J.W. Allan, “Abu’l-Qasim’s Treatise on Ceramics”, Iran 11 (1973) pp.111-20

To read about the application of Abu'l-Qasim's methods of pottery production in modern-day Iran,
go to this article by Alan Caiger-Smith: "Esfahan: an unexpected pottery workshop"

[Note: the numbers in square brackets indicate the page numbers of the original printed article]

[111] I. INTRODUCTION

Abu’l-Qasim’s treatise on the manufacture of tiles and other ceramic objects is part of a larger work by the same author. This exists in two manuscripts, dated 700/1301 and 991/1583 respectively, and the chapter in question has been published in its original Persian with a German translation and notes by ??? Ritter, J. Ruska and R. Winderlich in Istanbuler Mitteilungen III (1935). The present writer has for some time felt the need for an English translation of this treatise with a more detailed commentary, not only for the sake of the art historian or student interested in the history of medieval Persian minor arts, but also for the sake of potters interested in more particular aspects of ceramic history - details of the manufacture of bodies, frits, and pigments for example. The commentary is not intended as the final word; analyses of Persian pottery bodies and glazes recently undertaken by Dr. Francis Schweizer of the Research Laboratory for Archaeology and the History of Art in Oxford, while they have provided extremely valuable evidence on the nature of some of the materials used, have also highlighted how little is actually known about these materials in general, and a great deal more research will be required before definitive answers to many of the outstanding questions can be found. It is hoped that publication of this English translation and notes will further stimulate such research.

The author is particularly indebted to Mr. Alan Caiger-Smith, whose practical comments helped to elucidate many obscurities; to Miss Rosemary Brewer, whose experiments gave an insight into the problems of reproducing the compounds and objects described in the text; and to Mr. George Morrison for help with the more obscure passages of the Persian text.

Variant readings or additional comments from the later manuscript of the treatise are put in square brackets in the translation; round brackets indicate words added by the translator to clarify the meaning of the sentence where English grammar requires it. Numbered paragraphs have been inserted for convenience and clarity.

II. TRANSLATION

§1. The final chapter about knowledge of the art of kashi-gari, which is also called ghadareh. This art is in truth a type of “philosopher’s stone”. This chapter is divided into three parts:

i. Introduction - knowledge about vessels, ingredients, materials, and their requisites.

ii. Knowledge of how to dissolve these substances.

iii. Knowledge of how to compound these substances.

§2. Introduction about the materials, vessels, ingredients and their requisite raw materials. The vessels, ingredients and materials which serve as raw materials for these people are many.

§3. The first is hajar-i maha, known in Arabic is hasat, and by the craftsmen as shukar-i sang. It is a white, clear, shiny stone, less clear than rock crystal, but clearer than white marble. It is very hard, and many sparks come off it by means of the fire iron. Its deposits are in many places. Rock crystal would be equally useful, but it is rare and therefore not used. [Shukar-i sang is the basis of transparent agents.]

§4. Other rocks are similar to maha, one variety of them being called sa-ishkineh. It most commonly comes in large pieces, and is less clear and less solid than shukar-i sang, but is found in more places [or: mountains, hills and high places] than that stone.

[112] §5. The third is a finely-divided white stone [of a uniformly dirty colour] which comes from mountains of the villages of Fin in the district of Kashan. It is the colour of lime [or: quicksilver], and is called bataneh by the craftsmen. This is the basic ingredient for pots of two firings.

§6. The fourth is qamsari stone, attributed to a village (of that name) [in the district of Kashan]. It is burned and broken up until it becomes a powder like white sugar.

§7. The fifth is shakhar, which they call qali, which is made by burning pure, fully-grown ushnan plants, not mixed with shureh, which is like ushnan. [If it is (mixed with shureh) it makes it rotten.] The best shakhar is that which has a red-coloured centre when broken, with a strong smell. [Ushan is found everywhere, and the value of qali is that it combines the stones together.]

§8. The sixth is the stone lajvard, which the craftsmen call Sulaimani. Its source is the village of Qamsar in the mountains around Kashan, and the people there claim that it was discovered prophet Sulaiman. It is like white silver shining in a sheath of hard black stone. From it comes lajvard colour, like that of lajvard-coloured glaze etc. Another type comes from Farangistan and is ash-coloured and soft. And there is a red kind found in the mine which is a deposit on the outside of the stones and is like the red shells of pistachios. This kind is very strong but is a fatal deadly poison.

§9. The seventh is a stone which is absolutely black and murky, like kohl. It comes out of the fire shiny black. Its mines are in Khurasan, in the mountains of Jajarm, and it is called muzarrad. [Its benefit is the black colour with which they paint objects.]

