African Textiles: the Karun Thakar Collection

My new book, out now. African Textiles: the Karun Thakar Collection, (Prestel), available now from good bookstores and the usual sources. Main texts and captions are by me, brief foreword by Bernhard Gardi, text in the North Africa section by Dr Miriam Ali-de-Unzaga, and an introduction to his collecting experiences by Karun Thakar. The book is a substantial (320 page) introduction to one of the largest and most important collections of African textiles, with particular strength in West Africa and a number of previously unpublished styles illustrated.

“Arkilla kunta / arkilla kerka: patterns” by Bernhard Gardi

“Written descriptions of looms, weaving technologies, and their histories often prove difficult for both author and reader.” Doran Ross. 1998. Wrapped in Pride: Ghanian Kente and African American Identity. UCLA Fowler Museum of Cultural History. p.75

These three motifs (figs.1-3) from an arkilla kunta-blanket are of outstanding finesse, quality and of a certain age. The vegetable colours have not faded and match very well, the yarn had been finely spun with great regularity. As arkilla kunta-blankets are very rare, their ethno-historical background is of interest.

The arkilla kunta is the old form of the better known arkilla kerka woven by the maabuube, the weavers of the Peul in Mali. Something like 200 years ago two groups of people left the Upper Delta of the river Niger in c0entral Mali and eventually settled down south of Gao in today’s Republic of Niger. These two small groups are known as Kurtey and Wogo. Their language is Songhay. The Kurtey claim their descent from the Peul, the Wogo see their mythical origin in a small place called Tindirma south west of Tombouctou.

Among other Malian traditions, the Wogo kept a specific symbol for their marriages until the early 1960s, namely this huge tent-like blanket kunta being completely woven with sheep wool. Not so the Kurtey; they had no such arkilla-blankets.

Literally, arkilla has an Arabic root meaning ‘mosquito net’.

There are significant differences between an arkilla kunta and an arkilla kerka. A complete kunta should have five rather wide strips (32 cm), the fifth or lowest one being simpler woven than the other four. [Fig.4] A kunta is tent-shaped. Hung over the bed it gave warmth and shelter during the cold season. Made of 95-100% wool the whole blanket is brittle and much rougher to touch than a kerka. A complete kerka should have six strips (22 cm wide), the sixth or lowest one being simpler woven than the other five. [Fig.5] In some areas there is a seventh strip in black and white called sigaretti to hang the kerka as a curtain in front of the bed. Well over 50% of a kerka (including the warp) is made with cotton yarn.

However, their ‘look’ has striking similarities. For instance, all the motifs and weaving techniques are the same, and on both types of arkilla we have a comparable rhythm of successive plain weave (cannyudi), weft float patterns (cubbe) and patterns in tapestry weave (tunne). While a kerka has just three dominant weft patterns with white tapestry on red grounds – the largest one being in the centre of the blanket - a kunta has four of them. And as a kerka of the first quality has twice six rather narrow weft float patterns, a kunta has five of them, the largest one being in the centre of the blanket.

On both, kunta and kerka, there are two major motifs as eye catchers in tapestry weave: the lewruwal (moon) and the gite ngaari (eyes of the bull), as the Peul say.

And for those readers who have followed us so far: a kerka has the lewruwal motif (moon) in the geometric centre - whereas a kunta has its lewruwal to the left and right of the centre – the centre itself being decorated by a huge diamond shaped pattern in weft float.

The moon motif lewruwal may be called differently in some regions – namely tiide eda: the forehead of the wild buffalo.

When we were in Wogo country in 1974 (Renée Boser-Sarivaxévanis and myself), we acquired three kunta for the Museum der Kulturen Basel. At that time the owners of the blankets or the villagers around there did not know the names of the patterns anymore. Names given to us were untranslatable for the linguists that we consulted in Niamey. Jean-Pierre Olivier de Sardan, the only anthropologist who describes an arkilla kunta, states that most names of the patterns are unknown to the weavers and to the old people in general or of unknown origin altogether (1969:108-109). As we did not stay long, our field data are poor.

When in Mali later, between 1978 and 1982, I put together some important documentation on the arkilla kerka - blankets that were still in use then. So all the terminology given here is in Fulfulde (the language of the Peul).

It is one thing to know the names of the patterns and it is another one to know how they are made. It is easier to understand these huge marriage blankets arkilla when looking at them through the eyes of a weaver – technically.

