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.
Weave Pattern: A Holistic Perspective on Rug Structure That May Be The Most
Widely Used But Has Not "Caught On" With
Rug Collectors, Scholars and Writers.
by R. John Howe
OK, let's look at four different types of Hamadan. These images were not chosen by Neff and Maggs but rather
by Willborg, attempting to use their approach in his catalog. Neff and Maggs include a Hamadan in their book and
we may put it up for discussion at some point.
Here are the four sets of images and text. Remember that Willborg also includes a closer technical description.
This first piece is an old Hamadan of the type perhaps encountered most frequently; one in which no village attribution seems possible.
"This runner is typical of the large majority 19th century Hamadan rugs that cannot be attributed to a particualr village or nomadic tribe. Grote-Halsburg (Band II Tafel 55) illustrates a first half 19th century runner that probably belongs to the same group. He calls it middle Persia (Hamadan - Mianeh). Ours is more geometric and a later example. Analysing the revers we again find a similarity to No 22 (the Bibikabad). The woolen foundation is beige, the pile yarn is Z2S plied and the weft shoot with medium tension, slightly depressing every second warp. This is a technique that verges toward Senneh - the back is a bit knobbly.
In Iran these rugs are called shotore - meaning from camel. As already explained one seldom finds real camel-hair in a rug - it is sheeps wool overdyed with walnut husks. This runner has 2 shoots of weft for a small portion of the rug. This practise of mixing 1 and 2 shoot is not unusual in rural Persia. Maybe it was done to correct the weaving into balance, or sometimes a second weaver (of different origin) took over.
The carpet has an interesting provenance. I found it in Marakesh (Morocco) where it had been brought from Turkey many years before. In Turkey it had been used in a staircase. The different areas of wear can easily be seen. The superb colours and wool point to a 19th century dating."
Size: 474 x 99 cm
Warp: wool, Z2S, beige, 0o displacement, slightly sinuous
Weft: wool, Z2S, beige, 1 shoot, c 192 cm with 2 shoots, 16 cm with cottonwefts Z2S
Pile: wool, Z2S, symmetric knots, Vert 26/10 cm. Hor. 34/10 cm - 88.400/m2. V/H ratio 1:0,76
: 1 cable (4 warps) overcast with brown wool originaly. Now overcast with brown wool over cable and 2 rows of knots
: not extant
: (19) ivory, yellow, pink, pinkred, windred, orangered, red, apricot, lightgreen, green, olivegreen beige, grey, brown, darkbrown, lightbrown, blue, navyblue, blackish-blue."
The second type Hamadan is an Enjilas.
"One of the most renowned of all the Hamadan villages is Enjilas. It lies about 20 km to the north of Hamadan, just about halfway to Malayer. The Enjilas weavers take great pride in their work and you never see a bad rug originating from there. If the quality is lower than expected, the rug is probably from Everu, a village lying to the south.
Characteristics of the Enjilas rug is; fine weave (often 200.000/m2) slightly smaller dozar format than customary, many narrow borders (i.e. 7 in this piece), high quality wool, medium high clipped pile. In principle one will only find two patterns, herati and boteh. A small lozenge medallion sometimes occur, the corner spandrels are exceptionally small often containing a leaf, a flower or part of herati. This old rug has a disjointed, loose version of herati. The patina of the colours and the woolen weft point to an early dating.
Size: 187 x 137 cm
Warp: cotton, Z6S, Z5S, ivory, 0o displacement, slightly sinuous.
Weft: wool, Z2S, Z3, black, red, 1 shoot, low to medium tension
Pile: wool, Z2S, Z2, symmetric knots. Vert. 42/10 cm. Hor. 43/10 cm = 180.600/m2. V/H ratio 1:0,98
Sides: 1 cable (5 warps) overcase with red wool (partly new).
Ends: beginning - 2 cm balanced plain weave in ivory wool, blue and yellow woolen weft twined thread along the middle; end - 2 cm balanced plain weave in ivory wool, blue and red woolen wefttwined thread along the middle.
Colours: (9), ivory, pale yellow, ochre, deep red, mid blue, navy blue, bluish green, yellowish olivegreen
dark tan (on dark wool)."
A third type of Hamadan is a Malayer.