§10. The eighth is gold and silver marcasite, male and female maghnisiya, yellow vitriol, [yellow and red] arsenic, mardasang, surmeh, tutty and lead, which will each be treated in their proper places. [The benefit of these is their red and yellow colours.]

§11. The ninth is a white, sticky, strong clay. It is found everywhere but the white one is rarer. [A condition of the manufacture (of the objects) is that it should be white.] The Kashani type is white and very strong, and the craftsmen call it Warkani and Luri after a village and the Lurs. One type of it is like white snow, and its mine is in the mountains of Na’in near Isfahan. It is mixed with plaster and used to whitewash the houses. [In the manufacture its benefit is the whiteness of its body and substance.]

§12. The tenth is that which is made from the seven metals. One of these is the form of tin called rasas. [Its mines are known in many places. The first is that from Farangistan.] In Farangistan it is cast in the form of snakes and stamped with a Farangi stamp [as prevention against adulteration]. Some is brought from China in large pieces, and some from the borders of the Bulghars in thin pieces, like sheets of paper, one inside the other. That is the best type of tin.

§13. The eleventh is lead. It is found in many places like Kirman, Yazd and Rum, and [further away] in the scattered territories of the Bulghars. This latter is the best type of lead, its substance being very white. The lead from Rum is the worst. From lead come mardasang, red and yellow isrinj, and the white lead of the painter. [The benefit of tin and lead is the uniform turquoise colour, and the benefit of lead is the varieties of its glaze.]

§14. The twelfth is roasted copper, and copper hammerscale. [Its mines] are in Rum and Dazzamar in the Azerbaijan region.] The best of all is the red, greenish-coloured, soft-natured one, from which a green colour is made. And also scales from burnt iron which make yellow [amongst other colours] upon firing.

§15. The Solutions. [or: The second division is the solution of materials.]

These materials are broken up and powdered like grains of kohl by means of beating, powdering, grinding, sifting and sieving. Some materials are broken up to the size of a pea with an iron hammer and ground by means of a mill. When this is done with a hand mill it is much better [and cleaner]. This is what is done with shukar-i sang, sa-ishkineh, and bataneh.

§16. The seven metals, tin, lead, copper, iron, lajvard, muzarrad and other substances, are ground for slightly longer on an oblong grindstone which they use [for grinding].

§17. Since the various materials are now known, they add one substance to another and make particular compounds depending on the compounds of the original materials. [or: they add one of these [113] substances to a cementing material such as potash and lead so that they mix the two substances together and make a single substance, and this exists either in the bodies or substances in relation to the compounding of the original materials.]

§18. This is done as follows: They take 105 parts of shukar-i sang which has been powdered, beaten, ground [to the size of a split pea], and sifted through silk, and 100 parts of shakhar in lumps the size of hazelnuts or almonds, and mix them and put them in a kiln, technically known as bariz. The pungency or weakness of the shakhar varies depending on the place. Hence they need for 1 man of stone 1½ mans of Tabriz shakhar or 1 man of Baghdad shakhar. This is cooked over a slow fire [for six hours], and is stirred from morning till night with an iron ladle made as large as the diameter of the kiln until it is well mixed [and becomes white] and is become one, like molten glaze, and this is the material for glass vessels. After eight hours they take out the brew by the ladleful. Below, in front of the oven, is a pit full of water, into which they put the glass frit. When water and fire meet there is a great noise and roaring like thunder, which for all the world could be real thunder and lightning [such that everyone who has not seen it and hears the noise fills on his knees shuddering and trembling]. The craftsmen call this mixture jawhar and store it, until the time comes to compound it, in a broken up, powdered and sifted form.

§19. Another such kiln is used for breaking up [and dividing] the tin and lead. This is done as follows. One takes 3 [or: 2] parts of good white lead and a third of a part [or: one part] of tin, or if one wants a better and finer mixture up to half (of a part) of tin. First they put the lead into the kiln for a time, and then they throw the tin in on top of it. They mix them at a high temperature until they are well melted. When (this mixture) brings up an earthy substance on its surface it is completely ready. They then make the fire smaller and seal the furnace door with mud. The earthy substance which collects on the melt is taken of with an iron shovel until in half a day it has all gradually changed to earth. The craftsmen call this sirinj and if one wishes one can make sapideh-yi zanan from it. What I mean is as follows. If they leave it for one hour layers of good white are formed, and they take them out with a shovel and separate them [slowly] from the sirinj [or: earth]. If they want it all to be glowing white they leave it in the furnace and make the fire hot for a time. They leave it there for 24 hours until it is completely white. It must not be disturbed with a ladle or shovel. After this one roasts one part of qamsari stone and one part of shakhar together, leaves it to cool, and for every [two or] three mans of the above-mentioned frit one adds one man of this sirinj. One then puts it back into the frit kiln for 12 hours brewing until the frit becomes a uniform [turquoise-white] mass. When it has become properly melted it is put into water by the ladleful and is stored, broken up and ground.

§20. If one wants to use tin alone to get a glowing white colour, one uses two earthenware vessels. Tin is put into one and beaten with an iron pestle until it all becomes earthy and black in colour. It is cooled and sifted with a fine sieve on the end of a ladle, and put into a second vessel [which has been baked]; they leave it until the fire catches it and it rises nicely from its place [or: slides about in an oily manner]. When it is cool it is a white earth and it is used for making [sapideh-yi zanan and] turquoise preparations.

§21. If they want to make white out of lead, they put the lead in an iron ladle and throw clean ashes on top of it together with pure linseed oil, and this is mixed with a pestle until it all becomes like millet. It is washed in clean water, put in a [bulky] linen bag among sour grape husks, [and put in a damp place] for some days [or: for one or two weeks] until it has all become [white] lead. This white lead is used in eye medicines and for painting.

§22. From lead comes ithmid, tutty, [abarl, mardasang, and also shangarf. What I mean is that they make lead into mardasang, grind and sift it, and put it into new earthenware vessels for 48 hours in a furnace, and it becomes red shangarf.

Everything described above is to do with Solution.

§23. The Combinations of all these components is the sum of the aforementioned substances, each one according to a determined measure. Thus if they want to compound a body out of which to make pottery objects and vessels such as dishes, basins, jugs, and house tiles, they take ten parts of the aforementioned white shukar-i sang, ground and sieved through coarse silk, and one part of ground glass frit [114] mixed together and one part of white Luri [or: Warkani] clay dissolved in water. This is kneaded well like dough [with Luri clay] and left to mature for one night. In the morning it is well beaten by hand and the master-craftsman makes it into fine [or: vessels of all attractive design] on the potter’s wheel; these are left standing tin they are half dry. They are scraped down on the wheel and the feet are added, and when they are dry they [or: the tile surfaces] are washed with a damp linen cloth in order to smooth over the lines on them so that they disappear. When they are dry again they are rubbed with a wool cloth until they are clean and smooth.

§24. The material for house tiles and writing is made of bataneh and sa-ishkineh mixed to a dough - I mean together with glaze frit and clay. There are two types: transparent and opaque. The transparent form is also of two types: one for painting on a white ground, and one for painting on a green ground. To paint on a white ground one takes muzarrad for black, Sulaimani for lajvard colour, maghnisiya for red, roasted copper for green, and ... [burned iron] or hammerscale ... [for yellow], ground, pounded and sifted. They paint, having mixed each of these with a little hasa [and they coat with white frit]. For a green ground they paint with plain muzarrad.

§25. The vessels are then coated with a glaze frit which has been ground up, finely sifted, and dissolved in water, and are stood on top of a broad-meshed sieve, which is the lid of a trough, so that excess of colour drips away. They are dried in the sun. If they want a green ground they coat on a mixture of ten parts powdered glaze to a quarter of a part of [a mithqal of] roasted copper. The craftsmen call this tini. It comes out of the firing transparent green, like green glass. If they use one part of [brayed] lajvard to forty parts of glaze frit it becomes transparent blue like a sapphire. If for every ten [or: two] parts of glaze frit they add 1 part of maghnisiya it comes out black as shabeh, and if they add less it comes out a red the colour of an eggplant. If they want an opaque colour such as turquoise they add for every man of ground tin ten dirhams of ground roasted cooper [sic: copper?], and coat this on. If they want lajvard colour they add [to the glaze frit] ten dirhams of Sulaimani lajvard and daub and coat the vessels with that. If they want a greyer tone they put in less lajvard and add a small amount of red sirinj. If they use an absolutely plain colour the vessels come out of the heat white.

§26. For each vessel is made an earthenware case with a fitting lid. These are placed in the kiln, called in Arabic shakhureh and locally dam [and dasht]. This is like a high tower, and inside has row upon row of fired earthenware pegs, each an arsh [or: a dhira’l and a half long, fitted into holes in the wall. The vessels are placed on them and fired for twelve hours with a hot even fire, with this stipulation: that no wood be put on until the smoking has stopped, so that the smoke does not ruin or blacken the pots. In Kashan they burn soft wood [like hyssop and walnut], and in Baghdad, Tabriz and other places the wood [of the willow] is stripped of its bark so that it does not smoke. The vessels are removed from the kiln after a week [after they have cooled].

