| Dear all,
Please forgive this long post, but I wanted to share some information
that ought to be meaningful to parts of this salon.
The issue of the capacity of natural dyes to bleed continues to be
discussed, with conclusions seemingly being drawn from anecdotal
testimony. If we had before us an actual example of a rug with bleeding,
simple dye testing could tell us whether the running was from natural or
synthetic dyes.
While I can provide no personal expertise myself, in an earlier salon I
cited an experienced local weaver and dyer to the effect that natural dyes
in wool essentially do not bleed. I may have perpetuated a possibly
incorrect belief by quoting Elena Tsareva as saying that she had "heard"
that the excessive use of cochineal in the dye vat can result in bleeding,
even though she was merely repeating something she had been told but had
no personal experience with this effect.
Because of the continuing controversy, I asked Paul Mushak, a North
Carolina physician and toxicologist as well as a recognized authority on
the subject of dyes, for his comment. Here is his reply, without my
further comment:
"Dear Wendel:
I have a few comments and conclusions in response to your e-mail query
about the problem of whether natural dyes would ever bleed into adjacent
areas of oriental rugs, trappings, etc. You may pass along these comments
to whomever or in whatever venue where they may be useful. I have been
occasionally tempted to write on the topic but I am always pushed for time
with my regular professional work. I have not considered any formal
write-up paper or the on-line version of ORR.
Context and Historical Comments
There has been considerable confusion about the matter of whether
natural dyes will or will not "bleed" into adjacent areas, bleeding being
most discernible and most objectionable as to aesthetic appearance of a
piece when bleeding is to ivory or other light color areas. Bleeding has
also been claimed as a crude visual criterion for sorting out natural from
synthetic dyes. To be more technically precise, however, this sorting out
would have to include both natural AND FAST CHROME DYES as non-bleeders
and potential bleeders as synthetics which are not chrome-mordanted. That
is, natural and chrome-mordanted dyes on wool in rugs would not bleed,
while synthetics other than chromes would bleed. However, absence of
bleeding is, obviously, not evidence of natural dyes being present. Pieces
may simply contain dyes that would run if given the chance but were never
washed or otherwise made damp, i.e., not given the chance. I have seen
numerous cases of easy-to-bleed red synthetics in the more recent Turkmen
rugs, specifically the Afghan "Mauri" or "Sarooq" pieces from the late
1960s and 1970s, but where the whites and creams are crisp and no bleeding
has occurred simply because the rugs were never washed or ever became
damp. These pieces therefore seem O.K. Taking a damp cloth to any red area
in these pieces, sad to say, gives immediate running with even gentle
rubbing.
This confusion over whether natural dyes will run onto adjacent areas
traces, at least in part, to the appearance of an off-color in the
cream/ivory portions of Turkmen pieces otherwise assumed to be of
considerable age. In some cases, the off-color appearance approaches a
discernible tan-pink. This was the focus of Steven Price's letter in HALI.
Also, certain very fine and presumably early dozars from the Fereghan
(Zili-Sultans) and Malayer Hamadan (Mishan Malayer) areas carry an
apparent early synthetic red that has typically bled. Similarly, there's a
type of Senneh kilim with a garish red and grey-olive palette with this
run problem. Here, too, the red appears to the eye as early synthetic
since the pieces seem to be from the early 1900s.
A seemingly well-known example of the problem in Turkmen work is a rare
example of white-ground asmalyks, i.e., the allover flower heads in a
field of ivory and a simple border, sold by Sotheby's in the early 1980s
for 44K (a record at the time for Turkmens), and mentioned again in print
more recently. It had a pink cast in the ivory. That led one Turkmen
"expert" to reputedly postulate that the pink was (presumably natural)
field yarn ivory but dyed pale pink to symbolize the bleeding attending
loss of virginity during the honeymoon by the bride. This would serve as
proof of a bona-fide virginal status to the newly-minted husband. Sheesh!
