Home > Turkmen Rugs >RK reviews rageth's New Perspectives Part 4
Wed, Apr 27th, 2016 11:11:35 AM
Topic: RK reviews rageth's New Perspectives Part 4

Detail Arabatchi MC; private collection but probably hans sienknecht; Cat. no. 127

There are basically three areas of inquiry claimed to provide the ‘new perspectives’ in the title of rageth’s Turkmen publication.

The first and primary is c14 dating, which we have already discussed in part 2 and 3, and will revisit again in a later part of our review.

The second is dye testing and identification, the subject of this part.

The third, rageth’s trying to trace the origins of the iconography found in old Turkmen woven products, will be tackled in a future part of our extensive review.

Dye analysis, like c14 dating, requires expensive scientific lab equipment and, of course, expert operators.

But there is a distinct difference between these procedures, one that unfortunately rageth’s book completely overlooks and ignores. This error seriously torpedoes any broad credibility it could ever establish.

C14 radiocarbon dating for old carpets is chancy at best even when competently done, and at the worst totally unreliable. While the results from dye analysis are absolutely reliable when done properly.

We have already discussed the problems utilizing c14 analysis to date old carpets, and they are major ones.

RK’s position results cannot be trusted is long on the record.

Formerly few others agreed with us but now, two plus decades on, more and more voices have joined our position.

It is clear jurg rageth’s is not among them, and his publication suffers the consequences – again chiefly because old carpets have been contaminated through use by a variety of substances that skew c14 results, as well as the fact results can be and are effected by the decontamination procedure itself.

There is no denying: It is far more factual to distrust the reliability and efficacy of dating Turkmen carpets with c14 than to trust it.

We will also return to this issue before our review is finished, but for now let’s look at the dye testing information jurg rageth and two other authors have contributed to this publication.

Detail S group MC; collection George and Marie Hecksher

These short essays have been placed after the large color plates and much too small black and white ones in the back of volume one.

Leading off is Jan Wouters, who is acknowledged as the leading expert on dyestuff identification and analysis. He did most of the dye testing for rageth’s project and his essay is titled “Dye Analysis: A Generator of Knowledge Beyond Science Alone”.

Unfortunately his title, like that for this publication, is enticing but it does not deliver what it suggests.

More than a decade ago RK spoke with him about our plans, which still have not been actualized, to build a scientific database of information on the dyes in historic weavings. We were impressed.

There is no doubt Wouters knows his stuff and rageth could not have chosen a more capable scientist.

That said, Wouter’s essay is basically say-nothing work, we read it several times to see if somehow we missed anything of import.

We can report we didn’t, but why this is the case we cannot say. Perhaps had Wouters been paid to write something extensive he would have.

His prices are very high, we remember being quite taken back by the prices he quoted for dye analysis work.

But perhaps there is another reason -- maybe Wouters had nothing to really say about the sampling and the results of the tests he did for rageth.

Considering this quote from him, that might actually be the reason: “Although dye analysis as such may be exciting from a scientific point of view, its usefulness for the study of historic objects research will depend greatly on its complementarity(sic) to results from other studies and on the way all data is compiled and interpreted.”.

In other words don’t rely solely on any dye test results.

And it’s not up to him to supply insights and conclusions, leaving that job in the lap of jurg rageth. Who we will show has not in any appreciable way used the results to discover any new ground-breaking information.

Wouters continues “This project(ed. rageth’s book) is a seldomly(sic) encountered opportunity of having so many samples so generously provided with the clear objective to collect as much as possible data on the production technology of Turkmen weaving. And thanks to this policy dye and mordant analyses have much contributed to historical, geographical and cultural observations of the weavings studied, as clearly outlined in the book as a whole.

Sorry, Dr. Wouters, but after carefully reading this book, and continuing to refer to it, RK can honestly state what you have said is nothing but hype and publicity that is not fulfilled by reality, as readers of our review will soon see.

One of the few real tidbits of information all those dye studies returned is the specific finding of Mexican cochineal dyestuff in a number of believably early, ie pre-1800, Turkmen weavings.

About this dye’s use Wouters states “The first European reports date from the second quarter of the 16th century. It was commercially imported in Antwerp since 1549, but early in the 17th century its use was not widespread. The dyestuff was also marketed in Asia by 1580. In 1700 it was known in China”.

This information, while interesting and indicative of the apparently wide geographic trading of dyestuff found in Turkmen weavings, does not really help us to understand the major questions surrounding Turkmen weavings: When they were made, Who made them, and Where were they made.

We will come back to discuss this when we examine rageth’s essay.

Wouters mentions another type of cochineal from Poland that was found in Pazyryk carpet.

The consideration of Polish cochineal is justified by this fact (Pazyryk) as well as by its detection on 4th-6th centuries silk ribbons of Egyptian or Persian origin.

He then mentions cochineal from Armenia, which one would think would be the major suppliers to Turkmen weavers. But, according to the testing Wouters did this is not the case.

