Home Page Discussion

Salon du Tapis d'Orient

The Salon du Tapis d'Orient is a moderated discussion group in the manner of the 19th century salon devoted to oriental rugs and textiles and all aspects of their appreciation. Please include your full name and e-mail address in your posting.

Between ‘Classical Carpets’ and the ‘Pazyryk Rug’:  A Sassanid Era Nestorian-Christian Design and its Transformations

by Horst Nitz


A flat-woven rug unearthed by the author in the far southeast of Turkey some time ago has been identified as to its origin with the Mountain Nestorians in their former retreat area in the High Kurdish Taurus (1). The central medallion of the rug bears an un-iconic, non-idolatrous representation, a composite symbol of Jesus Christ that represents a stylistic tradition as it has evolved in the early days of Oriental Christianity. While the rug has not been C14-dated and the actual age of the textile remains uncertain (2, 14), it appears to carry the key to a new research area, i.e. the textile  heritage of the Church of the East (3), the great missionary church of Asia during late antiquity and European Middle Ages.

The author proposes that early Nestorian Christians (4) were functional in the creative process and in the distribution of textile patterns in Asia; that this took place in the context of their missionary activities in the early centuries of the new religion; and that this process is traceable in the textile heritage of the nations that lay on their missionary path.

 A first reconnaissance focuses on the region between Mesopotamia and the Caucasus, encompassing Anatolia east of the Euphrates and the Roman-Persian border running roughly at 190° from the eastern shore of the Black Sea, and Azerbaijan to the east. These areas had traditionally been or had come under Persian suzerainty and Nestorian influence in the first centuries AD. Also, the progress of the mission in this region is fairly well documented, which provides a raw matrix of the distribution path of a particular  rug design and its subsequent transformations.


The discovery of the Pazyrky rug by S. I. Rudenko in a frozen tomb in the High Altai in 1949 left Kurt Erdmann in a dilemma. The prevailing concept with the rug world that he had formulated, of Turkish pastoral nomads as the originators, and their westward stride as the force behind the migration of pile rug designs and technique from Turkestan into Anatolia did not allow for an ancient pile rug showing up as far northeast as Siberia, and neither for one originating as far southwest as Asia minor and Azerbaijan as Rudenko suggested. To Erdmann, Turkestan was the cradle of pile carpets, and the Seljuk had introduced them into Muslim culture (5). He must have been aware of the Pazyryk rug’s potential to invalidate his theory. Apparently, without having seen the rug himself, in a somewhat winding statement he argues that the Pazyryk rug can be no pile rug, and that it must be a cut-loop fabric (Erdmann 1975, page 13). In spite of such inconsistencies and implicit fundamental errors in Erdmann’s theory it is still going strong with the rug world (6).  If this is so for want of an alternative model of developmental rug history, there is good news. It centres on a rug design commonly known as the ‘Gashgai Göl’, which is not what the term suggests. In fact, it is a transformation of the probably oldest known complex rug design after the Pazyryk carpet, and its geographical origin may be very near to that of the most famous rug of all. This will be explained in more detail in the following chapters.

The literature on Caucasian Carpets is extensive (7).  Generally, it seems agreed upon that the oldest rugs from the Caucasus region can be dated to the16th century and belong to one of the ‘classic’ groups, of which the ‘dragon’ carpets are probably most prominent (Opie 1992). However, it has also been argued that there may not be a clear enough distinction in every case between Caucasian Carpets and such from neighbouring Anatolia and NW-Iran, and that there may even exist some older Caucasian carpets that have been attributed to those neighbouring weaving cultures (Azadi, Kerimov and Zollinger 2001). In any event, this is about prime carpets made in workshop with a likely attachment to a local aristocratic court or similar. The situation with village carpets is quite different. Not only that they account for the far majority of rugs, they also are considerably younger, mostly dating from the 19th century. However, questions have arisen, whether the motifs in those ‘folk’ rugs may not be much older than the designs of the ‘courtly’ carpets (8).  This question mark may be attached with equal justification to the neighbouring regions in Anatolia and in Iran, and it is going to guide us in the exploration there as well.

Almost thirty years after a rug reconnaissance tour by the author into the far south-east corner of Turkey, that at the time had been considered a failure, it had turned out to have unearthed a rug, which seems to be the key to the understanding of important early developments in the genre.

In the autumn of 1980 that tour came to a premature halt at a military post on the road from Van to Hakkari (9). Back in Van an opportunity arose to look at a number of old flatweaves attributed to an hitherto unheard Kurdish group, that of the Herki. According to local informants, the textiles had shortly before emerged in connection with a military operation on the frontier to Iran and Iraq, the target area of the intended survey. One of the pieces especially resisted all attempts of attribution within a Kurdish pattern catalogue (10).

Historic sources of the late 19th century relate that the assumed weavers of the rug, the nomadic Kurdish Herki, on their annual migration between summer (Turkey, Iran) and winter pastures (Iraq) regularly crossed Nestorian settlement areas and committed robberies and other acts of violence on the sedentary population (11). Since they were forced into a sedentary life by Turkish and Iranian authorities on their respective territories in the 1930s, they have been dwelling in the Turkish-Iranian-Iraqi triangle, the former settling area of the mountain Nestorians (12).

The textile is at hand. It is a tightly woven, carefully executed piece of work measuring 75 x 172 cm, is in a good condition and appears to be the left one of originally two flatwoven panels of the sumac type (13) that were stitched together along the middle. Warps and wefts consist of wool and goat hair. The age is uncertain (14). In colour scheme and style (general layout, secondary motifs, borders etc) it resembles a number of rugs dated to the 13th - 16th centuries in the keep of Istanbul museums Vakiflar and TIEM as well as other international collections (15).

At a first glance the rug appears to be laden with symbolism of a complex and unaccustomed kind. This gives way to a sense of understanding once assessment is carried out within the cultural and historic context of Northern Mesopotamia and of the special Nestorian Christology and liturgy: the two natures of Christ; Mary as birth-giver but not mother of God as in other churches; Christ’s assumed true presence in the Lord’s supper ceremony etc.) all is represented in symbols. The complex representation of Jesus Christ on the white-grounded medallion (16) in the form of the construction rhombus of the 'vesica piscis' refers to the early Christian period and a late-antique period international style; also a marked old-testament influence is apparent, that has as yet not been unravelled in all aspects (17). Antithetic mighty horns as attributes of the divine make use of an ancient image language that places the rug firmly in a Mesopotamian tradition (18). 


