Ali Aydin, On Restoration and Conservation of Rugs and Other Textiles

On November 17, 2012, Ali Aydin,


conducted a Rug and Textile Appreciation Morning program, here at The Textile Museum, in Washington, D.C. in which he discussed “appropriate measures to take in conserving, protecting and repairing rugs and textiles.”  He also talked about some of the things to consider when choosing among the various repair/conservation alternatives.

Ali is a long-time restorer of, and dealer in, rugs and other textiles.  He is associated with Mark Keshishian and Sons, but also deals and repairs, independently.  He is active at collector shows, visible in Hali, and was one of only two U.S. dealers to exhibit at ICOC XII in Stockholm.

He said that if one has a damaged textile one was considering, a first question is whether it would be best only to conserve it rather than repair it.

During his talk he treated a number of things that affect repair/conservation consideration:

o  The seeming value of the piece.

o   How unusual it is.

o   The degree and character of the damage.

o   What can reasonably be accomplished.

o  Whether a restoration should be when completed “invisible” to those other than experienced experts.

o   How much one is prepared to invest in the conservation or restoration.

o  How the owner plans to display or use it.

This listing is not exhaustive, but does provide a good introduction to some of the issues involved.  It should also make clear, that, and why, sometimes, it is better to conserve than to restore.

Ali had brought a number of examples to illustrate such considerations.

The first piece he treated was the one below.



This was Ali’s example of an older, flat-woven, bag face in excellent condition.  Its edges are all there and in good shape, and there is no visible damage to the interior areas of the piece.

Here are some closer details of A1.



A second piece was the one below.



This is another, older,  flat-woven bag face, but one that is not in as good condition.  Even in the comprehensive image above, one can see damage on the lower left edge and to the multi-colored slit panels at the bottom.

Here are some closer details of A2.




There may be minor damage in the brown part of the center of the gul-form above.


There is also some damage along the top edge of this piece, but it is older, has good color and graphics and its interior areas seem mostly sound.  Ali said that, in his judgement, this is a piece to invest in restoring.

The next piece was a Caucasian pile rug with Lesghi guls in its field.



At first glance, this rug seems to be in pretty fair shape.  All its edges seem undamaged and there are no readily detected difficulties with its field.

But this is an example of a rug that has been badly repaired.

Here are some closer details of A3.




Those of its back make clear that has been repaired at a low level of quality.


This rug has apparently been repaired using what is called “Kashmiring.”  Kashmiring is a method of simulating pile knots on the front by using a “wrapping” approach (much like that used in weaving in sumak).  No effort is made to tied individual knots on the warps in the areas being repaired.  Instead, the warps in such areas are wrapped with pile yarn of the same color as that used in the knots in sound pile areas.  The result can look quite good on the front side.  But Kashmiring is always readily detectable on the back of a rug on which it has been used.


Kashmiring is not always an inappropriate repair technique for damaged pile rugs.  If a rug is worthy of repair on other grounds, but has pile worn down so that only the “collars” of the knots are visible any longer, Kashmiring may be the only way to produce a repair that looks like the surrounding areas.  Tied knots can, sometimes, not be clipped closely enough to make their “brushy” ends essentially disappear.

But Kashmiring can be done faster and less expensively than can a pile repair in which knots are tied on the warps.  So one needs to be alert, when one encounters it (or has it proposed) to whether it is being selected for sound reasons. 

The use of Kashmiring in this particular repair is seen to be inappropriate because the pile of this rug is not worn enough to make it necessary.  The repair person who used it here was likely “cutting corners.”

The next rug was Chinese.



Ali said that this is a younger rug with “old” repairs. 


Chinese wool and been used but the original colors have been changed in the repairs.


Here are some additional detail images of A4.



The next piece was half of an Anatoian kilim.



Let me rotate this image 45 degrees so you can see it a little closer but comprehensively.


It has good colors and some unusual design features.

Here are some closer details of A5.





There is some work to be done here on ends and edges, but Ali said that he would recommend repair of this kilim.

This is a good place to talk about one of the most difficult aspects of sound repair; obtaining and using the right repair material.