Armenian cochineal “…may have also been an important dyestuff in Asia Minor in the Middle Ages” but he leaves this issue and then mentions another insect dye, lac. “Lac has been used since early times in India and China….In Medieval times, Persia and Arabia imported lac from Indonesia and Indochina. Despite its widespread use in Asia, and although it has been identified in textiles from Coptic Egypt after the Arab conquest (8th century C.E., it does not seem to have been of regular use in European textiles before the end of the 18th century. This is radically different from Mexican cochineal.”

Wouter has this to say about kermes, another insect dye. “Kermes was a highly esteemed dye in Classical Greece and Rome. It was abundantly found on early Medieval silk weaves in Europe. After the discovery of America, Kermes was gradually replaced by Mexican cochineal.”

About madder, a vegetable source of red dye he says “On historical and geographical grounds, it is logical to expect to find red dyes prepared from the latter sources (different types of madder) in Central Asian textiles, but this does not seem to have been the case with the Turkmen weavings studied.”

Now please remember this quote, it is perhaps the most important one he imparts. And we’ve added the bold type face to help.

On redwood, another vegetable source of red colorant Wouters writes “The main coloring matter of redwoods occurs in the wood of many species.”

Yes, that’s it from Wouters essay, and while it has some additional very general information it has, as we wrote, nothing more to say on the dyes found in Turkmen weaving.

The next author to contribute an essay is Ina Vanden Berghe who starts off with the following “Up to the last quarter of the 19th century, various kinds of biological sources were used for dyeing…In fact, these dye compounds are present not only in madder but also in the roots of many other plants belonging to the Rubiaceae family, in very different or non-distinct ratios. One of the principal and at the same time most complex dye sources for red dyeing is madder…Not only for vegetable red sources, but even more so for the scale insect reds (cochineal, lac and kermes) -- the other major source for red colors in Turkmen weavings – the relative composition of the dye compound mixture is of major importance for the specification of the applied insects.

She was also responsible for doing dye testing for rageth and her commenting on finding madder seem to contradict what Wouters wrote.

RK has carefully perused both their essays, as well as what rageth has written and nowhere did we find any mention of this, let alone any attempt to resolve this discrepancy.

She continues“The actual color of the fibers is not determined only by the dyes used. Most natural dyes are able to attach directly to textile fibers. To overcome this, people used to work with organic and inorganic mordants with which they treated the fibers before or, eventually at the same time as the dyes in the dye bath.

If you are an astute careful reader you will, like RK did, question her using the word overcome. Surely she did not mean "overcome this" but rather to 'increase this'(the efficacy of the dye).

Poor editing is clearly responsible for this gaff making it into publication.

Next she outlines an important experiment she instituted “To determine the type of cochineal dye source of a particular sample it was decided to undertake experiments to dye wool with dyestuff to more accurately determine the sources of the samples provided from the historic Turkmen weavings in this publication…So far, the experiments have proven the extraction procedure and the applied mordant have an influence on the relative composition of the characteristic dye component found after analysis. Despite this, the results did not resolve the issue of identification of the cochineal species.

RK added the bold type to highlight this important drawback of the testing procedures that are presently available. More on this later.

Continuing she adds“The dye experiments improved the understanding of the major external influences on dyeing with cochineal species. For the majority of the Turkmen wool samples, it resulted in the identification of the applied cochineal source. For the silk samples, it is still not possible to specify what insect has been used among the Turkmen to dye their silk for pile of their carpets.

One could theorize Wouters and Vanden Berghe’s essay to set the stage for jurg rageth’s comments. However since both of them wrote nothing to amplify these findings, and their essays are quite general in nature and lacking reports of any other new discoveries, some of what he writes just redundantly repeats the general information they described.

Here’s the beginning to rageth essay “The preference for the color red among the Turkmen is beyond any doubt. Reds in many varying shades are present in all their weavings. Most of these reds are dyed with madder, a vegetable dystuff which, depending on the dyeing method, can produce a color range from light orange to a deep purple.

Now remember what Wouters wrote about madder? Who is right, clearly they both cannot be. Again we added the bold typeface to assist comparison.

The central thesis of rageth’s dyestuff essay, which is much longer than either of his two co-authors, is “It therefore made sense to compliment the insights from radiocarbon dating by examining carefully selected fibre samples for their unusual reds.

There were 221 samples from 118 weavings dye tested and “The majority of the samples were taken on the basis of visually suspected dyestuffs, a smaller group to identify suspected synthetic colorants. In this context a hint from Amy Butler Greenfield turned out to be very helpful. In her book “A Perfect Red” she describes the amplifying effect of tin in connection with dyeing red with insect colorants, especially cochineal..”

Not to discount Bulter Greenfield’s idea but jurg rageth has overlooked the important fact siawosch azadi wrote about the use of tim as a mordant in his 1971 Hamburg Volkerkunde Museum Turkmen exhibition catalog.

