Ethnographic Arms & Armour

Ethnographic Arms & Armour (
-   Ethnographic Weapons (
-   -   FYI Brief history of crucible damascus steel (

Ann Feuerbach 23rd March 2005 04:07 PM

FYI Brief history of crucible damascus steel
Origin and Meaning of the Phrase Damascus Steel

There are four varieties of swords or steel to which the adjective Damask, Damascene, or Damascus is applied: pattern-welded, inlayed, preferentially etched, and crucible. The use of the adjectives crucible, pattern-welded, inlayed or etched together with Damascus, Damask or Damascene should be used to avoid confusion. Using these adjectives, the method used to form the pattern (if known) is clear and free from ambiguity. Unless otherwise noted, the term Damascus steel is used here only to denote the crucible variety.

The lack of proof of Damascus ever having actually been a centre of sword production has been argued by Elgood (1994, 103-108). The origin of the name Damascus steel is frequently attributed to the crusaders, who, as the legend goes, were introduced to these blades in Damascus and brought the word and the legend of the steel back with them upon their return to Europe (e.g. Sherby and Wadsworth, 1985, 112). Although this assertion is common, no reference to crusaders having used the term has ever been reported in any of the literature. There are more credible roots for the origin of sword names. The Islamic writers al-Kindi and al-Beruni name swords based on surface appearance, place of production or forging, or the name of the smith. There are three likely sources for the term Damascus in the context of swords. The word for water in Arabic is, damas (Sachse, 1994, 13) and Damascus blades are often described as exhibiting a water-pattern on their surface. Al-Kindi called swords produced and forged in Damascus as Damascene (al-Hassan, 1978, 35). Additionally, al-Beruni mentions a sword-smith called Damasqui who made swords of crucible steel (Said, 1989, 219-220). Any or all of these may have inspired the term “Damascus steel” swords but it certainly were not crusaders who coined the term.

The terms Wootz and Pulad

In Central Asia the term pulad is always used to denote crucible steel. The word pulad can be traced back to the Avesta, the sacred book of the Zoroastrians (Allan and Gilmour, 2000, 7) However, in the 6th century AD Sogdian became the language of the Manichean church in the east (Gignoux and Litvinsky, 1996, 417), therefore we can assume that the text was written before this time, making this text the earliest literary evidence for the term pulad. Variations of the word pulad can be found in New Persian (polad or pulad), Mongol (bolat), Russian (bulat), as well as in Tibetan, Armenian (p’otovat’), Ossetic, Grusinian (poladi), Ukranian (bulat), Chechnian (bolat), Turkish, and Modern Arabic (fūlād) (Toussaint, pers. com.; Abaev, 1985, 265). Additionally, “in Urdu the word is farlād for steel. But in Hindi itself the word exists as phaulad meaning steel ” (Toussaint, pers. Com)

Abaev (1985, 265), during his search for the history of the Russian word bulat, proposed that the word may have come from Sanskrit. It can now be argued that the word does indeed come from Sanskrit or one of the many Sanskrit related languages. The word pulad can be viewed as the conjunction of two words pu (also transliterated as fu, phu) and lad (or ladh). In Sanskrit pu means cleaning or purifying (Cologne Digital Sanskrit Lexicon, 2001). There is no direct translation of lad or ladh, however, there are over a hundred words for iron in the various Indo-Aryan branch dialects that use variations of the word lōhā, including lauha (see Grierson, 1928, 77). The similarity between pu-lauha meaning purified iron, and pulad, meaning refined or purified steel should not be overlooked and strongly suggests a possible etymological origin for the word pulad. It should not be assumed that the word originated in Sanskrit proper. The Avestan language of Central Asia was very similar to Sanskrit and the possible forerunner of the word may equally be found there or in one of the languages which has a similar root.

There are many words in the different Indian languages for the word iron including lauha and loha. According to the Cologne Digital Sanskrit Lexicon (2001) there are also scores of words for steel including the following transliterated as piNDA7yasa, abhraka-sattva, cIna-ja, and tIkSNa4-loha. A variation of pulad also appears in Sanskrit as po1la1wade1n, and po1la1wad (Cologne Digital Sanskrit Lexicon, 2001).