The early Christian period in Northern Mesopotamia has not left much evidence of its art. This makes comparable objects extremely scant. Although no exact equivalent to the complex symbol of Christ in the rug is known, obvious parallels of a related content and / or with similar formal solutions exist in some works of art. We are lucky that we have the ascension scene in the gospel book of Rabbula, 586, that has survived and is now kept in the Medici-Laurentian Library, Florence. The book is written in Syriac (East-Aramaic) language and contains magnificent illustrations that bear witness to specifically Syro-Mesopotamian art (Zibawi 1995). However, to make things as complicated as indeed they are, it needs to be mentioned that the book has a (West-) Syrian Orthodox context from which the Nestorians were divided by schism. Two other objects are from further afield: a 9th/10th century gilded silver plate from the Talas river valley in Kyrgystan, that originated in a Nestorian context, now in The State Hermitage Museum St Peterburg (Baumer 2006); and Ratchis Altar, 8th century Lombard, Christian Museum Cividale del Friuli, Italy. The ornamental border of the altar is made up from s-forms that are typical for textiles that share the geographical context of the rug. As textiles, a number of Egyptian Coptic roundels with themes in the tradition of the Alexander myth exist. They were dated to the 8th/9th centuries; some may originate elsewhere (Syria), with one showing Sassanian influence (Berliner 1962, 1963; Herzfeld 1927).

Such references across church boundaries may be indicative of an early Christian international style with its roots going back to a time before the great schisms (also see endnote 11).

This may also be said of another element in the composite form, which has not been mentioned before, i.e. those dangerously looking angled jigsaw forms within the 
vesica piscis’. They may contribute least to the importance of the overall symbol, but have taken longest to identify. For some time, it was considered that they are derivates of the Akkadian god Shamash’s emblem, the arc-shaped pruning saw. Shamas was also worshipped in Assur (19). He is also identifiable by the rays emitting from his shoulders, which is quite appropriate for a sun-god (Black & Green, 1992 p.182-184). The white field of the rhombus, the ‘vesica piscis’ was thought to express the same in an image language better adapted to the technical requirements of the medium carpet and as a result of a transformation process from an earlier cultural context into a later one. Maybe Christ has inherited the symbolic white colour of the sun at midday from Shamas together with the Mesopotamian type mighty horns as a symbol of divinity. However, the angled jigsaw forms may also represent a simplified palmette motif. In a more naturalistic form, such palmettes are flanking the figurative busts of saints in a chapel LIV fresco at Bawit / Egypt (Zibawi 2004; page 74). This may be another indication of the existence of an early Christian era international style. For the time being, the question has to be left unanswered, whether it is the palmette or the pruning-saw, and which one precedes which (if any).

The objects in the comparison discussed so far are more or less naturalistic interpretation of their respective theme, including the Talas valley plate from a Nestorian context. Where now can the composite symbol be positioned in relation to those and other major art works of the wider region, i.e. Dura-Europos (Synagogue, House Church) and Edessa / Urfa (Gospel of Rabban) on a scale of figurativeness vs. abstractness? The traditional classification rule - Christianity: images; Judaism and Islam: no images – has recently given way to a more flexible approach as it has been advocated by modern scholars. However, the Oriental Orthodox Churches i.e. Nestorianism have not been in the focus of this discussion – although they ‘play on the same grounds with Judaism and Islam’. Any answer to this question should address the social situation in the first few centuries of the Christian era as well. There were Jews that had turned to Christianity but maintained their old ways to varying degree; others had submitted to baptism and had given up Judaism altogether; further, there were the religiously homeless, outcasts of a Jewish society, deported tribes of captives, resettled populations, stranded soldiers etc. When these circumstances are taken into account, a broad spectrum of interpretations and practices of the Mosaic ban on images seems probable. The ‘You shall not make for yourself an idol’ offers some leeway that can be used or not: the representation of epic scenes or everyday events do not necessarily meet the criteria of idolism - or challenge them. The Jews of Dura-Europos on the fringe of one empire and within the reach of another, may have felt that they want to accept the challenge thrown at them by the Christians (who had just before decorated their church with frescos – more correctly, temperas) and show what they can do themselves. The amount of money that has gone into the project seems to speak for such an interpretation. How Jewish of a strict type were the Christian originators of the symbol of Christ that is being discussed and, were they the predecessors of those Mountain-Nestorians in the Hakkari region? For the first US missionaries to reach them in the 1830’ies, they were the ‘protestants of Asia’ because of a complete absence of images from their churches (Perkins, 1834; Grant 1835). Both authors address the question of descent, because they were impressed by the observation that the Nestorians can converse with the Jews in the region in a common language and that they follow religious practices, that are described in the old testament and that are unfamiliar to other Christians. The Nestorians, that Grant has lived among, described themselves as descendants of the Jews that were taken into Assyrian captivity (20). This could mean, that the Hakkari Nestorians may have been made up from the deported Jews that were settled in or near the mountains in Assyrian times and had adopted Christianity early, and later Christian refugees indigenous to the Assyrian plain who moved up into the shelter of the mountains in the 14th and 15th centuries, when under pursuit by Tamerlane’s army.

Expatriates often hang onto traditions of their old home country with verve, and foster conservative ideals. This may have been the case here, with a mixed fraction of Jewish descendants at the starting line of Christianity, that abode to the Mosaic law more fervently than their brethren at an intersection of major trade routes at the great border town of Dura-Europos, where rules may have been somewhat looser. Christians seem to have benefited as well in that they were tolerated and could practise their religion rather openly, whilst elsewhere in the Roman Empire – we are still writing the polytheistic era – they were persecuted.

The abstract nature of the composite symbol in the rug may also be due to a tradition even older than the Dura-Europos paintings, which are dated 235-244AD. A nomenclature offered by Baumer (2006) knows three phases within the orthodox community, (1) an emerging allegorical phase towards the end of the 2nd century, ‘in which animals and objects such as the lamb, the dove, the fish, the vine and the anchor referred to symbolically suggested figures. Then the first depictions of people appeared around the mid-third century in the catacombs, where Old and New Testament figures and the Good Shepherd could be seen on the walls. The murals of the Christian chapel of Dura-Europos date from the same period… At the beginning of the fourth century the depiction of Christ, the apostles and the saints began to become more widespread; but it remained controversial’ (p. 164-165). If going by Baumer’s nomenclature, it would be phase one for the rug.

