One unusual aspect of Sotheby's sale was that it included a group of rugs in which the auction house held an interest. In March of 1995, Sotheby's and two European art galleries (one in London, the other in the Netherlands) collectively purchased a number of old master paintings, Rembrandt etchings, Dutch silver, art nouveau, 17th/18th century glass, and 39 oriental rugs from Joost (Joseph) R. Ritman, a Dutch collector. Thirty of the rugs were offered at Sotheby's London sale in October. All 30 were sold, including a star Kazak and a late 18th century Caucasian rug that brought $88,910 and $50,670, respectively. The nine pieces that weren't offered in London, seven Caucasians, a Turkish rug and an Ottoman embroidery, were included in the New York sale. I thought all of them were of very high quality, and was not surprised to see that they all sold.
Sotheby's sale opened with a Central Asian silk embroidered item, catalogued as a vestment. It was shaped rather like a saddle cover, but the size (6'3"x4'10") suggests that it had some other function. It was elegant, with beautiful colors and workmanship. Despite some condition problems (actually, not severe considering that it probably dates to the early 19th century), it sold for $2,645, around the high end of its very conservative estimate ($1,500-$2,500). It will make a spectacular wall hanging. The only other Uzbek piece, a suzani, went unsold. One of the more interesting pieces at the sale, Lot 71, was catalogued as an early 19th century Central Asian carpet (9'9"x4'6"). The field consisted of 11 rows of three large rosettes, the border was not typical of any of the weaving areas with which I am familiar, but was probably closer to northwest Persian than to any other. The long, glossy wool, the goat hair selvedges, the palette of only seven colors, and the remnant of a kilim end led me to believe that it is a Baluch weaving, probably made around 1900. It was an extremely attractive and unusual piece, and in quite good condition. Estimated at $8,000 to $12,000, it sold at $12,650.
The sale included eight pieces catalogued as Baluch, of which four were sold.
There were 11 Turkoman pieces offered, seven were sold. These figures do not include Lot 256, an early 20th century workshop carpet of Turkoman design. Lot 11 was a Yomud ukk-bash, of the seldom encountered zigzag stripe variety. It had wonderful wool and a nice, draping handle. It sold for $3,450 (estimated at $3,000 to $4,000), which seemed quite reasonable despite some breaks in the foundation. Lot 99, an odd size Ersari bagface (2'9"x1'9") with excellent wool and in quite nice condition, sold at $8,050. I had thought the $7,000 to $9,000 estimate was far too optimistic. Obviously, I was mistaken.
The stars of the show, to my eyes, were Lots 17 and 45. These were both splendid little (around 5'x4') Qashqa'i rugs, finely woven, beautifully drawn, and with similar palettes and related field designs. Lot 17 seemed artistically superior, Lot 45 was more of a tactile delight, but both were outstanding. Lot 17 failed to sell against its estimate of $8,000 to $10,000. Lot 45 sold for $17,250, right at its low estimate ($15,000 to $20,000). This piece sold for $18,700 at Sotheby's in December 1990. It is interesting to reflect on the reasons why the price should be somewhat lower five years later. Perhaps the market is simply softer now; perhaps there is less demand for outstanding Qashqa'i work now; perhaps the 1990 underbidder was the 1995 purchaser and finally got it for exactly what he was willing to pay in 1990!
There were seven other south Persian village or tribal rugs, none particularly distinguished. Only two of these failed to find buyers.
There were 48 Caucasian weavings, mostly pile rugs. Of these, only 12 were bought in. Lots 35 and 105, two arresting Karachopt Kazaks from the Ritman collection, sold for $12,650 (estimated at $10,000 to $15,000) and $17,250 (estimated at $14,000 to $18,000), respectively. A Sewan Kazak from the same collection, with a wonderfully abrashed lime green field, sold for $51,750 (estimated $25,000 to $30,000). The most interesting flatweave in the sale was Lot 9, a Caucasian verneh with a black field decorated with columns of multicolored men on the backs of multicolored horses, each with an accompanying goat and dog, alternating with columns of pack-laden camels. Estimated at $10,000 to $15,000, the piece sold for $14,950.
I had a brief encounter with Lot 44, a soumac khorjin face. It was lying on a countertop, and the glow of the colors caught my eye from 25 feet away. When I picked it up, I realized that the glowing colors were the back of the piece, the front actually being kind of grubby looking. I put it down without paying much attention to it; the estimate of $700 to $900 seeming about right for a drab but once beautiful piece. It sold for $4,025. Evidently, I put it down too quickly.
Lot 78 was a classic, ivory ground floral lattice Shirvan prayer rug inscribed with the date, 1892. It sold for $4,600 (estimated $2,500 to $3,500).