Were rageth truly knowledgeable about Turkmen carpets and their body of research he surely would have known this. To not know is just another example of his far from expert knowledge.

Leaving the tin mordant issue aside rageth then asks, rhetorically, “Is there a way to differentiate material dyed red with insect dyes from material dyed with madder?

He then goes on to say “…a fine quality of spin” is one major clue.

This clue and what follows is one of the very few solid contributions this publication makes to Turkmen studies.

To make knotting material of the correct thickness out of the more finely spun insect dyed yarn requires the use of more than two pieces of yarn(the usual method in Turkmen weaving). 4-, 6-, 8-, or in one case, even 18 plied yarns [9(Z2S)] has been found. The single case mentioned, the torba cat. no. 112, shows 18-plied [9(z2S)] lac dyed woolen yarn of unusual fineness in very small areas only.

cat. no. 112

Since rageth maintains the presence of insect dyes is usually found only in the earlier pieces, and this plus using c14 radiocarbon dating results to determine the age of Turkmen weavings is a fundamental part of his working thesis, why was cat. No 112 with its highly unusual perhaps unique 18 plied yarns not C14 tested to see how old it ‘might be’?

Again this is an omission a competent researcher would have avoided. Considering all the faith rageth places in radiocarbon dating his failure to test here is quite surprising.

Concerning these high ply yarns rageth states “It seems likely that such extremely fine insect dyed woolen materials were purchased rather than being produced by the weavers themselves.

It does?

This is just more rageth guesswork, it is far more probable just the opposite was the case.

For if yarns of such a high ply count were purchased from an outside professional source the likelihood more weavings would have such yarns is obvious.

And the fact no other weaving has been identified with them points more likely to a single weaver rather than any professional source.

It was already well known insect dyes are found both on wool and silk, and rageth repeats this “Even with the technical progress regarding the analytical methods, there are still limits to the specifity to which we can identify the insect dyestuffs on silk in the field of Turkmen weavings...Though they can be identified as a type of cochineal, a more precise identification is currently not possible.”

But the following is a supposed contribution of this publication that in fact isn’t.

The use of insect dyestuffs on wool among the Turkmen has been assumed for some time, but had not previously been proved by chemical analysis

Amazingly, in the next sentence rageth writes “As early as 1973, Thompson pointed out a presumed insect dyestuff in the pieces of the Salor, then still described by him as S-Group. Describing the characteristic features of his S-Group, he wrote ‘use of wool of a special pinkish-red in which the dye is corrosive, causing increased wear on the wool’. Commenting on Bogolyubov’s cat. No. 8 Thompson goes even one step further in connection with the just mentioned ‘pinkish-red’: ‘It is interesting to speculate what this dye could be – Kermes perhaps? We await the results of dye testing which is now in progress. Three years later, in 1976, at the first ICOC in London, Mark Whiting introduced the results of his first dye test on Turkmen carpets. The corrosive pinkish red on wool turned out to be dyed with lac, and not with kermes, as suggested by Thompson…Mark Whiting was the first who chemically identified the insect dyestuffs cochineal and lac in Turkmen weavings, although largely only in the pieces of the 19th century…

Rageth then goes on to explain Whiting was unable to differentiate between the different species of cochineal and it was not until much later that scientists, particularly Jan Wouters using HPLC (high pressure liquid chromatography), were able to identify them.

But these results are only reliable on wool and not silk dyed with cochineal.

So, in fact, rageth’s publication is not the first to prove insect dye on wool in Turkmen weavings but actually to determine the type of cochineal used to do the dyeing. This is a big difference and while both are important discoveries rageth should have gotten it right which one his testing proved.

He also mentions research done by Dominique Cardon has identified a number of different species of cochineal, now known as ‘root cochineal’, and it is possible these are the dyestuff used by the Turkmen to dye the silk highlights found in certain weavings.

Why his testing did not try to identify them is not mentioned. We think it should have been.

Also, claims rageth makes about dating certain Turkmen weavings on the basis of their having Mexican cochineal dyed fibers are far from conclusive. “It is unusual to find wool dyed with Mexican cochineal in Turkmen carpets dating from as early as the 16th century. There is no doubt that this occurs by the first half of the 17th century. In our study, six Turkmen weavings (ed. 16, 36,110, 117, 127, and 157) radiocarbon dated to before 1650 contained Mexican cochineal on wool.

cat. no. 16


It is easy to believe all of these weaving are early ones, but it is a totally unsupportable to maintain all of them are older than 1650, especially considering several returned other, later, c14 results than the time period rageth mentions. Also art historical comparison does not support all these could be circa 1650.

Fact: the presence of wool dyed with Mexican cochineal cannot in any regard convincingly date a weaving to the mid-17th century based on any c14 result.

Don’t get RK wrong here, we totally believe there are 17th century, and even earlier, Turkmen weavings we just do not believe rageth’s evidence for this group is as conclusive as he tries to make it appear.