Supporters of images like the early Cappadocian fathers thought that pictures might help to educate the ‘spiritually poor’ and referred to an invisible spiritual reality emanating and passing on from them (Baumer 2006; Döpmann 1991). If this goes for images, why not for symbols as well? It would have strongly recommended rugs for missionary work sic! Ease of handling and availability of materials to reproduce such woven symbols and multiply the message could be added to the list.

Concerning the function of the rug one may first of all think of the Lord’s Supper ceremony (Eucharist). To Eastern Christians the Eucharist is not just symbolic, it is the central part of the service to them, as they believe it truly becomes the real body and blood of Christ, and through their partaking of it, they see themselves as together becoming the Body of Christ, that is, the Church. In one of the liturgies used by the Mountain Nestorians, as Percey Badger observed, at the hight of the ceremony ‘the priest shall fold his hands upon his breast in the form of a cross, and shall kiss the centre and the two horns of the altar’ and by doing so physically wellcomes Christ to the assembly (Badger 1852, p. 234). In the absence of other two horns on or at the altar, this tells us where the rug has probably served (some of the lent cloths’ and altar blankets at Kloster Lüne in Northern Germany echo this usage in form and function).

However, rugs of this type may have had an additional function as a woven catechism, carried by monks and traders on their journeys to far-away and illiterate societies. The ‘facilitators’ may even have been illiterate themselves. Textile backed visualisations would have crossed language barriers easily, and teaching the technique how to do them should have been an effective prompt in opening hearts to the new religion - of the ‘spiritually poor’ on the missionary path beyond frontiers. Redrawn, expanded, simplified or otherwise processed in tune with the tides of history, in any case mostly unrecognised and somewhat ruffled, the central motif of the rug has survived and made or makes an appearance on textiles of a wide array of people from Anatolia to Central Asia (21).

If the rug had passed through the arts trade, it almost certainly would have been attributed a Qashqai rug from the Iranian south-west province of Fars, as has happened in a known instance (see plate 11). The symbol of Christ appears in a more or less digressed form and in many variations on rugs from that region. It is generally assumed in the rug literature, that the Qashgai Federation was founded on the order of Safavid ruler Shah Isma’il (Housego J, 1978) some time after the Persian defeat by the Ottomans at the battle of Chaldiran, or by Shah ‘Abbas. The federation was made up from tribes that had settled near to what after Caldiran became, and still is, the border with Ottoman Turkey, before they were expatriated south. However, the literature that specializes on the Abbasid era or on the Qashgai does not seem to know about this expulsion (Boyle J A, Marsden D J, 1976; Newman A J 2006; Canby S R, 2009). The Christ symbol, in its digressed form on rugs from the southern Fars province is usually termed ‘the Qashgai göl and is thought to have migrated south with Turkmen members of the newly founded federation.

This is the frame of reference for Eagleton (1988; plate 63) who wonders how the Qashqai emblem may have found its way into a rug from the remote Barzani area in the Kurdish Taurus (in the neighbourhood of the Mountain Nestorians). Luckily, the textile featuring in this paper could be secured at or very near to its origin. This made it possible to build hypotheses on fresh facts and not on preconceived ideas (22).

The Advance of Christianity in the Caucasus

The mission progressed from the south via Armenia, that at the time stretched much further in all directions; and possibly, through antique Albania. The beginnings in Armenia are thought to reach back to apostolic times, but this may be a biased account (Hage 2007).  A first bishop with an Armenian name being mentioned is a certain Meruzhan of around the middle of the 3rd century. The foundation of the church is recorded a few decades later as having been affected by king Thrdat IV the Great (Tiridates, 298-330), who had spent his youth in Rome. Doing as he had experienced there, back home he suppressed Christianity during the first years of his reign.  His attitude changed with Grigor Lusaworisch, the Enlightener (257-321), who had come to Tiridates from Caesarea in Cappadokia (Kayseri) and baptized him at around 313 or 314 according to Hage (2007).  In the account of the Armenian Church, this baptism happened somewhat earlier, in 301. In any case, Armenia would have been the first country with a baptized Christian ruler. Grigor became the founder of the Grigorides, an hereditary priests’ dynasty that remained tied to the metropolitan of Caesarea and to the influence of the Greek church from the west, that brought the written Bible and liturgy (in Greek) to the still illiterate early Armenians.

There had also been an Syrian-Aramaic influence from the early mission days in the south of the country – with Syriac bible and liturgy - represented by another priests’ dynasty, that of the Aghbianos (Albianos) from Man(t)zikert, some 50 km north of Lake Van near the modern town of Malazgirt.  Mounting pressure on Armenia from the east and west by the Persian and Roman empires lead to internal strife, as a result of which the western oriented Gregorides lost their power to the Aghbianos well before the end of the 4th century. When Armenia was divided between the two adjoining empires, the greater eastern part came to Persia. The border ran in a somewhat crooked line from the eastern Black Sea via the modern cities of Erzurum and Mus down to a point somewhat east of Diyarbakir.  The Sassanides who had replaced the Parthians in the third century had adopted Zoroastrism as a state religion. Converting from Zoroastrism to Christianity was a capital crime and warranted death to all parties involved; to mission among non-Zoroastrians in the new territories of the empire, however, was sanctioned (Baumer 2006). Sassanid general attitude towards Christianity was volatile, phases of appreciation are known; more usual were restrictions, accompanied by severe persecutions from time to time. With these political parameters acting as performance conditions in the background, the stage was set for the Christian expansion into Asia for centuries to come. The patriarchate then had its seat at Seleucia-Kthesiphon.

To the south of the Great Caucasus, two kingdoms established themselves in the first centuries AD. Lasika (Colchis) in the west (Imereti) was an ally of Rome, and Kartli (Iberia) in the east was contested between the Parthians (from 3rd century onwards by the Sassanides) and the Roman Empire.  That the apostle Andreas may have missioned here is thought of as a pious legend by independent experts.  Another legend involves a Caucasian Jew who allegedly travelled to Jerusalem, from where he took home an unfinished gown that had been made for Jesus before his death (Hage 2007).  The realistic content of the legend may be the that it was the Aramaic speaking Jews who provided the first stepping stones to the mission.