Two Turkish "Yuruk" (the word simply means "nomad" in Turkish) rugs were among the highlights, although for quite different reasons. Lot 51 was an early 19th century rug from the Ritman collection, with a deeply corroded dark brown field that set off the relatively formal arrangement of motifs within it. It sold for $16,100 (estimated $15,000 to $20,000). The other, a smaller piece (Lot 101; about 6'x3') dated 1871, had a single row of human hair woven into the pile about a third of the way up from the bottom (someday, somebody will find a rug in which this technique has been used to place a moustache on a face). I'd describe this one as interesting, rather than beautiful. It sold for $2,300, the low estimate ($2,000 to $3,000). Lot 27, a Konya runner, had what I imagine to be marvelous colors hidden by considerable dirt. At least two bidders recognized the treasure beneath the murky surface, and it was knocked down for $10,350 (estimated at $5,000 to $7,000).
There were six 17th century silk velvets and embroideries offered. Lot 57 was a green silk and metallic thread Ottoman velvet pillow. It had to be viewed from several angles to fully appreciate how beautifully it captured light, although its elegance was obvious even from a distance. Estimated at $10,000 to $15,000, it sold for $11,500. Lot 107 was another silk velvet, this one from Safavid Persia. With an unusual design consisting of rows of pairs of robed men around nude female figures, alternating with rows of pairs of men around flowers, it was striking, indeed. Estimated at $4,500 to $5,500, it sold for $21,275. The most outstanding 17th century silk velvet item, though, was Lot 23, a Mughal India piece of extraordinary quality and aesthetics that sold for $51,750 (estimated $25,000 to $35,000).
I seldom find Chinese carpets to be interesting, but clearly there are many folks out there to whom they have great appeal. Of 13 offered in this sale, nine were sold. Lot 28 was a most unusual 18th century carpet. The buff field is strewn with an assortment of artifacts, mostly drawn in blue and yellow. The design is known as "100 antiques," although I counted only 63, many of which appear two or three times each. It sold for $34,500 against an estimate of $10,000 to $15,000.
There were, of course, a very large number of Persian carpets in the sale. I thought Lot 111, a 6'x9' Gerus Bijar wagireh, particularly handsome. So, apparently, did at least two bidders, and it sold at $12,650 (estimated at $5,000 to $7,000). There were a surprising number of 20th century Persian rugs and carpets in the sale, including six Isfahans dating from after World War I, and four Nains from after World War II. Of these 10 pieces, only one was bought in. There were also two Bezalel carpets, made in Jersualem around 1920, that both sold. One of these, Lot 243, was particularly attractive and harmonious, and unlike any of the better known genera of oriental rugs. It sold for $18,400 (estimated at $14,000 to $18,000).
Sotheby's had nine pieces from the Ritman collection; Christie's had ten from Arthur Sackler's. All nine of the Ritman pieces sold, as did nine of the ten Sackler consignments. It would be very interesting to know how much of the success of these collections reflects a willingness to buy a distinguished provenance. Selling 18 of 19 pieces at sales where overall percentages sold were 63% and 45% suggests that this was an important element. However, nearly all of the lots in both collections were truly excellent, and I suspect that the provenance significantly increased the prices but that the pieces probably would have sold even if offered anonymously.
Christie's offerings included more than their usual proportion of collectible pieces. Their catalog included 15 Turkoman weavings, of which only four sold. While many of them were rather pedestrian, a number of those that were bought in struck me as particularly nice. For example, Lot 156 was a handsome Beshir carpet with excellent colors and no serious condition problems. It seemed reasonably estimated at $4,000 to $6,000, but failed to sell. Another, Lot 126, was a nice dating to the first half of the 19th century. From the Jon Thompson collection, it sold at Sotheby's two years earlier for $2,300. Since then, it had received considerable high quality restoration and looked terrific. It was bought in (estimated at $4,000 to $6,000), proving that distinguished provenance, good age, and a reasonable estimate are no guarantee of a sale.
The most impressive of the Turkoman pieces was Lot 107. This was a Yomud carpet, plausibly attributed to around 1800. The colors, the drawing of the kepse guls, and the excellent border and end skirts would set the heart of any lover of Turkoman work aflutter. It was estimated at only $6,000 to $8,000, probably reflecting the fact that the condition was a little rough. Not awful, mind you, and this is a piece with considerable age. Just a little rough. I thought it was estimated far too modestly, and the buyer got a real bargain at $7,475.
Lot 39 was a fragment of a Tekke "three Salor gul" juval. It had excellent pile and color, but consisted of only the field, parts of three borders, all of the fourth border and about an inch of the end skirt. Estimated at $2,000 to $2,500, it sold for $2,300. Lot 1 was a Yomud mafrash in the ak-su pattern, generally only seen on torbas. Very unusual, although not to my taste, it sold for $4,830 (estimated $2,500 to $3,500). Seven rugs from East Turkestan were included. Three of these sold. Lot 101, from the Sackler collection, was an especially attractive Khotan carpet. Although dating to around 1800, the pile was in rather good condition except for being heavily matted. I suspect that a good bath will correct that. Estimated conservatively at $7,000 to $10,000, the $20,700 selling price seemed reasonable.
Christie's had 48 Caucasian pieces, and I thought most of them were at least pretty good and reasonably estimated. Only 17 of these were sold (35%), a remarkably low figure for Caucasians, which are generally in high demand.
Lot 111 was among the more interesting of these. It was a Fachralo Kazak prayer rug, attractive, with a good strong green and in nice condition. It had an inscription and a date woven into it and an estimate of only $4,000 to $6,000. Despite having "all the right stuff," it failed to find a buyer. Perhaps the problem was that the inscribed date, 1930, was too recent for the Caucasian collectors in the crowd. There did not appear to be any synthetic dyes in the palette and, if the date had not been inscribed, I would surely have attributed it to around 1875. The catalog attribution was "late 19th century," but I see no reason to disregard the inscribed date and would cite this rug as evidence that a few really nice Caucasian rugs with natural dyes were still being woven in the second quarter of the 20th century. I suspect that many (perhaps most) of the Caucasians that we attribute to the 19th century were, in fact, woven in the 20th century.
The most expensive Caucasian that sold was Lot 133, a Shirvan runner. It brought $25,300 against the $15,000 to $20,000 estimate. No other Caucasian had a five figure sale price.
Lot 27 was a long rug from the South Caucasus. It had lively, cheerful colors and sold for $6,325 (estimated $5,000 to $7,000). A rather informal Talish long rug, Lot 69, had a boteh-covered field and nice, fleecy wool. Estimated at $8,000 to $10,000, it sold at $8,625.
Lot 73 was a Genje long rug with a field of "dice" laid out on diagonal stripes and a border with a Turkoman flavor. One unusual feature of this handsome rug was that a corrosive black was used for some of the diagonal stripes and as the center of some of the "dice", rather than for outlining only. It seemed a good buy at $8,280 (estimated at $8,000 to $10,000).
Lot 139 was an East Caucasian soumac rug, in rather good condition. I thought it a little busy, but the selling price of $5,750 (estimated at $5,000 to $7,000) was certainly reasonable. Speaking of busy, Lot 223, a Karagashli long rug that did not find a buyer, may be the busiest Caucasian rug on record. There is scarcely a spot on it that is more than half an inch from a little animal, a rosette, or some other small motif, and the impression of busy-ness is further emphasized by the color juxtapositions.
I thought Lot 195 to be an outstanding Karachopt Kazak and expected it to easily draw the $12,000 to $15,000 estimate, but it failed to sell. My wife noted that the central motif had the appearance of a face staring at the viewer and commented that she didn't like rugs that look back at her. Perhaps the potential buyers agreed.
Baluch rugs fared very poorly, indeed. Five were offered, not a single one was sold. None of them was truly noteworthy (except, perhaps, Lot 214, at 14 feet the longest Baluch I've seen) and all seemed rather optimistically estimated. The lowest estimate on any of the five was $1,800.
Turkish tribal and village rugs did relatively well at this sale, although none drew terribly large sums. Of 11 pieces, eight were sold.
The best of the group was Lot 5, a good looking Kurdish rug with long pile, braided ends and a simple design of large hooked lozenges in the field. It sold for $6,670 (estimated $3,500 to $4,500). Lot 32 was a Ghiordes prayer rug attributed to around 1800. Its rough condition was reflected in the estimate of only $1,000 to $1,500 but, by the time it was hammered down, the selling price was $4,370. Lot 170 was an 18th century Konya long rug with some large areas of wear but otherwise quite a nice specimen. It sold for $5,175 (estimated at $5,000 to $7,000).
South Persian rugs, too, did reasonably well. Six out of nine were sold, and the unsold threesome included a Qashqa'i kilim that looked as good as the catalog photo. Lot 180, a Qashqa'i "millefleurs" rug in very good condition, while not an outstanding example of its type, seemed like a good buy at $6,900 (estimated at $6,000 to $8,000). I expect that Lot 7, a small Afshar rug, has excellent colors hidden behind that schmutzig surface. If it does, someone got a real bargain at $2,530 (estimated at $2,500 to $3,500).
As expected, Christie's had many workshop carpets, mostly Persian but including 10 from India. The most striking of these, a turn of the century 19'x14' Agra carpet with a spacious, large scale derivative of the Herati pattern on an ivory field failed to sell (estimated at $80,000 to $100,000).
It's really difficult to make sense of the relatively poor results of these sales. The offerings seemed to be of decent quality overall, the estimates were generally reasonable. The dollar is at very low levels against the major western European and the Japanese currencies, so many of the prices were at bargain basement levels for foreign buyers. South Persian and Turkish village and tribal weavings did relatively well, but there were few other bright spots. I suppose that, like everyone else, I'll just have to wait to see whether this is a trend setting off or just one of those hiccups that markets get every now and then.