can get to bedrock with respect to Turkmen rugs. The Akhal oasis,
however, may provide an opportunity for finding a few brass tacks.
Although nowadays largely ignored in the lexicon, rugs of this small
area were a discrete type in the pre-revolutionary Russian Turkmen
Embrace of the Bear
correctly bounded Akhal as a 240 verst by 2030 verst
(a verst more or less equals a kilometer) strip of land flanked
by Caspian Sea, desert, and mountains. 1 As such it was
quite isolated, had little by way of hinterland, and was not on
a trade route as was Merv.
land fringe immediately north of the mountains, colloquially known
as the Skirt (Atek), was watered by rivers and
streams coming down from the mountains and creating a series of
oases. The Russian conquest started from a base on the Caspian and
proceeded west to east through the Akhal, then peacefully to Merv,
the Mervis wisely having decided not to fight. A late 1880s
map (Figure 1) shows the Skirts major features. The solid
line running through it is the railroad; with this the isolation
abruptly ended. The curving dashed line is the outer limit of Akhal
usually English, had been in the area, but these for the most part
were military reconnoiters having to do with terrain and routes,
and largely uninterested in the socio-economic setting. Not all
had tunnel vision. One 1839 trip noted the presence of carpets of
quality as well as the sedentary agricultural population. 2
But it is Russian sources which are genuinely descriptive. To be
expected, for a new colonial fiefdom was of interest.
point of the Russian military operation was the defeat of the Akhal
Tekke in 1881 at the fortress Geok Tepe. The advance (8,000 troops,
58 cannon) proceeded via construction of a railroad soon extended
(1888) to Bukhara and Samarcand. Looting took place after the slaughter
at Geok Tepe, rugs very much in evidence. An Armenian sutler with
the Russian force participated in the second day of pillaging and
acquired a large number of valuable carpets, subsequently
confiscated by the authorities. 3 A later -- and likely
less than accurate observation (1895) -- had the looting going on
for a week with the value of things stolen as three million rubles.
4 Armenian merchants arrived on the heels of the Russians,
obtaining rugs in quantity for resale. 5
By 1885 the Askhabad commandant, Komaroff, was collecting them.
6 Since the initial administration of the new province
was based in Tiflis, some rugs found their way to Caucasia; in 1889
Turcoman carpets from Merve somewhat improbably were
on the walls of a Daghestan khans home. 7
surveys of Akhal and environs identified its peoples, places and
products. 8 One written in the wake of a 1879 Russian
campaign noted the pronounced degree of cultivation in the area.
9 A latter 20th century Soviet journal article sums and
references most of these compilations. 10 The area was,
in brief, arable, irrigated, cultivated land with a large sedentary
population; both clan and family had nomadic and sedentary wings.
In 1888 a traveller
trekked from Kizil Arvat through extensive truck gardening and its
various crops -- cucurbits, grains, animal fodder, etc. -- and a
nascent silk industry, trees only recently having been planted.
It was at the 130 kilometer mark that rug and palas (kilim)
weaving began in earnest, activity in virtually every household,
building up to a dense concentration in Merv. The point where rug
making began to intensify would have been somewhat east of Geok
Tepe. Also noted was that rugs here, unlike those of Persia, were
made with vegetable dyes. The author went on to say that carpets
up until now [are] not in use by us due to the absence
of good roads, but because of the railroad, will not be slow
to appear in European markets. 11 Turn-of-the-century
government reports said the same, adding the facilitating role of
account is frustrating. Henri Moser went to Akhal, the only place
he visited, before 1885, having crossed the desert from the north,
an unusual route. While there he remarked on the fine embroideries
and beautiful rugs of Akhal. 12 Mosher purchased
some of these and returned to Switzerland. Dying in 1924, he donated
his many-faceted collection of objets de virtu to the public.
But not the rugs, left to his wife; there seems to be no way to
consequence of one of the early Russian survey expeditions was a
recording of Akhal designs: 13 Tekke Ornamental Designs
together with their application for Carpets, Embroidery etc.
For is a key word. This is a straightforward book consisting
of (a) illustrations of possibilities for carpets and carpet-like
products to serve as furnishings in European settings, and (b) drawings
of motifs to serve as patterns for rugs and embroidery.