§27. Those that come out of the firing white they paint with the enamel of two firings, or with lajvard, or with pure turquoise. [Or they are translucent and require no enamel painting.] The enamel is composed as follows: Take one and a half mans [or: parts] of red and yellow arsenic, one man [or: part] of gold and silver marcasite, one batman [or: half a part] of Tisi [or: Tabasi or Cypriot] yellow vitriol and a quarter [of a part] of roasted copper, and mix it to a paste and grind it. A quarter of this is mixed with six dirhams of pure silver which has been burned and ground [with sulphur] and is ground on a stone for twenty-four hours until it is extremely fine. Dissolve this in some grape juice or vinegar and paint it onto the vessels as desired, and place them in a second kiln specially made for this purpose, and give them light smoke for seventy-two hours until they acquire the colour of two firings [which is like gold]. When they are cold take them out and rub them with damp earth so that the colour of gold comes out. Other people add certain preparations like sirinj and zanjar to this enamel. In fact, shadanaj stone with roasted silver serves the same purpose. That which has been evenly fired reflects like red gold and shines like the light of the sun.

§28. If they want to gild transparent or opaque pieces they hammer a mithqal of red gold into twenty-four sheets, putting paper covered with plaster between them. They cut them carefully [into the required pieces] with scissors, and stick them with a pen onto the vessels with dissolved glue, and [115] smooth them with cotton. For red they mix qamsari [or: bukhari] with ground glaze and embellish with it. For white, they paint with tin white dissolved in glaze; for black, they use muzarrad mixed with glaze; and for yellow, iron scale dissolved in glaze. Each of these is put in an earthenware capsule into the gilding furnace, specially made for the purpose, and fired from morning till night at a low heat. When they are fired the fire is made smaller and the firebox sealed with clay. From time to time samples are tested. All gilded pieces such as writing, etc., are made in this way. Should they want also seven colour vessels, which have passed into oblivion in this time - well, God knows what is right. [or: Should they want also the seven colour vessels, which have passed into oblivion in this time, which are called in the technical terminology of the artificers lajvardineh, odd ones of them are found in townships and shrines and they make them after the same pattern and bake them; but God knows the truth of things.]

III. COMMENTARY

§1. Kashi-gari: lit. “tile-making”, but since ghadareh means “earthenware” or “pottery”, it is clear that the author is writing about the making of all types of ceramics.

§3. Hajar-i maha: lit. “crystal stone”; hasat: lit. “pebbles”; shukar-i sang: lit. “sugar stone”. This stone is presumably quartz, and the Arabic name hasat probably indicates the quartz pebbles found in dry river beds. Quartz is silicon dioxide, and rock crystal is the purest and most transparent form of quartz.

§4. Sa-ishkineh: no literal translation is possible, but the word could be explained as meaning “that which a stick breaks” or “crumble stone”. Chalkstone would fit this description, but see §5.

§5. Bataneh: lit. “putty”, “filler”, or “lining of a cloth”. Abu'l-Qasim offers one further piece of helpful information about bataneh in §24: “the material (for underglaze pigments) for house tiles and writing is made of bataneh and sa-ishkineh ... together with glaze frit and clay.” The problem lies not only in identifying these two substances for certain, but also in understanding the term “pots of two firings”. If §§5 and 24 are to be taken together -and there is no reason to think that they should not be - then presumably the term “pots of two firings” means pots decorated under the glaze, and hence pots which have been biscuit fired and then painted, glazed and fired a second time. (It would also seem probable, however, that pots destined to be painted over the glaze were fired twice before they were actually decorated, particularly those with tin glazes, since tin glazes tend to crawl if fired on the leather hard body: Wulff describes how the modern potters still use three firings for overglaze decorated tiles.) Hence, the last sentence of §5 could be read as: “This is the basic ingredient for pigments on underglaze-painted pots”. The original editors of the text suggested that bataneh was felspar, but there is no evidence to suggest that felspar was known or used in medieval Persia. However, Naraqi says that the spring of Sulaiman at Fin boils up from the ground at the foot of the rocks called Kuh-i Dandan and the rock slab gulley of the mine of gypsum (gach) and limestone (ahak) situated six kilometres west of Kashan, and on the basis of this information it is tempting to suggest that bataneh is in fact limestone. When burnt, the limestone would make quicklime which could be converted into hydrate, and then be used as a “filler” or ground for the underglaze pigments. It must be admitted, however, that there is little reason for including both chalk (sa-ishkineh) and lime in the underglaze pigment material, and equally conceded that Abu’l-Qasim does not mention any burning of limestone to produce quicklime.