It's hard to envision any likely situation where this explanation would
apply. In theory, the virginal bride in the wedding camel cupola and on
the way to her wedding would not use a camel festooned with flanking
asmalyks already showing symbolic evidence of activity before the
honeymoon. This howler shows what people will say to defend forking over
big bucks for a rare asmalyk with what probably is an early synthetic dye.
I can offer a number of conclusions on the topic, based on available
scientific evidence and historical information.
Conclusion 1. Naturally dyed wool yarns in any rug-producing tradition
and area would simply not bleed as a general rule. Nor would modest or
moderate bleeding occur into undyed or lightly-dyed adjacent areas.
Dyeing of yarns with natural dyes for rugs and trappings in the Middle
and Near East entailed natural dyes being applied to wool (or in some
cases silk) yarns which had first been heavily mordanted by any one of a
small group of metal mordants (sometimes with minor use of additional
fixatives but never without metals as main mordant). The metal mordant
worked by binding first with the wool during mordanting, and then
connecting the wool surface with the dye through further binding when
placed in the dye bath.
A metal atom functioning as a mordant forms a binding "bridge" between
the dye molecule and binding sites on the wool surface. That is, we have a
metal mordant atom in the middle, between a fiber-mordant binding surface
on one side and dye-mordant binding sites on the other. Each mordant atom
has multiple binding sites, some of which hook into the fiber surface, the
rest hooking onto the dye. The dye can't release from this binding bridge
because it's held tight by binding with the mordant metal atom. Normal
washing and use of rugs and trappings would not release the dye molecule
from its binding with, say, aluminum ion in the case of alum mordanting.
In fact, dye release from testing fibers in the lab requires use of strong
lab chemicals such as acids to liberate the dye for subsequent analysis.
Conclusion 2. Overdying with more dye than mordant to accept the dye
would not occur. It is rarely the case that one can overwhelm or saturate
the metal binding surfaces with too much dye. Despite talk of too much dye
in proportion to mordant being possible, giving free dye to bleed to
adjacent areas, this rarely if ever technically happens. This is for
several simple technical reasons. First, because the amounts of metal
mordant used were huge. Any American craft dyer working, say, with alum
(potassium aluminum sulfate) as mordant and natural dye baths can tell you
this. Specialized dyers in rug areas would mordant the yarn, wash the
mordanted yarn and then add the yarn to the dye bath. No free mordant
existed at this point. Dyed yarns were typically then washed. Mordants
were not typically applied in rug areas in the same bath for yarn as the
dyes. Secondly, any excess dye in a bath would simply not stick to yarn at
all, never mind hanging around to then bleed later, or would be rinsed off
in the yarn rinse step. It would logically bleed off into the original
water dye bath or the rinse water. This simple bit of logic has usually
been lost on the scientifically naive.
Conclusion 3. Dyeing of yarns during the pre-synthetic period was a
highly-specialized craft practiced by specialists. It was also a
highly-developed craft in rug-producing tribal and village areas. Dyeing
was never practiced by just anyone who tried their hand at it while also
making rugs or trappings. Villagers and nomads typically bought or
bartered for the yarn from the skilled dyers. One cannot argue that
natural dyes would bleed if the process was done by unskilled or
amateurish dyers.
Conclusion 4. Once a rug was made, it is unlikely dyes might bleed if
the rugs or dyed areas got wet and stayed damp for a while. That
occurrence would require migration of dye and mordant complex from one
area of yarn surface binding to another. It would be highly unlikely.
Secondly, the types of pieces showing dye running are usually those from
arid areas, e.g., Turkmen, Central Asian pieces, not those from damp ones.
Conclusion 5. People here in the States who are proficient craft dyers
and quite familiar with traditional dyeing in the major rug-producing
areas should be able to readily show the above experimentally for one and
all. For example, if yarn skeins are unmordanted, does any dye uptake at
all occur when these are immersed in, say, a madder bath? If unmordanted
yarn skeins can be discolored without mordants present, does the
discoloration readily wash off with rinsing, as might occur with typical
dyed yarn processing back in early times in Central Asia and elsewhere?
Hope the above is helpful.
Best regards,
Paul Mushak" | 