About the Antique Rugs of the Future Project

Sheep Breeds of Azerbaijan

Sorting, Washing, Carding, Spinning

"The advantages of handspun yarn to machine spun yarn"

Rediscovery of Ancient Natural Dyes
Our Natural Dyestuffs


Difference between synthetically and naturally dyed rugs

Weaving and Finishing Steps

Galleries of ARFP Caucasian Azerbaijani Rugs



The Marby rug, "Peacocks on either side of the Tree of Life", Turkey. XIV-XV century. Location: Museum of National Antiquities
Stockholm, Sweden (Statens Historiska Museum). Size: 109 x 145 cm. (112x160cm according to the other source)


Material(s) / Technique(s): Wool; knotted.
Date of the object: Hegira 700–823 / AD 1300–1420
Period / Dynasty: Ottoman
Provenance: Anatolia, Turkey.

Description: One of the few examples surviving in Sweden from a tradition of carpet–making in the early Ottoman period in Anatolia. In 1925, the so-called 'Marby rug', was found, cut in two pieces, at the abandoned church of Marby, a village in the province of Jämtland. Belonging to the group of animal carpets, it shows red, stylised birds standing symmetrically on either side of a tree set within octagons on an ivory ground. The birds and tree motif has been long known and very common in Central Asia.

To judge from their presentations in Italian paintings of the 14th century, animal carpets with a tree flanked by two birds already seem to have been popular at this time. However, they reached the peak of their production and circulation during the first half of the AH 9th / AD 15th century. The provenance of this rug from a village in Jämtland in Sweden is important as it shows that the export was not restricted to Italy, but also reached the Baltic region. Animal carpets disappeared towards the end of the AH 9th / AD 15th century.

Apart from the Marby rug, some fragments with the motif of birds flanking a tree from Fustat in Egypt are known and there are two completely preserved examples. The first was found in a church in Italy and is now in the Museum of Islamic Art in Berlin, Germany, and a second was acquired by the Metropolitan Museum of Art in New York in 1990.

How object was obtained: Bought by the Museum of National Antiquities in 1925.

How date and origin were established: The rug was dated by radiocarbon (Carbon-14).

How provenance was established: The patterns of the border and the guard strips are characteristically Anatolian; the guard strips also appear in another group of early Ottoman carpets discovered in the 'Ala al-Din Mosque at Konya, Turkey.


Original Owner: Church of Marby, Jämtland
Museum Inventory Number: SHM 17 786

Selected bibliography:
Ådahl, K., Den Orientaliska Mattan i Sverige, Stockholm, 1998, cat. no. 1.
Erdmann, K., The History of the Early Turkish Carpet, London, 1977.
King, D. and Sylvester, D., The Eastern Carpet in the Western World: From the 15th to the 17th Century, London, 1983, cat. no. 2.
Lamm, C. J., Carpet Fragments: The Marby Rug and Some Fragments of Carpets Found in Egypt, reprint, Uddevalla, 1985.
Sterner, M. and Kind, J., Antika Orientaliska Mattor i Sverige, Stockholm, 1937.
Citation of this web page:
Friederike Voigt "Marby Rug" in Discover Islamic Art. Place: Museum With No Frontiers, 2014.


Warp: goat's wool. To 1 dm. about 50 to 60 threads (=12 to 15 to the inch).
Weft: strong red, twisted threads, 16 to 18 to 1 cm. Between the rows of knots generally two shoots; in one case 5 shoots of pale red, untwisted thread.

In the end-webbing, two twisted or untwisted threads taken together, 8 to 10 shoots to 1 cm. Knotting: thick woollen thread, two-fold. To 1 dm. about 30 to 54 rows of knots (= 7 to 8 to the inch), in the length, and about 22 knots (= 5 to the inch) in the width. Knotted over 2 warp threads. Scheme 1b. The pile thick, short, upright, unevenly cut.

Colours: the ground yellow, the inner and outer contours of the inner field red; the spaces filled in yellow, green and blue, the small triangles in the pattern brown. The borders of the octagons blue and brown: in the main border a brown pattern on a yellow ground; in the border-stripes white pattern with blue triangles on a red ground.

State of preservation: the carpet has been cut into two; it has now been put together again, and a narrow stripe between the two panels has been lost. Small injuries; a hole at the upper edge of the lower field. (The information for this description has been given by Miss Vivi Sylwan of the Röhsska Konstslöjdmuseet at Gothenburg in Sweden.)

Design: The inner field has, like the carpet in the Department of Islamic Art at Berlin (Plate 1), two large oblongs, approximately alike, converted into octagons by cutting off the corners. Each of these octagons is filled with the same design, consisting of two confronting birds in a tree. The only variation is that the bird to the right in the upper panel (green inner space with a brown triangle) is to the left in the lower panel and the other two birds are also alike in colour. This ancient motive of the bird and the "tree of life" is treated in the same geometrical fashion as the contest between the dragon and phoenix in the other carpet. As in that instance, the bodies of the creatures are transformed into spaces bordered by straight lines and filled with colour, with very little resemblance to nature. The conventionalisation of the feathers into hook-forms is noticeably similar. The low, wide-spreading top of the tree is changed into a symmetrical system of triangles, a horizontal stem and two angular spirals. For decorative reasons it is made to consist of an upper and lower part, both alike, joined by a stem on either side of which the birds face one another. The bordering stripe of the octagons is patterned in hook-shapes in blue and brown, in counterchange fashion. In the corners there are rows of red hook-forms on a green ground, and a yellow triangle with a rectangular projection in the colour of the ground.

