11.3.3 Evolution of Pattern
||We still have three questions to deal
The last questions merits its own
sub-chapter. I will deal with questions No. 9 and 10 right here.
- The history of the making of pattern
welded swords. Who made the first ones when and where? How did the technology
develop and spread?
- The history of the discovery of the
pattern welded swords. How many have been dug out when and where. Where are
they now? What kind of investigations took place? Why are the Danes so pissed
about the Nydam treasure in Schleswig?
- Most important: The metallurgy of the
pattern welded sword. What do we know and what do we guess?
||I have put a lot of emphasize on the
pattern welded swords found in Danish bogs. Almost all of them were from Roman
work shops that were supposedly located somewhere in South Germany / France or
along the river Rhine. The "Romans" who made them might have been
Romans in the same way as the guys who produced the atomic bomb or rockets were
Americans. The "Romans" in South Germany used to be members of Celtic
tribes before they were coerced to become Romans. Later (after 200 AD, say)
they gradually turned into Alemanni (more or less forcefully) before they
became (very forcefully) Merovingians and then members of the Carolingian
A number of bog-swords from around 300 AD where made with pattern welding
skills that could not be improved upon then or now - provided you work with
bloomery iron / steel and a manually wielded hammer. So let's consider 300 AD
as the point in time when pattern welding technology climaxed.
at the history of pattern welding we thus need to ask two questions:
- What happened before 300 AD? How did pattern welding evolve?
- What happened after 300 AD? How did pattern welding go on and eventually
||Besides the history of pattern
welding, there is also the history of discovering the pattern welding of old. It is one
thing to find an old pattern welded blade, and quite another thing to recognize
it for what it is. The basic concept of twisted striped rods is not so obvious
if you (and everybody else you know) has never heard of that. I will give that
topic a quick look, too.
Pattern Welding Evolved Before 300 AD
||How did pattern welding evolve? Who
knows. There are no written records and what has been dug out of bogs and
graves does not mirror the distribution of pattern welded swords 2000 years
ago. There are far more Roman pattern welded swords in museums in Northern
Europe than in Italy or Southern Europe because the barbarians up there
sacrificed a lot of swords in bogs where iron was preserved and they put swords into the graves of the warriors.
The Romans did neither. We need to go for educated guesses based on the few
things we do know.
It is helpful to realize that here is a continuous transition from
structural piling to pattern welding of complex patterns. Since all
cultures working with bloomery iron had to do some
piling for making
products, it was unavoidable that:
Eventually striped rods became standard, producing not only a pleasing pattern
but preventing sudden and complete fracture, especially in
cold weather, for
all swords employing phosphorous steel for the harder parts. Many in-between
situations are imaginable and that means there is no definite beginning of
pattern welding. The Celtic (La Tène) swords bear witness to this. Some
swords had their iron / steel parts piled in such a way that a pattern might
have resulted, possibly without the smith being aware of that. Contrariwise,
another smith might have attempted to produce a certain pattern and failed.
- Structural and compositional piling
developed, e.g. by using hard steel on the outside of a blade for the edges.
The Roman "Thames" sword
from the 1st / 2nd century AD offers a good example for a thoughtful way of
structural and compositional piling that did not produce a pattern. It was essentially made from
one broad striped rod.
- Random patterns were produced on
occasion and must have made people aware of the aesthetic potential of
At least one Celtic smith making a "La Tène sword" around 300 BC (No. 12 in
the picture accessed by this link)
produced a complex pattern welded sword that looks suspiciously like one of the
the more complex pattern welded swords made about 600 years later. Far be it
from me to doubt Pleiner's analysis of this sword. But this sword just doesn't
make sense. It is possible that a particular talented La Tène smith was
500 years ahead of the rest, it's just not very likely.
Then again, in 2015 (after I had written this module), two Celtic swords (with
anthropoid hilts) and pattern welded blades turned up in an auction catalogue!
The pattern resulted from simple striped rods that might have been employed for
increased fracture toughness but it sure looks like a pattern was clearly
intended. Here is a
detailed picture of one blade and
here are the
descriptions from the auction catalogue.
Yet more amazing is the celtic blade from the middle La Tène period (2nd
century BC) that came up somewhat later in an auction in April 2016. It
definitely employs two very well made twisted stripes rods.
