The boteh carpet design
The boteh carpet design
The motif is basically in the shape of a pear, with the top bent over. In some designs there is a little leaf or tail at the bottom; in others this tail becomes detached from the adjacent motif above and attached to the one below, in the form of a ‘nose’. The motif is thus asymmetrical and it is no doubt this quality which makes it fascinating as an all-over repeating pattern it represents a calm and restrained image without the dryness or rigid formality of plain repeating flowers.
The diversity of form which the motif evinces even in antiquity is a point taken up by S. V. R. Cammann. In a series of lectures printed in the Washington Textile Museum Journal in 1973 he makes out an interesting case for the religious or mythological origin of many carpet motifs. Other writers, too, have suggested an animal origin for the motif, drawing attention to the strong influence on the ancient world of the art of the Scythians, which was dominated by animal forms. Another idea sometimes put forward is that the motif represents a flame, which again may have religious significance as a symbol for immanence and eternal life. This concept can be found in the Caucasus and also among Zoroastrians, the adherents of the ancient pre-Islamic religion of Iran which was established by Zoroaster (Zarathustra) and survives today, its present headquarters being in Yezd.
None of these theories about the boteh takes account of how the motif found its way into carpet design. It does not appear in any known carpet that can be dated to much before 1800, and it seems likely that the early nineteenth-century floral textile designs of India inspired the botehs of Khorassan and Genje, Senneh and Serabend.
The version most commonly found in carpet shops today is the Serabend or a finer type called Mir Serabend. Serabend is a region in Persia located about 2,000 m (6,500 ft) above sea-level in the mountains between Arak and Borujird. The word Serabend seems to mean a cold place, but there is also a river Saravand just to the north of the carpet-making area, which may perhaps be the origin of the name.
Red is the most common ground colour used today, but blue and cream are also found. The Serabend is an all-over design, but a diamond-shaped medallion is sometimes found, especially in rug sizes. The knot is usually Turkish, and most sizes are made.
The Serabend of course employs a fairly basic version of the motif, but at some point in the nineteenth century (possibly earlier) a liner version was created, perhaps in villages nearer to Arak, using the Persian knot. This is called the Mir Serabend (known in German-speaking countries as Saruq Mir). The reason for this choice of name is uncertain – possibly it derives from the name of the village of Mal-e-Mir in the centre of the Serabend region and how the development of this finer version came about is shrouded in the mists of history. Some museums have pieces which seem to show the process happening, as it were. One such example is in the Victoria and Albert Museum, London (fig. 119). The picce was acquired by the museum in 1918, but must already have been quite old even then. It has much of the feel of the Serabend about it, and uses the Turkish knot, but stylistically it foreshadows the present Mir.
As our example shows, the version found in these nineteenth-century rugs can be totally different from the Serabend form in character and expression. There are other versions again, the existence of which makes one wonder whether the Mirs and the Serabends are not in fact a development one from the other, but have perhaps always co-existed, derived maybe from a common ancestor. It is perfectly conceivable that some villages of the area may have specialized in weaving fine, large pieces for the local gentry, for example, using a more elegant style than the rugs of their peasant neighbours. The source of the design may, in fact, be Indian; the bedcover shown in fig. 122 is older than any known Serabend and may have served as the basis for a carpet design, copied from it in the Arak area in the early nineteenth century.
In today’s carpets the Mir can be distinguished from the Serabend not only by the knot but also by the structure; the Serabend is coarse, usually has a stiff back and a thin pile, the Mir varies from medium-coarse to quite fine, has a more flexible back and a thick, luscious pile. The colourings of the latter are usually more subtle, the best pieces having a superb rosy ground shade which results from the use of a natural dye derived from madder root. Another fine feature of the Mir Serabend is its wide border, consisting of a small main border (usually cream) surrounded by a large number of narrow guards. Indeed, the border, as is clearly seen in the illustration, can be so rich that one occasionally comes across carpets with the Mir border that have no design at all in the ground. Like the Serabends, Mirs are also sometimes sound with a diamond-shaped medallion. Persian Mirs are made in all sizes from small mats right up to large oversize carpets, and including runners and squares.