If one wants a repair that is largely undetectable, one needs to find old wool with the colors one needs.  The colors of new wool will shift differently than those of old wool and repairs made with it may become visible after a few years. 

Old kilim fragments are a frequent source of such old wool.  The kilims are unraveled to permit this wool to be used in high quality repairs.  You can see with the kilim above that a wide variety of shades can be harvested.

The next piece was a Tekke engsi with good age.



Notice in the upper right of the image below that a patch from and entirely different rug has been used to fill in a missing corner.


Here are some additional details of A6.







Ali said that he would fix only the edges and ends on this engsi.

Many Turkmen pieces present a special challenge, on the “old wool” dimension, because the materials from which they were woven are often so fine.  That difficulty might be a strong reason for resorting, on more a few occasions, to conservation rather than to repair of a weaving for which more repair material is be needed.

Someone from the audience asked whether restoration adds value to a piece.


Ali said that this is a good question, but in his experience the answer is, usually, “no.”  Restoring a piece (even some instances of conserving one) will, mainly, usually, allow its owner to enjoy it in a condition closer to that it had when it came off the loom.

The next piece was another Chinese one.



Again, let me turn it so you have a larger comprehensive image of it.


We can see in the comprehensive images that this rug is missing some from both ends.

Let’s look more closely at some details of it.


Here we can see that it’s missing some pile in the left side of the image above.


This detail suggests that the sides are intact.  There is a transfer of red in one area, but that is likely from another rug since there are no reds in this one.


The detail above confirms that the sides (this detail is with the warps horizontal) and the ends “bobtailed.”




There is, in the detail above an instance of loss of pile.

The piece is nice, but not really unusual.  Fixing the ends is going to require the extension of two sets of warps across the entire width of this piece and then the reweaving of approximately this much of both ends (acknowledging that this is a detail of a side and selvedge).


I’m not sure that Ali said this, explicitly, but the quality of the rug seems not to merit what would be required for restoration of it.  Instead, it might be best to secure both ends (my notes say “binding the with cloth” an inexpensive procedure) and consider whether any repiling at all should be undertaken (probably not).

We took a little side trip at this point to deal with an example of conservation.



The piece used for illustration is one of my own, so we can talk frankly about it.  This is a piece I bought long-distance out of a Jordanian flea market.

It is, it may not be too redundant to say, fragmented, worn in places and features holes.



It is no surprise that is was, from the beginning, a candidate for conservation only.  The conservation needed was of two sorts: 1) to secure all it edges (including some interior ones) so that no more of it is lost, and 2) to minimize the holes in it.

The conservator chose a backing cloth close in color to the red ground of the piece.  Here is the comprehensive shot again.  Check your sense of whether the “minimizing holes” objective has been achieved.


The couching of the piece onto the backing achieved the second “securing” objective.

I recently had it up on a wall, and a visitor noted that it seemed to sag a bit, and it would have been better to use a “stiffer” backing cloth. 

I think this observation is correct, but it draws attention to the fact that,  often, a number of variables would, ideally, be simultaneously addressed in either conserving or repairing a piece, and it is frequently impractical to do so.

I give this conservator high marks for choosing a ground cloth the color of which minimizes the holes in this fragment.  It seems unreasonable to me to ask that she also take on the task of finding a backing cloth of an ideal stiffness.  In fact, if I had to choose between a good color selection and “stiffnes,” I have no problem choosing the color match first.

This not to argue against the likely correctness of the suggestion that a little stiffer backing cloth would likely present this piece a little more crisply.  It merely acknowledges the complexity of the decision-making entailed in repair and conservation undertaking and the fact that compromises with the ideal almost always have to be made.

The next piece was a Caucasion pile rug from Shirvan with a niche design.



The comprehensive image above suggest that this rug is in good condition. Let’s look at some closer details of it.


The color is quite gorgeous.



The cotton selvedges are classic, but look chalky white.   Restored?




Don’t see any obvious problems. 

Ali said that this is an  example of a good repair.  The lower end had been restored and some edge work had been done to it.

The next piece was another Caucasian with “Memling” guls in its field.



Nice size, good color.  Some damage visible on the right side in the image above.