Here is a picture and
the description from the auction house.
||Let's see what seems to be definitely
known as we ascend upwards in time:
|Sword Data; Time /
||Pattern / Structure
|Middle Latène (250 BC - 150
- Unknow location
- Unknow location
|Structural piling; face welding
- 2 striped rods (and
anthropoid iron hilt)
- 2 twisted striped
First indications for twisting. Not in sci. literature
Lang & Ager
- Antique trade
- Antique trade
|Late Latène (150 BC - 0)
- Heiligenstein / Speyer (Germany)
- Llyn Cerrig Bach/ Anglesey (England)
|Three striped rods all through the blade. -
Lang & Ager
Sword with full pattern welding
- Cuvi (Italy)
|First known example with twisted striped rods and
complex structure in the sci. literature.
This is the fully pattern welded No 12
sword of Pleiner
|Latène swords found in
- Wachock / Ilza (Poland)
- Saône / France
- Museum Rouen / France
- Port Switzerland
|Some pattern welding.
|No Roman pattern welded swords from the 1st century AD.
||As stated by:
|However, from the 1st half of the 1st century AD
Pugios with stripe patterns from
- Munich, Germany
- Leuven, Nijmegen, Velsen (Netherlands);
- Mainz (Germany)
| Most Pugios, however, are not pattern welded. The ones that are seem to have
pattern welded stripes only for decoration.
Typically the "classic" 4/3 iron/steel layer package was used.
|Spathae, around 200 AD
- around Oslo; (Norway) with Victoria/ Mars incrustations
- South Shields / Durham (England)
- Lauriacum / Linz (Austria) spatha
|Fully pattern welded
- 5/4 low/high phosphorous steel plates; some twisting.
- 2 counter twisted rods.
- 6 torsion strips on both sides!
|Around 300 AD
Danish bog spatha
|Climax of pattern welding technique
|500 AD - 1000 AD
|Percentage of pattern-welded blades rose
dramatically after » 500 AD, peaked to
» 100 % during the 7th century, and fell again
during the 9th / 10th centuries.
After 800 AD swords with
as far as investigated by
Lang & Ager
||That is not particularly
illuminating. But what, exactly, is it that we want to know? It is quite
simple. We want to know how the following three techniques originated and
How, where and when? From the above one might guess that pattern welding
evolved in Celtic regions north of the Alpes. But we simply know too little in
general, and next to nothing about the use of faggoting in particular. And let
me say it once more: If low / high phosphorous steel was used for the bright /
dark parts, no clear pattern could result without first faggoting, i.e.
homogenizing the phosphorous concentration, since phosphorous is always
distributed rather inhomogeneously in the "raw" iron /steel. And
phosphorous steel was used a lot!
- The making of striped rods; including
the selection of "bright and "dark" steel.
- The twisting / grinding of the striped
- Faggoting of the two materials before making a striped rod.
||Did you appreciate that I have hidden
a contradiction in terms in the statements above? No? I thought so. Let me
explain: If you are able to select the right steel grades and you can do faggoting, you do not need to go into
pattern welding at all. You could make a superior all-steel swords right away.
That pattern welding nevertheless dominated sword making (but not sax making)
for about 3 centuries then tells us that:
The compromise was the full-steel sword with a pattern welded
"veneer" that we find so often around and after 300 AD.
- either the conditions above were not fully met, then pattern welding might
have made sense,
- or they were met and pattern welding made no sense from an utility point of
view - but people liked to have "pretty" swords.
||It is not clear to me when the first
striped rod was made. The "Hermann Historica" sword shown
here definitely shows
two striped rods running the length of the blade. A (better preserved) second
sword looks like it also incorporates striped rods but the catalogue picture
are not conclusive.
||Unfortunately no provenance is given
for these swords. They are supposed to be Celtic and from the middle La
Tène period (3rd century BC). While there can be no doubt about the
Celtic origin of these remarkable swords, the dating is not beyond reproach
since Celtic swords with an anthropoid
hilts are typically assigned to the late La Tène (1st century BC)
We thus can be sure that Celtic smiths understood and used pattern welding with
striped rods as early as about 100 BC, possibly earlier. We don't know,
however, if Celtic smiths invented pattern welding with all that implies or if
they imported the technology from somewhere else. While I'm not aware of
pattern welding being practiced outside of Europe in the first few centuries
before the common time, that doesn't mean much. We may not have discoverd the
relevant artifacts or we may not have given them proper attention. Having said
all this, I nevertheless like to state:
Pattern welding was invented by
Celtic smiths sometime between
300 BC and 100 BC.
||The Celts used the technology, and so
did the Romans. Around 300 AD pattern welding was routine and "Roman"
blades with incredibly complex strucures were routinely made. However, we do
not know much about the time between 100 BC and 300 AD.
||Let's stop here and look at the
second point from above: What happened to
pattern welding after 300 AD?