There is also quite a wide range of ground shades available, including light greens and light blue, but the bulk of the production is in red or dark blue, and cream grounds are rather rare.
Another carpet of this group is the so-called ‘town Serabend’ – a manufactured carpet made in and around the market town of the Serabend area, Borujird. The quality is much better than that of the average Serabend and is comparable to that of the coarser Mirs, which has prompted some dealers to coin words like ‘Mirabend to suggest the superior quality.
Mention must also be made of the Indo-Mir, since there is a huge production in the Mir style in India, in the Bhadohi-Mirzapur district of the state of Uttar Pradesh. Although the version of the design most commonly made there today was copied directly from a Persian original, these new Indian carpets are entirely in tune with a strong local tradition. Some authorities indeed assert that the boteh motif originated in India; whether or not this is true the pattern has certainly been used in the textile industry of Uttar Pradesh, and indeed the whole of India, for centuries. Three Indian variants are illustrated, both from southern India, and an eighteenth-century cotton bedcover, shown here, which is in a style used throughout Uttar Pradesh.
The range of Indo-Mir carpets produced in Uttar Pradesh today is very wide: the worst are cruder than the poorest Serabends, but the best are finer than any Mirs produced in Persia today. A typical naksha, or loom-drawing from which an Indian weaver works, is shown here together with a carpet produced from it: every knot of the repeating pattern is set out in the naksha, and the woollen tags attached to the code-panel at the bottom indicate which shades are to be used. These are readily changeable to suit particular needs. Note how the essentially geometric
character of the design is retained despite the fineness of the weave.
Within the Indian Mir output a much greater variety of designs is found than in Persian goods. In particular, the Indian manufacturers have appreciated the usefulness of the Mir design in producing medallion carpets, either by setting a traditional Borujird-arca medallion against a Mir background, or by superimposing on an all-over Mir pattern a skeleton medallion outline of the type used in the Mashayekhi Saruq illustrated in fig. Indian Mirs, with or without medallion, are available in almost every conceivable size and colour.
By way of contrast with the Indian naksha, a Persian loom-drawing (prior to painting) is shown: this represents a boteh from a Kcrman carpet made in the 1920s.
Here one can see the other extreme: the motif has been subjected by master designers to the most convoluted elaboration, one boteh being placed within another and the whole consisting of a mass of finely drawn detail. Between the two extremes of the rustic Serabend and the refined Kerman lies a wide range of interpretations,
geometric, floral, large- and small-scale all-over ground patterns, individual boteh motifs as part of a larger design, boteh motifs grouped in set patterns to form a new design, linked botehs in borders – and each one bearing the typical ‘handwriting of the style of the area or village in question. Some of the most commonly found
types are illustrated on the following pages.
The village of Everu, near Hamadan in north-west Persia, weaves rugs and runners of medium quality, using the small boteh design illustrated. Everu rugs are often sold as Enjilas, which description is inaccurate Enjilas, a neighbouring village, makes pieces of a very much higher quality (see p. 96). The design layout resembles that used in the Serabend region, but the Everu rug is easily distinguished from the Serabend because the former is a single- and the latter double-wefted.
The border is also quite different. Other distinguishing features are the strong deep red, rather heavy dark blue and the regrettably vicious greens and turquoise blues; Everus have fine silky wool with a quite high pile. Most rug sizes are made and also runners, but no carpets. The output in this boteh design is small, the bulk of Everu’s production being woven in the Herati design.
Borujird (south of Hamadan) is the market centre for the Serabend rugs, as well as for single-wefted Hamadan-type rugs in various designs, including several variants of the distinctive large boteh version illustrated. The typical features are: a single-wefted structure, coarse weave with thick grey or brown wefts, fine yarn, long pile. The rugs are often very sombre when new (over-heavy dark blue and dark red) but can be very attractive when made in softer or lighter colours. Borujirds are found only in zaronim and dozar sizes, and runners. They have much in
common with the traditional old single-wefted Malayirs, a feature which is more obvious in other designs woven in this area, such as that of fig.
Senneh (properly: Sanandaj), the principal Kurdish city of Iran, uses a boteh similar in shape to the one used in Borujird, but the goods are easily distinguished from each other by the respective weaves, Senneh’s being very much finer — indeed Senneh makes one of the finest single-wefted weaves anywhere in the Orient. The Kurds often use a highly-twisted yarn, which gives a knobbly appcarance to the knots on the back of the rug. In the finely-woven Senneh pieces, the back of the rug feels almost like sandpaper. Pushtis, zaronims and dozars are made, plus occasional strips and carpets, the latter often being long and narrow. The colour range is extremely conservative, being largely dominated by red, blue and ivory. Many other colours occur as subsidiary shades, but usually only in very discreet quantities.
The other principal Kurdish town in Iran, Bijar, does not have a design of its own it does in other Kurdish rugs. The item illustrated here is not a carpet but a vagireh, that is a weaver’s model. Vagirehs (also known by other terms in other areas) may be found anywhere in the Orient, but they are most common in north-west Persia and usually come from areas where there is semi-organized village manufacturing, and here they take the place of the graph-paper loom-drawings used elsewhere. This Bijar vagireh which is very old indeed, shows a large number of design elements of a rather wild, rustic nature. There are four sets otbotehs involving three different shapes. The largest pair are clearly meant as fruits or pine cones on the end of a branch. The others seem to have a purely abstract function as filler-elements in the design.
To the south of Bijar and Senneh there is a large Kurdish area in which many designs and styles are woven, the principal markets being Sonqur and Kermanshah. As in Bijars, boteh motifs occur mainly as subsidiary elements: a good example may be seen in fig. The rug illustrated here, however, is one of the rarer examples of a Kurdish all-over boteh. It is only just a boteh design – in fact the design may well be a derivative of the gol farang. The Kakaberu piece has all the unmistakable features of this tribe’s work: dark, almost sombre colours, including a blue that is almost black and dark, browny red; a medium-coarse weave, but a structure as solid as a board, the kelleyi format, and, above all, a wild indomitable expressiveness highly suggestive of the rugged independence of this fierce mountain people.
Note that both the Senneh rugs illustrated are on cream grounds, which is unusual; Sennehs in other designs are more on blue or red grounds. The weave here has the same characteristics. The hashtguli idea is also used in other areas which weave the boteh motif, such as the province of Khorassan. Indeed, old pieces from this area now in museums suggest that it may have originated there.
The next carpet, from Ardebil, is typical of the tribal kelleyis made in the southeastern part of Persian Azerbaijan. Dealers in Tabriz might call it Ardebil, Meshkin or Khalkhal, but from its appearance the name Shah Savan would perhaps be better. The Shah Savan tribe does not seem to be Kurdish, as some carpet dealers suppose; it is a confederation formed at the end of the sixteenth century for political reasons, and many of the clans seem to be of Turi origin. However, it may be that their carpet weaving is influenced by that of the Kurds, just as the Beluchis of Khorassan are influenced by the neighbouring Turkomans. The Shah Savan inhabit large parts of southern and eastern Azerbaijan and spread as far south as the Saveh area west of Tehran. The Azerbaijan origin of the carpet illustrated is easily distinguished: the flat selvedge and the very tough, tight, though not necessarily very fine, double-wefted structure are unmistakable features. The weave, the kelleyi format and the bold uncompromising design might suggest Kakabcru, but the flat selvedge would preclude this origin; more important, however, is the colour combination – the red is too bright and the other colours are too light for a Kakaberu piece. True, the Kolyais, close Kurdish neighbours of the Kakaberu, often produce a similar red, but the Kolyais always use a single-wefted construction.
Another typical clue is the Caucasian-looking border, for present-day Persian Azerbaijan and Russian Azerbaijan in the Caucasus have common traditions dating back many centuries. In this kind of Ardebil only kelleyi sizes are made.
There are many hundreds of villages around the town of Hamadan, all with distinctive designs and weaves. Generally the boteh is used only as a subsidiary motif (cf. fig. 729, from Mehriban), but occasionally one comes across a rug where the weaver has produced a boteh design of her own. In such cases the design is no guide to the origin the key factor in such cases are the weave and the colours (which are dealt with in detail in the relevant sections on Hamadan designs). The piece illustrated here is from the Victoria and Albert Museum Textile Study Rooms collection and shows a fascinating interpretation from the Hamadan region dating from perhaps a century ago.
In the nineteenth and early twentieth centuries the boteh motif was used by weavers throughout the Caucasus, in a diversity of form and richness of imaginative invention unsurpassed by any other weaving area. Several examples are shown here, and another two appear in figs. Old Caucasian rugs like these are available only from specialist dealers, but are worth illustrating here because of the different interpretations they reveal within a fairly small geographical area. The Marasali prayer rug from the Baku area includes a strictly geometric form of the motif whose jagged edging seems to suggest a flame; the figure at the base of the botch could represent the cup of a torch. By contrast, the Genje botehs shown in figs. 138 and 136 are clearly interpreted as plant forms, the one sharply geometric and the other fully curvilinear, but the carpet shown in fig.
treats the motif in a highly abstract form in the ground, in a pattern known as chibukli, the Turkish word for a (smoker’s) pipe. This type is woven in both floral and geometric forms in many parts of the Orient, but especially in Khorassan, Qum and Genje. In this example the boteh appears in the border as well as the ground; in the border many of the botehs take the form of birds. This may be accidental, but there will certainly be theorists who will interpret the ground as a Modern Caucasian rugs are made under controlled conditions similar to those that
prevail in India or China. Unlike Indian practice, however, the designers of the Soviet State manufactories keep strictly to the traditional design elements of the Caucasus (although not to the traditional designs of their own respective region).
Their re-interpretation is often rather mechanical and stereotyped – the standardization of qualities and the separation of weaver and designer tends to produce rather soulless results, the dullness of the designs being surpassed only by the insipidness of the colours. The boteh is used in several forms, but always as a subsidiary
motif, as this illustration of a new carpet from the Kazak district shows. There is none of the wealth of elaboration found in old Caucasian rugs, and the famous flame of Baku, where oil and natural gas have sprung from the ground for centuries, has apparently been snuffed out. More details of the typical features of modern Caucasian rugs will be found in the chapter on Geometric Designs.
As in north-west Persia and the Caucasus, the tribal weavers of Fars province produce a wide range of designs, again including several versions of the boteh, especially in the style of rug known as the Gebbeh (literally ‘undipped’, a reference to the shagginess of the pile.) One version has been shown in; a second is
shown here. Genuine Gebbehs are rugs made by nomads for their own use. Rusticb simplicity and indeed a certain wilfulness of design (and colouring) give the rugs an unaffected freshness which is much sought after, especially for use in a modern decor. Most Gebbehs measure about 220 X 110 cm (7′ 3″ X 3′ 9″); a width of 135 cm
(4′ 5″), as seen in this rug, is rare.
The Turkoman tribes, who are located mostly in Afghanistan and Russian central Asia, make little use of flower or plant motifs in the Persian style and do not usually include the boteh in their carpets at all. The two principal exceptions are illustrated here. The first shows a distinctive version found in the Beshir rugs of both Afghanistan and Russia. Notice here the suggestion of a tree design, although it is sometimes claimed that this is not a boteh design at all, but a representation of the two ancient Chinese motifs of the dragon and the phoenix (see also fig. 174). Early examples of this design support this interpretation, and it would certainly accord with Grote-Hasenbalg’s assertion that the Beshirs are not a true Turkoman tribe, but that the name is corruption of’Bokhara. This theory is highly speculative, but it is certainly possible that Beshir weavers developed this design from motifs introduced into the area by Timur’s Chinese artists towards the end of the fourteenth century.
The second Turkoman boteh design is one used in the borders of carpets from all over northern Afghanistan, although it is not normally found in Turkoman goods from Persia or Russia. The three botehs grouped together in the Kaldar example form an unprepossessing but subtle and effective motif, which, once one is made aware of it, seems to occur in almost every Afghan carpet one looks at; but here again it must be doubted whether the motifs are intended as botehs in the usual sense (the Turkomans call the motif ‘judor’, which means ‘almond’). The Heriz and Shirvan examples illustrate a similar design much used in the Caucasus and Azerbaijan (where there are also many weavers of Turkoman origin): although the boteh shape is hinted at in these two rugs, it is more likely that the motifs represent opening flower buds linked together in the way that bud or palmette motifs are usually treated in carpet borders. It may thus be pure accident that the Afghan version looks like a boteh. It is cases like this which must make us very suspicious of attaching universal labels to motifs; quite often there are all sorts of influences at work to determine the particular shape which a motif acquires, and these influences can make nonsense of too dogmatic theories.
Bakhtiari carpets are made by Armenians, Kurds, Lurs and other settled tribes in the province of Chahar Mahal va Bakhtiari, south-west of Isfahan. They are not made by the nomadic Bakhtiari tribe, who live further west and have no significant carpet production. A very large range of qualities is made, from the coarse, singlewefted products in geometric designs from villages like Borujen, Farah Dumbeh and Boldaji, through medium-fine doublc-wefted carpets from Shalamzar and Shahr Kurd, to the fine double-wefted products, generally using floral designs, of Saman and Chahal Shotur. A wide diversity of designs are used, too, but one in particular predominates, the so-called ‘garden design’, consisting of rectangular panels containing a variety of different ‘garden’ motifs, from summer houses to grape vines and from individual stylized flowers to fully fledged cypress or weeping willow trees. More details will be found below, in the section on panel designs.
In the panels, the boteh often figures as one of the plant forms — in various shapes and styles, depending on the stitch and character of the particular village’s production. Between them, the two Bakhtiari carpets illustrated show five different forms Shirvan of the motif A particular feature which distinguishes these botehs from those illustrated in the preceding pages is the three-dimensional effect created by the placing of one boteh outline behind another with the hooks at their tops bent in opposite directions. To the author this clearly suggests one of those plant forms which opens at the top a leaf-bud, perhaps, or a hazelnut husk or a pine cone.
Not all Bakhtiari botehs have this characteristic, but it is more common there than in any other geometric version of the motif. The finest Bakhtiars arc more or less curvilinear – often one can sense the influence of the neighbouring city of Isfahan but they always retain a certain angularity as a reminder of their peasant origins. The character of the botehs in this carpet from Chahal Shotur (in the panel and in the border) lies halfway between the geometric style seen in the Caucasian rugs shown on the preceding pages and the truly floral version used in the Qum panel design in fig.
A rug from Quchan provides another example of botehs used as subsidiary motifs. Quchan is a small mountain town north-west of Meshed in eastern Persia. The town is the ccntre of a fascinating and unique carpet-producing area, since it is surrounded by an area occupied by Turkoman and Beluch tribes, with their own very significant
carpet output. Quchan and the neighbouring villages, however, are Kurdish. The Kurds have lived here since the beginning of the seventeenth century, but the designs of their rugs are often strongly influenced by the neighbouring Beluchis and Turkomans. But the true Kurdish character manifests itself in many details – from the names of the villages, like Kolyai and Shirvan, to things like the flat selvedge (typical of Caucasian Kurdish rugs). The border of the rug illustratedalso shows a clear link with the Caucasus. Most rug sizes exist, but small pieces are very rare. The most common types are kelleyis.
The Afshari tribes, settled mainly around the three centres of Sirjand, Shahr Babak and Rafsinjan in the region of Kerman in south-east Persia, weave perhaps the widest range of different designs produced in any one area of the Orient. A visiting buyer could easily find a hundred different designs in current production, maybe many more, and certainly enough to produce a book devoted to Afshar designs alone. The author of a more general survey is obliged to select a few typical examples and simply disregard the rest. The boteh appears in many variants, two Shahr Babak examples being illustrated here. The geometric form of the motif seen in the second of these is the more typical of the Shahr Babak region indeed the first might well have been woven not in Shahr Babak but in Pariz village, on the road to Rafsinjan. A fascinating feature of fig. Lies, however, in the links it suggests with north west Persia and the Caucasus. In this case the historical connection is well documented, for the Afshars are of Turkoman origin and for many years lived in the Lake Urmia region. Several Persian rulers from the seventeenth century onwards
deported rebellious Afshar groups at various times. Most notably, when Nadir Shah, himself an Afshar, seized power in Persia in the early eighteenth century after the Afghan invasion, he smashed the opposition of the Afshars of the northwest by transporting thousands of tribesmen to the inhospitable desert area between Shiraz and Kerman where their descendants still live. Many Afshars still remain in northern Persia, however, and in Turkey as well, and their influence on carpet design in die north-west is strong. Some of the finest Bijars are made by Afshars, for example. Is it pure coincidence that the rather odd shape of the botehs in fig. is the same as that of the main pair of botehs in fig.? In the case of our second example the botehs are of course used as a background pattern. This is a common feature in several Shahr Babak designs, more details of which will be found in chapter IV. Note here the ‘barber’s pole’ binding of the selvedge, and in particular the general colouring of the rug. The rosy-brown shade is the key clue to the recognition of the Shahr Babak rugs, as this colour is not used anywhere else in the Orient.
Another distinctive feature is the subdued grey-green used as a subsidiary colour. The Bcluchis constitute one of the great tribal groups of Persia; their territory lies on either side of the modern Afghan-Iranian frontier. Among their many designs are several versions of the boteh, one of which is illustrated. This version is perhaps
not particularly distinguished the Beluch designs shown in figs. 435-7 have more flair, and it may be for this reason that the boteh design is not encountered very often in Beluch rugs, but it does illustrate how the asymmetry of the motif gives life to the repeating pattern. For a general note on Beluch rugs see pp. 139 and 193.
One of the most attractive of the tribal versions of the botch is the old Qashqai design illustrated here, though it is more often found nowadays in the coarser form illustrated in the Luri piece from the Shiraz region. Note how in the Qashqai the boteh takes on yet another different form. One of the fascinations of this version, too, is the link it indicates with the Caucasus, spotlighting the unity of the culture of the Turki-speaking races despite their wide dispersal and subjection to greatly varied historical forces. Note also that in both these rugs the top of the motif in most cases inclines in both directions, producing a version which is much more nearly symmetrical than usual.
For general notes on Qum cf. fig. 691. Figs. 158-60 all illustrate typical different treatments of the boteh as a subsidiary motif. It is these versions which perhaps come closest in modern Persia to the elaborate nineteenth-century Indian designs, well known in the West from copies made by manufacturers of fine textiles at Paisley in Scotland.
A full illustration of the Qum version of the Bakhtiar ‘garden’ design will be found in fig. 340; here, fig. 159 shows the boteh panel which is almost always found in it (in the Bakhtiar original the boteh occurs only from time to time). The design is very well drawn here – compare it with the Chahal Shotur version in fig. but most carpet lovers may well declare that the Bakhtiar version has more character.
Similar motifs will be found in panel designs from Birjand, where they are made in an even finer weave than in Qum, sometimes with strikingly beautiful results.
Like Qum, Kashan and other towns in central Persia weave versions of an all-over floral boteh design. In many cases there are derived from nineteenth-century rugs produced in Khorassan. Today the centre of fine weaving in eastern Persia is Birjand, but in the nineteenth-century Qain and the surrounding area (known as the Qainat) were also important. It seems fitting to close the boteh section with a detail of one of the fine Qainat originals, which have inspired many of the Persian manufactured versions of today. This rug was presented to the Victoria and Albert
Museum, London, in 1877 by the Shah of Persia.