The damage on the right side seems restricted to the selvedge.


Ali looks at the back in this area.

Some details of aspects of this piece.



Ali points to something wrong on the right side of one gul.  Looks almost like a patch.


But there is rather little serious damage.  This rug is a candidate for repair.

The next piece was another Tekke engsi.




It appears to have a complete outside parameter, but the field is very worn.



A look at its back reveals a backing apparently needed to reinforce this worn center.



Estimates in the room suggested that this piece is younger than the earlier one.  Tekke weaving is pretty fine.  Damage is extensive.  Would be very expensive to repair and it’s not that unusual.  Probably best kept as a “study” piece.  No repair or conservation recommended.

The next piece was yet another Turkmen engsi, this time a Yomut, with a design sometimes associated with age.



Here are some closer details of A11.


Minor damage on the right side of the image above.  Looks to be mostly selvedge.



A restricted look at its left side seems OK.  Color in this piece seems very good.

Let’s look more closely at its back.


More damage, old repairs visible here.


More of the same.  This piece not in the condition the first images suggested.


Problem is general: little bare spots all over and lots of old repairs.

This piece could be repaired, but the extent of the damage to it, and the fact that repairing it might not raise its value, makes it one difficult to decide about.  If retained no real conservation work is needed.

The next piece was a Yomut Turkmen asmalyk.



Ali puts his finger on a problem area.



Some repairs in several color areas visible here.


Outside perimeter seems sound here but a few more repairs visible.  Note that weaver changed the white-ground border as she moved up.

A13dThis part of the bottom edge seems OK.


Possible repair on the left side of the central blue diamond.  Possible bare spot on the lower right narrow, diagonal, “barber pole” stripe.

This is an attractive asmalyk.  Repair might not be extensive.  Consider for repair.

The next piece was Anatolian with a Melas niche design.



Its outside perimeter seems OK at this distance.  There is a color change under the niche that seems too irregular to be abrash.  Also a brighter yellow area in the left border.

Here are some closer detail images of A12.




A large bare area.  Structural level seems pretty good.  Mostly repiling work.




Ali puts his finger on a bad repair.

There is quite a bit of work to do on this piece, but it is an old type and the work is mostly repiling.  Consider for repair.

The next piece was half of a sumac khorjin.




Very good color.  It has damage on it sides.  But not a lot of work.


Some repair of sumac in the field visible in the image below.


Repair this piece.

The next piece was a classic six-gul Tekke Turkman torba.



Initial impression at this distance is quite good.

Here are some closer details of A14.


Ends a little irregular and ends show slight damage.


Something suspicious in the lower left part of the upper gul in the image above.


Yes, some damage here.  Closer image also shows that there is some additional moth damage.

A16cMore moth damage visible in the image above.

This is a well-drawn classic type with good articulation of the interior of its major guls.  Robert Pinner once told me that he considered the apricot orange this piece has to be an indicator that it was woven before 1850.

Ali said that he would fix the ends and edges and leave the holes and the moth damage.

The next piece took us back to China.



This little mat is in pretty bad shape with both edge and field problems.

Here are some closer looks at parts of it.




This is a piece to conserve, not to repair.  Ali said that he would all of its edges with cloth.

Ali held up the next piece, small and in pile..



Although it may not be apparent, initially, this is the front of a large-ish Anatolian yastik.


This piece is unusual in several respects. 

It is yastik size, but lacks the top and bottom lappets that many yastiks have (although quite a number do not). 

Its field devices are a version of the so-called “baklava” design, although the diagonal use of color with the baklava devices is very unusual.  Morehouse provides several examples that use this device, but all of them turn it 45 degrees from their orientation in the image above.

Its simple, compartmented border is similar to usages in several Eastern Anatolian yastiks and all of Morehouse’s “baklava” examples are in his eastern Anatolian section (although he admits the this device was widely used).  The yellow ground and its reddish wefts may press its attribution back towards central Anatolia.

Here are some closer details of A16.


Notice that the reddish areas are not color runs, but rather the reddish wefts of this piece visible where pile is missing.  It seems to invite repiling.