Looking back again at the first three photos above, Fig.1 shows the motif lewruwal (moon) with two stars kode to the left and right. Usually, there are four, six or eight stars. ‘Moon’ and ‘star’ are tunne, woven in the technique of tapestry. The two diamond shaped motifs above and below the lewruwal are called jowal meaning ‘rhomb’. As the two jowal are woven in a special kind of a weft float (in French “broché”) and not in tapestry they have another technical name, namely cukke. So, besides plain weave (cannyudi) we have three tunne (1 moon and 2 stars in tapestry) and two cukke. That is how the weaver thinks. After all, for him the weaving of an arkilla is a mnemotechnical challenge. “What I weave”, a weaver would say, “I have found with my fathers.”

That means, he repeats what he has learned. He doesn’t care whether a pattern is called ‘ears of the second wife’, ‘teeth of an old woman’ (black-white-black in reference to gaps between the teeth), ‘hoofs of a gazelle’, ‘upturned calabash’ or ‘tracks of a rabbit in the sand’. He is the artisan who masters the techniques and he rather thinks in terms of quality of the yarn, the colours, or his deal with the bride’s father and the bride’s mother who cooks his meals.

He will equally think about his salary: the more tunne he weaves and the finer and more eye catching they are the more additional income (cattle, millet, cash, kola nuts – a mix of all of them) he generates.

Arkilla kunta weavers were real master weavers. It would be wrong to think that their weaving was just repetitive. Now, in writing this contribution, I went through all the data I have on kunta blankets – 16 altogether – discovering, that no two kunta are exactly alike. There is a great range in designing the over all aspect of the blanket just in varying distances between patterns, the size of motifs and the combination of cukke, cubbe and tunne.

Fig.2 in fact is another jowal (rhomb), however of another kind than the two small jowal on fig.1. The weaver will call it cubbe, as the pattern is woven in weft float (in French “lancé”). This very cubbe here was probably in the middle of a kunta although, as it is rather small I am not quite sure.

The twisted fringes in yellow and red are a decoration. However, they serve as an important help to knot the newly woven strips together so that at last – after four or more weeks of weaving – the whole surface or the kunta can be admired before the sewing together of the strips starts. The very same fringes are on the kerka-blankets, too.

Fig.3 is another tunne extremely finely woven. It is called fedde (fed = finger), as all the weft threads forming the tapestry pattern have to be pushed through the warp threads with the fingers during the weaving process. On most kunta-blankets there are four weft stripes with the fedde-motif embracing several kode (stars) and a cubbe (woven in weft float pattern).

Throughout the old world – Middle East, North Africa, Europe – tapestry weaving was done by women on an upright loom. Not so in Mali: the Fulani of the Upper Delta of the Niger are an exception to the rule. Tapestry weaving is done on their horizontal treadle loom with one pair of heddles, and the weavers are men. That makes tapestry weaving interesting. It even seems that tapestry is a cultural marker, that wherever we find tapestry on West African textiles, there are Peul/Fulani around or the weaver has learned his profession with a maabo-weaver (singular of maabuube).

So, let us have a second look at that lewrual-moon-fig.1: it is a rather rectangular moon. In this case, all together, there are seven concentric layers or colours: light indigo blue, white, red, yellow, light indigo blue again, red, and white/blue. We can compare this kunta lewruwal with a kerka lewruwal (fig.6) having equally seven layers. This is just one way of talking about quality. The more layers, the more work involved, the better quality it is.

The striking difference, however, are the ‘antennas’ sticking out at the kunta lewruwal-fig.1. repeated below On kerka-blankets there are never such antennas. On arkilla kunta sometimes … or even often - but it depends where. These ‘antennas’ might exactly be one of these details a master weaver plays with and which might be in relation with the arranged ‘salary’.

On fig.1 we have 11 ‘antennas’ – quite a lot. Of the three kunta in Basel only one has such antennas, namely 10, on red ground. The other two have such ‘antennas’ on the smaller lewruwal which are towards both ends of the blanket on yellow ground, having 5 or 7 ‘antennas’ only.

That logic seems to be systematic: Either a kunta has its ‘antennas’ on red ground or they are on yellow ground. Not both together. Or there are no ‘antennas’ at all. And as the lewruwal on red ground is more to the centre of the blanket it is bigger and more important. Of the 14 kunta in my file where such details are visible only 3 have ‘antennas’ on red ground.

