Overview Luristan Project | Part 1 | Striations and heads |
Part 2 | Pommel and hilt | |
Part 3 | Animals and blade | |
Large Pictures | ||
The Luristan Project - Results from Cut Sword; Part 4 | ||||||||
Discussion of the "Cut Sword" Findings | ||||||||
The Findings | ||||||||
Most everything of interest has already been stated in the three preceding modules. For starters I will therefore give a description of how we envision the sword was made. I base this on our findings and on some straight-forward logical deductions. Further down I will discuss how that ties in with what is known from the literature. | ||||||||
Raw Material | ||||||||
The raw material used for forging the swords must have been some bloom from an 800 BC or so (and thus relatively early) bloomery. Those blooms might have been compacted and thus purified to some extent right after smelting but they still must have contained a lot of slag and possibly some other inclusions. Their carbon content strongly varied (from ferrite / wrought iron to hypereutectoid steel) even so the bulk of the bloom was probably low in carbon. | ||||||||
The only raw material that is known to me from the time period in question (ca.
800 BC – 550 BC) are the Assyrian
"double pyramid bars" found in the palace in Dur Sharrukin (present-day Khorsabad),
the capital of Sargon II from 717- 705 BC. Khorsabad is close to Ninve and thus not all that far from Luristan. In fact, during its largest expansion around 650 BC, the Assyrian empire may have contained parts of Luristan. To quote myself: "Sargon II, who ruled the Assyrians from 722 BC – 705 BC, was in possession of a tremendous treasure of iron that was stored in the palace of his capital Dur Sharrukin, present-day Khorsabad. One Victor Place, resuming excavations started by Paul Botta in 1843, found 160 tons of iron just in storeroom 84. Most of that iron was in the form of bipyramidal bars weighing 4 kg - 20 kg. Metallographic investigations by well-known Radomir Pleiner showed that this iron was rather non-uniform stuff, a mixture of wrought iron, mild steel and hard steel steel, with plenty of slag inclusions.". Fits the Luristan iron like a glove. Just a coincidence? |
||||||||
If an average bloom weighs in at 10 kg, Sargon's treasure involved 16.000 smelts.
Considering that this was just the strategic reserve and that far more iron was most likely out there in the form of weapons
and tools, Sargon's Assyrians must have made and forged iron on an industrial base. This implies that plenty of experience
and highly skilled smiths were around then and certainly also before Sargon's II reign since this kind of industry does
not come into being over night. However, besides the Khorsabad treasure, we have no iron whatsoever from the Assyrians (or Babylonians). In contrast, we have the 100 or so Luristan iron swords (plus a few other iron objects) - but nothing whatsoever is known about smelting and working iron in Luristan. | ||||||||
Starting Material | ||||||||
From the results we can state with certainty that the ancient smiths did not
start to forge a mask sword directly from the bloom. They rather prepared a range of intermediate semi-products as starting
material for making the various parts of the sword. It appears from our results that they first forged thin sheets. The
elongated slag inclusion always show a large length-to-width ratio, just what you would get if you forge a cube into a sheet
that is much longer than the cube and accordingly thinner. Double-pyramid bars by the way, are a good shape for drawing
out thin and long strips as evidenced in this picture
By cutting these single sheets to suitable dimensions followed by stacking them in a way that approaches the desired final shape, fire-welding these layers then allows to produce billets that already assumes the basic shape of the various parts needed. That is particularly attractive for the blade and the hilt; especially if only one piece was used. Fire-welding might have been done by forging the stack of pre-shaped sheets or, if the shape attempted was simple, by stretching and folding, i.e. by faggoting. Here we propose something new that you will not find in the literature. Mind that we do not claim that this technique was used for all mask swords. Mind also that around 800 BC iron technology made a big step forward and that composing blades by layering / faggoting was used by the old Celts as demonstrated by the "sword from Singen" | ||||||||
However - the ancient smiths were not too careful in doing this since the finished
stacks, as shown by all our pictures, contains very bad (many large slag inclusions) and very good iron pieces in a seemingly
arbitrary manner. Maybe they couldn't do better or - maybe once more - they didn't care since the mask swords were definitely
not designed for fighting but only for showing off. Only the outside had to look good, the inside quality didn't matter at all- in stark contrast to a fighting sword!. The outside quality mattered for two reasons.:
| ||||||||
Basic Forging | ||||||||
The artisans making the Luristan mask swords were experts at their trade. They
knew not only how to fire-weld but in particular how to make an object with a complex shape out of a (by now layered and
possibly pre-shaped) billet) of iron / steel. The next step involved forging the various pieces as closely as possible into