Pattern Welding Matured Between 300 AD
- 600 AD
||Good headline - I just don't know
much about the topic. All I could figure out so far is that "late"
pattern welded swords were found in graves all over (Northern) Europe. I have
no idea about statistics (percentage of graves with swords, percentage of
pattern welded swords, where and when) and thus will give you only a few
numbers from one tiny area for the beginning:
|Area of "Sindelfingen swords"
||Sindelfingen is a town about 15 km
south-east of Stuttgart, sporting a large Mercedes plant. It is likely to be an
settlement like all towns in South-West Germany with names ending on
"-ingen". The Allemani graves in the Sindelfingen area are by no
means special; there are plenty similar ones all over South Germany. I only
focus on them because Dorothee Ade compiled a detailed analysis of what has been
found in the Sindelfingen area in her 2012 PhD thesis. Most of the stuff is
from ancient cemeteries, some from single graves.
Starting about 1860 people payed some attention to artifacts from old times
other then treasure that were occasionally dug up, and some of the stuff ended
up in museums. Nowadays the location of several ancient cemeteries are known
but the graves are left alone for future generations of archaeologists. We
already know of another Alemannic cemetery in
Pleidelsheim, just to the
North in the map above, and unexpected finds are made all the time, too (see
||I'm talking mostly about Alemanni
graves here. The Alemanni,
as pointed out
before, coming from the North (and East), crossed the Roman Limes, the
fortified boundary "wall", in 260 AD and occupied parts of what is
Germany, Switzerleand and France. After the official end of the Roman
empire in the second half of the 5th century, the Alemanni were more or less
dominated by Ostrogoths in the East or the Frankish Merowingians in the West.
The Alemanni were nevertheless a culture that has left deep traces until their
nobility was wiped out for good at the "blood court at Cannstatt" (Blutgericht zu Cannstatt) in 746. What happened was that
Carloman, the eldest son of
Charles Martel, Charlemagne's grandfather,
invited all nobles of the Alemanni to a council at Cannstatt (close to
Stuttgart). Carloman arrested the several thousand noblemen who had followed
his invitation, accused them of taking part in the uprising of Theudebald, Duke
of Alamannia and Odilo, Duke of Bavaria, and summarily executed them all for
high treason. The action eliminated virtually the entire tribal leadership of
the Alemanni and ended the independence of the duchy of Alamannia, after which
it was ruled by Frankish dukes. Things like that happen when you neglect to
invade Gaul (now
called "France") from time to time!
||Anyway, the Alemanni buried their
mighty ones with their weapons, and that included pattern welded swords. Here
is an especially interesting burial in Niederstotzingen from around 600:
|Alamanni grave in Niederstotzingen / South
Germany (about 600 ±)
Large picture plus a
picture of the real things
|Source: Old drawing shown in many places,
e.g. here or in
warriors, each fully equipped with a big sword, a sax, and many smaller items.
And one of those warriors was - surprise! - a woman as a relatively recent DNA
analysis proved 2).
Here is what those swords might have looked like:
||You don't want to see
drawings but the real thing? OK - here are some of the newest ones:
| Arian (left) und Tarik Wissendaner show what
in 2012 while perusing an Altingen construction site.
|Source: Local newspapers
||Altingen is on the edge of the circled area in the map above.
While South-German museums are full of (pattern welded) Alemanni swords, there
seem to be very few pictures around. Manfred Sachse showed a
good one in his
book, and this link hub
leads to a few more.
I have also started to take pictures myself;
here is one (large
scale) example. Even more pictures can be found
in this link
||So what do we know about the Alemanni
swords found in the "Sindelfingen" area? Here is a list of what was
found in just a few places in the area investigated:
||500 - 525
||500 - 525
||550 - 600
||450 - 500
|Sums / Averages
||450 - 600
||Even if half the graves
were for (unarmed) females, it is clear that not all men and women had a sword.
