Marla Mallett

Hi Folks,

On another thread there are lengthy speculations on the use of ensis as yurt door covers, and the extent to which they were used for that purpose. Since that discussion then veered off in another direction, I'd like to revisit it here, with a technical aside.

A few years ago when I was doing technical analyses of Turkmen rugs and bags, I encountered a detail on Yomut and Saryk ensis that seems relevant to this discussion. Occasionally people have mentioned that unusual wear would result from continual pulling at the sides of an ensi if it were hung as a door curtain. I agree that this would be a serious matter for the weaver to consider.

I was investigating several unusual knotting practices in Yomut rugs and bags, when along with many other kinds of irregularities, I discovered extraordinary strengthening along the entire selvages of several ensis. I'm not talking about the ubiquitous "edge" knots that occur on so many other pieces, and that have been written about by others, but instead, series of 6 to 9 overlapped knots used at the end of every knotted row--along the entire length of both sides of some ensis. This strengthened the edges immeasurably. I was interested at that time to know if this practice only occurred on older examples or if it appeared also on later, more likely commercial pieces. I must admit that I never followed up on this idea to reach definite conclusions. I don't remember now how many ensis I checked for this feature, but I don't believe I looked for this in examples from other tribal groups. In any case, this unusual detail seems like a significant clue to the use of these rugs. I would enjoy hearing from Turkotekers who have ensis that they could examine.

I wrote about this and illustrated the knotting detail toward the end of my first website WOVEN STRUCTURES UPDATE page: www.marlamallett.com/updates.htm.

Best,
Marla Mallett

Steve Price

Hi Marla

I have a Tekke ensi, probably late 19th or early 20th century (the type made by the thousands for sale to the west). The knotting at the edges is quite ordinary.

Regards

Steve Price

Note added July 9: Correction. There is double knotting along the entire length of both sides, just inside the selvages.

Paul Smith

Hi Marla--

I have an Ersari ensi, I guess 4th quarter, 19th c. I'm not sure if the selvage matches the characteristics in your update, because I'm not entirely sure what an Ersari selvage would normally look like (my other Ersari rug has been re-selvaged, alas).

What struck me about this ensi is that it has damages in the upper right part of the rug that could be associated with being hung. There is a nick along the top towards the right side, and an oddly-repaired tear in the right selvage towards the top; these damages along with evidence of torn-off cords in the upper corners seem to point to a rough life of hanging at some point in this ensi's existence. It would seem to me that the reinforcement you describe would be to prevent this sort of thing.


Regards,

Paul

Martin Andersen

Hi Marla

Interesting and detailed point

I have an Ensi which I would guesstimate to be just before Russian commercial period.



It has losses in top and bottom, which perhaps could point in the directing of hanging wear, as it is otherwise in okay condition.

Must admit i'm no good at the technical descriptions, so here are a few closeup photos:







I suppose this not what you are looking for, it doesn't really match your description. The 2 knot towards the overcast (which I suppose is not original) look a bit thicker and more messy from the back than the rest of the knots, but I suppose its part of mingling with the original (red?) overcast.

best Martin

Marla Mallett

Steve, Paul and Martin...Thanks.

Paul,

The reinforcement I was describing was NOT in the construction of the selvages, but in the ends of the rows of knotting themselves--6 or more OVERLAPPED KNOTS actually within the body of the rug. In your back side photo detail , it appears that there is no disruption of the ordinary knotting sequence near the edge of the rug. In the examples I was describing, the knotted reinforcement occurred down the entire length of each side of the rugs.

If we were to find extra reinforcements at the top corners, that would also be interesting, of course. On occasional kilims from eastern Turkey, we find little woven tabs that extend from the corners of the pieces for the purpose of hanging. Am I remembering that a while back someone turned up an ensi that had such tabs at the top corners? Or am I dreaming?