Continuing with the discussion of finding Mexican cochineal in those 6 Turkmen weavings rageth states “Whether we are dealing with already dyed imported wool or with imported cochineal from Mexico as a raw material is still not clear, although the latter appears to be more likely.”

cat. no 110

cat. no. 117

Considering the fifteen years rageth spent working on this book he could easily have inspected all those 6 weavings (16, 36,110, 117, 127, and 157) to determine if there are, in fact, any visible physical abnormalities or differences between knots dyed with Mexican cochineal and the other knots to demonstrate imported yarns. To have missed doing so, when such a simple exercise would really have provided some truly important information, is inexcusable. And to make a guess instead even worse.

So much for rageth’s overblown reputation as a Turkmen weaving researcher and scholar.

cat. No. 127

cat. no. 157

One of the few genuine breakthrough findings contained in this publication is the discovery the earliest Turkmen weavings with insect dyed woolen knots have ones with much higher ply counts, from 6-8z versus the normal 2z count. Later weavings where insect dyes on woolen knots appear only have normal 2z count plying.

The conclusion rageth reaches is that “No longer was cochineal dyed wool purchased, rather the dyestuff itself (ed. was).”

Again, here is another rageth assumption presented as fact. While other explanations -- like by the 19th century weavers were just too lazy to spin and ply far finer threads, or if specialists and not weavers were producing the insect dyed woolen yarns they, too, just got too lazy to bother -- are equally as viable considering the deteriorations in quality and skill 19th century weavings exhibit.

Interestingly enough rageth reports finding both high ply count knots and normal 2 ply count ones dyed with insect dye in the same piece.

This feature was also noted by Mark Whiting in his published work.

However, rageth conclusion “This leads to the conclusion that all three examples (ed. cat. no. 39, fig. 5 and cat. no. 61) likely date from the mid-19th century, when the enormous drop in price of the dyestuff resulted in the shift from the purchase of pre-dyed wool to the local dyeing on local wool.” is far from believable.

In fact, it is nothing but sheer unsupportable speculation and an attempt to bend the facts to fit into his dubious conclusion

cat. No. 39

fig. 5

cat. no. 61

Far stronger arguments could be made for cat. no. 61, the Tekke chuval, and cat. no. 39, the tent band, being at least first half of the 19th century. Though this is not much earlier, it is a significant enough time considering the differences weavings ‘like’ them that actually are middle 19th century exhibit.

For instance cat. no. 61’s sparing use of the insect dye, its rare and very well articulated lower elem’s ring-tree icon, and the larger than normal comb motifs in the upper elem are never elements found in later mid-19th century Tekke chuvals of this type.

But it is the really rare inner and outer minor borders, which harken back to some of the earliest iconography found on Tekke trappings (see photos below), as well as other very rare and early types of Turkmen weavings, that seals the reality cat. no. 61 is decidedly earlier than rageth’s questionable guess.

Left: minor border detail Tekke chuval cat. no. 61; Right: minor border detail ancient large format torba(LFT), RK collection

This minor border is exceptionally rare, we know of only a handful of appearances and absolutely none of them dating to the middle 19th century.

Although a case, though weak, might be offered to show the tentband, cat. no. 39, is as late as rageth claims we seriously doubt it. Lacking a photograph of the entire tent band we will not try to prove our point, but when it comes to fig. 5 it would be impossible to make even such a case.

According to everything RK has learned in 45 plus years of studying Turkmen weavings it is a slam dunk his idea it is circa 1850 is grossly incorrect.

We have included a close-up detail, above, to show the unique iconography sequestered in ‘heads’ at both ends of what rageth calls a compound palmette-tree on fig.5.

This ain’t no middle 19th century weaving, we’d opine it is at least circa 1800, if not earlier.

Also notice the ‘weeping’ branches on the palmette-tree, a design convention never seen on later examples. And thanks to the knife-like sharp edges the weaver was able to articulate, added to an overall design that flows smooth as water down an incline, we can assure readers this tent band is not middle 19th century.

To explain why there is little evidence for the use of Armenian cochineal in Turkmen weavings rageth states it is the “much higher efficiency of the cochineal from Mexico” as if this was the only conclusion.

There are others, one being indications the majority of the Turkmen weavers who produced weavings with significant insect dyed wool and silk fibers must have lived in or very close to urban centers where both far reaching international trade contacts and organized workshops, where luxury weavings were produced, existed.

This idea is supported by the homogenous character of such weavings with large, lavish, amounts of insect dye colorants. The S group trappings and main carpets, along with its variants, are the main examples of these workshop weavings.

Were this not the case logistics and cost of those insect dyes, or fibers already colored with them, except in minute quantities would have been beyond the reach, financially and geographically, of weavers working in far more isolated and less connected environments.

On the other hand very little is known about the logistics of trade in Armenian cochineal, or the amounts produced there, compared to that from Mexico. But it is not hard to imagine the geographic difficulties of supplying the market in Turkmenistan must have played a major role.

As time goes on it appears very few Turkmen weavings, even early ones, were made in extended family/clan encampments or isolated small villages.

The image hordes of wild Turkmen and their weaving women were living far from relative urban civilization is, from studying the weavings themselves, one that appears more and more to be nothing but old-fashioned prosaic illusion.

The systematic use of the dyestuff known as lac by what rageth calls ‘the Salor’ presents a quandary for any savvy reader because he does not make clear actually which are the weavings of the ‘Salor’ he means. Are they S group, the open right variants, or both?

In a quote above, rageth mentions jon thompson referring to a weaving that is“… Salor, (ed. but) then still described by him as S-Group”.

With this offhand comment rageth seems to discount the specific and important validity of the existence of an S-group in favor of the more general and totally mythic term ‘Salor’.

The identification of S-group by jon thompson is not only his best and most significant contribution to rug studies, particularly Turkmen rug studies, it set the stage and demonstrated the importance of technical analysis to categorize groups of many seemingly ‘similar’ Turkmen weavings.

S-group defines one of the few solid facts of Turkmen studies by identifying a group with an asymmetric knot open to the left done on warps which are on two levels, a feature often called a ‘depressed warp’. These features are uncommon enough to imply these weavings had a very special cultural history and origin.

Whether or not this group was, in fact, woven by the Salor tribe is unknown, nor is it necessary to know to appreciate the importance of its identification.

Another group of very ‘similar’ weavings does exist, the only difference the asymmetric knot on two levels of warp opens to the right. These are therefore not S-group.

Are they Salor?

Frankly RK does not care because it is impossible to know, but for rageth to not explain exactly which type of weaving he is referring to when he calls them ‘Salor’ is both confusing and careless.

Technical analysis has also been used by Annetta and Volkmar Rautenstangl to classify several different groups of weavings they named ‘eagle group’. It is, so far, the only reliable methodology to identify and classify Turkmen weavings into specific groups, and rageth’s avoidance of mentioning information gathered from technical analysis in his text discussions of the Salor is inexcusable.

One of the major ‘findings’ rageth hangs his hat on is the use of lac dye among the ‘Salor’.

The Salor adherence to the use of lac, when using an insect dyestuff on wool, is extremely consistent…Our investigations proved the systematic use of lac dye in considerable quantities exclusively on wool among the Salor…Whether this special use of lac dye represents a local tradition and can be traced back to the pre-10th century Sogdians or reflects some influence from 16th/17th century Safavid Persia, or both, has not been examined thoroughly yet.

In a future installment of RK’s review and critique we will discuss rageth’s belief Turkmen iconography is nothing more than a reflection based on earlier western Asian art, or that of the Persian Timurid and Safavid dynasty.

His position, of course, begs the issue what and where were the origins of those iconographies.

There is no doubt the Turkmen weaving style and certain specific iconographic figures are related to both these traditions but to not recognize the possibility they all developed from an even earlier, yet unknown, cultural tradition makes no sense.

And, in fact, any idea this was the case can be supported by the specifics of the complex abstract diversity the earliest Turkmen weavings display compared to the totally predictable naturalistic – flower, plant, tree, shrub and animal – iconography used in western Central Asia and Safavid Persia.

It is this symbolic content that imbues the best and earliest Turkmen weaving with an appearance that seems far more ancient and archaic than the pretty flowers and heraldic animals which grace the best and earliest Safavid and western Asian weavings.

RK cannot prove our assertion but rageth’s old, worn-out, adherence to the long held but absolutely unproven paradigm Turkmen weavings are nothing but trickle down descendants of large scale society, ie Safavid and western Asian, cultures is even more unlikely.

We will discuss this further when we review rageth’s ideas of Turkmen design origins.

Back to the Salor and their use of the exotic lac dye.

Far from the possibilities rageth offers up – ie it was an old tradition inherited from the Sogdians, or a reflection of one from Savafid Persia, let RK present another.

Because the weavings rageth calls Salor present the most homogenous, pre-commercial period, group of all Turkmen woven products -- in design, materials, and even technically (besides the open left, open right knotting difference) -- it is highly likely, and almost assured, they were produced in a only two or three very small geographic areas by weavers who shared great cultural, economic, genealogical and spiritual connection.

In addition, the difficulties to procure an insect dye like lac in great quantities, forget its cost, imply the weavers of these Turkmen products must have had awesome proximity to either the source of the lac dye, ie India, or extremely close relations with traders who did.

Obviously the more middle men between those at the source and the weavers, the higher the price the end user would have to pay.

For these reasons and others unmentioned, like the limited design pool of so-called Salor weavings particularly their borders and preponderance of large main carpet(MC), it is quite reasonable to assume these weavings were produced in town workshops, not in any other environment.