their final shapes. This included the rough shaping of the heads and animal figures but also of the pommel plate and the
"rings" around the hilt. Doing that would be a challenge to a modern smith today. It was important that the shape
after forging was as close as possible to the final shape because that made the tricky and laborious next step easier. Forging itself did take place at rather high temperatures followed by relatively fast cooling (water quench?) as evidenced by the Widmanstätten structures and other observations outlined in the preceding parts. That seems to be true for pretty much all swords because all investigations made so far agree on this point. High temperature forging requires a hearth of some sophistication and skilled helpers. There is, however, also evidence for low-temperature forging and even cold forging. Small wonder since the final steps of assembling the sword did need some high temperature but as little as possible. The chisel work for crimping obviously had to be done at room temperature. Then there might have been some short medium temperature annealing after all was done. | ||||||||
Final Shaping of the Parts | ||||||||
The various parts had to be worked into their final shape and that involved:
| ||||||||
Assembling the Parts | ||||||||
Attaching the Rings If you had made the core of the sword (hilt and blade) from one piece of material, I'm rather sure that in a first step the rings were attached. Looking at Luristan iron swords you will find that some rings were just wound tightly around the grip, while others were tightly wound etc. but also crimped into place. The following pictures show this |
||||||||
| ||||||||
| ||||||||
| ||||||||
There is no way to bend a partially elastic wire / ring around the hilt at room temperature in such a way that it snugs closely to the hilt at all places. You need to do that at a raised temperature, in particular if he material consists of hard steel as found in the rings. . | ||||||||
Attaching the pommel plate The pommel plate is kind of riveted on, using the thin extension of the hilt that fits through a hole in the pommel plate. A picture says more than thousand words: | ||||||||
| ||||||||
We guess that at this stage the head / animal figures are not yet attached tot the pommel plate. It is interesting in this context that some mask swords do not have an upper ring, while others have complicated constructions right below the pommel. Here are two examples: | ||||||||
| ||||||||
The upper "ring" in this example (also visible in some of the swords shown in one of the many Luristan modules) is more complex than the regular ones around the hilt, more a kind of decorative band. As the pictures above demonstrate, it takes some tricky crimping to make a fit that looked "wie aus einem Guß" (like from one casting) as we say in the true language. | ||||||||
Attaching the figures Early investigators laughed at the "primitive" crimping technique they encountered when they looked at Luristan swords. They assumed that the old smiths had not yet mastered the art of fire welding for putting iron parts together in a solid way. But crimping was the only option for the ancient smiths! As we have seen, at least some of the smiths could fire-weld quite well. However: | ||||||||
| ||||||||
Details would have burnt off at fire welding temperature, and banging on the figures with a hammer wouldn't have done them much good either. | ||||||||
Now what? There is no other way but crimping (if we discount modern
gluing or soldering) and that, if you think about it, is an exceedingly difficult thing to do. |
||||||||
First you had to produce the fittings or sockets for the figures (or the rings).
You needed to do some precision chiselling, producing the exact outline of the figures to be inserted. You can't do this
with the needed 1/10 mm precision if the the work piece is red.hot so you needed to do some heavy room temperatures deformation.
That is exactly what we see in one of the crimps I claim that you can do this only with soft ferrite / mild steel as a substrate. Hypereutectoid or just eutectoid stuff is hard and brittle and you just can't make those flawless (on the outside) crimping joints we find. Note that you ruin your intended product if the crimp wall you raise with your chisel disattaches, breaks, or gets otherwise disturbed. There is no easy fix if this happens. Then you insert the figures and affix them by "crimping", i.e. hammering the crimp rim over the figure base. This is not something you do on the side; it must have taken a lot of practice to produce the perfect crimping we still see today. |
||||||||
Optimizing the production process Not caring all that much for the "inside", i.e. slag inclusions etc., saves time and money. The same is true for the fitting between the backside of the figures and the body of the sword. You can't see if there is a lot of empty space between the heads / animals and the pommel plate or the hilt. Perfect fittings would have called for a perfect match of the two surfaces to be joined. We may safely assume that the artisans who made these swords would have had no difficulty to make perfect joints or fits, too - but why go through the labor and trouble? | ||||||||
I think it is likely that the old smiths also noticed that some of the sheets
they made as starting material for stack-welding were of inferior quality, containing large slag inclusions. But so what.