The book has come to notice a few times; perhaps the most visible
misstates its role and misses its relevance. 14 Its assertion
that the carpet applications were probably
in Russian homes for some decades rests on woefully inadequate
evidence. The point in time of the conquest and Russian comments
shortly thereafter about likely importation of Turkmen rugs make
it clear that Akstafev was not following a fashion, but creating
one. Elena Tsareva in her appreciation of S. M. Dudin got the Akstafev
role exactly right. 15
of course, are recognizable at a glance as being cartoons. These,
rather than the furnishings illustrations, are what is important:
a compendium of motifs then in use in Akhal. Its relative prior
isolation from both Yomuds and Mervi means less outside influence,
and a reasonable chance of motif fidelity to the past.
followed standard arrangements for managing a province (oblast).
One was creation of five districts (Figure 2) with Askhabad City
as capital. Imperial Russia was extremely good at cataloguing. The
great census of 1897 tabulates the Transcaspia population, its nature
and distribution, during the vintage weaving period, 1880
1915. Numbers of interest are: Askhabad district, 92,205, with 19,426
in Askhabad City; Merv district, 119,255, with 8,533 in Merv City.
The key data are those having to do with carpet makers: Askhabad
district -- 376; Merv district -3194. 16
interest in local craft activity began promptly with a kustar
section appearing in each provincial annual report. From about 1900,
functionaires were worrying about weavers movement away from
traditional designs. Standard kustar support measures
slowly came into being -- promotional exhibitions (Merv, 1908, 1909,
1910; Askhabad, 1912), establishment of wool warehouses, furnishing
of dyestuffs. But there were no training schools for weavers.
period 1883 1889 Askhabad rug production averaged 54 per
year, outputs rising toward the end. For palas the yearly
average was 385, tailing off the last three years. 17
Beginning in 1900 production data appear for both Askhabad and Merv
and reveal an interesting difference in scale. Statistics for 1900
show 114 rugs and 336 palas made in Askhabad, 1146 rugs and
26 palas produced in Merv. 18 Data of 1907 are
for rugs only, expressed in ruble value: Askhabad 9230, Merv 83700.
19 The 1910 numbers for Askhabad were -- rugs, 310; palas,
196; for Merv, rugs, 913; palas, 75. 20 Piece
counts in 1911 were: Askhabad rugs, 208; palas 200;
and, Merv rugs, 1115; palas, 101. 21
The data underscore
the point that home weaving was not only a domestic matter but also
an economic activity. Immediately prior to World War I (a back-breaking
blow to carpet-making) both Merv and Askhabad enterprises were hiring
weavers in order to produce for the commercial market, although
European buyers had not yet arrived . 22
A 1903 travelogue
treated rugs of Akhal as a distinct type. After noting the Merv
penchant for use of the Salor gol the author added
designs can most conveniently be distinguished as: Akhal, Ersari,
Yomud roses, and the patterns of namazliks and ensi
23 In 1914 Akhal products continued as a category: In
addition to the Salir design frequently used by Tekke-craftsmen
of the Merv area, there are the well known designs of the Akhal,
Ersari, and Yomud types.. 24
The Akhal carpet
remained distinct at least until 1930. Tekke. A completely
different type of rug. Its name is observed in literature. But in
all the markets, the name of Bukhara is used to cover all the varieties
of weaving techniques of the production-districts, for example:
Akhal-Tekke, Merv (double-wefted) and properly Tekke. 25
The early Russian
authors quibbled over provenance assignment of particular rugs between
Merv and Akhal. Semenovs commentary on Bogolubov followed
the not uncommon reviewer practice of devoting a great deal of ink
to what the reviewer thought, and not so much to the books
contents. (Semenov was a major scholar, the best of the early authors
in Dudins opinion, Semenov having been the one who spent time
in the area.) Dudin similarly had nit picks with some of both Bogolubov
and Felkershams attributions. This discourse took place via
artist illustrations and the argument was in terms of design.