§6. Abu’l-Qasim mentions qamsari stone elsewhere: roasted with soda it produces a frit (§19), and mixed with ground glaze it is used for overglaze red (§28). As will be seen, these two pieces of evidence have to be taken separately. A substance which when roasted with soda produces a glass must contain either boron or silica, and to the potter there are two obvious substances which could be used in [116] such a situation - colemanite (boro-calcite) or talc (magnesium sihcate). From Abu’l-Qasim’s treatise there is no way of telling which of these is the more likely translation for qamsari stone. However, our knowledge of the village of Qamsar and its geology is fortunately not confined to Abu’l-Qasim’s writing. Towards the end of the last century the then Director of Mines, A. H. Schindler, visited the village (which he calls Kamsar) and he later described what he found there: “the rocks are dolomites broken through by serpentines with an immense lode of iron ore, copper pyrites, sulphuret of nickel, cobalt bloom and earthy cobalt”. G. Ladame, during an extensive survey of Iran’s geology some forty years later, visited the same village (which he calls Khamsar), and noted dolomitic limestones with veins of haematite, one of which contained asbolane and erythrite, plus some malachite and azurite. From these two sources it is clear that the main rock there is calcium magnesium carbonate, but that magnesium silicate is also present, and it is this rock which must be referred to by Abu’l-Qasim by the name qamsari; Wulff’s suggestion that qamsari is boro-calcite thus seems unlikely. As to our author’s observation (§28) that qamsari stone was used for overglaze red, it is inconceivable that magnesium silicate could produce such a colour, and once again reference to Schindler and Ladame points to a solution to this problem. The former mentions iron ore at the village, the latter haematite, and it must have been this haematite which the potters used for their overglaze red, just as modern potters do. The way Abu’l-Qasim calls both stones qamsari is not unduly worrying when their common origin in that one village is understood.

§7. Shakhar means “soda”, and soda, according to Abu’l-Qasim, was the standard flux for glazes in his time (see §19). This has been confirmed by analyses of contemporary pots carried out at Oxford. Ushnan is the plant Salsola Soda, and shureh is Salsola Tragus. Olmer described the sodas made from these two plants: that made from the former, according to his analysis, contained 29.5 per cent carbonate of soda, that from the latter only 13.7 per cent, indicating the validity of the later scribe’s comment “if it is mixed with shureh it makes it rotten”. Olmer comments that he never saw any attempt to wash or purify these before use. Brill gives further analyses of Middle Eastern plant ashes, in which the soda content is anything up to 42.5 per cent. Because of the strong smell emitted the plant burning is done well away from the villages. Qali is the origin of our word “alkali”.

§8. Sang-i lajvard is the Persian for “lapis lazuli”, which was and still is mined in Badakhshan in Afghanistan. Here, however, it must mean cobalt ore, for lapis lazuli cannot be used as a glaze colourant. In view of the importance for ceramic history of Schindler’s description of the production and use of cobalt, and in view also of the difficulty in finding copies of his book, his text is given almost in full:

"Two miles below Kamsar some thin lodes of copper ore crop out in shales which dip 80º west, and strike north 22º west to south 22º east. One mile or less north of the copper lodes are the celebrated cobalt mines which have been worked in ancient times, and belong to some sayyids of Kamsar and Kashan. The rocks there are dolomites broken through by serpentines with an immense lode of iron ore, copper pyrites, sulphuret of nickel, cobalt bloom, and earthy cobalt (peroxide). The lode strikes north 7º west to south 7º east, and dips 80º west. Only the earthy cobalt is at present of any practical value; it contains about five per cent of metal. It is collected by the proprietors and washed with water, and the heavy sediment is made into cakes. The washing process is called saravabuna i.e. sar i ab va bun i ab (top water and bottom water). The cakes, under the name of lajverd i Kashi, are exported, principally to Kashan, Qum, and Isfahan, where they are sold at the rate of about one shilling and sixpence per pound. All the proprietors receive equal shares of the proceeds, and they have an agent (bonek-dar) who looks after the sale and keeps account for them for a commission. In order not [117] to? lower the price only a certain quantity, sufficient for the actual demand, about 1300 lbs. per annum, is put on? the market, and should there not be a demand for this quantity, the mine is closed and carefully guarded. The ore is reduced in the following way: ten parts by weight of the earth or ore (in cakes), five of potash (? ...iab), five of borax (bureh), are pounded together to a fine powder, and then made into a paste with grape ?cle (shireh), and formed into small balls or cakes. The balls are then put with pounded quartz into a sufar (?earthen pot with wide opening) and exposed to heat in a furnace for sixteen hours. The metal gained in this way ?amounts to about one twentieth of the weight of the cobalt cakes employed.

