I believe we have barely begun to understand or consider the implications of the Internal Elem configuration theory first posited first by Shiv Sikri. Part of the problem at hand is one John Howe will love. It is the language used to describe the occurrence of the markings that define the Internal Elem configuration in so many rugs. Yon's Salon uses words like "design irregularity" and "a tradition to insert imperfections" in describing these markings and suggests that the markings could be "inadvertant errors", "esthetic effect", "change of mind" or "traditional reasons" - the tradition being one in which imperfections are allegedly inserted because only Allah can be perfect.

I believe such words and descriptions misstate the theory and minimize the markings that define the Iternal Elem configuration and the significance of the theory as articulated by Sikri. If I understand the theory, the Internal Elem configuration hardly constitutes anything approximating an inadvertent error, a change of mind, an imperfection or an effect.

To the contrary, the Internal Elem theory depends upon these markings being both intentional and meaningful in their own right. As such the theory posits that these markings, however crude, arbitrary or random they might appear to our eyes, are in fact precisely and carefully executed and an integral part of the language of a carpet.

As I recall Sikri's argument, the markings he identified in their most articulated form should be regarded as an intentional elem type configuration internalised within the field of the rug that is potentially rich in symbolic value to the weaver and her people. The Internal Elem then is roughly a rectangular area with a perimeter described by a line extending across the second row, in most cases, of the major/minor design elements and the remainder of the rug below that line. In some cases the line is implied by two distinguished points in different color combination or scale. In others, the points are marked, two in the border/field for the top line and another single point at the approximate center of the bottom line. Other times it is implied by a subtle design variation. In this way, the relationship to the Turkoman Ensi, for example, and its three part design format including the elem familiar to us all is helpful only in that the three part configuration and the elem or skirt is so explicit in Ensis.

The Internal Elem theory and its markings are intriguing precisely because it is commonplace once you look for it in weavings from many areas and becasue it is part of a potentially much larger, all encompassing way to look at and read carpets. It has nothing at all to do with irregularity, inadvertance or imperfection. We see this in some of the images Yon has provided. In addition, if someone would be so kind as to post images of the Hilpp Baluch mina Khani rug (Chicago ACOR publication), the Hopkins Baluch (plate 315 in Atlantic Collections, the Haley main carpet from Pacific Collections as just three Baluch examples I would be delighted to make the points articulated by Sikri in 1996 concerning these carpets. From there it is a matter of self discovery.

For the record, the Internal Elem configuration and these markings also appear in many of the oldest Kurdish weavings.

Good luck, Michael Wendorf

Regards, Yon

I will adopt Henry Sadovsky's suggestion that we refer to the Internal Elem theory henceforth as the "Sikri hypothesis." As stated above, my understanding of the Sikri hypothesis is that the markers he has observed are intentional and precise. Further, the markers that make up a kind of internal elem in the bottom register of the rug are part of a larger systematic three part design scheme that can be observed in a large number of carpets over both distance and time in tribal type weavings. Finally, the Sikri hypothesis concludes that this systematic design scheme and the markers that articulate it are part of a kind of woven language - or the remnants of a language.

In Philadelphia, Mr. Sikri focused on Baluch group rugs. Among these is a well known rug belonging to Mark Hopkins that was exhibited and can be seen in the Atlantic Collections book that accompanied the ICOC.

Commenting on this rug together with a number of other related pieces, Sikri observed the opposed stripes which are distinctly colored and then stated that the unmistakeable marking forming the so-called internal elem is how the arrow motifs are dislocated in the second row.

Moving on, Sikri demonstrated this marking in several main carpets that are generally considered among the oldest and grandest Baluch group weavings.

Those of you who attended the second ACOR in Chicago will doubtless remember the so-called Hilpp Baluch carpet. Here Sikri observed the subtle but definite dimunition in the second row of the white flowers. This in turn was compared to several old Baluch mina khani bag faces where the same dimunition was observed.

Sikri then went on to another Baluch main that was exhibited at the San Francisco ICOC in 1990. The carpet, which is part of the Anne Haley collection, has distinct motifs in the bottom register of the field design.