Maybe this is just a European approach: counting and quantifying for lack of other data. However, including other factors I have the feeling that the lewruwal motif on red ground with ‘antennas’ means ‘old’ … whatever that means. I know photos taken in Ghana in 2013 showing kaasa-blankets (the ordinary 6 striped woollen blankets of the Peul in Mali) in a ritual. As the patterns of the kaasa-blankets have changed considerably during the 20^th century, it can be said that this very kaasa was woven well before 1930. That means the woollen blanket was kept in the tropics safely for over 80 years. The same is very possible for this piece of arkilla kunta-weaving.

Only one arkilla kunta-blanket is dated and has full information: the one at the National Museum of Mali, Bamako (inv.no. 2002.13.1) woven 1968. It is maybe the last one ever made.

So far, I have found no arkilla blanket in a museum collection dated before 1900. We have, however, one archaeological cotton cloth from the Dogon, dated to the 17th/18th century containing over 20 patterns strangely resembling the lewruwal motif of an arkilla kunta (fig.8 above - National Museum of Mali, Bamako 92-05-390 = H 71-2). Isn’t this an amazing cloth! It was found in a cave where all the human remains were female. So this is a female cotton shawl or wrapper. Originally, it had at least five strips.

Among the important corpus of published Tellem textiles (500 pieces, see Rita Bolland 1991) dating from the 11^th to the 15^th century, patterns with supplementary weft threads (“lancé” in French, cubbe in Fulfulde) go back to the 11^th century. In rare cases there are even tapestry patterns in wool. Are all of them imported textiles from North Africa? Probably yes – although there might be just one exception (C 19-1, see Bolland 1993: 174, fig.32). In any case, fig.8 is the oldest tapestry pattern known for sure and woven on a West African treadle loom with one pair of heddles.

We have come a long way from arkilla kunta and Wogo to arkilla kerka and Peul to tapestry weaving in cotton found in a cave in Dogon country and going back 200 to 300 years. In order to end this story, I would like to invite the reader to come with me to Ghana. There, the most sacred and the most venerated cloths are called nsaa. Sacred drums or horns with human jaws sewn on them may be wrapped in nsaa. Dead people before burying are put on their bed, the room being decorated with nsaa. The first textile layer in a palanquin in which Akan chiefs are carried around is often nsaa. Although several authors had mentioned the importance of these nsaa- (or nsa-) blankets (Rattray, Ross, others) no researcher really followed up that topic. Except, maybe textile researcher Brigitte Menzel. “They are not only attributed protective but also healing properties”, she writes, and she carries on, “my aged Asante informants were unanimous of the opinion that this woollen textile, which they all called nsaa, is the highest ranking of all traditional textiles…“ And “in the olden days“ – 40 head loads each containing 2000 cola nuts or five healthy male slaves were paid up north in Salaga or Dagomba for such a nsaa. Although the word nsaa embraces all kind of woollen blankets from Mali Menzel stresses that nsaa means mainly: arkilla kunta … made 1000 km to the north, south of Gao (Menzel 1990:83).

We all know that the Golden Stool is the symbol and “national soul” of the Asante, as Rattray puts it. When the Golden Stool “was borne”, he tells us, it was “sheltered from the sun by the great umbrella, made of material called in Ashanti nsa (camel’s hair and wool). This umbrella was known throughout Ashanti as Katamanso (the covering of the nation).” (Rattray 1927:130) [Please note: “camel’s hair” is wrong. It is always sheep wool however of mediocre quality. But sheep it is.]

Click on the photos to enlarge.

Captions

Fig.1 see text

Fig.2 see text

Fig.3 see text

Fig.4 Arkilla kunta-blankets in the exhibition of the National Museum of Niger, Niamey in the late 1960s. One kunta is hanging over the bed, two others are spread on the bed and on the floor (Photo by Pablo Toucet, Photocard, coll. B.G.)

Fig.5 Arkilla kerka hanging in front of the bed. The centre of the blanket with the lewruwal motif is to the right. The white tapestry patterns on red ground to the left are called gite ngaari: ‘eyes of the bull’ (N’Gouma 1982, photo B.G.).

Fig.6 Central motif lewruwal (moon) of an arkilla kerka, Mali (1980, photo B.G.)

Fig.7 A = fig.1 Duncan Clarke’s lewruwal on red ground

Fig.7 B Lewruwal on red ground with 10 ‘antennas’, 4 stars. Museum der Kulturen Basel: III 20456

Fig.7 C Lewruwal on red ground with 11 ‘antennas’, 4 stars. National Museum of Niger, Niamey.

Fig.7 D Museum der Kulturen Basel: III 20462 – lewruwal on red ground without ‘antennas’. Eight stars. Below: lewruwal on yellow ground with 6 ‘antennas’.

Fig.7 E Museum der Kulturen Basel: III 20455 – lewruwal with 7 ‘antennas’ on yellow ground.

Fig.8 Single motif in tapestry on an archaeological cloth, Dogon 17./18.Jh. (National Museum of Mali, Bamako : MNM 92-05-390=H 71-2-II)

Fig.9 kunta in Ghana: nsaa (Photo: Werner Forman 1964, published by Duncan Clarke. 1997: The Art of African Textiles, p. 68, 69)

Fig.10 A maabo weaver at work. To weave a weft pattern with supplementary weft (cubbe) he uses a wooden board to count the warp threads. Below to the right we see two stars tunne on white ground (N’Gouma, 1982, photo B.G.).

Quoted literature

Bolland, Rita. 1991. Tellem Textiles. Archaeological Finds from Burial Caves in Mali’s Bandiagara Cliff. Royal Tropical Institute, Amsterdam; Rijksmuseum voor Volkenkunde; Leiden; Institut des Sciences Humaines, Bamako; Musée National du Mali, Bamako. With contributions by Rogier M.A. Bedaux and Renée Boser-Sarivaxévanis

Menzel, Brigitte. 1990. Textiles in Trade in West Africa. Presented at the Textile Society of America, Biennial Symposium. September 14-16, 1990. Washington, D.C. Proceedings of the Textile Society of America, pp. 83-93

Olivier de Sardan, Jean-Pierre. 1969. Système des relations économiques et sociales chez les Wogo (Niger).Université de Paris. Mémoires de l’Institut d’Ethnologie III. Institut d’Ethnologie Musée de l’Homme.

Rattray, Capt. R. S. 1927: Religion and Art in Ashanti. Oxford, The Clarendon Press.

Ross, Doran H. 1998. Wrapped in Pride. Ghanian Kente and African American Identity. UCLA Fowler Museum of Cultural History.

Click on the photos to enlarge.

Coming this week: Arkilla kunta / arkilla kerka: patterns – by Bernhard Gardi

This week I am excited to be bringing you the first ever guest contribution to our blog, an extended consideration of the patterns in wool blankets from Mali written by eminent Swiss scholar Bernhard Gardi. He was for many years the curator of the Africa Department of the Museum der Kulturen, Basel and is the author of numerous book and articles on  African textile related topics. Among these are the catalogues of two of the important exhibitions that he curated  Woven Beauty: The Art of African Textiles (2009, Christoph Merian Verlag) and Le Boubou: C’est Chic (2000, Christoph Merian Verlag).

In his post Bernhard will look at woven patterns in a rare type of wool blanket called an arkilla kunta, shown below (MKB_III_20462 –click photo to enlarge)

and he will explore the similarities and differences with the much more common arkilla kerka (below, click to enlarge, photo by Bernhard Gardi, 1979.)

A fine Arkilla Kerka from Mali

l

After a special request by Bernhard Gardi for some wool cloths on the blog after he complained “always kente kente kente”, today we look at a notably fine wool and cotton wedding hanging from Mali, of the type called arkilla kerka.

In his catalogue Woven Beauty: The Art of West African Textiles (Basel, 2009 – incidentally the best book published on African Textiles) Gardi notes:

“The kerka is the ‘mother of all arkilla.’ They are no longer produced today. A kerka consists of six patterned strips, occasionally a seventh strip, striped black and white and called sigaretti (from the French ‘cigarette’), is added and used for hanging up the blanket. The white sections and the warp are made of cotton, the rest is wool. Black is never applied, only dark indigo blue. A kerka blanket of the highest quality requires between 25,000 and 30,000 metres of hand-spun yarn.”

This example was collected in Bamako by an English family in the 1950s and is now in the Musée di Quai Branly, Paris. It is an old well used piece with exceptionally fine weaving. Gardi commented “There are very nice noppi nawliraaBe motives, 'ears of the co-wive'. Looks very much as woven from a weaver who was equally weaving arkilla jenngo blankets. That means, that kerka does not come from the Guimballa area (East of Lake Debo) but from Niafunke or further North-West.”

Click on the photos to enlarge.

African Textiles in Close Up #2: a Sierra Leone robe.