“To use the cobalt for colouring pottery it is ground into fine powder with an equal quantity of quartz. This ?powder is applied with gum under the glaze and is therefore called zir rang i.e. “under-colour”. For painting ?over the glaze the metal is ground together with forty times its weight of rock crystal or old glass (that containing ?manganese is best) and twice its weight of borax, and the mixture, in an earthen pot, is exposed to heat in a ?furnace until the whole of it is deposited on the inside of the pot as a crust of blue glaze like glass. The blue ?crust is then separated from the pot, and is applied, ground into powder, to pottery with gum. Both these ?processes are very costly, and, calculating the cost of the cobalt cakes, of the borax, potash, quartz, fuel, furnace ?, cost of the powder, ready for being used as colouring matter, amounts to 28 shillings per pound, by the ?first process, and to 38 shillings per pound by the second process. A cheaper way of preparing a blue colour is to ?ground one part of cobalt (metallic), with four parts of the cobalt cakes, into a fine powder, and applying powder ?under the glaze, but the colour obtained in this way is not good, and is employed only for very cheap pottery.”

Ladame found the same cobalt compounds present in Qamsar. One must therefore assume that Abu’l-Qasim’s “white silver shining in a sheath of hard black stone” was asbolane, though such a ?description fits the sulpharsenide of cobalt, cobaltite, rather better. On the other hand, the ash-coloured ?d soft type from Farangistan (roughly Europe) sounds more like asbolane, which is essentially wad, consisting of up to 40% cobalt oxide. The red kind is clearly erythrite, which has a pinkish-violet ?colour, and being an arsenic compound (hydrated cobalt arsenate) is indeed a deadly poison.

§9. It seems that muzarrad was the pigment used for all black underglaze decoration (cf. §§24 and ??). In Europe black is often made by mixing ten parts of manganese with one part of cobalt and sometimes a little iron. Wulff mentions a mineral mined in the area of Na’in, and used today in Persia for ?this purpose, which contains about 85 per cent chromite, 10 per cent manganese, and 5 per cent magnesium silicate. The identity of muzarrad will only be ascertained for certain by analysis of underglaze ?black on pieces of this period.

§10. Gold and silver marcasite probably refers to two varieties of pyrites, and yellow vitriol to ?copper pyrites. Maghnisiya means “manganese” throughout the text; magnesium is never found in ?nature uncombined and was probably not known as a substance until much later. Perhaps the terms “male” and “female” apply to different manganese ores. Arsenic was mined as realgar (AsS; red) and orpiment (As₂S₃; yellow). Mardasang here probably means litharge, but see §§13 and 19. Surmeh is collyrium, a black-coloured eye cosmetic, but it is impossible to say in the present state of knowledge whether the collyrium known to Abu’l-Qasim had an antimony or lead base (cf. §22). Tutty is calamine or zinc oxide. The production of tutty is described by a number of medieval authors: thus the late tenth century geographer al-Muqaddasi describes its production in Kirman province, and this industry was still going strong when Marco Polo visited Persia in the late thirteenth century.

§11. This is clearly a white rich clay, but whether it is kaolin is impossible to say at present. Kaolin certainly exists in Iran, and the most significant deposits so far found are at Simiron, south-west of Isfahan. Wulff describes a very fine clay, “so fine that modern ceramists would call it bentonite”, which is found in a village 50 miles from Isfahan and used by the stone-paste potters. The name of the village, according to Wulff, is Shahriseh, but whether this is located near the kaolin deposits mentioned above, which are also some 50 miles away from Isfahan, is not clear. However, the difficulty of manipulating the body which Abu’l-Qasim describes (§23) suggests that kaolin would not have been [118] sufficiently plastic to be included in it. Something like our white pipe clay or a fine ball-clay would have been much more helpful to the potter.

§12. The identity of the seven metals varied among medieval Arab and Persian writers, but included some or all of the following: gold, silver, copper, iron, tin, lead, and mercury. The ruins of the town of Bulghar are situated near the village of Bolgarskoye in the Spassk district, 115 km. south of Kazan, and 7 km. from the left bank of the Volga. The town was still flourishing in the early 14th century, and was in fact visited by Ibn Battuta in about 1332 AD. Hence tin must have been imported into Persia from the north, presumably by way of the Caucasus, as well as from the west (from Europe) and the east (from China).

§13. The existence of lead mines in Kirman province is attested by the anonymous writer of Hudud al-‘Alam and various authors note the lead mine at Kathah near Yazd. Rum here presumably means Anatolia. Mardasang is probably “dross of lead”, while red and yellow isrinj are the oxides of lead, red lead (Pb₃0₄) and litharge (PbO). White lead is lead carbonate. (See also §§ 19, 21, 22, 25.) “Uniform turquoise colour” must mean the opaque turquoise produced by the presence of both tin oxide and copper oxide in a glaze.

§14. “Red, greenish-coloured, soft-natured copper” indicates old pieces of copper which have started to corrode, and whose surfaces have become covered with cuprous oxide (red) and copper carbonate (green).