In studying the Haley carpet, please note also the elongation of the design in the very top design register of the field. Sikri also has observed this phenomena in many rugs and bag faces bearing an internal elem. The Sikri hypothesis suggests that this elongation is likewise something intentional and meaningful in these weavings.

I believe the Sikri hypothesis is worth studying and understanding. It is remarkable how these markings appear so consistently in so many of what appear to be some of the oldest and some of the most otherwise carefully and lovingly woven carpets.

Thank you, Michael Wendorf

Nor are errors in repetitive designs surprising. Anyone who has visited weavers working in dark cellar-like spaces or rooms with single small windows understands that instantly. It is surprising that errors are not more prevalent.

When weavers purposely inject a "surprise" or vary a repetitive motif to prevent monotony, where are they to put such an element? Certainly not in the center of a border, as that is disturbing. Instead, such a break is much more comfortable and satisfying nearer one end. Such aesthetic constants occur in works in all media--not just weavings. Long dissertations have been written on "The Golden Mean"...mathematical proportions that human artisans everywhere have found the most pleasing.

Also, I think we need to keep in mind that except for very small pieces, most knotted-pile objects are the work of more than one person--and sometimes strong-willed people who have different ideas about how a design should be executed. You think that a rebellious teenage girl and her mother working together should not occasionally produce conflicts?! How about a new bride working for her domineering mother-in-law? Or how about a young child's mistakes that a doting mother is reluctant to correct? I am not suggesting that there may not be some validity to the hypothesis being discussed, but that we should also allow for reflections of interpersonal relationships among the artisans involved. I've seen women haggling over such details, myself. We need to also keep in mind that during the course of a weaving, different people sometimes join the process; thus we can logically expect to find strange changes at any point in a rug.

As for "changes of mind" in the plan for a weaving, I've seen extreme example of this just recently in some Turkmen pieces--but in the plain color knotted lower border areas rather than in the designs themselves. In a Salor trapping that I have in front of me now, the weavers decided that the piece needed to be a little wider, and so they ADDED pairs of warps at intervals all across the weaving--a technically very difficult chore! In a Goklan chuval the weavers dropped out pairs of warps, a less difficult process, but one still prone to bothersome errors. So why such drastic alterations? Most likely, the weavers realized that the planned design required a different number of warps. Or perhaps someone decided that a different motif should be used. Did a customer or relative come along who wanted something other than what was planned? We certainly see rugs in which the borders have obviously been moved inward or outward to better accommodate the field motif. Whether or not this effect then became fashionable among some groups remains an open question. This "free-wheeling" approach at least became acceptable in some quarters. It is an interesting and provocative question.

Marla

One of Marla's observations concerning the Sikri hypothesis is that there is nothing very mysterious about finding design irregularities in the lower part of a rug with a repetitive design. The logical expanation is that, as Marla suggests, the weaver or weavers is/are merely experimenting with adjustments.

It is true that the Internal Elem configuration and its markings seem to be found mostly in rugs with a repetitive design. However, since formats such as the central medallion seem to be later design developments, I not certain what to conclude from this. More significantly, it is not simply that the markings appear in the lower part of the rug. The markings seem to appear rather consistently and precisely in the second row of the design. It seems to me that one would see a more varied and random distribution of what we are referring to as markings in these weavings and more commonly at the very bottom row rather than the second row if they were experiments with adjustments. Put another way, if the first row were successful, why experiment with the second?

I hope that the Sikri hypothesis has sent everyone back to their blanket chests and rug closets for another look.

Best, Michael Wendorf

This Salor trapping or torba was woven upside down--the pile inclines upward--and so we are seeing it here as it was on the loom. The top hem is at the bottom of the photo.

In the section shown in this scan, 7 pairs of warps have been added in the plain red band--14 warps in just 3 1/2 inches! This is repeated all across the piece! Alternate warps are severely depressed, and so we only see half of the knots clearly. If you follow knot columns vertically, you'll see where those columns branch within the red band.

Marla

There are a couple of reasons why a second row of motifs is often more problematic than the first in a rug with a repetitive motif.