Martin,

It's risky to make guesses from photos, but the two outside knot columns on your rug indeed appear to be heavier that the rest of the rows. I can't tell from the photo, but there may either be extra knots "stacked" one above the other on each row, or the two outside knots may be tied with extra yarn singles--more wool--to strengthen the edges. This is common on Tekke rugs, mafrash, torbas and chuval, so doesn't represent a practice unique to ensis. It is also common to find a couple of ordinary symmetrical knots used along the edges of these asymmetrically knotted Tekke pieces, as they provide a more stable edge... sometimes several such knots.

Best,
Marla

Martin Andersen

Hi All

I hope more people will put up some samples with photos of edge knots and selvages.

It is not that I dont think that the Ensi format has been used to cover the door of the yurt. But I do have trouble seeing it as a everyday door. The constant wear of pulling it up and down or to the sides, children in and out. And how about the wear of sun, rain and snow of the rough weather in Turkmenistan? Where are the sun blasted surviving old samples?

Looking at the drawing there is a thinner textile at the sides of the Ensi which perhaps could have protected the Ensi from the sun. But again if it were for everyday use, wouldn't it be rather irrational to put up a permanent door that needed permanent cover and protection? And looking at the drawing, a decorative thin protection with tassels, which looks a lot more fragile than the pile weave .



Even though Marla may have found a very interesting weave technical way of segregating later rugs that where made for purely commercial reasons, I would still think that the original door hanging of the Ensi where for festive or ceremonial occasions.
Well okay, I may underestimate the weavings ability to handle weather and wear.

best Martin

Steve Price

Hi Martin

There's another thread with lots of discussion about the points you raised, and it's my impression that nobody disagrees with any of them. Marla created this thread in order to gather technical information from our readers, intentionally separating it from the one in which ensis uses, significance, etc., is discussed. I ask that we handle this thread the way she requested - technical details only, especially about knotting near the sides.

Thanks

Steve Price

Martin Andersen

hi Steve and Marla

Sorry - didn't intend to digress the thread.

best
Martin

Wendel Swan

Hello Marla,

Your observation of reinforcement knotting on the edges of some ensis is astute. It’s such a simple and practical concept, but too often we don’t observe what it plainly there to be seen (I certainly never noticed this), nor do we find the time to pursue similar intriguing ideas.

Perhaps reinforcement might be found on the sides of other weavings with cords, such as asmalyks or chuvals, although you said that you had looked at some bags for this purpose. In particular, readers might examine Turkmen main carpets, especially Yomud, for this reinforcing structure or other evidence that cords originally extended from the four corners.

Below is an image of a Yomud main carpet fragment formerly owned by Robert Pinner and a detail of a Yomud ensi that I once owned.



Note the cord on the corner of the main carpet. In the ensi, notice the vestigal kilim from which the braided cord would have emerged. The bottom corners of the ensi were rewoven. The wear pattern and restoration always suggested to me that originally there were cords attached to the bottom edge of the ensi as well.

The results from such a wider re-examination of all pile weavings with cords might not be consistent, but my bet is that reinforcement knotting will be found on more than ensis.

Incidentally, we know that in the Caucasus both large flat weaves and pile carpets were often hung with the warps horizontal. Doesn't the presence of cords on main carpets indicate a similar used among the Turkmen?

My best to you,

Wendel

Frederik May

Hello Marla,

A really interesting point. I have a Yomut ensi which seems to have a thicker structure on the last knots along the entire length.







Maybe I am wrong? Please help?

Regards

Frederik

Marla Mallett

Frederik,

The detail of your ensi looks interesting. Would it be possible for you to take a detail photo of a small section that includes a portion of the heavier border area from the FRONT, with the rug folded horizontally along a weft to expose the knot collars? Then we could see if the knotting is overlapped.


Wendall,

Very interesting--that Yomut ensi detail with the little woven tab at the corner. I can't imagine any reason for such an extension, unless it were for hanging the piece. I can't tell from the other photo, whether or not the cord on the corner of the main carpet was formed with intact warps, or if it may have been a separate added element. If it has been added, there is probably no way to know whether it was done by the weaver, or by someone else, even possibly in the West, for hanging.