The question of whether not to restore this piece likely bears, in part, on how old we estimate it to be.  It has a bright orange that some would see as a likely synthetic, but there are 13th century Anatolian rugs with oranges that are (still) equally bright, so we might be justified in giving it an earlier date.

There’s a lot of repiling here, but it might be justified, if the piece is seen to be older, as well as very unusual for a yastik. 

Ali said that he and the owner have agreed that this large yastik, which is almost a fragment, would be too expensive to repair, so they are mounting it on a cotton fabric for wall display.

The next piece was this nine-gul Turkman torba.



Here are some closer details of A17.




This piece, Ali said, is an example of a pretty well-done repair, mostly for moth damage.

He said that older pile rugs with long, thick pile are another potential source of old wool for such repair jobs.  You can remove such pile, thread by thread, and use it to fix small areas of missing pile, such as those that limited moth damage can present.

He did note one poor color match in one border.


Here are two more close details of A17.



The next piece was an Anatolian nich design with detailed, delicate design elements.








Ali looked at one area of its edge from the back.  Selvedge damage is visible.


A check of the ends from the back shows that they have been secured using a species of “whip stitching.”


If the side selvedge repair is all that is needed, this piece is a definite candidate for repair.

The next piece was a large square-ish, pictorial kilm woven in Romania or Serbia.



Ali and a member of the audience  considered it.


Let’s look at some closer details of it ourselves.



There is lots of damage to the kilim structure.



And there are some color runs.



Ali said there is too much repair work here to undertake.  He would do “minor things” and back it with cloth.

The next piece was the fragment of a rug attributed to east Anatolia.



This piece has lost edges all around, has holes and was one medallion longer at its lower end.

Here are some detail images of A20.





It is a clear candidate for conservation on a backing.

The next piece was Caucasian with unusual large medallions.



Here are some details of A21.





Interesting designs but another candidate for a backing conservation.

The next piece was small.



Unusual and attractive design and color use.  Its owner thinks it is Kurdish.

Here are some closer details of A22.




This lovely little piece might be worthy of nearly complete restoration, but it would be expensive.  An alternative would be to repair the sides and make sure that the top and bottom are secure.

The next piece curved a bit on the front board.



It was attributed to east Anatolia in the late 19th century.

Unusual border.  Also a lot of damage.


Here are some more details of A23.




A27fAli said that the condition of this piece is too poor to restore.  He would fix the ends and put it on a backing.

The next piece was very colorful.



Ali said that it is Caucasian, but he is not sure of what part.  It projects very good condition.  Someone in the room wondered whether it might not be a reproduction.  Ali said subsequently, that he knows its provenance and that it is late 19th century.

I took more than a few images of it.


Top and bottom borders are different from the design used on the side borders.





Some signs of wear and of small repairs? 





Not much repair need or work here.  Remarkable condition for an old piece.

The next piece was a Borchalu Kazak.



Looks to be in pretty good condition.

Here are some details of A25.






Ali looked at its back.

An example of a rug skillfully repaired.

The next piece was another Tekke Turkmen engsi with an older look.


Large scale “candelabra” devices at the sides are unusual.



Ali took a look.



A little damage on the left here.



This is a rug with relatively little damage, it may have some age and is worthy of the repair needed.

The next piece was Baluch.



Some damage visible right away.  Looks like moth.

Camel field are frequent in Baluch piece but the borders in this one are unusual.


Borders, especially the one based on the “mina khani” motif, are interesting, but moth damage is extensive.

Here are some more details of A26.





A look at a back edge.


This piece is an interesting rug but not a candidate for repair.  Damage is too extensive.

About 15 years ago I bought a fragmented Kazak, thinking that I might experiment on it by attempting some repairs.  Time passed and no experimentation had occurred and so I sent it off to Turkey and had it restored.

I brought it to this session and asked Ali to assess its repair in unvarnished terms.

He was good.  He started by saying that the repair job was “not too terrible.”  🙂


Then he was specific about what he meant.  He looked at the top edge and said that the repaired area was clearly noticeable, in part, because new wool had been used.


You can see in the image above that in the rewoven areas, the “new” knots cover the under-lying structure more completely.



And, of course, warps have been extended at both ends and added at the sides.