Officially, church history in Georgia sets in with Nino, the legendary female ascetic apostle from Cappadocia.  Due to her actions, Iberia’s king turned Christian at around the middle of the 4th century, with Imeret to the west soon following him.  As in Armenia before, eyes at first had been turned west.  Under temporary Persian occupation in the first half of the 5th century, however, the church tied itself to the Apostolic Church of the East, whose Catholicos became the nominal head of the Christians in the Caucasus. The second half of the 6th century first saw Georgia becoming divided between Byzantium and Persia, and eventually coming under Sassanid rule altogether.  According to the records, Georgian Christians had taken part in the Apostolic Eastern Church Synode in 419.  Like their Christian brethren in other parts of the empire, Georgians suffered in the Sassanide anti-Christian pogroms in the second half of the fifth and first half of the sixth centuries.   In the following centuries, the Georgian church struggled free from East-Syrian Dyophysitism as well as from the Armenian Miaphysitism and joined the Christological position of Byzantium around the beginning of 7th century.   During the whole process, which outlasted Sassanid rule, the Church of the East retained its influence in the region; and it continued to do so under Islamic rule. Muhammad was allegedly identified as the prophet by an East-Syrian monk. In the advent of Islam the Church of the East had welcomed the change from the repeated measures of Sassanid suppression, and cooperated and thrived – although its members remained second class citizens. The Georgian Church consolidated itself as a national body that offered a focus for a growing national identity, an important step in preparation of national independence.

The now extinct church of the Caucasian Albanians, is as old as those of their neighbours and has been tightly bound to them. Its history exemplifies, how much politics were actually interwoven with church matters, even such as disputes over Christology. The volatility of change in Christology on the Albanian church seems to have superseded that of its neighbours even. Until the 11th century, Miaphysitism and Dyophysitism seem to have been at a constant tug of war.

When Armenia was divided in the late 4th century, the Albania of old with its capital and the seat of its Catholicoi at Tschoghay near Derbent in the north, had grown into New-Albania and more than doubled its size, because it had incorporated many groups of people with different languages in addition to those Armenians, who had settled in the lands between the Kura and the Aras (Hage 2007). Dowsett (1961) after Hage (2007) states, that in its great days, the Albanian church had missioned amongst the Turkish tribes (Huns) settling further north. Maybe, in the end, it was the lack of ethnic homogeneity as a result of expansion that hampered the creation of a focus of identity, as it successfully had happened in neighbouring Georgia. The constant church strives would have done the rest. Islamization had set in at the beginning of the 8th century, in the 11th century conciliar mosques existed in Partaw, Qabala and Shaki, the cities that were the creed of Caucasian Albanian Christianity (after Wikipedia org). The Albanian church is now extinct for nearly 900 years.

Epistemology and Method 

The composite symbol is a carefully balanced composite structure in which individual components carry meaning and add to the overall form. The author assumes that the delicate balance of the structure would be maintained as long as the spiritual charging engine behind it did not flaw or falter. But once this happened, due to change processes in the social community, estrangement, disintegration and transformation would have set in.

A nomadic or cottage weaver, in her individual life, is tied to the greater process by tradition and family bonds, but she also represents it and ‘translates’ it into something personal. This is what happens at the loom, the interface between the weaver and the rug with its symbols and motifs.

Looking at this interface from the perspective of the symbol, it could be described as a process of accommodation and assimilation (23), two underlying processes that act together and are as essential to one another as are warp and weft as the material fundament of the fabric. Accommodation describes the change process on the side of the symbol, ultimately its transformation from symbol to ornament; assimilation means the complementary by which the transforming symbol remains attached to the flow of a changing social and religious environment. Both sub-processes occur in increments. After many generations and life-cycles, the name and the form of the symbol will have changed to varying degree, but it has adhered to its central role, now as a significant motif  in the repertoire. It has become part of the heritage, while its earlier religious significance had become defused and is now inaccessible, after a new name or myth  were attached to it. This is what seems to have happened in the field of rug motifs on a broad front.

As has been mentioned earlier, the rug resembles others, in some aspects of its design, that are dated to the 13th - 16th centuries. These rugs are in the keep of Istanbul museums Vakiflar, TIEM and other collections.  In every discussion of carpets, those would be regarded as very old or ‘classic’ (Denny W B, 2003).  In an attempt to link any of those piled rugs with the Nestorian flat-weave at hand, a direct comparison from rug to rug on the level of main motifs, secondary ones and border ornaments, complemented by a structural comparison, is what normally would be undertaken.   One could call it the traditional approach.  Considering the immense time gap of more than thousand years between those ‘classic’ rugs and the symbol of Christ in the flat-weave, and the fundamental difference in technique, this approach seems intangible. It is also not exactly what the author hopes to be able to demonstrate. Instead, in a quasi-experimental and dynamic approach, the symbol of Christ is inducted right at the beginning of the proposed process: that early Nestorian Christians were functional in the creative process and in the distribution of textile patterns in Asia; that this took place in the context of their missionary activities in the early centuries of the new religion; that this process is traceable in the textile heritage of the nations that lay on their missionary path. The outcome can then be assessed as the degree of perceived concordance in a comparison of the composite symbol and exemplary rugs supposed to be related by a long chain of descent. Since the internal religious and value context remained rather stable within Nestorian communities, in contrast to external live conditions and political landscape, the composite symbol is assumed to have changed very little as well (24).

In other words, as the symbol at time zero had represented a significant Christology that distinguished it from other churches and, of course, from Islam, it should be traceable, distinguishable and recognizable, if it has survived at all. In this quest the author was looking out alertly for the following aspects in rugs to compare it with:

In the list of tables only such rugs have been included that in addition to (1-7) show a reasonable state of integrity, i.e. maintain a degree of spatial order in the sense of a proximity-distance relationship, that can be meaningfully related to the original symbol.

Map: The Roman-Persian border in the 5th century (Wikipedia.org):

Plate 1: Symbol of Christ in the centre of an ascending garland of rosettes encompassing the letters forming the word ΙΧΘΥΣ; with the stem or axis probably symbolising the genealogy of Jesus and the spiritual lineage of the Patriarch

Plate 2: Mary in a traditional cowering birth-giving position as Christotokos as opposed to Theotokos; Patriarch’s cross (top rosette)

Plate 3: Mesopotamian type mighty horns as an attribute of the divine, the ‘hypostatic union’ in the shape of a sceptre, fishes, birds on their wing forming a ‘mitre’ or ‘tiara’, root or stump referring to prophecy in Isaiah 11:1 and other attributes. In the Sassanid era heraldic birds (eagles) made their appearance. In biblical times, among the Hebrews and in Egypt, birds functioned as ‘space holders’, indicating the sphere of divine presence (Keel 1996). They may have a dual connotation here.

Plate 4: Cumulation of borders as known from ‘classical’ Anatolian rugs of the ’Holbein’ group

Plate 5: Enigmatic predecessor ‘God’ Alexander with ram’s horn on a coin of Lysimachus of Trace (306-281 BC)

Plate 6: In the epic of Gilgamesh, his goddess mother Ninsun offers herself to the sun-god Shamash in an act of pleading his protection for her son on his quest to the cedar forest. Dressing up to the event includes her putting on a tiara which may have looked similar to this one:

Plate 7: Cylinder Seal, Carnelian. North-Mesopotamia, middle-Assyrian time, ca. 1300 -1200 BC showing a winged goddess with horn-cap hovering over two antithetic horned animals.

© The Lands of the Bible Archaeology Foundation c/o Royal Ontario Museum  

Plate 8: Rolling of seal from plate seven. It amazes to find constituent aspects of the later rug symbol fully evolved at such an early age: god (goddess), wings, antithetic powerful horns are easily identified.

© The Lands of the Bible Archaeology Foundation c/o Royal Ontario Museum

Plate 9: At Iraq National Museum’s reopening in 2009: Protective God at entrance to the Hall of the Assyrian Kings

Plate 10: Detail from a German-Austrian Alpine Club Expedition Report Map (Bobek,1938) with seasonal migration route between summer and winter pastures of the Herki, crossing Christian (and Kurdish) settlement areas on their trek until the borders were closed in the early 1930’ies. This is where the flat-weave comes from, and possibly the Pazyryk rug as well - south of Lake Urmia and just outside the bottom line of the map (Schürmann, 1982).

Plates 11,12: (from left) Rug auctioned in Germany in 2005; Eagleton (1988) plate 63 ‘Barzani rug


Plate 13: A 17th century rug from Central Anatolia, Konya area. The white rhombus has been clipped and has become an octagon; the lining birds are recognisable and are in position, so is thesymbol of the trinity in its extended Nestorian version, incorporating the symbol of the hypostatic union. One of the divine horns has been lost, perhaps in an old repair. The staff / axis is recognizable and some other aspects as well. The image has been taken from Bayraktaroglu S and Özcelik S (2007); TIEM InvNo727.

Plate 14: A 17th or 18th century rug from Central Anatolia, Karapinar area. The image has been taken from HALI 4/IV p. 371. The resemblance is astonishing although  the birds seem to have metamorphosed somewhat. The religious symbolism is probably extinct.

Plates 15, 16: An East Anatolian rug dated to the 17th century by Balpinar and Hirsch (1988) and to the 15th century by Aslanapa and Yetkin (2005, 1991). Vakiflar Museum Istanbul, inv. no. E-1. Secondary winged creatures are flanking or sheltering the ‘main’ peacocks to the left and right of the stem / axis from below and above. Another significant association between the rug E-1 and the flat-weave exists: the rectangular compartments over the backs of the main creatures carry an emblem that is a minute version of the Christ symbol in the flat weave.

Plate 17: The ‘Marby Rug’  was discovered in the old wooden church of Marby in the Jamtland province of Sweden – an unlikely place for such a find one might think. Its origin lays in Eastern Anatolia or in Northwest Iran, 15th century (Lamm, 1985). As in the previous rug, secondary winged creatures are hovering over the peacocks left and right of the stem and axis. In this rug and the one before, the peacocks have substituted the fish flanking stem and axis in the composite symbol. The medallions rest on a white background, which always is a statement of exception similar to an aura. In the Christian age, peacocks were a symbol of the resurrection and of immortality. At least one incident is recorded, in which a textile with depicted peacocks served a symbolic purpose in such a context (25).

Plate 18: Detail of a Small Patterned Holbein Carpet fragment, 16th century, Anatolia, with the symbol of the trinity in its extended Nestorian version. Some aspects have been added to the symbol, but it remains recognisable and has become a motif in an all over design.

Plate 19: Fragment, TIEM Inv.no. 588 West or Central Anatolia,17-18th C (ICOC 2007)

Plate 20: East Anatolian Prayer Kelim, Erzurum area, 19th century. The symbol has become an ornament and has moved into the outer side borders. However, it has also asserted itself in the central position in the gable (see detailed plate 21).

Plate 21: The remnant of the former composite symbol in the gable carries a minute cross woven with spun silver thread.

Plate 22: The symbol of the hypostatic union as an all-over field motif. Shahsavan bag-face, NW Iran, late 19th century (Plötze, 2001):

Plate 23: A kelim displaying a simplified version of the former composite symbol as an all-over field design. Luri tribe, West Iran, ca. 1900 (Plötze, 2001) 

Plate 24: A splendid rug with what is generally thought of by rug experts as the ‘Qashqai-Göl’. Fars province, South-West Iran, end 19th century; plate 2 from Black and Loveless, 1979. This motif appears to be the nearest relative to the composite symbol outside a Christian context.

Plate 25: A 19th century ‘Moghan Shirvan’  rug (Eder 1990, plate 283). The divine horns are in proper place. The white background of the composite symbol has taken on an octagonal  shape. The symbol of the hypostatic union is recognisable in two of the medallions. What looks like anchors that have been added may meant to be plant shoots. The heraldic birds are simplified to a degree that makes them recognisable to the knowledgeable only. As if to make up for this, minor birds abound in all forms and degree of stylisation.

Plate 26: Also from the book by Doris Eder, plate 8, a 19th century Lori Pampak Kasak rug from the wider Tiflis area. The composite symbol is easily recognisable, somewhat less so the stem / axis. The crown from hovering birds over and beneath the hypostatic union, again, reveals itself to the trained eye easily. Four cute little beetles have taken over from the former birds.

Plate 27: ‘Borchaly’ -  a very similar rug in the nomenclature of L. Kerimov (1983; plate 77). Somewhat less colourful rugs of this type can be found on the Turkish side of the border, where they are called ‘Kars Kazak’.

Plate 28: A Kuba region rug according to Opie (1992), plate 16.9. The composite symbol shows considerable digression and has been outgrown by the symbol of the hypostatic union; birds appear as stylised wings, stem, axes and stylised horns are in proper place, an interesting outcome of the transformation process.

Plate 29: This is another Kuba district rug, plate 286 from the book by Eiland& Eiland (1998). The transformation shows an interesting dynamic: like an explosion drawing of machinery components, the elements of the composite symbol have moved away from the centre while maintaining their relative positions in relation to one another. Stem / axis, hovering wings, birds, divine horns, the symbol of the hypostatic union are all there.

Plate 30: This well known flat-weave from the ‘Bosporus to Samarkand’ book is easily identifiable as a transformation of the composite symbol. The rhomboid in the centre has stretched to become an octagon; the small rhomboids to the left and right remain unaltered, but have changed their relative position to the outer side of the bucrania (divine horns). The axis is obvious, the crowns at top and bottom made up from hovering birds are remnants of former glory. All else are additions that can be understood as vigorous interpretations by a succession of weavers, endowing sense to a tradition form -  from a perspective of a later age and context.

Plate 31: This Daghestan kelim, possibly from the Kumyk population ( Opie 1992; plate 16.6) is striking, and so is the transformation that took place. The hypostatic union has grown out of the central, rhombic medallion, each triplet containing a small version of the composite symbol similar to the border ornaments in the Erzurum area prayer kelim further up. Four birds are still lining the central rhombus, the shoots growing out of the axis at top and bottom are also recognisable.

Plate 32: The small versions of the former composite symbol in turn accommodate minute versions of the symbol of the hypostatic union, similar to those in the East Anatolian and West Persian kelims further up. The four birds lining the central medallion echo the Nestorian flat-weave.

Plate 33:  The Turkish tribes and the Hephthalites (‘White Huns’) were first missionized from Daghestan (Baumer 2005; 2006) and, the Byzantines, so it is reported, were amazed at the sight of crosses tattooed on prisoners’ foreheads, belonging to those tribes. It seems permissible therefore, to cast a glance east in the direction of their habitat. The primary göls on the cover of the Rickmers Collection book (Pinner 1993) share several details with the composite symbol, ie  the bucrania (divine horns) on the horizontal axis, and also the birds forming a crown at top and bottom on the vertical axis. This time they have taken on a somewhat estranged form, reminiscent of other Central Asian animal depictions with a typically backwards rotated head. Otherwise they are very similar to the ones in the Marby rug. The inner sets of bucrania are equipped with ledges or protrusions that give the impression of (horizontal) braids and prompt the association of birds’ heads. The colour schemata of this rug and in the composite symbol are a close match: red or brown-red, blue, black and white are the more or less sole colours. The centre rosettes in the göls look as if picked directly from the garland surrounding the composite symbol.

Plate 34: A similar theme and formal solution from further west; an ascension scene from the gospel book of Rabula (see page three of this paper) in West-Syrian orthodox style. Here, the four birds have become angels. Source for image: Wikipedia.org.

Plate 35: Same theme, further west still - Ratchi’s altar. It alone makes Cividale worth a visit, but there is more than that to the pleasant provincial town. Museo Cristiano di Cividale del Friuli, Italy.


The hypothesis of the author set at the beginning is that early Nestorian Christians were functional in the creative process and in the distribution of textile patterns in Asia; that this took place in the context of their missionary activities in the early centuries of the new religion; and that this process is traceable in the textile heritage of the nations that lay on their missionary path. This hypothesis can be quasi-experimentally tested in the available image material. Methodological considerations can be boiled down to the simple question, whether the composite symbol of Jesus Christ in a flat-weave from the Mountain Nestorian retreat area in the Kurdish Taurus, can be conceived as a prototype for rug motifs in those regions.

To the author, the answer is a clear yes. Considering the velocity of change and as many pitched battles as have hardly been witnessed elsewhere, the symbol of Christ, in some regions, has turned out as deranged as could be expected, and it has remained remarkably intact in other regions. No detailed assertions can be made regarding any conditions, that may have helped or hindered in maintaining the integrity of the symbol other than the strict compliance of early Nestorian Christians to the Mosaic law and its ban on idols, that knew few exceptions. A remarkably integrated motif from northern Daghestan demonstrates, that even a tight Islamic context is no adverse condition. Abstractness and degree of stylisation of the symbol of Christ may have prompted an early and successful assimilation with the result, that until now those ensued motifs have been perceived as always having been accommodated firmly within an Islamic tradition.

Whilst the composite symbol of Christ forms the basis for many later rug motifs, it itself  builds up on even earlier traditions, as has partly been demonstrated. As a consequence of this, the rugs and their symbols and ornaments as they present themselves to us should not be regarded as resting entirely in one cultural epoch, dynastic era or religious realm. Rather, they should be understood as transformations, reflecting developments and change processes that themselves are subject to the greater torrents of Near Eastern history and, of course, they are the results of individual achievements on the side of the weavers. All aspects combine in the lasting magic emanating from these rugs.


(1) Revised version of a paper presented at the 2011 Volkmann-Treffen at the Museum of Islamic Art, Pergamon Museum, Berlin 29th Oct. 2011. ‘Classical Carpets’ is the title of a book by Walter Denny (2002).

 (2) It looks very old, feels very old and shows no signs of heavy wear. The latter is not surprising, given its assumed function as an altar rug. Of one colour the author was not absolutely sure and a sample had been given to Harald Böhmer who identified it as cold dyed light aubergine. The sample was taken from a section of the stem or axis of the symbol that indicated a re-usage of still older yarns. See endnote 14 for this, last paragraph. Given the extreme adherence to tradition by the Mountain Nestorians, it might be possible that the rug (is a true copy of a true copy etc. going back many generations.

 (3) The identification of this rug may be a first step in the unearthing of this heritage. In their original homelands of present day SE Turkey, NE Iraq and South Azerbaijan, little in the way of textile artefacts has survived or is known that bears direct witness to the great time of the Nestorians, whose patriarch was seated in Seleucia Ctesiphon and Baghdad in the middle ages and had overlooked a sea stretching far wider than that of the pope in Rome at the time. The Nestorians’ history and special Christology, the schism that divided them (and still does) from the other churches, are somewhat difficult to access. Christoph Baumer (2005; 2006) provides a sound, thorough and amiably written synopsis of the theme.