The border is in three parts. The main band has a brown lozenge-pattern on a yellow ground, not adjusted at the corners of the carpet; each lozenge is divided by a thin brown diagonal stripe into two yellow triangles, each containing a spiral in the angular convention seen elsewhere in the carpet; the original form of this pattern is not clear. The inner and outer border-stripes are alike: a red ground with a pattern of double spirals, from which half-palmettes branch off, angularly conventionalised and set in a row. In the angle between the palmette and the spiral a small blue triangle is placed.

Professor Flemming has had the kindness to compare this carpet with the carpet in the Department of Islamic Art. The result of his investigations is as follows: "On the basis of the very precise information of Miss Sylwan about the carpet from Marby I have very carefully investigated the material and nature of the threads of the similar carpet in the Kaiser-Friedrich-Museum, and on account of the absolute similarity of the setting up of the warp, of the closeness of the wefts, of the knotting and of the material, I am quite convinced that the two carpets are not only of the same age, but also that they must have come from the same workshop. There are no essential differences. The deviations are so slight that they serve rather to confirm this view than to overthrow it. The peculiar knotting-in of thread-ends 7 to 9 cm. long at the back of the carpet from Marby led to a careful examination of the carpet at Berlin, when there were found on the back of that carpet, but only after the closest investigation, the remains of the red yarn of similar rows of knots. The ends of the threads probably at one time hanging down, are cut right off, so that now there only remain the parts knotted into the warp threads between the rows of weft. These knotted threads on the back differ from one another in the material and the method of insertion only to this extent, that in the carpet from Marby a thinner yarn was used than in the carpet at Berlin. Also the knottings are on four warp-threads (sometimes three) in the carpet from Marby, and on six warp-threads (sometimes five) in the carpet at Berlin, as may be seen in the diagram over leaf. These knottings are not always on the same warp-threads, but are shifted in irregular fashion, in order that the hanging thread-ends might cover the back evenly. On the other hand the intervals in these rows of knots are approximately the same in both carpets. In the carpet from Marby they come after 7, 8 or 9 rows of knots at the front, very much as they do in the lower border of the carpet at Berlin; while in the rest of the latter carpet they come mostly after fifteen rows.

The object of these knottings at the back is not quite clear, and I have not hitherto found them in any carpet, although I have examined a large number of old examples. I only know a similar arrangement in the Coptic stuffs with a looped surface. On this account I find striking evidence in favour of a similar origin and a similar age for both carpets. Assuming that the date of the Berlin carpet is established and recognised beyond dispute as the first half of the 15th century, the Marby carpet is to be dated likewise. I cannot in any circumstances agree with a dating in the latter part of the 17th century. Unfortunately the selvedges and the end-webbings are entirely wanting in the Berlin carpet, and these are so extraordinarily characteristic in the Marby carpet that they would probably have confirmed still further a similar origin and age."* The result of this technical examination, as well as the stylistic resemblance of this carpet to the carpet at Berlin, show that it is a second example of the Anatolian animal-carpets, formerly only known from representations in Italian pictures and in the example in the Department of Islamic Art at Berlin (Plate I). The representations in Italian pictures cited below, show that the motive of confronting birds in a tree is known in the range of these carpets (for example, confronting birds: Fra Angelico, "Virgin enthroned", Academy, Florence. — Birds back to back in a tree, the top of which is similarly conventionalised: Lippo Memmi, "Madonna", Kaiser-Friedrich-Museum, Berlin. — A similar conventionalisation of animals (birds?) is on a carpet in an "Annunciation" by Carlo Crivelli in the Stadel Institute, Frankfurt-on-the-Main), That examples of these Anatolian animal-carpets reached the Northern-European lands is shown by this, that they have been imitated in northern carpet-weaving (see H. Grosch, "Gamle Norske taepper", Berlin, 1889. — Riegl, "Altorientalische Teppiche", Leipzig, 1891, fig. 5. — Martin, "A history of Oriental carpets", pp. 142—144).

The carpet was first published by Miss Vivi Sylwan, "En orientalisk matta", Ur Fornvannen 1924, pp. 106—119.

* With regard to Professor Flemming's very interesting technical note, a long woollen nap is occasionally found at the back of old Oriental carpets, presumably for extra warmth in cold districts. The translator has elsewhere ventured to give a word of caution in regard to the dating of carpets with archaic patterns. Comparisons with representations in pictures are sometimes a surer guide to the antiquity of the design than to that of the actual carpet.