The repairer told me when he returned it that he “hadn’t made any money” on it.  Perhaps, he had to “cut corners” somewhat.

Readers of these posts will remember that I have said from time to time that I collect on a budget and as a result look in places and buy things that most collectors, who see themselves as serious at all, would not considered.  So I’m positioned well for participating in a session such as Ali’s here, especially with regard to conservations.

The next piece is one of mine that presses the question of “when should you conserve” a piece?



This piece is in a state of pretty advanced decay.  And it came to me without a backing.  BUT it has great color, so I invested in one.

Let me turn it so you can see it a bit closer but still comprehensively.


The conservator and I agreed on a blue similar to the top stripe of this piece (it would also, we thought, “pull out” the lighter analgous blue in the middle.

Earlier in Ali’s session the question came up of when to stop repairing a weaving.  Here, having decided to start conserving, the question became when to stop putting in additional “couching” stitches?  The conservator said to me, “Frankly, this is a piece where you could continue that almost indefinitely.”  She smiled.

Here are some closer details of A27.



This kilim is Anatolian and I asked Marla Mallett where in Turkey she thought it might have been woven.  She said that, these kinds of kilims were what were used to cover floors in Turkish homes and that they were woven everywhere.  She added that the strong yellow did suggest that it was not woven in western Anatolia.

The next piece is also mine and in very bad condition of another sort.



This, most will recognize is an Anatolian yastik face that has retained much of its basic perimeter but lost most of its pile.  It was likely woven in central Anatolia and, I think, it has some age.

Here are some details of A28.



Now the truth is that this piece could be entirely restored, and if I didn’t have limited disposable income, I might have it done.  I would enjoy seeing it more like it looked before it became largely bald.

But it would be very expensive, and most would say, instantly, that it’s too far gone to repair.  Reluctantly, I agree.

The next piece was a kind of curiosity: a scarf made in the structure and known design of a U.S. coverlet.



Coverlets are frequent, even quite good ones.  But I’ve never before seen a scarf in a coverlet structure and design.

There are two problems with this scarf.

First is has a stain in one area.


And it has some damage on one edge.


I’m talking to a museum quality conservationist about this scarf (she’s not seen it yet).  It is not an expensive item, so I think I will ask her to try to get the stain out and to stabilize the area of edge damage, unless it can be repaired very inexpensively.

My last piece, and it was the last of the day, is a kind of “hair shirt” of mine.

It is a rug likely woven in the Middle Amu Darya (we used to say “Ersari,” or if something looked more urban, “Beshiri”) with a field that features a “mina khani” design.


Ali looked at its, less than appropriate, selvedges.


I quite like the rendition of the “mina khani” and the devices on its elems suggest age to me.

But you can see that it has a lot of damage, and things get much worse when you look at its back, because there is a wide area of glue that continues across its entire width (the white spots are glued on fragments of newspaper).

Now some say the presence of glue means the rug is entirely ruined, but others say it can be removed (there is something called “Zip Strip,” a paint remover, that some have employed with some success to remove glue).  But the general condition of this piece has immobilized me and it has sat in my stack for a number of years.

I had about decided to cut up its “good” parts for pillows, when Elena Tsareva’s new book on the Hoffmeister collection came in the mail and, there, in it, was the piece below.

Now there is a great temptation to associate something one owns with published, perhaps remarkable material (Elena estimates the Hoffmeister rug as 18th century).  Mine is much younger (lots more borders); BUT the elems are nearly the same.  And the visual “feel” of the two pieces seems similar to me.

I’m not ready to restore my rug, but maybe I should consider trying to get the glue off its back.

Ali took questions,


and brought his session to a close.

Conversations and the surge to “pet” the rugs began.

I want to thank Ali for permitting this virtual version of his interesting session to be produced, and for his editing of my draft.

Thanks are due as well, to Ruth McDiarmid who took a useful set of notes for me.

I hope you have enjoyed this brief consideration, with the help of a professional, of some of the issues and decisions one often confronts when thinking about either repairing or conserving a weaving one owns.

Have fine holidays,

R. John Howe

Comments are closed.

%d bloggers like this: