum

Securing a sword with a peace band as shown also explains in a rather natural way the function of all those "sword pyramids" or other adornments that have been discovered in a number of Anglo-Saxon graves, lying beside sword scabbards. From the way they are constructed they were obviously meant to be attached to a strap of some kind.
But let's go on hypothesizing:

The rings advertises membership to some formal or informal warrior club, like the "club of Nobel prize winners" or the "club of warriors who saved the live of their boss" or the club of the "bearers of the Golden Honor Needle of the Kiel University" (guess who belongs to that club). The ring assembly then was awarded by some leader to a special person. It was just like receiving a medal today.
The rings are "oath rings" symbolizing the linking of persons, just like wedding or engagement rings. You put "the ring" on your pommel after you swore eternal allegiance to your leader. You removed it after you stabbed him in the back.
That seems to be the best explanation to me.

Enough. We simply don't know the exact function of the pommels with rings for for sure. But more and more people (including researchers) employ themselves to solving the riddles of iron steel and swords and new insights are certain to be made in the near future.

¹⁾	On the title page we read: AKADEMISK AVHANDLING SOM MED TILLSTÅND AV HUMANISTISKA FAKULTETEN VID STOCKHOLMS HÖGSKOLA FÖR VINNANDE AV FILOSOFIE DOKTORSGRAD FRAMSTÄLLES TILL OFFENTLIG GRANSKNING I HÖGSKOLANS LÄROSAL D LÖRDAGEN DEN 9 DECEMBER 1939 KL. 10 F. M. I'm rather sure that this means it is a PhD thesis.
²⁾	Svante Fischer and Jean Soulat: "The Typochronology of Sword Pommels From the Staffordshire Hoard. Symposium paper; no reference given (symposium is Greek and translates as "to drink together"). http://finds.org.uk/staffshoardsymposium/papers/svantefischerandjeansoulat

With frame