Just put them into the inside of a stack and nobody will know. Iron was very expensive, after all. Now we have an explanations why theses swords combine breathtaking workmanship with equally amazing sloppiness. The ancient smith had a modern attitude. Making money was more important to them then impressing archaeologists some 3000 years later. |
||||||||
Final Appearance
Luristan mask swords today are always more or less corroded and blackish to rusty in color. They are not a particularly pretty or impressive sight. That was certainly not the case when they were new. If polished to a high sheen, they must have been quite striking in appearance. | ||||||||
But polishing everything would have tended to obscure the fine details of the
figures. My guess is that the figures were painted to some extent, outlining, for example, details (like the rims around
the eyes) in black or some other dark color. They certainly must have made a (fashion?) statement for the fighting man that
was not rivaled by much else available 2800 yeas ago. In Luristan, that had a very specific and very large range of (bronze)
weapons and an iconic art style (exemplified, for example, in the "masters of
animals"), these swords must have appeared as marvellous implements, but rather exotic in appearance. Not even
remotely related to anything else around. My guess therefore is: | ||||||||
| ||||||||
I have no proof, of course. But consider:
| ||||||||
So why didn't we find Luristan mask swords in other places that once belonged
to the Assyrian (or Babylonian, if you like that better) empire? The answer is
simple and convincing: For the same reasons we found no iron whatsoever left over from
these empires (besides the Khorsabad hoard) even so we know for sure that a thriving iron industry must have been in place
(because of the Khorsabad hoard). Why did we find far more Roman swords in Denmark (never part of the Roman empire) then in the area the Roman empire (far, far larger than Denmark)? Well, the old Danes liked to go on tours, and like all good tourists, they liked to bring something special back to the old homestead. I rest my case. |
||||||||
Comparison With the Literature | ||||||
How do our results from the cut sword compare with what can be found in the literature? The straight answer is: | ||||||
| ||||||
However: almost all earlier investigations into the metallography of the mask
swords were restricted to very small samples making it hard to impossible to detect fire welding. Moreover, some of the
researcher harbored old and by now disproved believes (like the possibility of "carburizing" or "decarburizing"
bulk iron) that corrupted their interpretation. Our results shown here therefore do add considerably to our knowledge of
the Luristan swords. Moreover, the journals then (as now) printed only a few pictures at small sizes. Worse, print quality was not always very high and prints were only in black and white. Topping all that is the sorry fact that what you see now in (often badly) digitalized copies of copies... is often just a faded ghost of the original picture. That's why we "publish" here with lots of clear and often large pictures. | ||||||
I shall now progress through the key publications from this list. An earlier "literary guide" is given here.
Let's start with the oldest of these papers: | ||||||
1 | 1957 F. K. Naumann: Untersuchung eines eisernen luristanischen Kurzschwertes Archiv für das Eisenhüttenwesen, 28. Jahrgang, Heft 9, (1957) 575 - 581 | |||||
This is one of the earliest but still best papers to the subject. Small wonder,
it has been written in the true language by an expert metallurgist.
Naumann investigated the "Hamburg" sword and first dispels the believe that some iron casting was involved in
the making of the sword. Then he asks all the right questions about the making of the figures, the assembly of the parts,
the forging technique (fire welding?), the materials quality and so on. Since the sword had to remain mostly intact, he employed X-ray techniques, or better said, g-ray techniques since he used the high energy radiations of the isotope Ir 192. Here is an example of one of the many pictures he took: |
||||||
| ||||||
The picture is remarkable because it is the first "X-ray" image of a Luristan sword. It (and its brethren) show the individual parts and their partially "sloppy" attachment. We also see that the blade and the hilt were not made from one piece of iron (like our cut sword) but were "stitched" together from at least two pieces. | ||||||
Naumann could investigate only very small parts of the sword metallographically, like an area of about 0.5 cm2 of the blade. The results are shown below: | ||||||
| ||||||
I show this picture to illustrate the point made above: You don't see much anymore in present-day pictures of old publications. You do see , however, that Naumann's finding are quite compatible with ours. | ||||||
From the little he was allowed to do, Naumann drew many conclusions
that are also valid in our case. Here is a list
| ||||||
2 | 1961
Herbert Maryon;
with technical reports by R. M. Organ, O. W. Ellis, R. M. Brick, R. Sneyers, E. E. Herzfeldand F. K. Naumann: Early Near
Eastern Steel Swords, American Journal of Archaeology, Vol. 65, No. 2 (Apr., 1961), pp. 173-184 |
|||||
A key paper, reviewing the properties of 11 Luristan mask swords. The Naumann paper from above is prominently featured, including some of its picture in much better quality than what is left from the original. Here is an example: | ||||||
| ||||||
Compare to the picture above and you see the problem with pictures in old publications.