The Akhal design
lode is the Akstafev book with its 28 illustrations, all standard
Turkmen motifs; four are reproduced (from some very blurry Soviet
microfilm) here -- a pair of gol variants and a pair of border
motif variants. One gol is somewhat round (Figure 3), another
somewhat flat (Figure 4). While artist copies of motifs into cartoons
may distort (possibly the secondary gols), it is difficult
to think that either their shapes or the vertical and horizontal
connecting lines them are embellishments. Of the two lobed leaf
motifs, the one is naturalistic and curved, (Figure 5), the other
(Figure 6), geometric. That both were in use at one place in the
same time puts a necessary caution on Gogels 1927 hypothesis
that this particular motif evolved from the natural to the geometric.
26 Willy-nilly any such proposition, the motifs coexisted
in Akhal. In art, the new necessarily supplanting the old is scarcely
an iron law, something to bear in mind with respect to the alleged
flattening of the Salor gol over time.
It turns out
there is a fly in the Akstafev ointment. Semenov, 27
noting that the motifs were rendered with utmost accuracy,
went on to point out that some were copied from war booty carpets
(Yomud and Salor) hanging on yurt walls in Akhal. It is good
to find an instance of this type of rug migration, but it does mean
the Akstafev motifs are not all Akhal.
(1917) succinct Dudin essay on Central Asian rugs deals with Tekke
or Akhal products as a group, the better examples characterized
as only a little inferior to those of the Salors, being
on average a bit coarser. Akhal carpets are identified as being
somewhat wide and squareish with the reverse side having a lighter
cast than those of the Salors because the Akhal warp was made of
undyed white yarn. White similarly features in the pile colors often
a rather loud effect 28 as can be seen
in the 1901 photograph of Figure 7.
later and longer treatise (a posthumously published academic paper
of 1926) also discusses the Merv and Akhal rug types as a group.
Characteristics are straight, thin, and durable yarns
[presumeably warps] usually white and rarely grey, white wefts,
and dense knotting. There is again an observation about Akhal rug
pile colors employing considerable white, more than Salor, which
permits their recognition even at a distance. 29
a consideration of Akhal designs, however, is Semenovs complaint
that Merv carpet-making had shown a great enthusiasm to imitate
the ornamentation of the Akhal Turkmens, something especially
noticeable in the dearth of Merv designs (almost absent)
in the Merv carpet exhibitions of 1908, 1909, 1910, with an overwhelming
majority consisting of imitation of Akhal-Tekke and Tejend
oasis carpet patterns. 30
The 1908 exhibition
is described in great detail by a provincial newspaper its
80 rugs, khorjin, torba, and chuval; districts and
villages of origin; names of the eight prize winners. 31
The first Askhabad exhibition, of 100 rugs (14 from Merv, 8 from
Tejend, and one Yomud from Krasnovdosk district) was held in 1912.
Some rugs were in beautiful Pendeh designs; another
although woven in Merv had a Pendeh pattern. 32
in the literature have to do with Akhal dyes, beginning with the
1888 comment use of vegetable dyes. At this same time,
dyes were imported from Khiva and Persia; only yellow was made in
Akhal (1890). Synthetics had come into use by 1900. The recent
use of synthetic dyes and hasty work were an official
complaint of 1903. In the Merv exhibition of 1908 only eight
rugs (the prize winners) had all natural dyes. (One of the 80 had
a natural field and a synthetic border.) Amid the complaints about
synthetics there was a positive comment that some weavers were learning
to give them up. A ground rule of natural dyes only for future competitive
exhibitions was under consideration by the authorities.
In 1911 the
Askhabad warehouse purchased 26 tons of spring shearing wool for
re-sale to rug makers and distribution gratis to poor weavers. Dyes
were imported from Bukhara. 33 At least by 1911, imported
vegetable dyes were being stocked by the Askhabad warehouse. In
1914 the Akhal Tekkes were making only red and yellow dyes; other
colors came via yarns from the city. 34
noted a surprisingly rich warm-cherry shade in antique
carpets including those of Akhal, by his time lost because the old
method of preparing madder is almost forgotten. 35
of design, workmanship, and materials began to be noted early on
and by the latter 1890s was perceived as a problem. By the
end of the first decade of 20th century there were complaints that
it was difficult to see sharp outlines in new rugs.