The next logical assumption might be; Are these even Turkmen weavings, or ones made for the Salor by other weavers, and hence having the unique for Turkmen woven articles a warp on two levels (found by the way mostly on Persian weavings) and the different knotting, asymmetric open left and right?

We will leave this question hanging for the time being and continue to examine what rageth’s New Perspective Turkmen book says about dyestuffs.

RK has noticed other instances where rageth’s text becomes confusing, not only his failure to explain exactly what weavings he is referring to as Salor.

Whether this was due to the German language to English translation, or some other more inherent cause we do not know.

For instance in the discussion about Salor use of lac we quoted rageth saying “The Salor adherence to the use of lac is extremely consistent.” but several paragraphs later he appears to contradict this “Our chemical tests proved the usual bright red ground of all Salor pieces to be dyed with madder”.

So which is it? Adding to the confusion the following appeared several paragraphs before “Madder as an ingredient was found in very few cases and in traces only.” Remember his contradictory quote we pointed out earlier.

Since there is no possibility to resolve this let’s move on.

After listening to an online presentation of a talk rageth made in 2015 at the deYoung Museum in San Francisco, RK posted comments that can be found here:

His talk centered on a tentband with an unusually bright red color.

Upper: photo of tent band detail from rageth’s de Young talk; Lower: same tend band published as cat. no. 157

According to the dye testing this bright red color was cochineal with a tin mordant.

Long ago, in fact in 1971 siawosch azadi wrote in his catalog for the exhibition he organized in the Volkerkunde Museum in Hamburg, Germany about tin and other metals that were used to mordant Turkmen colored yarns.

Regrettably he never mentions azadi’s work, but instead rageth presents use of tin mordant as a new discovery, which it surely is not. That rageth’s dye testing proved azadi’s idea is correct, that this discovery is rageth’s is not.

He also uses a dubious chain of facts to establish a 17th century date for this tent band. And while we have no quibble believing this tent band is very old, is it late 16th to early 17th century?

We do not know but we do know we have serious concerns about the explanation for this dating rageth has concocted.

Here is what we wrote, with a few additions, when we critiqued rageth’s lecture and online presentation in 2015:

“In his presentation rageth explains the SEM/EDX test proved the use of Mexican cochineal as the source of the band’s unusually bright and luminous red (the arrow in the picture of the tent band above points to that color).

This is good science, but his conclusions go downhill from there.

According to rageth, based on the research of others, Mexican cochineal came to Spain no earlier than circa 1520, and was probably not available in Central Asia until circa 1550AD.

We have no argument with this logic but we do with rageth then “narrowing” the dating of the tentband to 1570 – 1670 based on the “tin mordant factor”.

To do this “narrowing” he cites Dutchman Cornelius Jacobszoon Drebbel’s, circa 1610, accidental invention/discovery of tin mordanted cochineal achieving the unusually bright and luminous red seen in the tentband.

By the way Drebbel was quite the inventor and is credited with devising the first submarine, among other interesting inventions.

This, Drebbel’s tin mordant discovery, is the “scientific evidence” rageth presents for “narrowing the dating window to 1610-1670”.

Here rageth again is presenting opinion as science because this fact only sets-up a valid terminus ante quo, ie tin mordant cochineal could not be found in any textile made before 1610.

However, it surely can be found in ones made after that date, and what proof does rageth present that this is not the case with the tent band?

Answer is none.

Let’s be clear here, and again restate, we believe the tent band is that old – we just don't credence the supposed “psedo-science” rageth advertises as scientific “proof”.

Also: Isn’t it possible some enterprising and brilliant Turkmen, Persian, Anatolian or Jewish dye specialist discovered the process of mordanting cochineal with tin before Cornelius Drebbel did? And therefore the tent band could be even older than the “narrowed” range?

The answer of course is, yes, it is possible, which together with the points above establish a major sink-hole in the “scientific evidence” rageth uses to “narrow the dating of the tent band to 1610 – 1670AD”.

Science is not opinion or conjecture, it must be like math 2 plus 2 equal four.

Unfortunately for rageth and his supporters, c14 dating for old carpets and weavings has serious problems and drawbacks.

His talk, this publication or his earlier on Anatolian Kelim might look good on paper but, in reality, fall far short of reliable science.

Proving the tin mordant with HPLC is science, but all the rest of his talk is nothing but, as we wrote, a pseudo-scientific charade."

Now in his new publication rageth goes further out on that shaky limb by maintaining “The use of tin (ed. mordant) as a ‘color amplifier’ in connection with insect dyestuffs apparently was not known before the 17th century in Central Asia and appears to have been abandoned in the mid 19th century. This is clearly the conclusion we are led by our test results. The early dated Salor khali cat. no. 16 still does not contain tin (ed mordant), while the majority of all pieces dated somewhat later to the 17th and 18th century do.

To base such an opinion – there was no one using a tin mordant in Central Asia before the 17th century – on the limited number of dye tests and the corroborating unreliable c14 dating of a handful of Turkmen weavings is one more example of rageth’s fantasy research conclusions.