Everyone,

Not trusting my memory, I've just now looked through my notebooks of analyses of approximately 400 Turkmen weavings I did a few years ago. Among other things, I was interested in knotting anomalies back then, and made careful notes of what I found. Most of the strange anomalies were used by weavers to clarify and adjust designs, or to straighten uneven, crooked weaving. But some were done to strengthen the edges.

In ASYMMETRICALLY knotted pieces, including Tekke, Saryk, Yomut, Goklan, Arabatchi, Chodor and Ersari bags and rugs, it was very common to find SYMMETRICAL KNOTS USED TO REPLACE ASYMMETRICAL KNOTS along the edges. This sometimes occurred along just the left edges on pieces having asymmetrical knots opening to the right. This substitution occurred sometimes methodically, sometimes very erratically. There were sometimes just single knots substituted now and then, sometime pairs, occasionally several. This practice occurred equally on bags of all sorts, and on rugs. Sometimes these substituted knots were heavier, i.e., using more singles together, than in the rest of the piece.

On symmetrically-knotted Yomut and Saryk pieces--bags of all sorts and rugs-- I found OVERLAPPED KNOTS occasionally in border areas, but this tended to be sporadic. Only in the ENSIS I described in the website notations did I find CONSISTENT, VERY CAREFUL USE OF 6 TO 9 OVERLAPPED KNOTS all along the edges, next to the selvages.

On Salor pieces I found none of these irregularities near the sides.

On assorted Goklan pieces I found symmetrical knots substituted along the selvages, with occasional pairs of knots overlapped.


Best,
Marla

Patrick Weiler

Marla,

Here are some pictures of a Tekke ensi with what appears to be two rows of heavier knotting along each side:

The selvedges are not extant along both sides. This second photo is of the top, left corner. The top several rows of knots are missing, so this is not the original top of the weaving:

And here you may be able to see better the two rows of thicker knots a bit farther down the piece:


Patrick Weiler

Steve Price

Hi Marla

I have to correct what I said in my post about my Tekke ensi. I examined it again, and it does, indeed, have double knotting for three or four rows of knots along the full length of each side. Like the others in this thread, the double knotting goes from the selvage to the the external edge of the border.

My apologies for not recognizing what I was seeing.

Regards

Steve Price

Marla Mallett

Hi Steve,

Thanks. But what exactly do you mean by "double knotting"? Are there extra knots stacked one above the other in each row? Or are the knots just thicker because they are made with more yarn singles? Are these symmetrical knots, rather than asymmetrical knots as in the rest of the rug?

Best,
Marla

Wendel Swan

Hi Marla,

You posted:


I may have seen the fragment, but I can't recall details. However, the image seems to show the cord emerging from 4 or 5 bundles of warps. To me, everything about it looks as if it came off the loom with extended warps for braiding.

While it may be rare to find main carpets with cords on the corners, Robert's fragment is not unique. I have seen images of others, but this is the only one I could find.

The existence of these cords and the reinforcement knotting that you have noticed raises the larger question of the alternative uses to which ensis, main carpets and other pile weavings may have been put.

Best,

Wendel

Richard Larkin

Hi Marla, et al,

Sorry, nothing very interesting to report, but I do have a Tekke ensi with a feature that may be instructive to others examining their merchandise. As you will see, it is a pretty typical "late nineteenth century" example.



The notable feature is that in some parts of the rug, the weaver employed symmetrical knots in the three vertical rows (all red) nearest to the selvage overcasting. However, in other parts, she, or her sister, went back to asymmetrical knots for the whole business. The right hand detail image shows the symmetrical knots in the upper part and the other technique in the lower. There is a mixed approach to this aspect of the matter on the other side of the rug, with the number of visible rows of symmetrical knots changing to two or one, depending on where you are in the rug. (I'm not sure whether there is still another row of knots covered by the selvage wrapping.)