First, with a drop repeat--that is with motifs staggered so that the second row of figures is positioned down partially in the spaces between the first--the weaver must deal not only with forming the motifs themselves, but with the connecting spaces, elements and the way they all fit together. Adjust one part, and another is askew. Very difficult sometimes.

In some instances, especially with major projects, the most experienced weaver in the household is likely to be the one who gets a pattern started; then it is turned over to younger weavers who may have difficulties in articulating the motifs. There are so many variables when more than one person is involved in producing a single object.

Marla

Regards, Yon

The first posting in the discussion of your Salon was one I put up suggesting, among other things, that there are a number of different things going on in the different rugs and that the same explanation doesn't apply to all of them. I still think this is so.

The jogs in border width or changes in border design that occur within the first inch or two of the origin of the field seem most easily accounted for as corrections of initial judgments, and (in my opinion) are in a category all their own. Furthermore, I have offered criteria by which this notion could be disproved if it is incorrect.

I cannot even imagine criteria by which the Sikri hypothesis can be disproved if it covers all irregularities occurring within the first or final third of any rug, whether unilateral or bilateral, whether in the field or in the border or both, whether conspicuous or nearly invisible. Indeed, the position seems to be that when they are present they have the special (albeit mysterious) meaning, no matter when, where or by whom they were made; when they are absent the weaver simply decided for some equally mysterious reason not to include them.

I find the topic fascinating, and have a fairly strong notion that there's some real stuff to be mined from it, but that the "lumping together" approach is sort of like supposing that every disease has the same cause.

Regards,

Steve Price

Regards, Yon

I am not suggesting that every one of these things is an accident or adjustment, and am fairly comfortable with the notion that many of the examples are irregularities introduced intentionally for some purpose. On the other hand, everyone with experience in crafts in general and weaving in particular acknowledges that errors do occur from time to time, so it seems reasonable to suppose that some of them can be explained that way.

I think it very unlikely that the stray single knots in your Yomud juvals and the odd motifs in the border of the Calachi carpet (to select two examples) have a common explanation. The border jogs that I believe are adjustments - that is, changes made after the original plan once the outcome was visible - appears only in your first piece (the Belouch), but it is a fairly common occurrence in village and tribal rugs from many sources and seems unlikely to be the same thing as the stray single knots on your Yomud or the border feature in the Calachi. That is, I would posit not less than three distinct phenomena from those examples alone.

The Luri bagface that Barry O'Connell sent to you also has a jog, but on one side only. I don't know what that's all about, but it would be a strange way to represent an elem. Marla's explanation for that one makes as much sense as any I can think of: two weavers, one on each side, the one on the right messes up at first and then straightens out. The difference in warp color between the right and left of that piece is also very strange.

Michael Wendorf has pointed out the frequent occurrence of irregularity in the second row of motifs in rugs with overall designs. This seems like a coherent category to me, different from the other three.

I don't think we can hope to come up with explanations for every one of these phenomena, but we should make the effort to do so in the hope that we can succeed with a few.

Regards,

Steve Price

Regards, Yon

I am concerned that you are both becoming mired in minutia and shifting focus away from the core of the Sikri hypothesis. This seems evident to me in your discussion of the Luri bag face that was posted in another thread and picked up here. Suffice it to say that I am personally familiar with the Luri bag face that is being discussed. In my judgment, the Luri bag face (and others like it that may exist) is a red herring that adds little or nothing to our efforts to understand or test the Sikri hypothesis at this point in the discussion. Breaks such as that found on the bag face may turn out to have a function or place beyond mere adjustment but they are not really consistent with the Sikri hypothesis.

My judgment is based on the following: First, I am aware of many Luri rugs and bags that do not contain such a break. In my judgment, many of these pieces show greater care in execution and age than the piece posted which in my view is clumsy. Second, some of the pieces that contain no break show an internal elem consistent with the Sikri hypothesis. In short, I tend to agree with Steve that the break in this particular piece is an adjustment.

The most important reason I ask that you not mire yourselfs in the Luri bag face is that it does not clearly fit into the Sikri hypothesis which is based upon the existence of a rectangular area that can be described as "elem" like. A single break on one side does not really comport with this concept.