Frankly, it’s laughable and RK does not know how anyone reading this book could believe such a conclusion, or other similarly baseless ones it forwards.

There is absolutely no doubt positive determinations of what dyes and mordant were used to color the wool found in historic Turkmen weavings is important. But how one uses this information even more so.

It is obvious discovering certain combinations, especially ones that are infrequently found, can be highly indicative of unrecognized connections certain weavings maintained.

Grouping examples with such rarely seen combinations could give researchers help solving the mysterious myriad of attribution questions now present in Turkmen rug studies.

RK has been talking about the importance of dye analysis for more than twenty years but we have always said there is another related study that also must be undertaken to truly begin to solve these questions.

Both dyestuff samples and water samples must be collected in the historic areas now known to have been home and pasture for the Turkmen.

Once this essential work has been done, the samples carefully analyzed and compared to the dye testing of Turkmen woven products, RK believes we will have a far more open window of understanding.

For the 15 years rageth spent working on this publication, and considering the considerable amount of money he raised, such a project to collect samples in Turkmenistan could well have been initiated.

Too bad rageth was not capable to both understand this relationship, and to have made at least a start towards gathering this type of data.

In the final section of his essay, titled “Tribal Attribution by means of Dye Analysis”, rageth attempts to make practical use of the dye testing results he collected.

Of course being able to correlate the results of dye testing from historic Turkmen weavings with physical evidence of water, plant and insect dyestuffs samples collected from their now known locations would be the best and most efficacious way to propose credible tribal attributions and locations where a particular weaving was made. However, the limited scope of rageth’s investigations not only precluded securing such information, it actually prevented generating anything better than the guesswork which is already prevalent in Turkmen carpet studies.

So let’s look at what rageth did with the dye testing data he was able to collect.

The new scientific findings on the use of lac dye on wool among the Turkmen gained through this study can be helpful in connection with a tribal attribution. This is particularly the case in regards to differentiation between the Salor, Sariq and Teke. In most cases, a differentiation of objects like khali and chuval of these groups is not too difficult, but looking at tent bands, things become less clear. In the first case, of a group of nine tent bands distinguished by a particular stylized ‘compound palmette-tree’ design (figs 19-21) an unambiguous tribal attribution was not possible.

figs. 19, 20 and 21

It was unclear whether Teke, Sariq or even Salor people made them. The results of our dye tests in our study add considerable clarity and insight to the attribution process. Based on the frequent and systematic use of lac dye on wool amd cochineal dyed silk, in conjunction with its general appearance, one of the tent bands (fig. 19 cat. no. 4) can now in all probability be attributed to the Salor.”

Cat. no. 4

This tent band’s color sequence – dark blue, lac dyed red on wool, dark blue, cochineal dye red on silk etc. – very much resembles the Salor kapunuk cat. no. 3, which shows the same sequence in the upper horizontal panel’s curved leaves, and also in the vertical drooping panels to the left and right.”

Cat. no. 3

In addition, the tent band uses this sequence in what seems to be the most characteristic Salor tent band design: the stylized compound palmette-tree (figs. 19-21). All these facts speak for a Salor ttribution for this specific object.”

They do? And are these facts or merely supposition and opinion?

The only fact here is the dye test findings of lac on wool and cochineal on silk.

There is absolutely no connecting bridge of evidence to say this was ONLY done by the Salor. Nor is there any evidence to connect the so-called compound palmette-tree to the Salor, nor even a shred of documentation to say it was exclusively a design belonging to them.

Then there is the larger over-riding issue of whether most tent bands were actually made by Turkmen weavers, even those working in workshops, or were they produced by specialist non-Turkmen weavers for the Turkmen?

Essential issues like these are overlooked by an amateur ‘researcher’ like jurg rageth, his presenting suppositions and opinions as facts an even a greater indication of a failure to properly place what little information this book contains in a realistic and proper framework.

Another point rageth brays about is the adoption of science to Turkmen rug studies, yet he continues offering up much that has nothing to do with any “new perspective” based on science.

For instance in discussing the two other tent band illustrated above, figs 19 and 20, rageth states “So, at least one of this group is a Salor weaving. But what are the others? One of them is the tent band from the Textile Museum in Washington, DC (fig 20) and the other is cat. no. 39, the tent band from the collection of Francois Ang in Paris.”

Cat. no. 39

Several samples examined for insect dyestuffs from both bands turned out to be dyed with Mexican cochineal, and no lac dye was detected. Comparing the complex ‘compound palmette-tree’ designs (fig. 19-21) of the three tent bands, clear differences can be observed; in the band now recognized as Salor (fig. 19) not only the design, but also the palmette is more sophisticated than in the two other objects (fig. 20-21). Although the first two meters of the Ang band (cat. no.39) indeed show what one could call ‘Salor quality’ colours, this is not the case for the rest of the band; the palmette changes to darker, more somber shades. Instead of bright red we now find brownish purple. This color change is even more visible in the lower left corner of the colour illustration (cat, no 39). All this speaks more strongly for an attribution to the Sariq or Teke. Much of the same applies to the band from the Textile Museum (fig. 20). In the 19th century it was not unusual for the Sariq and Teke to copy Salor designs.”

