"When I was seventeen, it was a very good year", sung The Kingston Trio from the radio in the dashboard, and I wholeheartedly agreed. Now I was eighteen, and it was an even better year. The summer of 1966 saw me in a small band of friends roving the higher latitudes of Scandinavia in an old Volkswagen Beetle, enjoying life under the midnight sun and the company of nice Finnish girls. No thoughts of rugs at this time. But I had been working on it. The year before, in Belgrade, I had acquired my first kelim.

Back home Hannah Erdmann was pulling loose ends together of unfinished works and was writing the foreword to a posthumous publication of her husband Klaus Erdmann, former director of the Islamic Arts Department at the "Dahlem Museum" as it was called, in Berlin (West-). The book was published in the autumn (Erdmann K, 1966, Siebenhundert Jahre Orientteppich,. Busse Verlag, Herford; English Edition: 1970, Seven Hundred Years of Oriental Carpets. Faber & Faber, London). It includes a chapter in which Erdmann undertakes a survey of the stock of oriental carpets in both Germanys that had survived WW II - not an inappropriate undertaking although it being twenty years after the war by now, considering that the Berlin Wall had been up only a few years, and cold war was going through an especially frosty period, restricting contact on most levels. One of the most outstanding pieces in this survey was the so-called "Synagogue Rug" in the Pergamon Museum in the East, being one of the oldest rugs in existence, being the oldest Spanish rug, dating to the 14th century or earlier and being the one and only known rug with a Torah Shrine motive. Before it was acquired for the museum by its founder Wilhelm Bode in 1880, it allegedly was kept in a Church in Tyrol/Austria. One wonders how it got there. A full account of it has been given by Friedrich Sarre, former director of the Islamic Art Department of the Berlin Museum (Sarre, F. & Flemming, E., 1930, A Fourteenth-Century Spanish Synagogue Carpet. The Burlington Magazine for Connoisseurs, Vol. 56, pp 89-95). The rug is still there and open to visitors in the redecorated exhibition rooms of the Pergamon Museum in Berlin, now hosting the reunited collections of both former Islamic Arts Departments.

The design represents a tree, from which rather thin stem branches spring off on either side, terminating in oversized blossoms, each representing a Torah Shrine.



I quote from the early description by Wilhelm Bode in his "Handbook" (Bode, W., 1902, Vorderasiatische Knüpfteppiche, Leipzig). These (blossoms) are of a remarkable form. In the centre is a closed door surmounted by a pyramid; on either side is a hook-like leaf, much conventionalised, and angular in its outline; the interior of these figures is filled with a variety of small birds, stars, zigzag lines and similar ornamentation.” The border according to Sarre contains purely decorative Kufic characters meaningless in themselves. Bode refers to the peculiar method of knotting - unlike any other Oriental carpets - discussed in detail in the mentioned article by Sarre & Flemming and which gave rise to the term "Spanish Knot" (see below). In the form of the blossoms growing from the branches of the "Tree of Life", Sarre and Flemming perceive a decorative rendering of the Jewish "Ark of the Law": "The Aron-Ha-Kodesch", or holy shrine, which was (is) affixed to the chief wall of the synagogue, and contained (contains) the Rolls of the Law, enjoyed from remote antiquity the profoundest veneration. It always has the same form - a rectangular chest with double doors - each with four recessed panels - surmounted by a rectangular gable-end.

A full-size coloured image of the rug is depicted as plate 91 in Felton, A., 1997, Jewish Carpets, London.

The authors support their interpretation by relating the design to similar representations in architecture and decorative arts of the period, as well as to historic predecessors: the decorative marble slabs on the Great Mosque at Cordova (10th c.); a 10th or 11th century capital in the museum in Saragossa (B); a stucco ornament from the Casa de Mesa in Toledo, 14th or 15th c. (C); the mosaic pavement of the ancient synagogue at Bet Alfa in Palestine (D); on a door-post in the ruins of the synagogue at Tel Hûm, the ancient Capernaum; on figured gilt-glasses of the first centuries of our era (F); a picture of the Ark from a Pentateuch manuscript, presumably Egyptian of the middle of the tenth century, preserved in the State Library at St. Petersburg (E):



The congruency of those designs of various origins and backgrounds is remarkable. It extends to details, which may easily escape one's perception, i.e. the three-piece construction of the floor in front of the shrine as depicted on the gilt-glass, and its very similar representation in the construction of the base of the blossoms in the carpet.