§18. The subject of medieval Persian weights and measures is a most confusing one. Ghazan Khan [dates] attempted to unify weights and measures throughout his empire by making Tabriz weights the standard ones. In 1926 a law was passed trying to equate Persian measures with the metric system, and at that time the Tabriz man was equivalent to just under 3 kilograms and there were 640 mithqals in one man. It is thus possible that Abu’l-Qasim’s man and mithqal correspond to about 3 kilograms and 4.7 grams respectively. According to Ghazan Khan’s decree, there were to be 260 dirhams in a Tabriz man, which would give a figure for one dirham of about 11.5 grams.

Two small ovens, probably used for producing glaze, were excavated at Siraf, and a modern frit kiln of more complex design used by potters in Maybod has also been published. Jawhar means “frit”.

§19. Tests on glazes by X-ray fluorescence have shown that lead usually appears in glazes of this period in conjunction with tin, bearing out this description of the joint production of both metal oxides. Both sirinj and sapideh-yi zanan (lit. “ladies’ white”) appear to refer to the mixture of the two oxides; presumably the second is a purer, more totally oxidised form. A similar technique of tin and lead oxide production is described by Ustad Ali Mohamed.

This paragraph contains a description of a second frit used by medieval potters (cf. the description in §18). From the body analyses carried out in Oxford, it seems that this second frit was not used in the making of ceramic bodies.

§20. The details of this paragraph are not at all clear.

§21. The sour grape husks would presumably react with the ashes to produce carbonic acid, which in its turn would react with the lead to give lead carbonate (white lead). In Europe in the Middle Ages, the process was different: earthenware pots containing vinegar and lead were embedded in fermenting tanbark or dung, with the result that acetic acid vapours from the vinegar acted on the lead and produced lead acetate, which then reacted with the carbon dioxide, heat and moisture from the tanbark or dung to produce lead carbonate.

[119] §22. Ithmid, though usually translated as antimony, may well in this case mean galena (lead sulphide). AI-Dimashqi, for example, writing at about the same time as Abu’l-Qasim, describes ithmid as coming from rasas stone and being powdery and sulphurous. Rasas presumably means lead here, hence ithmid means galena. Httle is in fact known about the rise of antimony as a colourant in medieval Persian ceramics. J. Boisier, in his chemical study of the ceramics excavated at Lashkari Bazar, showed that the usual pigment for yellow was an arsenic compound, although in one case an orange colour showed traces of antimony. The antimony pigment used for producing a yellow on pottery has to be a mixture of both antimony and lead oxide, since antimony requires a flux.

Tutty is mentioned here presumably because the ore of zinc, zinc blende, is almost always found in nature alongside galena. Abar or abbar appears to be another word for lead. Mardasang might here be translated as “massicot”, an obsolete English usage to describe the litharge used as a raw material for the manufacture of red lead or minium (shangarf).

§23. The recipe given here by Abu’l-Qasim produces a body closely akin to that of ancient Egyptian faience. Experiments carried out by Miss Rosemary Brewer, previously lecturer at the West Surrey College of Art and Design, showed that this Persian ceramic body is extremely difficult to throw on a wheel, and the skill of the medieval Persian potter in throwing such a wide range of shapes and such delicate ceramic bodies is all the more to be admired. The modern Persian stone-paste potter uses almost identical materials in almost identical proportions.

§24. This paragraph describes the pigments (“material”) used for underglaze painting, and the expression “painting on a ground of such-and-such a colour” means “painting under a glaze of such-and-such a colour”. By “white ground” is evidently meant “clear glaze”, and the colour green presumably indicates the turquoise variants so common in medieval Persian glazes (though it should be noted that Abu’l-Qasim does use a Persian word for turquoise, firuzeh, elsewhere in the treatise). As one would expect from surviving examples, the only colour used, according to our author, under a turquoise/green glaze was black. The compound made of bataneh, sa-ishkineh, frit and clay, constitutes the base for the colour - the chromophore (see §§4 and 5), and the hasa (quartz) must have been the basis of the paste used as a fixing agent for the chromophore; the two would be ground together before being painted on to the body of the pot.

§25. For dirham and mithqal see §18. The earlier manuscript’s figure for the ratio of manganese to frit to produce a black colour is the correct one. Shabeh is a black stone.

§26. Possible pieces of saggers were found at Takht-i Sulaiman, and kilns with rows of earthenware pegs are known from a number of sites e.g. Siraf, and Takht-i Sulaiman. The pegs found by ?Naumann were originally 55-60 cm. long and 5.5 cm. thick at the widest point. Such pegs are still used today. A dhira’ is apparently equivalent to about 41 inches, which makes these pegs about five feet long: this seems unlikely, unless the pegs went right across the kiln forming a complete grid. However, it seems that dhira’ was also used for the distance between the elbow and the tip of the middle finger, and hence could be somewhere round about 40-45 cm. This would point to pegs of 60-70 cm., which fits better with Naumann's excavated pegs.