"Malayer is not only a famous small town, but also lends its name to many surrounding villages. Hence it is always difficult to exactly pinpoint the village of manufacture. Nenej and Jokar, already mentioned in no. 27, are two such places that sometimes weave an indistinguishable product. Qualities run from coarse (many of those rugs are probably made in Hosseinabad to the northwest) to extremely fine. Some of the finest old rugs are sometimes called Mishin. Many designs incorporate herati, both in the field and in the medallion. Many Malayers made between 1900 - 1920 use a strong synthetic red (as in this example) that does not fade but is prone to bleeding in water.
Size: 200 x 156-62 cm
Warp: cotton, Z6S, white, 0o displacement.
Weft: cotton, Z3S, white, 1 shoot low tension
Pile: wool, Z2S, symmetric knots. Vert. 43/10 cm. Hor. 34/10 cm = 146.200 m2. V/H ratio 1:1,26.
Sides: not original
Ends: beginning - 1 cm balanced plain weave in white cotton, thereafter 3 cm frringe; end - not extant
Colours: (12), ivory, redbrown, hot synthetic red, rosered, pink, dark blue (oxidised) midblue, light blue,
black, brown, greyish brown, mustard yellow."
The fourth type Hammadan is a Borodjert. I know I have seen one only because I recognize its very distinctive back.
"This is a more typical Borodjert compared to the previous example. It has an overall palmette, rosette
and leaf design that can be traced back to 17th century classical Persian carpets. The pattern in light blue and
pinkish tones stands out well against the midnight blue ground. The wool is glossy and the dyes appear organic.
A later version of this design can be seen in Ford (fig. 462). The white specs of warps on the reverse are very
apparent here. Borodjert is the southernmost outpost that can be included in the Hamadan group. It lies about 120
km due south of Hamadan, just to the north of the Saraband area. It is probable that some singlewefted rugs with
the Saraband boteh pattern originate in Borodjert.
Size: 231 x 143 cm
Age: 1910 - 1930
Warp: cotton, Z8S, white, slightly sinuous, 0o displacement.
Weft: wool, Z2S, beige, light brown, 1 shoot.
Pile: wool, Z2S, symmetric knots. Vert. 37/10 cm. Hor. 31/10 cm = 114.700/m2. V/H ratio 1:1,19.
Sides: 1 cable (5 warps) overcast with black wool.
Ends: beginning - not extant; end - not extant
Colours: (13), ivory, grey, beige, pink, greenish beige, olivegreen, lightred, red, brownish red, blue (abrashes),
greenish blue, midnight blue (abrashes), dark brown."
And here again, courtesy of our good friend Vincent Keer, are the four backs of these pieces so that you can more directly compare the weave patterns they exhibit.
The key for these four types of Hamadan is: ham2 = Old Hamadan without village attribution; ham6 = Enjilas;
ham10 = Malayer; ham14 = Borodjert
I think that there's no question that the Borodjert is quite distinctive. The old Hamadan without village attribution is rather muddy for me in this example. Perhaps its weave pattern would be more visible in a piece with different colors. The Enjilas seems to me to be different from the Malayer and both from the Borodjert, so it may be that weave pattern would be useful in distinguishing many Hammadans.
But let's move to the Salon tasks.
First please re-examine (in my summary above) the language used by Neff and Maggs (it might be useful to print it off so that you have it in hand as you examine examples).
Then look more closely at the weave patterns in these eight examples.
The task of the Salon is to consider what we think about "weave pattern" as an aid to attribution.
Are Neff and Maggs right when they say that if we do not use it something is being left out and in fact that we expose ourselves to error? What counter arguments might we offer them?
I also propose that we look about at the rugs we have at hand for backs that have either very "typical" weave patterns for a given type or for those that seem rather unusual.
For example, what does a Laver Kerman look like? Is the weave pattern of a Laver Kerman distinctive from those of other Kermans or is this a design distinction?
Do Turkish rugs (that often seem very similar to me) exhibit useful distinctions of weave pattern?
What does the back of a Talish look like? Is it distinctive enough to help in that attribution?
As we go along, I will put up some additional examples that Neff and Maggs provide and ask some questions about them.
OK then, what do we think about the notion of "weave pattern?" (With as many interesting direct scans as we can manage.)
R. John Howe