(4) More correctly, Assyrians, East-Syrians, or in its own modern diction, members of the "Holy Apostolic Catholic Assyrian Church of the East". However, here the term Nestorians is adhered to, because of the common understanding attached to it.

(5) To Erdmann, the 'Classical Period’ of pile rug production comprises the 13th to 15th centuries, i.e. the Seldjuc and Mongol eras (1974, 1977). Erdmann does not mention the Ottoman at this stage. The 16th century demarcates the culmination and is post-classical to Erdmann. Walter Denny (2002) includes more recent carpets in his catalogue of a Textile Museum exhibition in 2002/3.

(6) Merv, on the silk road, was the principal town in Khorassan and in size second only to Baghdad at the time of the Sassanids and the Caliphate. Its first Nestorian bishop was consecrated before the middle of the 4th century. Later, wandering bishops and metropolitans reached the Hephtalites (‘White Huns’) and migrated with the Turkmen tribes of whom many eventually adopted Christianity (among other sources, Baumer Chr 2006). The strong resemblance of several Turkmen Göls of the composite symbol discussed in this paper suggests that rugs and rug symbols had had a function in the mission process. In this sense, of the Turkmen groups later migrating into Anatolia, some were probably travelling up to the source of their rug designs. This worked into  Erdmann’s theory constitutes a necessary major revision of it.

(7) Azadi S, Kerimov L und Zollinger W, 2001; Bennett I, 1993; Eder D, 1990; Opie J, 1992; Schürmann U, 1990

(8) This view has been explicitly expressed by Opie (1992) and has been implied by Eder (1990) and Bennett (1993), with the latter authors having contemplated the presence of Avar Thrones, animal hides and other archaic symbols as central motifs. These observations should be taken with some caution. They appear to be unconnected to identifiable religious or historical circumstances and no approximate expressions in objects belonging to other art forms are known that warrant such an interpretation. In a ‘tapiologic Salon’ hosted by the author on the Turkotek server, he had introduced the Jewish Ark of Covenant into the discussion. This, appears as a motif on rugs on the Caspian side of the Caucasus, where once a substantial, and now greatly diminished Jewish minority settled (Nitz  2007).

(9) The planned route was to branch off east at Hakkari from the one taken by Anthony N. Landreau (1973) Kurdish Kilim Weaving in the Van-Hakkari District of Eastern Turkey. Textile Museum Journal Vol III 4, 26-42.  Reliable contacts in Van that were helpful in planning and organising had been established from 1974 on. In the summer of 1980 first signs appeared that the trip might no longer be feasible. From the mid 1970’ies onward, Turkey had experienced a swirl of increasing political violence between the left and the right. In the summer of 1980, civil war seemed to be a definite possibility. Parallel to this, the PKK had launched a nationwide attack on public and military institutions that interrupted civil life severely. Iran’s Islamic Revolution was about to spill over into Turkey. When national flags were torn down and burned in Konya and the green banner of Islam was hoisted, the military may have thought that it had little option but to move in with force.  As a consequence, several eastern provinces were put under martial law and were closed to foreign visitors. Already granted permissions and promised support became vain. In the spring of 1984 it was again possible for the author to travel to Hakkari. For a revival of the project however, it was too late for a variety of reasons.

(10) Eagleton (1988) plate 63 - depicts a more recent pile rug that is very similar in its principal design. He wonders how a Qashqai emblem could have found its way into a Barzani Kurdish rug from NE Iraq, 1000 miles away from Qashqai settlement areas. Yet another pile rug of this type has passed through auction in Germany in 2005. It was advertised as a Qashqai rug. Both pile rugs are of a more recent date, they are coarse weaves and lack the symbolic depth of the flat-weave.

(11) Barzani aga was the lord over a mixed Muslim Kurdish and Christian people. He was nicknamed the ‘Christian Aga’ because of his tolerant attitude towards his Christian vassals (Wigram& Wigram 1914,  p. 153). He also seems to have had his own chicken to pluck with the Herki, ‘those hostes humani generis … this horde of wandering robbers, the bane of all settled communities.’ The authors relate how he quite cunningly managed to retrieve some two or three thousand sheep and more from the Herki, which they had previously lifted from his subjects. At this outcome ‘all the country was jubilant to see the original biters so badly bit’ (p. 149 ff).

(12) This probably is the material basis for the inflationary use of the Herki attribution, that James Klingner (1999) is unhappy about in his Hali article on rugs and flatweaves of the Northern Zagros

(13) All over weft-wrapping technique; coloured wefts are wrapped around one or two warps.

(14) By the time of Timur Leng’s death in 1405, the Nestorian Christians who he had persecuted so severely, also had ceased as a functional and influential body in social life in all but a few townships and rural communities. The patriarchs in office had begun to change their positions frequently for security reasons and communities drew back in order to find protection in or closer to the mountains or they converted. Cultural exchange with the outside world became greatly reduced. Accordingly, in the flat weave, no later design principle is apparent, than that of the ‘Holbein’ type and of an early combination of a central medallion with a cartouche form. Although some design aspects pay reference to much earlier periods, on the whole, the rug echos designs not earlier than 14th or 15th century.  In comparison with other flat-weaves for profane use that the author has access to and that come from the same region, the actual rug however, could be as late as mid 19th century. Obviously, this is too wide a gap to be satisfactory. It is the result of a complete cut-off from outside developments after Timur.  All who have written about the Nestorians in the 19th century agree in the observation of their very traditional customs and their extreme poverty.

The rug may have been reproduced from respective predecessors more than once with painful accurateness in the way early scriptures and illuminations were copied in monasteries. Reports mention very old scriptures with the Mountain Nestorians but also narrate losses due to exposure to the elements and inadequate keeping.  To make matters still more complicated, there is an indication that the rug may have been made with batches of wool retrieved from an earlier rug. The stem or staff that runs right through the middle of the rug in its upper section contains a number of small compartments, some of which show no more than a faint hue of violet (see plate 2). This impression is caused by a small amount of a very light and fluffy wool dyed violet in standard dye density that is spun around a thread of natural light wool. Behind this may be piety, because a Christ’s rug cannot be discarded in an ordinary way, or it reflects on the impoverished autarky of the Nestorians that had forced them to make a little last as long as possible; or for both or still more reasons.