As far as the metallurgy of Luristan swords is concerned, Maryon reports on Naumann's finding and on metallurgical results from the "Toronto" sword. Only small parts (the "beard region" of one of the figures) was investigated in some detail, and the results are in line with what we found: High-carbon regions, some Widmanstätten structure - an example is shown here, |
||||||
Maryons paper is interesting because he speculates about the origin
of these swords. Here are a few quotes
|
||||||
Maryon might have been the first one addressing the problem of making the figures:
"The formation of the lions and of the human heads would have been effected first by forging, then the finer details would be added by means of chasing tools and punches, for there is no indication of the employment of cutting tools upon them." He also realized that we are looking at supreme examples of craftsmanship: "Just as the earliest books printed with moveable type are in many ways unsurpassed, so here, the sword handles forged in the new metal, steel, by these pioneer smiths of the Near East, exhibit skill of a high order, and no comparable steel sword-hilts have been found in any other land before the time of the Renaissance in sixteenth century Europe." | ||||||
3 | 1964
Kate
C. Lefferts: Technical Notes on Another Luristan Iron Sword American Journal of Archaeology, Vol. 68, No. 1 (Jan., 1964), pp. 59-62 | |
In this short paper the issue gets confused. Two major if wide-spread errors
are introduced in the context of the New York Metropolitan Museum
sword.
| ||
Well, Kate Lefferts obviously is not a metallurgist, as also evidenced by statements
like: "This gives us a hardness softer than air-quenched steel and again indicates, as does the
absence of pearlite, that the blade was not quenched." For you non-metallurgists out there: Hardness increase by quenching only works if there is sufficient carbon in the iron (typically in the form of pearlite) and it is the absence of martensite that indicates a lack of quenching. |
||
4 | 1966
K.
R. Maxwell-Hyslop and H. W. M. Hodges: Three Iron Swords from Luristan Iraq, Vol. 28, No. 2 (Autumn, 1966), pp. 164-176 | |||||
An important paper describing the first sword cut into two parts like ours. From
todays point of view this cut sword was rather atypical in its construction. Since nobody could know this at the time, its
unusual construction was assumed to be typical and dominated the structural discussion for a while. Look here and below for some details. We investigated a rather similar sword and found that it was much simpler in construction, rather more like our cut sword (just without figures etc.) Here is a comparison: |
||||||
| ||||||
Note that the hilt of the Hyslop sword actually consists of a stack of "thin" sheets. Fire weld the stack and you have what we found. | ||||||
Beside the cut sword several others were investigated, all of them but one without figures and therefore probably earlier than our cut sword. The set of swords investigated is shown below: | ||||||
| ||||||
The sword with the figures led to some insights:
|
||||||
| ||||||
However, the authors completely misjudged the sword production process:
| ||||||
5 | 1968
Vera Bird
and Henry Hodges: A METALLURGICAL EXAMINATION OF TWO EARLY IRON SWORDS FROM LURISTAN Studies in Conservation, 13 (1968), 215-223 215 | |
This paper contains in essence the metallographic study to the Hyslop-Maxewell
paper above. It contains more micrographs of etched surfaces but does not add much new insights. Two swords were investigated,
one was made of "wrought iron", the other of steel (including hypereutectoid steel). "There is, however, an amusing inconsistency :On the one hand it could be argued that the second sword was made from deliberately manufactured steel, and that during working the surfaces it becomes decarburized ". So far, it was the other way around;: The wrought iron, supposedly, became carburized. Both processes are impossible anyway as pointed out many times already.. | ||
6 | 1971
C. S. Smith:
The Techniques of the Luristan Smith Science and Archaeology (Ed. Rober H. Brill), Cambridge, Massachusetts, 1971 | |
Smith actually purchased "Luristan" iron swords so he could take them apart for analysis. However, only one of his objects can be classified as an iron Luristan mask sword. At least one of the others is an akinakes, usually associated with the Scythians. Smith knows his metallurgy and the paper is rather interesting. He found and discusses at length spheroidized carbon (actually cementite) and found and discusses nitride precipitates for the first time. Unfortunately, the pictures are by now so bad that one can't see anything anymore. | ||