But whoever had a sword was very likely also in possession of a sax. Some,
however, had only a sax. By the way, the town names ending in "ingen"
signal Alemannic origin. In contrast, the ending "Heim" (=home)
signal Merovingian dominance. From some towns it is known that the name changed
after the Merivigians took over, e.g. from Hessingen to Hessigheim.
||Pretty much all of
the swords in the table above were pattern welded. The first blades without
pattern welding were from the beginning of the 7th century and their number
increases towards the end of the 7th century. As far as one can tell,
exclusively twisted striped rods were used for pattern welding, producing
simple herring bone patterns (like in the picture above or
here) or more complex
ones like in the "Ingersheim sword"
(almost in the area discussed above). No complex
chevron or palmette
patterns have been found.
So much for the larger Sindelfingen area. But what about the rest of Europe?
And for times after about 600 AD, when Sindelfingen ends? Tall questions with
very short answers:
- As far as sword blades are concerned,
the technology appears to have been about the same everywhere in (Northern)
Europe. Pattern welding was the standard. However, I have yet to see blades
that surpass the 350± Nydam / Illerup blades. Around 600 AD the first
sword blades that are not pattern welded but made by laminating pieces of
supposedly faggoted steel appear in some areas (like South Germany). In other
areas (like England) that transition might have happened somewhat later.
Pattern welding, however, never went quite out of style for another 400 or 500
- We need to be a bit careful about the statement above. A corroded pattern
welded blade might just appear not to be
pattern welded. Look a the one right below or the
ones in the link. Was
pattern welding involved in the one below? It doesn't look like it but only
X-ray pictures or metallographic investigations will tell for sure. Indeed,
more recent X-ray
investigations did produce evidence that pattern welding was used for
blades that did not show it. Contrariwise, a blade with a clearly visible
pattern could have been made from solid steel while the pattern was only
encrusted, inlayed or "damascened" into the surface of the blade.
- In contrast, while there are some pattern welded saxes, the majority of that "utility
weapon" seems to have been made without pattern welding. To what extent
piling and faggoting was used I can't tell.
- What had happened with respect to making pattern welded blades between -
very roughly - 400 AD and 700 AD was and is totally eclipsed by what had
happened to the hilt. That was true then and is true now. There
are far more papers in the present literature about hilt details than about
blades for that time span.
- In other words: For the top-of-the line "show-off" swords the
hilt became the distinctive part; it also must have been far more expensive
than the blade. That doesn't mean that the blade was not top; it just means it
wasn't used for fighting. You don't use your Porsche, Mercedes or Bentley
four-wheel drive luxury off-road car to actually go off-road either. For menial
jobs you have other cars, not to mention people, for God's sake. But your
Porsche would be up to the job! We just don't have all that many fighting
swords from that period, not to mention that warriors might have used a sax for
||In other words: As far as steel
technology and pattern welding is concerned, the technology was established and
didn't change much for several centuries. That the focus shifted to the hilt
was then to be expected. Basic car technology with respect to the
"mechanics" of cars became an established technology around 1980.
Gone are the times when young men (like me and my buddies) discussed at length
and with a lot of emotions the relative advantages and shortcomings of
suspension systems (Längslenker, Querlenker, Schräglenker,
Raumlenker... there are no English terms because the American car industry has
yet to discover those techniques). Nowadays the sound system of your car is far
more interesting than the suspension system.
In other words:
|Gold hilt spatha, Stotzingen, Germany;
|Source: Left: Ellwangen / Germany museum
Right: "Photobucket", Matt Bunker's (medicusmatt) picture
||Is that a pattern welded blade? It
doesn't look like one but one must be careful in making such a judgement. It
was certainly a sword for showing off. Gold was and is in short supply in
Northern Europe, and the little that was panned around the Rhine must have been
very precious. I doubt very much, however, that a sword with such a hilt would
have been any good in a fight. It has no pommel, must be badly balanced, and
easily slips out of your hand.
||Gold hilt spatha
were in use from about 450 AD 490 AD and are seen by some as derivatives
of Romano-Byzantine designs. Alemanni mercenaries, serving in the army of the
Byzantine empire, might have brought them home as a kind of memorials and
show-off piece. after they This link,
this one, and
in particular this one gives a few more
pictures. King Childeric's sword
from about 480 AD embodies the culmination of the gold hilt sword and also
leads over to the lavish use of
almandine (red garnet)
and gold adornments in the time span from (roughly) 500 AD - 700 AD.