Another point worth mentioning is that in the area of the symmetrical knots, the rug feels distinctly thicker. It is very soft and supple in texture, and running the fingers over the back in this area discloses a distinct raised area where there is symmetrical knotting. A careful examination of the yarn singles (thanks, Marla), as well as the underlying warps, indicates that this effect is not the result of extra material. Evidently, the difference in knot formation accounts for it.

I have another Tekke ensi that is finer in weave, and it also has some symmetrical knotting near the selvages. The differences in the feel of those areas can be discerned, but much less clearly than in the posted example.

Rich Larkin

P. S.: One of the benefits of being directed to the Marla site is that one finds one is regularly hitting on approximately three out of five of the proprietor's pet peeves. As painful as such an exercise is, it must be done, and can only be of benefit in the end.

Frederik May

Hi Marla,

I was not able to fold the rug properly as it is increrdibly tight and thick there and it was almost impossible to shot a good photo. But I was able to see that the knots were a lot thicker or bigger than in the inner parts and I think there were double knots. I must admit that I am not trained enough to judge properly. But I was able to make an other observation: Folding the rug vertically which is a lot easier I could measure (front side) that the outer vertical rows are a lot wider (almost double width). To me specially the strucure of these lateral rows looks like the one of Rich's Tekke rug. So maybe only thick symetrical knots. I think I will try to make some pictures in the daylight this weekend.

Regards

Frederik

Marla Mallett

Everybody,

Thank you all for your comments and photos. These are rather difficult details that I've been asking about, and so it is not surprising that there has been little or nothing in the literature about them. I hope you don't mind me following up with questions. We may be able to uncover practices that I've not seen before.


Wendal,

I'm still puzzled by the corner braid on the Pinner main carpet. It's a shame we can't see this detail more clearly. The braid appears to be at least 12 to 15 inches long or so doesn't it? The process of braiding, would have shortened the group of yarns considerably, suggesting that they must have been close to 20 inches long or so before the braiding. I've never known of an instance when primitive weavers would have wasted so much warp! (15 to 20 inches of extra open warp yarns, all across the full width of such a rug is a LOT.) Only 8 or 9 inches of "extra" warp length is needed on a rug loom to allow for either tying on or opening a shed. Western weavers are stingy about not wasting warp; I would expect anyone using hand combed, handspun, hand-plied yarns to avoid such extravagant waste. At any rate, I would be astonished to learn of such a practice occurring on more than an isolated example. If it does indeed appear on other pieces, I would have to conclude that the weavers were warping their looms with separate (perhaps weighted) narrow warps at the edges--a difficult, unwieldy practice I have not heard of, or seen evidence of on Middle Eastern or Central Asian weavings, and one even more difficult on a loom staked to the ground, as are most Turkmen looms used for main carpets.


Pat,

Do the knots you say are "thicker" look like those in Rich's detail photo? Symmetrical knots replacing the asymmetrical knots used in the body of the rug?


Rich--and all,

Thanks for your detail shot. That shows the shift from one knot type to the other very clearly. It also demonstrates the erratic way that this practice was followed. I have squinted upside down, backwards, and sideways at that photo, but can't determine if more singles have been used in those symmetrical knots or not.

This photo of yours, Rich, illustrates quite clearly exactly WHY the substitution of symmetrical knots was practical: When an asymmetrical knot opens to the right, the left side of the knot is exposed, without any pile sticking up to cover the small collar. Of course it's also true that symmetrical knots are more secure. It's for these reasons that this practice occurs on all kinds of asymmetrically-knotted pieces--bags as well as rugs.

Thanks also for your comment on my short list of Rug Book Pet Peeves! I'm not aware that the particular tantrum displayed at the end of my WOVEN STRUCTURES UPDATES, Page 1 has changed anyone's use of weaving terminology.


Frederick,

From your description, you may have an example of exactly what I'm searching for, so if you don't mind, please try again to do a close-up from the front, with the rug folded back along a weft. I'm sorry that I don't have an example of such a thing at hand to photograph, as such a photo would clarify for everybody what a series of overlapped knots at the edges looks like.


Best,
Marla

Rich Larkin

Hi Marla,

This is the back of the same section I posted yesterday.



Referring to the three vertical rows of red knots between the selvage wrapping and the brown vertical line, the knots above the white marker line are symmetrical and those below are asymmetrical. As near as I can tell, all of the knots consist of a pair of yarn singles, although scrutinizing this image seems to turn up a few containing three yarn singles. In any case, at the end of the rug in this section, one or two symmetrical knots that are unraveling themselves definitely contain two yarn singles. Looking at the tops of the knot collars of the symmetrical knots (e. g., yesterday's image), one finds two singles per knot consistently. There appears to be no doubling of knots in the symmetrical section, i. e., no cramming of extra knots there in the vertical stack.

Regarding the Pinner Yomud posted by Wendel, it certainly looks as though that cord is the extension of two or three inches worth of original warp ends gathered at the edge of the rug. Your reflections on the matter from the point of view of a weaver are interesting and illuminating. This type of comment always makes me wonder what else we are overlooking because we aren't weavers. Wouldn't it be possible to independently stake out these extended warps in order to make the cord possible?

Wendel, I suppose if you knew where that Yomud went, you'd mention it. I would imagine it would be possible to compare the material of the corded section with the rest of the warp material and make a judgment whether they came from the same kind of animal. In any case, I have a sense of having seen other Yomud main carpets in particular with such extended cords, but cannot cite any specifically.

Rich Larkin

Marvin Amstey

'morning Marla,
Awhile back, I also had a Yomud main carpet with the braided corners just as in the Pinner piece. I also had an Ersari and Yomud engsi with these braids. Unfortunately, I have no remaining photos of these.
Marvin

Rich Larkin

Hi Marvin,

Can you recall whether the braided cords were integral to the rug, as extensions of the warps?

Rich Larkin

Frederik May

Hi,

I saw those braided cords on diferent Jomud main carpets. There are some pictures in the Bausback books too. Like a Tekke main in the 1977 book. Two Jomud ensis in the 1979 book and so one. I unfortunately don´t have the time to scan the pics today. A known expert once told me that these braided cords were comon on many old Tukrmen rugs. They just were too fragile and so there are not many left. He said that they were the parts of the structure which secured the rugs during the making and sometimes were decorated nicely maybe to secure the rug somewhere afterwards too.

Tomorrow I will try to make some pictures from the knots of the ensi outside in the sunlight it is not possible to make proper pics late in the evening.

Regards

Frederik

Marla Mallett

Hi all,

The back side detail of Rich's ensi shows us pretty clearly one of the reasons that symmetrical knots were substituted at the edges on some many rugs and bags: In the lower section (below his white marker), the outside knots shows some disruption, as they are less secure.

It is interesting, indeed, to hear of all these braided straps on rug corners. Short braids are easy enough ...Long ones would have required separate narrow warps at the rugs' sides. This would have been difficult and troublesome enough in terms of maintaining even warp tension, that it would not have been done casually, without very good reason. Done "to secure the rug" during weaving? I don't think so. And why don't we find such braids (that are extensions of the warps) on chuvals and torbas that we know from many, many reports were hung inside the yurts?

The presence of these braids should convince skeptics that ensis with such accoutrements were most likely hung.... likely as door curtains. So what would have been the purpose of such braids on main carpets? Do these braids appear on just the corners of one end? Or are they on just the two corners of one side (one on each end)? In other words, if these carpets were hung, were they hung horizontally or vertically? Hung in yurts sideways perhaps as a space divider? We do know that some kinds of weavings were hung to separate a part of the yurt from the rest of the space. But large, relatively heavy pile carpets? Or were they hung in villagers' houses?

One more thought about ensis hung as yurt door curtains: Must we assume that they were hung with the pile side outward? Or maybe with the pile side inward? I don't recall seeing many ensis that are badly faded. But have we paid enough attention to notice that some may be faded on the BACK side? Or if they were rolled up and tied in place (as in one photo on another thread), should we expect limited fading--in a section across one end?

Best,
Marla

Martin Andersen

Hi All

Just to keep the general level of scepticism: As the cords are present at other weavings that we know where not door rugs, then the cords on Ensis only shows that they where meant for hanging - they might as well has been intended for indoor hanging as trappings (or whatever)
And the knot reinforcements seems to be a rather frequent, though variable, on both main carpets, bags and trappings. I think I can see them in more or less systematic forms on almost all my bags and main carpets. Perhaps not the specific knot structure Marla has discovered, but some kind of reinforcement.
I would still say the question is open on whether the Ensi was a utilitarian everyday door - or a trapping for the door (and perhaps even other things) on special occasions.

best Martin

Wendel Swan

For another example of vestiges of corner cords, see this detail of a Yomud flat woven main carpet published by Lefevre in 1976 (only one corner is illustrated in the catalog), with this comment: “At the corners plain tapestry has been extended and the warps left long to form a grab-rope.”



Another example is from the catalog of the Yomud exhibition of the International Hajji Baba Society, also in 1976, with this entry: “The imposing reddish brown kilim aprons … are in a perfect state of preservation – even including the hanging straps at the upper corners of the rug.” There is no evidence of such vestiges on the bottom of the carpet.



I know that the kilim extensions or “tabs” on my ensi were original. They are essential identical to the extensions found on the IHBS and the Lefevre examples. Notice that in all three the kilim extensions are gathered, probably in anticipation of a braiding process.

Other published examples must exist. As to Frederik’s comments on the Bausback examples, the cords on the Tekke are without kilim extensions (as with Pinner’s) but the Yomud (if I am seeing the same one) has only parts of the kilim extension remaining.

Also, I know that I have seen others “in the wool.” Both Marvin and Rich report similar sightings, but their memories are no better than mine for specifics. These cords may be rare (especially since they were probably snipped when put into service on Western floors) but Robert’s is not unique.

We (certainly I) lack sufficient information to isolate a single function for these corded pile weavings, especially since one format could well be used for multiple purposes. Impractical as it may sound, rugs were often hung horizontally and some practices in village homes often did not materially differ from those of the tents of the pastoral nomads.

Wendel

Rich Larkin

Hi all,

Can any clever person think of a plausible functional purpose for the corner straps on the main carpets other than as a means of hanging the piece? Wendel alluded to LeFevre's "grab strap" idea.

Rich Larkin

Patrick Weiler

Marla,

I knew there would be homework when I posted earlier.
The outer two rows of knots appear asymmetrical like the rest of the rug. They are so closely spaced that it is nearly impossible to discern their construction.
I took the rug down and attempted some closer pictures, but it is rather difficult to pinch the rug with one hand and get a decent focus while holding the camera with the other. I would have asked Vanna White to hold the rug while I was taking the photos, but she was busy.
I will send some of these pictures to Steve and post them soon.

Patrick Weiler

Rich Larkin

Hi all,

I have come across some examples with corner braids, or the vestiges of same, in some old Nagel auction catalogs from the '70s, including a dyrnak gul main with a treatment similar to the flatwoven piece Wendel posted at panel #25; and a Yomud ensi with the cord intact. If there is interest, I will cheerfully scan and post; or maybe we've seen enough?

Rich Larkin

Filiberto Boncompagni

In Uwe Jourdan’s "Turkoman" there are several examples of Ensi and main carpets with braided cords.
Here’s a tekke small rug, from the on-line version of the book (the caption related to the scan is #43 - image of caption #42 is missing anyway)

The scan is quite bad and only the right cord is visible but the plate on the book shows both cords and it looks like they were “integrally” woven with the original kilim.

Regards,

Filiberto

Filiberto Boncompagni

After all, better put the scan on our server too

Patrick Weiler

Here are some more photos of the piece I posted earlier. From the rear, the outer row of knots is much thicker.
I am inclined to think the overcasting at the edges was redone at some point.

From the front, they do not look any different than adjacent rows of knots.