A few amendments on the observations and conclusion by Bode and Sarre seem appropriate.

It is a characteristic of the design of nearly all representations of the shrine that usually two or even three pentagonal constructions are boxed into one an other, as in the case of the carpet and the Pentateuch. As said, the Torah Shrine invariably comes with a pentagonal outline, many synagogues seem to have picked up this shape and show a pentagonal facade themselves, in this way demonstrating the principle. There also appears to be some confusion as to the significance of the Torah Shrine vs. The Ark of the Law or Ark of Covenant; the terms seem to be used interchangeably. The Torah Shrine is the most sacred space in any synagogue up to the present day. In it the Torah Rolls are kept from which members of the community read to the assembly. Usually the Torah Rolls are concealed from view by curtains. I don't know of any instance where a carpet has taken this function. Neither do doors seem to be a regular feature. On the other hand, not being a Jew myself, I have not seen many synagogues from the inside. The Torah Shrine on the gilt glass, looking much like a mixture of a cupboard and a temple, clearly contains objects that allow being associated with Torah Rolls. However, there are no rolls on the other images and, possibly, what looks like doors on them may really be something quite different.

The gilt glass dates to the 1st century AD, which is quite early in terms of the history of Torah Shrine as such. The representations on the rug and the Pentateuch and the Bet Alfa mosaic appear to stand in a much older tradition: that of the Ark of Covenant (5. Mos. 10, 1 ff). It has been suggested that the Torah Shrine is not the direct successor of the Ark of Covenant, after its disappearance from history in 6th century BC, although containing the Torah Rolls in much the same way as the stone tablets inscribed with the Ten Commandments were kept in the Ark, but the successor of the Second Temple. This could be the explanation for the "box principle" referred to before: in the synagogue (outer pentagonal box or shell) the Torah Shrine (second box/shell) modelled on the facade of the Temple (third box/shell), in it a representation of the Ark with two stone tablets inscribed with the Ten Commandments.

There may be another reading of the box principle. According to an entry in www.JewishEncyclopedia.com in the Rabbinical literature it is recorded "that Bezaeel made three arks which he put inside of one another. The outside and inside ones were made of gold ... while the middle one was of wood." This shows what it may have looked like in the fantasy of a later artist:



A long time after the destruction of the First Temple (Solomon's Temple) by the Babylonians in 586 BC and the loss of the Ark, representations of the stone tablets appear to have gained a new meaning and have become "doors", possibly the representative doors of the later Second Temple.

The following images show representations of the stone tablets in different contexts:



In the following part II of the original article, Ernst Flemming discusses the peculiar technique in which the carpet was being made (scan from the original article):


In the preparations for the "ORIENT" exhibition some years ago, the carpet had undergone restoration. Like some other early carpets, i.e. the "Ardabil" in the V & A (Victoria and Albert Museum) in London, it had been assembled from fragments of the same rug. This and other flaws had been put right in the course of restoration, slightly altering the carpet's measures.

Eder, D., 1980, Kaukasische Teppiche (and the English edition of the book, Bennett, I., 1982, Caucasian Carpets, have for a long time been among the cornerstones to our understanding of 19th and early 20th century Caucasian rugs. Therefore it is surprising that in the more recent book by Azadi, S.U., Kerimov, L., and Zollinger, W., 2001, Azerbaidjanisch-Kaukasische Teppiche, no reference is made to a group identified and described by Doris Eder, the so called Tachte Shirvan.