The gold hilt swords could be seen as the first step into the new fashion of
emphasizing the hilt. This is not so interesting to us here (we are into steel,
not gold, after all), so I only give you a (long and lavishly illustrated)
special module on the hilt business.
development, however, is good for something:
dating swords! Hilt
fashions changed more quickly with time than blade fashion, just look at the
years before about 300
AD. Certain types then are seen as typical for certain times and areas.
Fancy hilts, containing gold, silver or bronze, corrode far less than iron or
organic materials and often these parts are about all that has been left.
The reference for hilt fashions is still the
of Elis Behmer. He
sifted through much of what has been found in connections with swords, in
particular the metal pieces left from hilts and scabbards.
Pommels, cross-guards, chapes, lockets and so on in later "migrations
period" years. Behmer's successor, up to a point, is
Wlifried Menghin with his book "Das Schwert
im frühen Mittelalter" (Swords during the Early Middle Age).
It's all in the special module. We now look at:
History of Discovering the Pattern Welded Swords of the 1st Millennium
||People must have unearthed 1st
millennium swords all the time without looking for them. Farmers unearth
things, and so do builders, peat diggers and children. But interest in old
things other than treasures was not large before about 1750, when serious
archaeology started by dedicated digging in
Herculaneum. It took another 100 years before old rusty iron found some
interest, and yet another 100 years needed to go by before the science of iron
and steel had developed to a point where one could try to figure out what old
iron / steel artifacts could tell us.
Let's play the old game and ask:
Who first noticed that he was
at the result of pattern welded twisted
striped rods when he contemplated an
old sword? When was
||I certainly don't know. I have sifted
through 100+ old papers but couldn't figure out who exactly had made the first
decisive step in the right direction. But as we shall see, it will be
sufficient to look a the work of just a few prominent old researchers to arrive
at a surprising insight:
Something went spectacularly wrong!
||Let's start with
Conrad Engelhardt and the Danish bogs.
Systematic digging there started in 1859 - 1863, look up the
special modules for
details. Pattern welded swords were found and recognized as such by Engelhardt.
So what did Engelhardt state with respect to pattern welding?
published his findings
in 1886 (in English) and referred to the swords unearthed in Nydam as being
"damascened". This was quite natural if unfortunate. Now let's see
process of "damascening" in 1886.
|"The Nydam swords are of
iron, long, straight, and two-edged; the blades are for the most part - ninety
out of a hundred - richly damascened in various patterns, and afford good
illustrations of the poet's sword, " the costliest of irons, with twisted
hilt, and variegated like a snake" (Beowulf). Iron wires, arranged in patterns,
have been laid in grooves made in the surface of the blade, and then the whole
has been welded together, so that the surface must originally have been smooth.
That we now see the patterns raised is probably owing to unequal oxidation.
Among the many elegant and ingenious patterns represented on
Plates VI and
VII, l would call
special attention to Fig. 5 and 5a, with borders of flowers freely rendered in
twisted iron wire."
||That is a description of "tauschieren", and there is no proper English word
with exactly that meaning; words like inlaying, incrustation come close. The
proper word would be "damascening" but that is now misleading since
it was misused as another word for "pattern welding" (mostly by
Germans). Whatever - we note that Engelhardt in 1886 was obviously not aware of
the twisted striped rod technique.
||Now lets move up to 1939 and look at
what Behmer has to say. He studied the literature
quite carefully, and if the "twisted striped rod technique" was
generally known by then, he would most certainly refer to it.
||Here is the decisive paragraph from
|"Im Zusammenhang mit den
Blutrinnen tritt so gut wie stets Damaszierung auf. In jeder Rinne sind dann
auf dem Grunde längs der ganzen Klinge lange, parallele, einander
kreuzende, gebogene oder wirbelförmig gewundene Drähte, vermutlich
aus Stahl, eingelegt, die in das Schmiedeeisen der Klinge eingehämmert
sind. Klingen, die keine Blutrinnen aufweisen, sind oft in ähnlicher Weise
Der Zweck dieser unechten" Damaszierung, die von der
morgenländischen echten" Damaszierung, mit der das Abendland
erst in der Zeit der Kreuzzüge bekannt wurde, wohl zu unterscheiden ist,
dürfte der gewesen sein, durch Einhämmern von Stahlstreifen in ein
weicheres Material dieses stärker und elastischer zu machen."
|"In connection with fullers
we find pretty much always damascening. Each fuller groove contains hammered-in
wires down its whole length, probably made from steel, running in parallel or
wavy, crossing each other or forming eddies. Blades without fullers are often
damascened in a similar way.