Yon has asked Steve not to get hung up on the connotation of an "internal elem". Henry Sadovsky has also suggested that we discuss these markings in the context of the Sikri hypothesis rather than an "Internal Elem". I agree with this generally. However, we cannot discuss and test the Sikri hypothesis unless we understand exactly what that hypothesis is and what it says.

So what is the Sikri hypothesis? Rather than simply rely on my own recollection and understanding as articulated in previous posts, I called Sikri. He reiterated to me in very explicit terms that the markings he has identified "in their most articulated form might be regarded as an elem type configuration internalised within the field of the rug." I believe that it is critical to understanding the Sikri hypothesis that we agree that not all markings form such an elem type configuration. In Sikri's words, improvisation is integral to tribal weavings and some markings may be seen as just that. However, when one compares examples and the existence and consistent presence of certain "mannerisms" he concludes there is something more going on. To reiterate, the configuration that is at the core of the Sikri hypothesis is explicitly "a rectangular area with a perimeter described by a line extending (usually) across the second row" of a rug's major/minor motifs and the remainder of the rug below it. This is the Sikri hypothesis.

I think the Beshir rug in Yon's Salon and posted in a thread by Henry Sadovsky is an excellent example of this configuration. Likewise the pieces depicted in this thread that were among those originally cited by Sikri. We need to study these and other pieces that clearly fit this hypothesis and the even larger theory expressed by Sikri in order to advance the ball in this Salon.

Best regards, Michael

You do not need to remind me what the topic of this Salon is. The topic is the Internal Elem: Myth or Reality. Of course, the term "Internal Elem" is precisely that coined by Sikri to describe what some others are now calling the Sikri hypothesis. Moreover, I believe your Salon makes the point that not all "irregularities" occuring in a variety of weavings fall into the Internal Elem or Sikri hypothesis. I am merely suggesting that we make clear what markers or "irregularities" do form an internal elem under the Sikri hypothesis because it is fairly clear to me that many people who follow this Salon are not clear on this point.

Of course, you are welcome to discuss whatever you want. My view is now clear. In short, I believe that if you wish to have a meaningful discussion concerning whether the "Internal Elem" is myth or reality, you have to make sure that everyone understands what you are and what you are not describing when you refer to the "Internal Elem" - to do otherwise invites a rather obtuse discussion of "irregularities" that, in my judgment add nothing to the intriguing issue of whether the "Internal Elem" is myth or reality.

Happy holidays, Michael Wendorf

Seriously, though, I feel that these other types of irregularities are just as interesting and just as significant in terms of being possible manifestations of traditions practiced by weavers in various regions. And, for the record, I have observed many of these, which appear to be common among (but not confined to) Turkmen weavings, before I knew anything about Shiv's observations.

Regards, Yon

I don't see how anyone with eyes can fail to see that the rugs you present (and many others) can be logically divided into two or more rectangles arranged vertically. Likewise, it is obvious that many of these divisions are created intentionally.

That much is common to the Bard Hypothesis and the Sikri Hypothesis, and I suspect that most of our readers are comfortable with it.

At this point the two hypotheses diverge, and this is where problems arise. Let's take the Sikri Hypothesis first. It asserts that every rug in which it is possible to delineate two (or more?) rectangles has that feature for the same reason. It doesn't specify what that reason is; perhaps it is the dictum that only Allah can make perfect things, perhaps it is to introduce irregularity that distracts the Evil Eye, perhaps it is something else. But, and this is important, it is always the same reason and that it is never unintentional. It is the elements in this sentence that so many of us consider to be fantasy.

Now for the Bard Hypothesis. It asserts that those instances in which two (or more?) rectangles can be delineated are sometimes adjustments, sometimes errors, and sometimes intentional. The intentional ones can often be recognized, and might have been put there for any number of reasons. The reasons are not the same in every rug, but each reason has some foundation in the traditions of the weaver. I find this hypothesis easy to live with, and it leads very naturally into trying to define the various categories of "internal elem" in the hope that we can understand some of them (even if that understanding turns out to be discovering that some are "typographical errors"). Yon's example of the stray white knot in some Yomud borders is, I think, a perfect example of one kind.

NOTE: This message board is set to display 20 messages per page. If and when this thread (or any other) exceeds 20 messages, a new page will begin and can be accessed by clicking on the "next page" text at the bottom of the board.

Regards,

Steve Price

I do not think that Steve has it quite right.

I do not understand the Sikri hypothesis to state that every rug that contains the internal elem configuration, as defined by Sikri, to have it for the same reason although I believe it is true that Sikri argues that a true internal elem configuration is intentional. Moreover, I do not believe that the Sikri hypothesis specifically states why the internal elem configuration exists, only that its precision suggests a meaningful process that may or may not have something to do with a kind of old woven language or meaningful, evocative internal structure.

Steve's description of the Bard hypothesis is not all that much different except that the Bard hypothesis focuses equally on markings that may be and probably are errors, adjustments, changes of mind and the like that Sikri ignores - whether intentional or not. Sikri's hypothesis by contrast with the Bard hypothesis trys to see the internal elem as a process that is part of a larger description of the carpet and its design structure.

The yastik above, the cover of the book documenting the exhibition at the Philadelphia ICOC is another example of this process and hypothesis. Please note that it is here in a non-repeating design format. Note also how the bottom border cartouche is indented to express the internal elem together with the compression in the spandrel area.

To understand this process would take a lot more work and may ultimately be fruitless. However, I find the basic hypothesis and the expression much more intriguing than that proposed by Yon which strikes me as more of an insect collection than a hypothesis.

Best, Michael

I was at Sikri's talk at ICOC a few years back, and so was Yon. I wouldn't want to hang for being wrong about Sikri asserting that the reason behind every "internal elem" was identical, but that is the way I recall it. Perhaps Yon can confirm my recollection or point out that I misunderstood or am remembering wrong.

As for Sikri's rejection of the notion that some "internal elems" were errors, he made that absolutely clear. A specific question was asked about the border jogs that nearly everyone agrees are most likely to be adjustments, and he responded that they are positively intentional, they are internal elem definition points, and that making the observations that might eliminate that explanation would be a waste of time.

As for the relative levels of excitement that the two hypotheses (Bard's and Sikri's) might elicit, I'd say this. Sikri's, if correct, would represent a completely new and unexpected expression of a hitherto unknown cultural practice common to all central and western Asian peoples and dating back to at least the time of the Pazyryk rug. That, of course, would be a blockbuster. The problem is that it is unsupported by evidence, so for the moment it is just an interesting fantasy. Bard's hypothesis is much more mundane in its implications. It simply asserts that some "internal elem" configurations are not there by accident. On the other hand, it has one great advantage over Sikri's. It is almost certainly correct.

Regards,

Steve Price

Best, Michael

Thanks for correcting my misunderstanding about whether Sikri maintained that all internal elems were a reflection of the same cultural tradition. If I now understand that part correctly, it is identical to the Bard Hypothesis in the matter of what is meant by "tradition": it is any practice that the weaver didn't do by accident or invent herself, and different "internal elems" can reflect different traditions (using the word in that sense).

That leaves the issue of whether any internal elems are the result of accidents or changes in plan. Yon acknowledges that some are. Sikri insisted that none are, or, at the very least, that jogs in borders are not. I think he missed the ball by a mile on that one.

Regards,

Steve Price

We are like two ships passing in the night.

You are correct, I think, to conclude that Sikri believes that no internal elem is the result of an accident, except perhaps where someone is merely copying something. However, the question is whether an internal elem exists in a particular example. Her is our problem. I am not certain that a jog in a border always constitutes an internal elem under the Sikri hypothesis. In fact, I am fairly confident that he does not make this argument.

Remember, in fact etch it on your forehead, the internal elem is a rectangular area with a perimeter described by a line extending across, in most examples, the second row of major/minor motifs and the remainder of the rug below it.

If you internalize this, I believe it is fair to assume that Sikri would not consider a jog as creating the perimter of the rectangular area unless the jog appeared precisely at the second row of major/minor motifs. One of the keys to these theory is the precision of the marker, this must be absolutely clear.

Regarding tradition, as previously stated I do not believe Sikri has articulated it the way Yon has. He has indicated that where the internal elem appears it is so precise as to suggest the weaver intended it and that it may be part of a larger system of design that contains a kind of woven language. I am not sure it goes much beyond this at this time. But you are correct that in concept the internal elem hypothesis advanced by Sikri would be a blockbuster. That is one of the reasons why I am so insistent that we understand clearly what it is and what it is not and that we not confuse it with all irregularities or improvisations that appear in rugs.

Best, Michael Wendorf

Check the markers in the cartouches of the main border. Do you see the indentation on both sides? - not a knot count issue.

Best, Michael

Now, I know that I'm what the feminists call a linear-thinking male, so I take it that someone means what he says when he says it clearly. Sikri specifically included the border jogs in his presentation in Philadelphia. He was questioned about whether those weren't most easily explained as adjustments and a specific test of that alternative was proposed. He dismissed the possibility that they do not comprise "internal elems" as self-evidently incorrect and not worth the time it would take to apply the test.

Yon was present, I'm sure he remembers it that way, too. In fact, he was the one who asked the question.

Perhaps Sikri has now modified his position on that; I hope so. All I have to go on is the public presentation he made in Philadelphia at ICOC. My recollection of some of the details is, no doubt, faulty and inexact, but the business about the border jogs is one about which I am certain.

Steve Price

1. Steve, much as I like and respect Shiv, he is not the prophet Mohammed. Let us not turn the question of he did or did not say into a theological proposition.
2. Michael, you over-stress the importance of the second row of medallions. Many of the examples are otherwise, or relate to rugs with small-scale designs.

Regards, Yon

I do not know whether I over stress or not, all I am doing is stating the Sikri hypothesis as a basis for explaining some of these irregularities. Perhaps there is nothing more to say.

Best, Michael

One of the problems I'm having with this discussion is that the topic is a moving target.

I have, in several previous postings, stated what I understand to be Sikri's Hypothesis and what I understand to be Bard's Hypothesis. Yon has not contradicted my statement of his in my most recent iterations, so I assume that it represents his position. It is, in simplest terms, that many (but by no means all) irregularities in rugs, particularly in the lower thirds of rugs, were put there with forethought by the weavers. Some groups of these planned irregularities may have special significance to the weavers who made them, some may represent what we might call force of habit. He uses the term "tradition" to cover anything not accidental or done as an adjustment for an error in judgment. This is one reasonable meaning to the term, and as long as we understand the language being used, there's no problem. I find the hypothesis unassailable.

I thought I understood Sikri's Hypothesis, but have learned in the past day or so that I did not. However, one of the things that he made crystal clear at his talk was that he did not accept the notion that any border jog or change in border pattern in the lower part of the rug was donewithout forethought. Michael now objects to taking this statement as reflecting Sikri's Hypothesis. But if it doesn't, he must have changed his idea. I applaud that decision if it's so, but if the Sikri Hypothesis has changed since the talk it really ought to be given another name. It's confusing enough having to keep track of whose definition of "internal elem" is being discussed - one demands at least two points to define it, the other allows just one point to do so. Having to keep track of which hypothesis is being discussed under the title "Sikri's Hypothesis" makes discussion unwieldy.

I will add this. If Sikri's Hypothesis is essentially identical to Bard's except for Bard's acceptance of single points as defining the "internal elem", and if he described it that way in his Philadelphia presentation, I cannot imagine why it met with skepticism. To deny it makes about as much sense as to deny that rugs with borders have them because the weavers put borders on them purposely.

Regards,

Steve Price

Yes, that's why.

Is it an elongated rug or a compressed rug.
The way it's presented it looks elongated.
But looking at the corner fields, the compressed border design in the lower border design, it could well be a compressed rug.

You're telling me the way you should be looking at the rug's design. But that doesn't mean it has been knotted like that.
If this discussion needs to be specific, in order to get us any where, I think this is essential.

Best regards,
Vincent Keers

Now you really are testing my patience on this border jog issue. Have you read post 29, the second paragraph? Read it again please. If we cannot get this far there really is nothing left to say.

Best, Michael