Tachte in Turko-Iranian language means ‘throne’, in a semantic as well as in a figural sense. For instance, Tacht-i-Sulayman  (throne of Salomon) is a hill in NW Iran 110 km west of Sandschan and northwest of Bidjar, carrying an excavation site of a once fortified city from the Sassanian and Mongol periods. The significance of Tachte Shirvan rugs rests in the peculiar large symbol on those rugs, for Doris Eder representing an Avar throne. The Avar were probably related to the Huns and had settled in an area between the lower Danube region in the west, and the north-eastern shores of the Black Sea. For a hundred years they appear to have been a military force, giving the early Byzantines much trouble. Those were the old Avar. There also are the younger Avar, settling in Daghestan to the present day. On grounds of their language, to some writers they are probably unrelated to the old Avar (www.wikipedia.org). The two or three rugs with an Avar label I have come across in recent years in the trade or in the internet seemed unrelated by design and left no lasting impression with me.

So, if it is not a younger Avar throne that is depicted on those rugs, might it be an old one? Probably neither. A Google based survey, following up a substantial number of entries on Avar, yielded no results that could be linked to this particular (textile) design (entries followed-up included data on a major Austrian Avar exhibition a few years ago). After this, it is not surprising, that the Avar throne attribution did not find resonance in the mentioned publication by Azadi, et al.

The Avar throne on those rugs very likely is a representation of a Torah Shrine or of the Ark of the Law (also: Ark of Covenant).

Comparing the main motive on the rug on plate 09 with the much older Torah Shrine or Ark images in part I of this salon, it becomes immediately apparent that all the significant elements are shared, i.e. the gabled roof over a pentagonal structure, the box principle, the threefold base or rest; the entrance doors appear as an arrow, expressing the principle of transcendence in a different image language (for a very interesting discussion, among other topics of the symbolic meaning of the "Sky-Door" see Cammann, S.V.R., 1972, Symbolic Meaning in Oriental Rug Patterns I-III, TMJ III/3 pp 5-54).

In the discussion of some of the rugs in the Zollinger collection Azadi, et al express their wonder on that strange and most peculiar symbol ("Das äußerst eigenartige Fünfeck", Azadi, S.U., Kerimov, L., Zollinger, W., 2001, a.a.o., p 362) in their high level technical analyses, but offer no concept. On one occasion they call it an animal-throne composition ("Tier-Thron-Komposition", a.a.o. p 332). Unlike Eder, who reserved a group-label for those rugs (Tachte Shirvan) they simply call the rugs by the name of their geographical area: i.e. Marasali, a village in the Shirvan area.

Doris Eder gives no indication as to who had introduced the term "Tachte-Shirvan", or whether she was the originator. I can quite see why "ruggies" - in the absence of a more appropriate interpretation - would have jumped on the concept of "tachte" as a throne, allowing imagination to run wild. There is another, less evocative and more technical translation of the word "tachte", that is "arch" or possibly "gable". As has been discussed, "gable" is of a particular significance here. Anyway, this "cooler" translation does not answer any questions. It is less distractive, though.

"Ark of the Law" - Rugs can be allocated to four different groups:

(1) This group features a repetition of the Ark of the Law, all else are decorative additions (plate 09). The     arrangement of the arks and the usage of space, although much less sophisticated, echoes the "Spanish Synagogue Carpet" in the Islamic Museum in Berlin. In that sense it is a "key rug" to me. If it had not been for its crisp, clean drawing that was giving me a deja-vu, perhaps the likeness to the motive in the Synagogue Carpet would have gone unnoticed.

It is not a common type, the only other one to this group I can think of passed through auction at Nagel's in Stuttgart in 1981 (lot 173, 291st auction).

(2) One of the two most common types, featuring the Ark, tablets depicting the ten commandments and figures who according to Azadi, Kerimov & Zollinger a.a.o. entertain in the group-dance, "Jally". If one has accepted that the main image is that of the Ark, the Jally-interpretation becomes unconvincing (as does an interpretation based on the assumption, those tablets were Muslim tombstones). I know a man who grew up in Baku and who spent much time in his youth with his grandparents in the Kuba area, where his grandfather was a rabbi. Now, in an official position, he is fostering the integration of Jewish immigrants from the territory of the former Soviet Union into the German ways of life - quite a delicate task in my opinion. According to him, the Jewish community does not dance Jally, it being a dance of the Muslim population only.

There is a special significance attached to those figures. If I got it right, ten men are needed for full prayer according to Jewish tradition. In most rugs of this type, especially the ones that appear to be older, ten figures are depicted. As far as Jally is concerned, no significance is attached to the figure ten.

Also, according to an entry in Encyclopaedia Britannica, once ten Jews (men) had come together forming a community, a Synagogue had to be built, a task that may have stretched the means of some - could a woven synagogue or ark have been an easy way out?

The following rug is from the Doris Eder/Ian Bennett book (plate 11). It is a splendid example of the type and illustrates the bondage of those figures to the Ark in a kind of "tilted image" (figure-ground-separation being the underlying perceptual principle). If you look carefully, you can identify the shape of the Ark between those figures. Their shape is determined by the representation of the Ark, which serves as a kind of "independent variable" in the process.


 And two more recent ones:



             

(3) This group is characterised by a very unusual integration of Muslim and Jewish  religious representations, the Ark under the gable of a prayer rug:


 
 
 

 

As to the origin of this type or the question, what may have prompted this integration, nothing has been heard or published (see further down for a discussion).

(4) Rugs of this type (plate 19) are also uncommon I believe, the one presented by Austin Doyle at the Keshishians's party and introduced to us by John Howe on Turkotek in January this year is the only one I have seen so far. The Ark seems to have been left out along with the tablets, only iconographic representations of human figures remain (see type 2). Only if one knows what one is looking for, in between individual figures the shape of the ark can be made out.


Here are a few hypotheses that may prompt further research and discussion:

1. Jewish converts to Islam may have kept the symbol of the ark as a token of  their ancestry. It may or may have not lost its religious significance over the centuries. According to some (www.wikipedia.org), the Tat may have been Jewish altogether before the majority (were) converted to Islam at the time of the Mongol rule. The so-called 'Mountain-Jews" in the Kuba and Daghestan areas speak a Judeo-Tat language related to Tat language of the Muslim Tat (www.wikipedia.org). According to Felton, A., 1997, "Jewish Carpets", the language of the "Mountain-Jews" is "basically Persian written in Hebrew characters"(p. 27). There is some saying that the Mountain Jews are descendants of the Babylonian captives, led away at the time of the destruction of the first temple in Jerusalem and the loss of the Ark.
   
   There were later incursions and persecutions by Christian-Orthodox powers before and after the "Russification" of the region that may have prompted Jews to look for shelter under the umbrella of Islam. An old formula of survival in the Diaspora comes to my mind: "Where you can't get over something, you may always try to get under" (Singer, I.B., 1977, A Crown of Feathers and Other Stories. Penguin, Harmondsworth, England).
   
   
2. Besides the Ark, all four groups of rugs share design and technical characteristics of other rugs in the Shirvan-Kuba-Daghestan and Karabagh regions. This seems to coincide with the (earlier) distribution of the Jewish population. Also, quite a few Kurdish communities exist, Kyurdamir in the Shirvan area (Marasali not being far off) being a district capital. The Kurds, being amongst the most prolific weavers wherever they settle, may be responsible for these rugs; some of them may be or may have been Jewish at the time those rugs were made, or long before. In any case, there is a lively Jewish community in Kyurdamir.
   
   According to a dealer from Tabris in Iran, there, rugs of the type as shown in plate 09 he claims, are called "Kurd Kasak". It makes one wonder.
   
   
3. We have all learned at some time or other that "more sophisticated" designs were handed down from court to village and tribe; an assumption that may be right as often as it is wrong. However, if one wants to look out for a court that could be given the credit, there was the Khanate of Khazar. The Khazar settled from around 600 AD on north of the Black Sea and the Caspian, later expanding and becoming the major force in the area, a buffer state between Christian Orthodox powers and expanding Islam, collecting tributes from most of the Caucasus. Those people were Turks, they were Jews and had put up an oasis of civilisation. One of their rulers, Joseph, was in correspondence with an high ranking Jewish official at the Moorish throne in Spain, serving two subsequent califs, Abd-al-Rahman III (912-961) and his son Hakam II (961-976). The correspondence survived: The Letter of Rabbi Hasdai, Son of Isaac Ibn Shaprut, to the King of the Khazars (http://www.fordham.edu/halsall/source/khazars1.html).
   
   We do not know which exact route the letter went. But most likely it would have travelled on the ancient route that has seen much traffic in the course of the millennium before, the northern branch of the silk-route, crossing the Caucasus at the Pass of Dariel - Pompey once left a garrison there (Stark, F., 1966, Rome on the Euphrates, John Murray, London) - reaching for the Pontic at Trapezunt. Where letters could travel, there may also have been space for gifts, for rugs and perhaps drafts or blueprints of designs in decorative arts?
   
   Eventually, the Khanate was overrun by Djinghis Khan, and its people dispersed over a wide area. According to Fenton, A., 1997, Jewish Carpets, some settled in the western Trans-Caucasus. The way down along the Caspian shore and across the Albanian or Derbend Pass (Albania was the name of this country at the times of the Romans) to the eastern Trans-Caucasus would have been easier. Anyway, once across the pass, below the Khazar refugees would have spread the valleys of the Phasis and the Cyrus rivers, linking the Black Sea and the Caspian Sea. In Roman times it would have been a four days journey on a paved road, according to Freya Stark. If they travelled east via Tiflis, as some undoubtedly would have done, down the Cyrus (Kura) and Araxes (Aras) rivers, they probably would have met others Jews already settling there, and still others to arrive at a later time.
   
   The gathering and cohabitation of those groups, some of them Kurdish, Armenia just behind the next range of mountains, would have created - and indeed it has created - an ideal breeding ground for rug production, a kind of "Silicon Valley" for rug technology. All requirements needed were at hand, good sheep, water, plants and minerals for dyes, famous already in Roman times.
   
   
4. When the Spanish Jews (and Muslims) were forced to exit Spain by Royal Decree in 1492, many took up the Sultan's offer to choose Ottoman Turkey as a safe haven. Others settled in Kashan, Persia, where they inspired arts, philosophy, science and medicine (Azadi, S., by word of mouth). Whilst the Caucasus at this time was not exactly part of the Ottoman empire, the Ottomans were the most influential if not controlling power in the eastern Trans-Caucasus. Jews had already been settling in that area for centuries and it seems likely, that some of the Sephardim refugees should have joint them there, bringing with them their own skill and knowledge.
   
   
5. In view of those hypotheses it seems likely, that the rugs were not actually made by just one group. A fair amount of exchange must have existed, this including Muslim neighbours, with whom relations were by and large on the relaxed side, sometimes even cordial. Is it thinkable, that Jews have adopted the idea of prayer rugs from their neighbours, the Ark under the gable oriented towards Jerusalem?
   

Sometimes rugs with the ark come from somewhat further afar.  Lot 164 at a Nagel auction on 16th May 1981 may be counted among those - a Fachralo Kasak:



The Eder/Bennett book includes several prayer rugs of the kind in group three from the Karabagh region (plates 106, 109,110):


Another rug from somewhat further afield would be plate 58 in Jewish Carpets - a dated Jewish Sarab Runner. Looking very carefully at the content of the white medallions, most apparent in the lower half of the third medallion from bottom, the shape of the ark appears in yet a different form but clearly related to what we have looked at before.



 Earlier this year in Istanbul I had a look into a rug dealer's library whilst visiting his shop. Having to deflect his enticements aimed at making me buy a silk-wefted Shasavan cicim, I was somewhat distracted and forgot the title of an old book that I thought interesting. In it was the picture of a rug with the Ark as the main motive in the style of a Sarköy kelim with curvilinear wefts. It was attributed to "Bessarabia".

We are coming to the end. Not all is lost, the woven Ark has been restored to memory. Some riddles remain, but we know better now what we are dealing with, field-research seems desirable.

Somewhat disturbing news reach my ear, of nuclear ambitions of a neighbour state, of military build-up and strategic airfield expansion in Kjurdamir, on the Aspheron peninsula and elsewhere, Azerbaidjan, the old buffer-state in the thick of it, anxious to define its role. It is not an easy time for the region. Was it ever?
 
The Kingston Trio is seldom heard on the radio these days. I might as well turn it off here and hope, it is going to be another good year, after all.