The purpose of this "untrue" damascene, that one
should strictly distinguish from the "true" damascene, that was only
known to the West after the crusades, might have been to render the soft
material more strong and elastic by hammering in steel wires"
||54 years went by since Engelhardts
publication, and nothing has changed! It is mostly but not completley pure BS, mind you, since some swords
have been "patterned" this way.
||It was Herbert Maryon in 1940 who coined the term
welding" and, as far as I can tell, had almost the right ideas
about the striped rod twisting.
Next, in 1954, the year Germany won the soccer championship for the first time,
J.W. Anstee and L. Biek published "A study in pattern welding"
3) and related their results of
attempts to replicate an old pattern welded sword by forging. Those were
probably the first attempts in "experimental archaeology" with
respect to pattern welding. They did use a kind of twisted striped rod and
experienced big problems with fire welding. They did not only use striped rods
but also wires. The idea was to fill the grooves of the screw-like spiral
obtained after twisting with a wire to obtain a smoother surface in order to
The essential picture of the paper tells it all:
|Anstee's and and Biek's geometries
||We do not need to go into details. It
is clear that Anstee and and Biek came close but weren't quite there yet. Here
are a couple of interesting quotes:
- "Alternate strips of carbon-free and low carbon
metal have been thought necessary to produce such a pattern, and up to 0.6% of
carbon has been reported in some material. As the present work shows, however,
patterns are obtained even with strips of the same, virtually carbon-free,
The role of phosphorous iron in pattern welding was not known by the authors.
Here we probably have the source of the myth that patterns could be produced
with one kind of steel. That a pattern was obtained probably relates to their
problems with welding:
- "Some of the difficulties inherent in this
natural and apparently simple method of construction were revealed when a
number of such rods were made in the Laboratory. No welding was then possible;
the twisted rods were merely flattened, soft-soldered and ground away to
varying levels to reveal the consequent changes in pattern."
What that means is that their weld seams, if "taking" at all, were
likely full of oxides and thus could have shown a "pattern".
- "Furthermore, the herringbone and other patterns
observed, either on the surface or (on X-radiographs) within this type of
blade, could evidently be produced by (the remains of) composite rods, each
pile-forged from thin strips of iron, and adjacent ones twist-welded in
That is the correct idea.
- "The answer to the problem of scale removal was
now clear. As during the work the oxide layer became thicker (and duller) it
was forced from all surfaces by the twisting stress. The best solution was to
allow the scale to form, twist a little, then tap the hot bundle with the
hammer. In this manner the strips were finally screwed up on to themselves, the
whole length being traversed in short sections altogether four times-twice up
and down. Even so, much fine scale was left in the joints; nevertheless, all
internal welds there appeared to be perfect, except at one or two points where
lumpy particles had been trapped."
- So twisting makes welding easier!?
||I do not know when finally everything was right - must
have been around 1960. I do know something else, however:
The twisted striped rod technique was
an industrial standard in 1850 and earlier!
||I have dealt with that
already - not so long ago. All of the guys named above must have seen pattern
welded contemporary objects like guns for hunting and military (dress) swords
in their time. In the larger German-speaking areas (including Denmark, etc.)
these pattern welded objects were called damascened and were standard issue stuff. While many
smiths or companies employed the technique, mass production of twisted striped
rod stuff took place in Belgium.
||How Engelhardt and Co. could miss
that is beyond me and that's why I say that something went spectacularly wrong.
However, they were men of the pen and not the sword and therefore might have
become early victims of the "damascene" confusion, believing that the
"damascened" gun barrels and swords they must have seen were made by
"tauschieren"=inlaying, encrusting; just as described by Engelhardt
© H. Föll (Iron, Steel and Swords script)