Along with these negative comments about the rugs of Akhal, however,
there was a positive remark concerning palasi made there:
are marvelous for their beauty and are
so tightly woven that it takes a long time for a pool of water to
seep through. 36
The key aspect
of structures, however, is a question about wefting.
The Adamov (1930)
typology casually mentions Merv carpets as being double-wefted but
does not say that about those of Akhal; the inference would be that
these are single-wefted. Adamovs contemporary, V. G. Moshkova,
a major field researcher, 1929 1945, said so explicitly. 37
Her text provides a great deal of descriptive information concerning
Akhal rugs and related products; this does not need repetition (except
for a reminder concerning the Akhal marker of equal knot nodes on
the back) because her material is well known from an annotated English
A three word
declarative sentence in the Moshkova/Galavin text states that Akhal
wefting is single, unlike that of Merv, with its typical
double weft. The OBannon commentary calls this
a glaring error. His criticism rests on review of a
group of published Tekke rugs (various attributors; no common definition)
of which 159 were double wefted, 57 single wefted. Given the 10
to 1 ratio of Merv to Askhabad weavers -- or the low Akhal share
of comparative production statistics (10% in 1900 and 1907, 34%
and 19% in 1910 and 1911) -- this result is hardly remarkable. His
speculation is that the observation stems from then contemporary
rugs in museum collections which Moshkova must have
seen. But any reading of her text destroys this speculation; throughout,
the words old and contemporary are regularly
used, and changes in materials, particularly dyes, over time are
speculation isnt necessary, since the Moshkova text is internally
inconsistent, saying elsewhere that at the end of the 19th through
the beginning of the 20th century double wefts were employed in
all Turkmen areas, and that in old Turkmen
rugs the single weft technique was much more widespread. This distinction
seems plausible. In brief, the matter is not glaring; rather, it
criticism has to do with design, the statement that Akhal fields
employed thin dark blue vertical and horizontal lines linking gols,
making uniform rectangles, and that such was not a Merv practice,
the gols being placed on a vacant (ie: empty)
field. The O'Bannon commentary sets up the straw man that the lines
cant only be present in Akhal rugs because it is rare to find
antique Tekkes (undefined) without them. The Moshkova/Galavin
text does not use the word only and makes the distinction
in a discussion of design. Good to remember that by 1908 Merv rugs
were using Akhal patterns. A plausible speculation in this instance
is also available as to what could have been going on: design spread,
uneven survival numbers, and uncertainties about time of manufacture.
a relatively weak instrument to describe a visual art, and care
should be taken about limitations of words. This is particularly
true for the Moshkova text, given its multiple authorship, its occasions
of joint authorship, and its posthumous publication without an evident
guiding editorial hand. Yet it remains a major source concerning
the rugs of Akhal.
So, too are
the snippets from various observations of the past posted here.
It may be of
interest to pursue the somewhat consequential palas manufacture
in Akhal, or the checked blue line of one of the Akhal gols
illustrated here. But these excavated snippets and the much more
substantial Moshkova research need to be tested against the great
deal of work done in identifying and linking the characteristics
of various Turkmen rug types -- physical attributes as well as stylistic
groupings. Both the old words witness the weaver numbers
distinction between Akhal and Merv -- and contemporary findings
when combined can shed more light on distinctions among Turkmen
rug types: in brief, an continuation of the OBannon Moshkova
record does, however, establish one thing. The weavers of Akhal
were key players in Tekke carpet design.
Felkersham, Baron A., Staryinnye kovry Srednei Azii, Starye Gody,
October December, 1914, p. 103.
2. Abbott, James, Narrative of a Journey, 2 vols., London,
1856, p. 17, p. 327, p. 47.
3. Marvin, Charles, The Region of Eternal Fire, London, 1888,
4. Shoemaker, A. A., The Sealed Provinces of the Czar, Cincinnati,
1895, p. 84.
5. Capus, Guillaume, A Travers le Royaume de Tamerlan, Paris,
1892, p. 381.
6. Cholet, Pierre, le Compte de, Excursion en Turkestan, Paris,
1889, p 43.
7. Abercromby, J. A., A Trip Through the Eastern Caucasus,
London 1889, p. 95.