A few quotes and comments, always in the context of the Luristan mask sword:
| ||
Smith acknowledges fluctuating carbon concentrations from smelting and the high
T - low T forging already mentioned several times. He also considers fire welding or faggoting but believes that, if at all, it happened when the bloom was processed. This certainly disagrees with our results |
||
7 | 1971 P. R. S. MOOREY: CATALOGUE OF THE ANCIENT PERSIAN BRONZES IN THE ASHMOLEAN MUSEUM | |
Interesting, in particular as far as the history of iron and the style of the Luristan swords are concerned, but no new metallurgical information is given. | ||
8 | 1973
E. Schumacher:
Eisenschwerter mit Maskenzier aus Luristan Kleemann Festschrift II Teil (=Bonner Hefte zur Vorgeschichte, n. 4), Bonn, 1973, p-97 ff | |
Describes the appearance of about 30 mask swords and provides with drawings (including Luristan type II swords). No new metallurgical information is given. | ||
9 | 1987 Claude Forrieres, Étude par microscopie électronique de structures de trempe d'une lame d'épée du Luristan, Étude par microscopie électronique de structures de trempe d'une lame d'épée du Luristan ArchéoSciences, revue d'Archéométrie Année 1987 11 pp. 17-29 | |
It's in French! Probably interesting but meant only only for Frenchmen. In addition,
the (many) pictures are of bad quality. We do see Widmannstätten structures, spheroidized cementite, and things with some similarity to what I've called "really weird" for our sword |
||
10 | 1988
O. Muscarella:
"Bronze and Iron", (Book) New York 1988 (pp 184 . 189) | |||||||
The "Pope" speaketh. O. Muscarella certainly knows about Luristan things
and has written about it at length. He is, however, not a materials scientist or metallographer. In the book referred to
here, he gives - among much other stuff - a detailed description of the special mask sword in the Metropolitan Museum in
New York. It is special because
| ||||||||
| ||||||||
Muscarella makes a point for the origin of these swords in Luristan (which I'm not so sure about) and mentions that he is aware of 88 of those swords. | ||||||||
11 | 1991 J. E. Rehder, The Decorated Iron Swords From Luristan: There Material And Manufacture; Journal of Persian Studies, Iran, Vol. 29 (1991) pp. 13 - 19 | |||||
This is a review without pictures. Rehder discusses the "Toronto" sword
and others, including the radiocarbon dating of the Toronto and MIT sword. He reviews the available information and mentions spheroidized cementite. He also points out erroneous or misguided stuff in older publications. Rehder marvels at the workmanship and concludes that the swords were not good as weapons and probably for representation. He discuses smelting techniques and iron / carbon systematics in detail and relates that to the making of the swords.. |
||||||
12 | 2003 Bruno
Overlaet, The Early Iron Age in the Pusht-i Kuh; Luristan; Acta Iranica, Vol. XXVI, 2003; selected parts of the huge volume Bruno Overlaet, The Chronology of the Iron Age in the Pusht-i Kuh Luristan, Iranica Antiqua, Vol XL (2005), pp. 1 - 33 |
|||||
The ultimate about Luristan graves. Excerpts from the detailed excavation report and so on - but no details abour mask swords. | ||||||
13 | A.
Hasanpur et al.; The Baba Jilan Graveyard Near Nurabad, Pisht-i Kuh, Luristan - A Preliminary Report, Iranica Antiqua, Vol. L (2015), pp 171 -212 | |||||
Some details about the only mask sword (now in Brussels) ever found during an excavation by archaeologists (if only in the detritus left by the grave robbers). No metallurgocla details, however. | ||||||
| ||||||
Conclusion | ||
There is a general agreement in the old literature about the following points:
| ||
Our findings generally agree with that but we claim the following points as new:
| ||
|
||
Part 1 Heads Part 2 Pommel and hilt Part 3 Animals and blade Large Pictures |
Critical Museum Guide: Metropolitan Museum, NYC
Scythian Special Large Pictures
Early Iron Making Empires in the Middle East / Mediterranean
The Luristan Project - Results from Cut Swords
The Luristan Project - Results from Cut Swords Part 2
The Luristan Project - Large Pictures of Cut Sword
Master of Animals Finials from Luristan
The Luristan Project - Literature Review
The Luristan Project - Results from Cut Swords
New Interpretation of Master of Animals Figures
© H. Föll (Iron, Steel and Swords script)