(15) Besides some rugs of the ‘Holbein’ group, or more specifically to their borders, an obvious association exists with some other rugs, namely the ‘Marby’ rug (Lamm, 1985), the ‘animal’ rug with the inventory no. E.1 in the Vakiflar Museum (Bayraktaroglu and Özcelik, 2007; Balpinar and Hirsch 1988) and a rug of the Karapinar group depicted in HALI 4/IV 1982 p.  371. The latter three rugs share a particular design feature: a stem or trunk as a vertical symmetric axis runs through the medallion(-s), which is (are) lined in all four quadrants by elongated animals, winged in most of the cases. Just to make sure that no misunderstanding occurs, what is meant are those secondary creatures flanking the ‘main’ or foreground creatures from below and above. In the case of the flat-weave, they form a kind of crown or mitre, and look  as if they are attempting three-dimensionality.

Another significant association between the rug E.1 and the flat-weave exists: the rectangular compartments over the backs of the main creatures (peacocks probably if one favours a more-down-to earth interpretation than Balpinar and Hirsch, who see a ‘flying dervish in the shape of a dear’ in it) carry an emblem that is a minute version of the Christ symbol in the flat weave. 

(16) John 8, 12 (Addressing the people at the temple - Again Jesus spoke to them, saying, ‘I am the light of the world. Whoever follows me, will never walk in darkness but will have the light of life.’ ); Matthew 17, 2 (The Tabor-Light - And he was transfigured before them, and his face shone like the sun, and his clothes became dazzling white).

Jesus’ suffering at the cross apparently did not feature quite as strongly with the Nestorians, as it did in the western churches and, in fact still does. To the Nestorians, he was most prominently, the Christ of the Resurrection. The white background of the composite symbol is, in accordance with the above quotations, all about divinity and glory in the resurrection. Interestingly, across church boundaries and schisms, in Byzantine frescos as well as in Russian icons, we usually encounter Jesus Christ in a white gown in the resurrection theme (Ouspensky L (1962) pp 74 ff in Hammerschmidt E, Hauptmann P, Krüger P, Ouspensky L& Schulz H-J (1962) Symbolik des orthodoxen Kirchengebäudes und der Ikone). It appears very likely, that the same passages of the Bible have become constituent to the respective art styles.

(17) The spatial arrangement of the symbol of Christ in the centre of an ascending garland of rosettes, with the Patriarchal Cross in the uppermost one, suggests a genealogical theme like the ‘Root of Jesse’ or ‘Jacobs Ladder.’ For the time being  it remains open, whether the rosettes are meant to represent Biblical figures and if, who could be featuring in the bottom section of the rug. The stylised birds on wings that line the inner flanks of the white rhombus, the ‘vesica piscis’, forming  something in the shape of  a mitre, also present a bit of a riddle. In the above mentioned works of art they appear as angels, which could speak for a old-testament Archangel tradition. Or could they be another loan taken from the Mesopotamian tradition? A Kassite cylinder seal shows the god Enki flanked by two sets of double-shaped birds and, with two flanking fish-men by his feet. This throws a surprising  light on the two fishes in the composite symbol (Pedde B (2009) Altorientalische Tiermotive in der mittelalterlichen Kunst des Orients und Europas; Cat.-No. 91, plate 22).

(18) The oldest known rug in existence, the Pazyryk Carpet, that was released from a permafrost burial mound in the High Altai by Sergei Rudenko in 1949 (Rudenko S I, 1970) is now being thought of to have been commissioned by a Scythian noble, based at Sakic, a major encampment at around 500 BC somewhat to the South of Lake Urmia in Iranian Azerbaidschan, according to Ulrich Schürmann, who had discussed the rug in an art-historic perspective. As the rug corresponds closely with stone floor ornaments at the Assyrian palaces in Nineveh and Khorsabad, the weavers of the Pazyryk Carpet and those of the ancestry symbol of Christ may have belonged to the same stock. Sakic and the Mountain Nestorian settlement areas in the Kurdish Taurus are situated less than 150 km apart from one another; the Nestorian villages being closer still to Khorsabad and Nineveh (120 km) than Sakic (180 km). The design of the flat-weave survived in the mountains, but its origin probably lay in the plain below, to the present day a Nestorian Christian or, better, Assyrian Christian settlement area. Their ancestors would make excellent suspects to have had a hand on the loom with the Pazyryk carpet.

(19) The remains of the city are situated on the western bank of river Tigris, north of the confluence with the tributary Little Zab river, in the Al-Shirgat-District of the Salah al-Din Governorate of modern day Iraq. The whole area had a long Christian tradition.

(20)The Northern Kingdom of Israel was invaded, conquered, and the population taken captive primarily by the Neo-Assyrian monarchs, Tiglath-Pileser III and Shalmaneser V  in the course of the second half of 8th century BC. The tribes exiled by Assyria later became known as the Ten Lost Tribes of Israel.

(21) One specific example, rug E.1 from the Vakiflar Museum has already been discussed in endnote eleven. Other examples from other regions will be given in the full paper.

(22) These preconceived ideas have led to a somewhat ironic intellectual detour: Pinner (1979) and others with him believed the ‘Qashqai’ emblem to originate from some unspecified remote location in Central Asia - via Turkmen members of the Qashqai federation and Oghuz predecessors; Eagleton (see note 2) on the other hand puts the central medallion of his Barzani area rug down to the Qashqai, settling one thousand miles south of NE Iraq, not realizing that its authentic origin in all aspects lay virtually on his doorstep.

(23) The interacting processes of assimilation and accomodation as they are used in this context are a loan taken from Jean Piaget’s conceptualisation of developmental processes. Jean Piaget has become a galleon head of modern developmental psychology, and has made important contributions to epistemology and science theory (Jean Piaget 1972a, 1972b, 1973, 1974, 1989, 2001, 2004, 2007).

(24) Every early visitor to the Nestorians had commented on their conservatism and adherence to old ways, as if living in a past age. 

(25) Cnut the Great (King Canute) ‐ King of England, Denmark, Norway and parts of Sweden ‐ visited the tomb of Edmund Ironside (Edmund II), his predecessor and former opponent, on the anniversary of his death in 1016 and laid a cloak decorated with peacocks on it to assist in his salvation, peacocks symbolising resurrection (M K Lawson (2004) Edmund II. Oxford Online DNB ‐ after Wikipedia.org).

Discussion   Bibliography