Indian/Sri Lankan crucible steel is commonly referred to as wootz. It is generally agreed that wootz is an English corruption of the word ukko or hookoo. The term wootz first appears in print in 1795 in Pearson’s Lecture to the Royal Academy on Indian steel (Hadfield, 1931). This was during a time when Indian crucible steel was being sent to England for laboratory analyses with the purpose of understanding what made it apparently tougher than steel made in Europe. A discussion of the early interest in wootz, those who studied it, and its etymology can be found in Hadfield (1931).

In summary, the term pulad has a true etymology, going back over 1000 years, whereas the term wootz is just over 200 years old, and is a corruption of the Indian word.

Archaeological Evidence
The majority of scholarly literature refers to all crucible steel made by traditional methods as wootz (e.g. Bronson, 1986; Figiel, 1991; Verhoeven, 2001). When they do acknowledge that it was produced in Central Asia, it is implied that it is a variation of the wootz process (e.g. Bronson, 1986, 42-45). Previously no attempts have been made by scholars to differentiate between the crucible steel processes used in India/Sri Lanka and Central Asia.

The following characteristics are typically found in the crucible steel process employed in South India/Sri Lanka, commonly referred to as wootz. In all the crucibles rice husks were used as temper. The shapes of the crucibles are conical or aubergine shaped (South Indian) or elongated, pear-shaped or light bulb shaped (Sri Lanka). The base of the crucible was either rounded or pointed. “Earth” or “clay” was put into the top of the crucible to seal it, but occasionally separate preformed lids were used. The crucible charge used was one type of iron and either wood and/or leaves. At most sites the steel solidified relatively quickly. The crucibles were stacked together in the furnace, often in a cone shape.

Remains from crucible steel production at Hyderabad, central India at Konasamudram, Nizamabad district, formerly called Golconda, Andhra Pradesh, have been studied by Lowe (e.g. 1989a, 1991) and reported by Voysey (1832). Bronson (1986) primarily uses Voysey’s account (1832, 245-247) as his source of information on the Hyderabad process. Bronson’s argument for differentiating the Hyderabad process from the South Indian Process is the crucible charge. The South Indian process used iron and carbonaceous material whereas the Hyderabad process apparently used two types of iron. Research by Lowe (1989a, 1989b, 1991) has provided additional information confirming that the characteristics of the Hyderabad process are distinct from those of the South Indian wootz process. When compared to the crucible steel remains from Central Asia, the Hyderabad remains have more characteristics in common with those remains than with the South Indian/Sri Lankan remains.

The remains of crucible steel production from Merv and Uzbekistan share many common characteristics. They are, however, not the same process. The physical characteristics of the crucible steel process used in Central Asia, based on archaeological remains, are made of refractory clay with quartz temper. The crucibles sat on the furnace floor and gravel like material was placed in between the crucibles, on the furnace floor. The nature of the crucible charge is uncertain but the best hypothesis is bloomery iron and plant matter, possibly with some manganese added. The crucibles were fired until the steel was liquid and then left to slowly cool inside the furnace. The result was a high carbon steel ingot virtually free from slag and non-metallic impurities. These are some of the physical and technological characteristics, which can, therefore, be used to define the characteristics of Central Asian crucible steel production.

In summary, in Central Asia and Hyderabad, the crucibles sit on the floor of the furnace and gravel-like material is placed in between the crucibles, indicating that the crucibles were not closely adjacent to each other, thus allowing hot air to flow between the crucibles. In contrast, all known crucibles from South India/Sri Lanka do not have flat bottoms and were in close contact with each other. This would have impeded the dispersal of heat around the crucible and probably accounts for the comparatively high number of failed products.