The stipulation about smoke is a good rule of thumb for avoiding reduction in the kiln; the stoker would wait until all smoke had cleared from the previous charge before stoking again. Heavy reduction will tend to turn lead glazes grey or even black.

§27. The first sentence and the comment of the later scribe may be translated in another way: “Those that come out of the firing white are painted with the enamel of two firings, either lajvard or pure turquoise. [Or: Those which come out of the firing white are painted with the enamel of two [120] firings. Otherwise they are lajvard, pure turquoise, or translucent and do not need enamel painting.]” In neither case is the original sentence very clear. The most coherent comment is the alternative reading of the later scribe. From the rest of the paragraph it is clear that “enamel of two firings” refers to lustre painting, and could be rewritten more exactly (see §5) as “enamel used to decorate pots which have already been fired twice”. Allan, Llewellyn and Schweizer showed that all six of the lustre pieces of this period analysed contained tin oxide, and it is tempting to interpret “white” in this paragraph as meaning tin-glazed (not including of course opaque turquoise etc.). This would then fit with what we know of Il-Khanid wares: those with dark blue or opaque turquoise glazes, if further decorated, have overglaze colours such as described in §28, those that are translucent have underglaze decoration, and none of these groups will therefore require “enamel”, i.e. lustre decoration. The fact that Il-Khanid lustre wares often have glazes splashed with blue does not materially affect the issue.

This paragraph contains Abu’l-Qasim’s oft-cited description of lustre-painting. A little elucidation is necessary. Lustre requires iron, otherwise the silver and copper do not reduce properly, and iron occurs in this recipe in the marcasite and vitriol (pyrites). The red and yellow arsenics (realgar and orpiment) would, together with the marcasite and vitriol, supply the sulphur necessary to aid reduction, while the silver and the copper are the two essential ingredients. The point of adding sirinj (lead oxide) to this mixture is not clear and obviously it was not really necessary; zanjar (verdigris) would increase the copper content and thus make the lustre redder. The identity of shadanaj stone is not clear; perhaps it is some sort of copper ore.

§28. Qamsari stone is here haematite (see §6). A specially designed oven or furnace was obviously used for overglaze decoration, but its exact nature is unknown.

It is from the last sentence in the later manuscript that our term “lajvardina ware” comes, but it should be noted that the later scribe was confused and confusing about this point. The seven colour wares in the early manuscript, it is generally agreed, refer to what we commonly call “minai ware”, but it is inconceivable that they should have also been called lajvardineh - none of thern have the dark blue cobalt glaze, and the colour produced by lajvard (§8) has nothing in common with the white opacity of tin oxide or the turquoise of copper oxide. On the other hand, while there is no contemporary evidence for the use of the term lajvardineh to describe Il-Khanid wares with blue glazes and overglaze decoration in red, white and gilding, the use of the term by the later scribe may well indicate that it was a usage which had been handed down rather than invented during his lifetime. It is in any case an extremely useful term for historians of ceramic art.

IV. ADDENDUM

The complete Persian edition of Abu’l-Qasim’s book, entitled ‘Arayis al-jawahir wa nafayis al-atayib (Tehran, 1345) came to my notice after this article had been set up in type. Noteworthy textual variants vis-à-vis the German edition are noted below, with other relevant comments. Page numbers refer to the Persian edition.

§ 8. Cobalt ore was called sang-i lajvard because (pp.137-8) it was used with shukar-i sang and shakar to make gem stones in imitation of real lapis lazuli.

§§ 10 and 22. Surmeh is identified with ithmid (p.189), an opaque, black, shiny stone, the best of which came from Isfahan. It contained lead (p.190). Ibn Rusteh also mentions Isfahani ithmid, and later authors such as Tha’alibi, Muqaddasi and Dimashqi agree that high quality kohl was made there. Abu’l-Qasim’s account of the production of tutty (pp.188-9) adds little to that of Marco Polo.

§12. The seven metals are given (p.208) as gold, silver, copper, tin, iron, lead and khar sini. The Persian edition gives pareh-ha (“pieces”) for marha (“snakes”).

§18. The Persian edition has barnar for bariz. Wulff records bariz but does not state his source.

§27. Shadanaj stone (pp.190-1) is mined, is black bending towards red, and when engraved produces a red water like blood. Hence it must be the ferric oxide mineral haematite. Since haematite and roasted silver alone would not produce lustre, one must understand some such phrase as “in addition to sulphur and copper compounds”.

To read about the application of Abu'l-Qasim's methods of pottery production in modern-day Iran,
go to this article by Alan Caiger-Smith: "Esfahan: an unexpected pottery workshop"