Folded over there appears to maybe be a bit more wear at this section.

But here it almost looks like the side knots are not on the same level as the adjacent rows.

These pictures do not seem to be definitive regarding structural differences. I may need to get a very sharp knife and start hacking the thing to pieces.

Patrick Weiler

Steve Price

Hi Pat

I've been told that professional rug assassins use scissors. If you're gonna do it, do it right.



Regards

Steve Price

Patrick Weiler

Steve,

I have successfully reduced the Tekke engsi in size using scissors.
Here is the result:

As you can see, it is actually one from a pair of Yomud dizlyks - camel knee covers.
I have a pair of these for the wedding camel for when my daughter gets married. I am still looking for a pair of asmalyks for that camel. My daughter may have to wait a while before getting married, as pairs of asmalyks are rather hard to come by.
As you can see from the back, the top and bottom have been folded over and sewn down. The cords and tassels have been attached and are not integral to the warp and weft of the piece. The red in the tassels and cord which they are attached to appears to be a later synthetic. It is not certain if this part was added later than the original piece, although the dyes are different than the pile areas.

Different from an engsi or asmalyk, these items were produced to be tied to an animal and needed cords.
I suspect that the pair were woven on the same set of warps, with a section of plain-weave between them and were then cut apart for finishing.
I have not looked closely at older versions of this type of weaving to see if the cords were made from the original warps as is apparent from some of the other pieces in this thread. Some salt bags also use the warps to make hanging cords.

Patrick Weiler

Patrick Weiler

Steve,

An asmalyk recently on e-bay also has the hanging cords extant and attached in a similar fashion, sewn-on rather than as an extension of the warps. A search of asmalyk images also shows this feature, although most older pieces are missing the outer edges all around.
The attached outer cord with hanging-cords and tassels indicates the hanging-functionality of these types of weavings. This surrounding, attached cord would have the advantage of imparting greater strength to these weavings when used on an animal during travel. The added cords would not put a strain on the pile part of the weavings.
The engsi does not seem to have this, although they also are not known to have tassels, either. So either the engsi was not traditionally used for hanging, or it was just not used as a trapping on horses, camels or donkeys as were dizlyks and asmalyks.
Richard Wright does not discuss this feature in his article on the engsi:

www.richardewright.com/0406_engsi.html

To bring the discussion around to the original intent of the salon, a Saryk torba, plate 29, in the Uwe Jourdan book Turkomen shows this feature of a cord attached at each upper corner which would indicate a use as a trapping.
He discusses the kejebe design as possibly panniers used on camels, the wedding chair or howdah on the wedding camel and a representation of ancient cult altars. He notes the added flat-woven strips and fringes are common on Saryk trappings.
There are Tekke engsi with hanging loops, but not the added cords, at the top corners, plates 57 and 58.
A Tekke torba, plate 78, "retains its added fringes, side cords and two-colored top cords for hanging".
Yomut engsi show the long warps at each top corner, possibly for cords, on plate 141 and plate 144 shows one remaining cord, but also no added cords.
Back to the Chodor, plate 228 shows the attached side cords, hanging cords and tassels.
It appears that weavings designed for use as trappings often had these added cords or strengtheners and those whose major use was different may have warp-cords, but not the added pieces.
These additions may have been removed when trappings were sold, or as they became damaged. There may be no way to tell at this late date whether your torba originally had these.

Patrick Weiler

Steve Price

Hi Patrick

I think you've got it right for ensis - at least some of them had warp extensions that were braided into handing ropes. Yomud asmalyks seem to have had braided red and blue wool ropes sewn to their edges, with hanging ropes extending from them and tassels fixed to the rope that's along the bottom.

I've seen a number of juvals with braided horsehair ropes sewn to the sides, with extensions for hanging and with ropes sewn onto the lips for closure. The thread called "That's a stretch" shows this one:


It's pretty clear that it was used as a container (notice the handholds on both sides), but most likely served as a display piece as well.

Regards

Steve Price