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.

Internal Elem - Myth or Reality?

by Yon Bard


At the Philadelphia ICOC in 1996, Shiv Sikri gave a talk entitled Internal 'elem' in Baluch Weavings. In his talk, Shiv demonstrated that many Baluch rugs had a design irregularity in the lower half of the 'main area' of the rug, consisting of the field and border system. Typically the irregularity occurs about one fourth to one third of the way up from the bottom. Shiv coined the term 'internal elem' to describe this phenomenon since the elem is an area outside the main area, often with unrelated designs; hence 'internal elem' for a part of the main area with deviant design features. Even though Shiv concentrated on Baluch rugs, the same phenomenon appears in many Caucasian, Turkmen, and other weavings.

While the existence of this phenomenon is beyond doubt (we can supply countless examples), its interpretation is controversial. It was Shiv's contention, and I concur, that this represents a tradition in many rug-weaving areas to insert imperfections in the designs of the rugs, perhaps to satisfy the adage that 'only Allah is perfect,' perhaps for some other reasons shrouded in mystery. I shall attempt to present the evidence for these claims below.

To get the discussion going, I shall show a Baluch balisht which was posted on Turkotek's Show & Tell forum earlier this year:

This rug exhibits a quintessential internal elem. In the discussion that followed its posting everybody appeared certain that this represents a 'change of mind:' the weaver looked at what she'd done, was dissatisfied, and changed design in midstream. Not a shred of evidence for this opinion was presented.


Possible reasons for design irregularities are:
1. Inadvertent errors, possibly (but not necessarily) due to weaver inexperience or incompetence.
2. Esthetic effect, e.g., to avoid 'stiffness' or to achieve a feeling of perspective.
3. Change of mind, because of dissatisfaction with what's been done or due to the need for making the rug shorter or longer than anticipated.
4. For traditional reasons, as stated above.

While several of these reasons may apply to any specific internal elem, the examples that I'll provide will show that only the last one could apply to all manifestations. Furthermore, in many cases this is the only plausible cause. While this does not constitute proof of our hypothesis, it is strongly suggestive. In my opinion, it is unlikely that irregularities occur at this specific location in so many rugs without there being a common underlying cause.

While I could supply examples from several weaving areas, I shall stick to my specialty-namely the Turkmen, with the exception of two surprises at the end. In the interests of brevity, I'll only show one or two examples of each kind, and in most cases I'll show only close-ups of the area of interest. You'll have to take my word for it that elsewhere in the rug the design is reasonably regular, and that I could produce many additional examples of most types.

The types of irregularities that occur within the internal elem framework can be classified as follows:
1. Global: The pattern starts in one form, then switches to another form. The Baluch balisht shown above is an example.
2. Local: The pattern is regular throughout except for a local disturbance at the proper spot. Note that local irregularities cannot be attributed to 'change of mind,' unless you believe that the weaver changed her mind and then reconsidered and returned to the original pattern. The Tekke ensi shown further below is an example.

Within each category, the irregularity can appear in the field and/or one or both side-borders. In Turkmen weavings, there is a preponderance of occurrences in the lower-right border, but examples of all the other possibilities also exist. Some weavings exhibit more than one of these types.

Certainly, not all irregularities fall within the internal elem model. For example, many Turkmen chuvals have guls whose height in one or more rows is different from the others. While in many cases the bottom row has guls that are taller or shorter than the others (which would appear to be an internal elem manifestation), there are examples in which the different row is at the top or in the middle, and even examples where each row is different from the others. I am inclined to attribute differing gul heights to esthetic considerations, but 'change of mind' due to the need to make the chuval come out the right size could also be a plausible explanation in some cases.


1. A Beshir main carpet, detail of lower half shown below. It has a global irregularity in the field: The bottom two rows of main guls are taller, and the secondary motifs different from the rest of the rug, which continues regularly to the top.

2. Chodor main fragment with global irregularity in border: main border starts with ashik motifs, then changes to a different motif. We don't know whether this also applies to right border. Note also peculiar irregularity close to bottom of field: after a few rows of knots, the weaver adjusts the pattern and starts anew. Perhaps this is a 'change of mind.'

3. A tekke mafrash belonging to Shiv. The global irregularity here consists of having three guls in bottom row and four in the others. The piece is so bizarre that I find myself unable to make intelligent comments about why it's so.

4. An animal-tree Tekke ensi with local irregularity. The meander forming the main border motif has a discontinuity in the lower right border, but nowhere else.

5. A Yomud chuval with local irregularity. Two knots are out of place in lower right corner. While appearing insignificant, this example is actually the most important of all. First, it is not alone: in my collection alone there are six examples of this type. The irregularity always occurs at this point: a quarter or third of the way up the right border. It cannot be a change of mind. It cannot be for esthetic reasons - it's hardly visible. It is unlikely to be an error - always occurring at this point and nowhere else. One possibility that comes to mind is that it is a marker that the weaver inserts for some reason; examination of all the examples that I have seen fails to reveal anything for which this could be a marker. Conclusion: it's an irregularity deliberately inserted by tradition. It's interesting to note that this particular effect appears only in weavings of the Yomud and related groups.

6. An Ersari chuval with both local and global irregularities. The latter is represented by the lowest row of guls having different center patterns from the rest. The former is shown primarily by the large amulet in the lower right border panel, which (along with the left panel) is otherwise decorated entirely with three-stemmed floral motifs. Note, however, that it's the top row of guls that's taller than the others.

7. A Yomud mafrash (Vanishing Jewels no. 23) with local irregularity in lower right border: The third motif from the bottom is different. All others are the same.

8. A Beshir prayer rug with a global irregularity in both borders and a local one in the right border. The border motifs start small at the bottom and gradually increase. In the right border only, this culminates in a differently shaped motif, but the normal kind proceeds from there on at constant size.

Woven from the top

All pieces shown so far were woven from the bottom up. What happens when a piece is woven from the top down? Does the weaver insert the irregularities in the normal place, or, ignoring the proper orientation, does she inserts the irregularity near the top, preferably in the left border? The answer, most of the time, is the latter. Examine the following two examples:

9. Ersari kap; local irregularity formed by broken boteh 'handles' in second row from top.

10. Yomud mafrash, local irregularity in left border: pattern in inner secondary border starts out differently; all other secondary borders are regular.

On rare occasions (this is my only example) the irregularity is placed in the bottom half:
11. Beshir prayer rug; local irregularity in both borders, with one kochak motif in outline, all others being solid.


To recap my main points:
1. A very large number of rugs exhibit the 'internal elem' phenomenon, i.e., they have design irregularities in the field or side borders, roughly a quarter to a third of the way up (from the weaver's point of view) the rug.
2. Many of these irregularities are too small to have an esthetic raison d'etre; many are too local to represent a 'change of mind;' the location of the irregularity is too consistent to be attributable to error. Unless we assume that this commonality of location is a coincidence, we have to conclude that the 'internal elem' is dictated by a widespread tradition among weavers.


How far back can this tradition be traced? Very far indeed; at least 2400 years, in fact. Yes, the Pazyryk has an internal elem! Shiv has just pointed out to me that in the picture of the main field below, all the square motifs have little black dots in their centers, except for the ones in the bottom row!

Here is another old example, this time from the classical era: a 16th century Isphahan, recently sold at Sotheby's London sale of the Calachi Collection (10/4/00, no.79). The picture speaks for itself!