8. For example, Kuzmina-Karavaeva, Col., Zapiska
Ateke 1885 g., Sbornik geograficheskikh, topograficheskikh i statisticheskik
materialov po Azii, VXXI, St. Peterburg, 1886.
9. Weil, Captain, Le Tourkmenie et les Tourkmenes, Paris,
1880, p. 9.
10. Orazov, A., Etnograficheskie ocherki khozaistva Turkmen Akhala
b XIX XX v., 1985.
11. Vascileva, P. S., Akhal-Tekinskii oazis,
St. Petersburg, 1888, pp.42-43.
12. Moser, Henri, A Travers lAsie Central, Paris, 1885,
p. 264, p. 330.
13. A. Astafev, Tekinskie Ornamenty c ix primleniem dlya
Kovrov, Vyshivok I proch., St. Petersburg, 1885.
14. Lownds, G., The Turkoman Carpet as a Furnishing Fabric
Turkoman Studies I, ed. Pinner, R. and Franses, M., 1980, p. 96.
15. Tsareva, Elena, Samuil Martynovich Dudin 1863
1926, HALI 7, No. 3:16-17, p. 17.
16. Pervaya vseobshchaya perepic naseleniya Rossiskoi imperii
1897, LXXXII Zakaspiiskaya oblast, ed. Troinitskii, N.
A., St. Petersburg, 1904.
17. Obzor Zakaspiiski Oblast, Kustarnaya promyshlennost
18. Obzor Zakaspiiskoi Oblasti za 1900 gode, Askhabad, 1902,
pp. 100 117.
19. Palen, K. K., Otchet po revzii Turkestanskogo kraya,
Materialy k kharakteristike narodnogo khozyzystva v Turkestan, Prilozheniye,
Supplement in three volumes, 1911 p. 405.
20. Obzor Zakaspiiskoi Oblasti za 1910 g., Askhabad, 1913,
p. 100 ff.
21. Obzor Zakaspiiskoi Oblasti za 1911 g. Askhabad, 1915,
pp. 207-- 212.
22. Semenov, A., op. cit., p. 93.
23. Dmitriev-Mamonov, A. I., Putevoditel po Turkesktanu
, St. Petersburg, 1903, pp. 11416.
24. Yampolskii, I. P., Kustarnoye delo, Asiatskaya
Rossiya, V. II, St. Petersburg, 1914, p. 398.
25. Adamov, A. K., Sovetskie kovry i ikh eksport, Moscow-Leningrad,
1934, pp. 18-19.
26. Gogel, F., Turkmenskii kover, vystafka muzeya vostochnykh
kultur, Moscow, 1927, p. 1213. A very similar but
not identical article appears in English in the Burlington Magazine,
V 1, No. ccxc, May, 1927.
27. Semenov, A. op. cit., p. 104.
28. Dudin, S. M., Kovry Srednei Azii, Stolitsa i
Usadba, No. 77-78, 1917, p. 13.
29. Dudin, S. M., Kovrovye izdeliya Srednei Azii,
Sbornik muzeya antrolopogii i etnografii, V. VII, 1928, p. 122-24.
30. Semenov, A., op. cit., p. 143.
31. Kovrovaya vystavka v Merve, Turkestanskie vedomosti,
N 46, 1 (14) March 1909.
32. Kovrovoe proizvodostvo v Turkestane, Turkestanski
vedomosti, 10 (23), January 12, No. 7,
33. Obzor Zacaspiiskoi Oblasti za 1911 g., Askhabad, 1915,
pp. 207 212.
34. Felkersham, A., op. cit., p. 104.
35. Felkersham, A., ibid., p. 157.
36. Felkersham, A., ibid., p. 104.
37. Moshkova, V. G., Kovry narodov Srednei Azii konsta XIXnachala
XX vv, Tashkent, 1970, pp. 139 157.
38. OBannon, George W., and Amana-Olsen, Ovadan K., Carpets
of the People of Central Asia of the Late XIX and XX Centuries,
Tuscon, 1996, pp. 208 211. The translation is somewhat
less than precise, beginning with the title. It is not a good idea
to use translators who are native speakers of the source material;
the translation should be the product of a native speaker of its
language. Fewer mistakes.