The cooling rate of the crucible steel ingots also seems to differ, although further research is needed to determine how significant this difference is in relation to the final product. The evidence so far indicates that in Central Asia the crucible, with the ingots inside, cooled slowly inside the furnace. Conversely, those from South India/Sri Lanka are usually recorded as being removed from the furnace while still hot. This would affect the solidification rate of the metal and would influence the final microstructure and ease of forging.

During the solidification of the liquid metal, the ingot shape is a virtual cast of the crucible’s interior. All known crucible steel ingots from Central Asia are egg shaped. These egg shaped ingots were also noted by al-Beruni during the 10th century (Said, 1989). The shapes of the ingots from south India/Sri Lanka, however, tend to be more varied but never egg-shaped.

It must therefore be concluded that the materials and techniques associated with the crucible steel process in Central Asia (pulad) and those used in India and Sri Lanka (wootz) are significantly different.
Other info and bibliography can be found at http://moltenmuse.home.att

Ian 23rd March 2005 04:11 PM

Thank you, Ann
This is most informative. It will take a little while to digest. Much appreciated.


Gt Obach 23rd March 2005 09:34 PM

wow... thank you Ann
this has made my day
the different forms of crucibles are fascinating..
I thought the different shapes of the ingot/button was for different purposes.... such as egg shaped for swords... etc...
- but I am corrected...thank you once again


do the crucibles with rice husks have a lower failure rate. ?

Andrew 23rd March 2005 10:54 PM

Another great thread! Thank you, Ann. :)

Ann Feuerbach 23rd March 2005 11:24 PM

Originally Posted by Gt Obach
do the crucibles with rice husks have a lower failure rate. ?

Good question. Unfortunatly the documentation/publication of the Indian sites is brief. If I recall correctly, the ethnographic references state a fair number of failures. However, the process obviously worked enough times to make it viable.

Jens Nordlunde 24th March 2005 03:07 PM

Thank you Ann, for your two posts. On a forum where oriental arms are discussed, posts like these are very important.
Also thank you to Andrew, for making them a sticky.


Rivkin 24th March 2005 04:53 PM

Thank you very much for your posts !

Sorry, but afaik:

1. In chechen (or more general in vainach) steel is "bulat", for example Haz-Bulat (beautiful steel), Bet-Bulat-Ka etc. It's assumed that the word was introduced by turkish tribes, either khazars or mongols.

2. I always thought that it's also Bulat for "forest" mongol tribes, like Bashkir and so on, and russian/ukrainian bulat is simply a copy of this.

Sincerely yours,


Jens Nordlunde 2nd April 2005 02:37 PM

Ann, do you have pictures of the different types of ingots, with measurers and weight?
If you have please show them, as I think not many have seen the different ingots earlier described.


Ann Feuerbach 2nd April 2005 03:38 PM

Unfortunatly Jens I do not and at the present time, can not scan either. :( There are very few published pictures of ingots (around 2), mostly descriptions, Craddock, 1998 illustates one from Gatihosahalli, Wayman and Juleff, 1999 illustate one from Sri Lanka. That is it from archaeological contexts. There are some modern ones around, but using replicated ones as models for ancient ones can be misleading.
Ethnographic references say cone shaped, puck or elongated. The only measurment is 2.3-3 cm in diameter for sone from Mel-Siruvalur. Lowe reports ingots from Hyderabad are cake like in shape and between 50-110 grams, whereas Voysey says 680-1260 grams.
We have two ingots from Merv, one egg shaped and one potato shaped. this is significant because the historican Al-Beruni said they are egg shaped.

Jens Nordlunde 3rd April 2005 12:57 PM

Gene/Ann, how big is Gene's ingot? From the picture shown on 'Magnetic' page 3, it looks as if it is about 5 cm (2 inches) in diameter, and about the same in hight, but what is the weight?
Somewhere I saw that the ingots were made in two sizes, one size big enough for one sword blade and another size big enough for two blades, but from what Ann writes, they must have been made in many different sizes.

All times are GMT. The time now is 06:54 AM.

Powered by: vBulletin Version 3.0.3
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.