Where to begin to shine some reality into rageth’s fantasy factoids?

First off, the tent band rageth pretends to prove is Salor (fig. 19) is far from proven, as none of the alleged evidence he presents goes beyond questionable supposition.

From the picture it definitely looks to be earlier and more accomplished than the other two, but that does not in any means equate to it being woven by a Salor weaver.

As for the other two, rageth tried to build a case, a flimsy one at that, they are Sariq or Teke. But here, as well, no real evidence is offered.

The only evidence is the discovery of lac in fig 19 and only cochineal in the other two.

This is interesting but that’s as far as it goes. No honest professional researcher would try to use this to assert provenance, and to do so is just more carelessness at best.

Same for calling the dyeing in the first two meters of fig.21 ‘Salor quality”.

Why didn’t rageth dye test a sample of the wool from that area, as well as some from the rest of the band to answer the obvious question what caused the color change. And of course what was the dyestuff to make that first two meters appear to be ‘Salor quality’? Considering rageth had 15 years, not doing this is unthinkable.

This would have been an excellent chance to have some facts to rest a case on, but clearly either rageth’s ability to understand the issues is too poor, or there was no one willing to give him the money for the testing.

Frankly, were the latter the reason RK thinks using the resources available to him here, rather than elsewhere, would have had much more merit, and then this could have been undertaken.

But, we believe the former to be the reason with monetary restraint not entering into the picture.

Next up for supposed ‘attribution based on dye testing’ is a well-known chuval, illustrated as both cat. no. 64 and fig. 22.

Cat. no. 64

RK knows this weaving well, we first saw it in the late-1980’s when it belonged to jim, aka windbag, allen.

In fact RK wrote up the story, no need to go over it again. It can be found here:

At that time RK thought it an example of the group with two level warp, open right asymmetric knotting that we view as an S-group variant. We still do.

The dye testing rageth did revealed no lac dyed wool but rather wool dyed with madder. This is not surprising for us, but the fact rageth attributes to the Teke is. The main reason being the two level warp construction, one absolutely no Tekke weaving has yet shown.

Basing that opinion, as rageth does, on the brownish red ground color is flimsy at best, and we definitely do not agree. Plus, we know this chuval well and it is clear to us the brownish red color has been caused by deterioration, perhaps excessive light or even a mild 5-10% chlorine wash. The latter was done to many brightly colored Turkmen weavings during the 1920’s and 1930’s in order to tone down their colors so they would not compete with the more somber decorative styles present in America and England at that time.

Again, were rageth a Turkmen expert he could have tried, with Wouters help, to see if in fact the color was artificially affected and changed.

In conclusion rageth’s believes “…our chemical tests of Turkmen weavings provided revealing results, particularly regarding dating and attribution, and are at least a stimulus for further research.”

RK has quoted all the passages where rageth has tried to prove dating and attribution, and we do not believe any sane reader could possibly agree any of it is proof.

Nor could RK honestly say anything truly revealing has been discovered.

Yes, a good number of examples with tin mordant, madder and lac dyed wool, and cochineal dyed silk were identified by rageth’s dye testing but all these factors were already long suspected and, since Whiting’s very initial chemical dye analyses more than 40 years ago, known to be fact.

The many dye tests proved the use of Mexican cochineal demonstrate the international trading contacts certain Turkmen weaving groups enjoyed, but none of those finding can be used to date any Turkmen weaving, regardless of rageth’s attempts to maintain they do.

To close out this part of our review the following quote from rageth makes clear the revisionist bent and illusionary importance he thinks his book’s information has brought to Turkmen carpet studies.
Finally, our study has demonstrated the value of dye analysis findings in constructing better supported attributions of specific pieces to specific weaving groups.”

Sorry, herr rageth, but first of all other people, RK included, have known for decades dye analysis using the newly developed techniques of SEM(scanning electronic microscopy), TLC(thin layer chromatography), and HPLC(high pressure liquid chromatography) would yield interesting results.

Granted, rageth’s book is the first to present a good array of them and for that he gets some kudos.

But results of scientific testing are only as good as how they are used for interpretation, and there is no doubt the results rageth has complied have not in any way, shape or form made any breakthroughs for Turkmen carpet studies.

Nor could they, as there was not a broad enough sample to build a big enough database from which far more detailed and significant interpretations could be drawn.

Plus we sincerely doubt jurg rageth could, even if he had such a database, properly and insightfully interpret the results to yield revolutionary conclusions and discoveries.

more to come, stay tuned…

Home   Buy/Sell at the Kazbah   Terms Of Service