Lot Essay
The attribution of the 'compartment' rugs to Syria is one that dates back a considerable time. In 1909 Konsul Bernheimer bought his example (sold in these Rooms, 14 February 1996, lot 27) as an "alter syrischer Teppich". Yet this attribution is far from certain. While the group has a clear homogeneity within itself, its combination of technical structure and design motifs make it very difficult to place. Egypt, Rhodes, the Adana plain and East Anatolia have all been proposed. The subject is discussed at length in various places, the fullest of which are Pinner, R. and Franses, M.: 'The Eastern Mediterranean Carpet Collection in the Victoria and Albert Museum', HALI vol.4, no.1, pp.37-52 and Spuhler, F.: 'Chessboard Rugs', in Oriental Carpet & Textile Studies II, London, 1986, pp.261-269.
The development of many of the motifs from the Mamluk carpets of the 15th century is clear. The continuous knot small medallion surrounded by inverted or radiating motifs is easy to parallel and the balance of colours, although slightly different in hue, gives a very similar overall effect. The structural analysis is however completely different, in particular in the way the wool here is given a 'Z' ply rather than the typical 'S' of the products of Egypt. A linking piece is a rug of the same structure as the present carpet, formerly in the Pietro Barbieri collection, which has the same field panels as are found here but arranged in a classic 2-1-2 formation, the central panel having a larger version of the design than the spandrels (Viale, M. and V.: Arazzi e Tappeti Antichi, Turin, 1952, pl.150). An attribution to Turkey however must also be discounted since the carpets are knotted with an asymmetrical knot open to the left in the same way as the Mamluk carpets but contrasting strongly with the symmetrical Turkish knot.
Most of the group shows field designs that are variants of that seen here. It is generally thought that most of those with larger central panels set within octagons, the corners filled with small triangles, antedate the more typical hexagonal panelled field as seen here. Other field designs are occasionally encountered, such as the overall cloudband design of Persian influence in Berlin (Spuhler, F.: Oriental Carpets in the Museum of Islamic Art, Berlin, London, 1987, no.73, p.66 and pl.p.216), the related example in the Vakiflar Museum (Balpinar, B. and Hirsch, B.: Carpets of the Vakiflar Museum, Istanbul, Wesel, 1988, p.130 and pl.59), one in the Keir Collection which shows the influence of the North West Persian medallion carpets (Spuhler, F.: Islamic Carpets and Textiles in the Keir Collection, London, 1978, no.40, pp.78-79), another recently on the London and New York markets with overall palmette vine (HALI 71, October/November 1993, p.100) which relates to two fragments in the Keir Collection (Spuhler, F.: Keir op. cit., no.39, p.77), and one of clear Ottoman design also in the Vakiflar Museum, Istanbul (Balpinar, B. and Hirsch, B.: op. cit., p.128 and pl.58).
Of the vast majority that have fields almost identical to that seen here, all the small examples (mostly with 3 x 2 medallions) share an identical standard border. Of the largest examples only one has the standard border; the others have differing more complex designs. It is particularly in these border designs that the relationship between the larger carpets with the standard field and those with field variants can be linked. The present carpet is a case in point; its border shares the same design elements as the second of the two carpets in the Vakiflar Museum mentioned above. The same border elements are also found in two other carpets in the same collection (Belkis, B. and Balpinar, B.: op. cit., pls.60 and 61). Both the last two carpets are symmetrically knotted and are ascribed by the authors to Syria or Southeastern Anatolia.
An other fragmentary example exists of a carpet with both the same field and border as here (Volkmann, M.: Alte Orientteppiche -- Ausgewählte Stücke deutscher Privatsammlungen, Munich, 1985, no.5, pp.24-25); the colouring in the border however is altered so that the addorsed arabesques are red, blending into the background far more than here. In his note on the German carpet Volkmann points out that a detail of another example was published by Sarre in the 1908 supplement to the "Wiener Werk". It is possible that the carpet referred to was one subsequently sold at Sotheby's New York, 1 May 1982, lot 295. That example also has the same border, with blue arabesques on a red ground, and with a related but more complex field design of different sizes of radiating roundels and quartered lozenges on a sandy brown ground. In its fluidity of drawing and in particular the power of the addorsed serrated split palmettes the present carpet has a border that makes it one of the most impressive of the entire 'chequerboard' group.
The development of many of the motifs from the Mamluk carpets of the 15th century is clear. The continuous knot small medallion surrounded by inverted or radiating motifs is easy to parallel and the balance of colours, although slightly different in hue, gives a very similar overall effect. The structural analysis is however completely different, in particular in the way the wool here is given a 'Z' ply rather than the typical 'S' of the products of Egypt. A linking piece is a rug of the same structure as the present carpet, formerly in the Pietro Barbieri collection, which has the same field panels as are found here but arranged in a classic 2-1-2 formation, the central panel having a larger version of the design than the spandrels (Viale, M. and V.: Arazzi e Tappeti Antichi, Turin, 1952, pl.150). An attribution to Turkey however must also be discounted since the carpets are knotted with an asymmetrical knot open to the left in the same way as the Mamluk carpets but contrasting strongly with the symmetrical Turkish knot.
Most of the group shows field designs that are variants of that seen here. It is generally thought that most of those with larger central panels set within octagons, the corners filled with small triangles, antedate the more typical hexagonal panelled field as seen here. Other field designs are occasionally encountered, such as the overall cloudband design of Persian influence in Berlin (Spuhler, F.: Oriental Carpets in the Museum of Islamic Art, Berlin, London, 1987, no.73, p.66 and pl.p.216), the related example in the Vakiflar Museum (Balpinar, B. and Hirsch, B.: Carpets of the Vakiflar Museum, Istanbul, Wesel, 1988, p.130 and pl.59), one in the Keir Collection which shows the influence of the North West Persian medallion carpets (Spuhler, F.: Islamic Carpets and Textiles in the Keir Collection, London, 1978, no.40, pp.78-79), another recently on the London and New York markets with overall palmette vine (HALI 71, October/November 1993, p.100) which relates to two fragments in the Keir Collection (Spuhler, F.: Keir op. cit., no.39, p.77), and one of clear Ottoman design also in the Vakiflar Museum, Istanbul (Balpinar, B. and Hirsch, B.: op. cit., p.128 and pl.58).
Of the vast majority that have fields almost identical to that seen here, all the small examples (mostly with 3 x 2 medallions) share an identical standard border. Of the largest examples only one has the standard border; the others have differing more complex designs. It is particularly in these border designs that the relationship between the larger carpets with the standard field and those with field variants can be linked. The present carpet is a case in point; its border shares the same design elements as the second of the two carpets in the Vakiflar Museum mentioned above. The same border elements are also found in two other carpets in the same collection (Belkis, B. and Balpinar, B.: op. cit., pls.60 and 61). Both the last two carpets are symmetrically knotted and are ascribed by the authors to Syria or Southeastern Anatolia.
An other fragmentary example exists of a carpet with both the same field and border as here (Volkmann, M.: Alte Orientteppiche -- Ausgewählte Stücke deutscher Privatsammlungen, Munich, 1985, no.5, pp.24-25); the colouring in the border however is altered so that the addorsed arabesques are red, blending into the background far more than here. In his note on the German carpet Volkmann points out that a detail of another example was published by Sarre in the 1908 supplement to the "Wiener Werk". It is possible that the carpet referred to was one subsequently sold at Sotheby's New York, 1 May 1982, lot 295. That example also has the same border, with blue arabesques on a red ground, and with a related but more complex field design of different sizes of radiating roundels and quartered lozenges on a sandy brown ground. In its fluidity of drawing and in particular the power of the addorsed serrated split palmettes the present carpet has a border that makes it one of the most impressive of the entire 'chequerboard' group.