In my last post I looked at two rare embroidered robes from Liberia or Sierra Leone in the British Museum collection. Today I turn to a third robe, also from Sierra Leone, that is in an even more unusual style. The vast majority of robes from the region were tailored from either plain or simple warp stripe patterned cotton cloths in shades of white, brown and indigo. They can be distinguished from other West African robes by their distinctive front pockets and their overall design – according to Venice Lamb there were two styles: a simple sleeveless tunic called in Sierra Leone  kusaibi, and a more complex sleeved robe called a duriki ba. A very few  surviving examples (two of which we looked at) were embroidered and fall into a group that Bernhard Gardi in his important book Le Boubou – c’est chic calls Manding robes.

However there remains an even smaller number of robes from the same region tailored from the elaborately patterned kpoikpoi cloths for which Sierra Leone weavers were so notable. This robe, part of the British Museum’s Beving collection, was collected at least by 1913 and most probably dates from the second half of the nineteenth century. It was apparently (Lamb 1984:136) collected in Bonthe region, Sierra Leone. It is among less than ten robes I am aware of that have been made from kpoikpoi cloths.

The structure is simple – an existing small cloth is simply folded in half width ways, a neck hole made, and a patch pocket added at the front. This results in a robe made up of seven strips of cloth, each around 18cm in width, and a total size of  124 cm width by 96cm length. The pocket is made from a square piece of cloth, one and a half strip widths in size, with the small corner fold typical of robes from the region. The lower half of each side is sewn up with the rest left open to create armholes. However in marked contrast to this simple tailoring the cloth used to create this robe is exceptionally elaborate. On the pocket detail above we can see blocks of thicker indigo dyed thread inserted as supplementary weft floats half way across the strip in sufficient quantities to distort the flow of the ground weft into a curved pattern. This is just one of the many variations used in a way that seems to me to suggest a deliberate echo of the embroidery patterns normally found on the pocket and chest areas of prestige robes. In particular when we look at the back of the robe we see that the small block of checkerboard pattern largely concealed by the pocket is repeated.

Elsewhere we see the use of tapestry weave techniques to create distinctive triangular patterns that are a feature of the more complex styles of Sierra Leone weaving.

The detail above from the lower left of the back also shows some of the remarkable variation in weft stripe placement and simple variants of the weft float patterning the weaver has utilised. To me though the most interesting feature, and the strongest evidence that this cloth was woven to order with its use as a robe planned is the contrast in colour between the front and the back. On the front of the robe white is the dominant colour, but on the back there is a preponderance of light blue indigo dyed thread. The pattern diverges at the fold in the centre of the cloth, yet the strips used are continuous, strongly suggesting that it was woven with this use in mind. We can see this in the photograph below where the two sides are shown together (please excuse the inept photoshop.)

Can anything useful be said about the ethnic origin of this robe ? Venice Lamb published it (1984:137) with a caption ascribing it to the Vai ethnic group, but in the text is much less certain noting only “It is possible that this garment is an example of Vai inventiveness in weaving.”  However I am not convinced that there is sufficient evidence to distinguish Vai weaving from that of the larger group of Mende weavers.  Easmon (1924:22) noted that kpoikpoi cloths were “essentially a Mende cloth, and is also made by the Gallinas [Vai]”. Very few of the small number of early Sierra Leone cloths in museum collections have any detailed collection data and where they do it is not generally sufficient to confirm that the piece was woven in the same place as it was collected. Prestige cloths and prestige robes were important trade goods and may well have been traded a considerable distance from their place of origin. Bonthe, where this cloth was collected, was mainly inhabited by Sherbro people . We might also note that the whole process of assigning a particular cloth style to a particular ethnic group is extremely problematic.

All photographs above by Duncan Clarke. Click on the photos to enlarge.

African Textiles in Close-up: two robes in the British Museum

At the end of last month I had the opportunity to spend a morning at the new textile store for the British Museum in Blyth House, West London. With the patient assistance of curator Julie Hudson and Textile Centre manager Helen Wolfe I was able to look closely at a number of robes and cloths from the British Museum’s vast collection that had attracted my attention either in publications or via their online database. (The entire collection is now online here and provides a hugely important resource for those interested in studying African textiles.)

Although I am lucky enough to spend most of my days surrounded by stacks of old African cloths there is in my view always more to be learned and more details to be grasped by paying close attention to threads, weaves, patterns, constructions, layouts, textures, etcetera. Some of these can be learned from photographs but actually seeing and handling cloths reveals more again. [Any of the cloths, or indeed other items, in the collection of the British Museum can be viewed by appointment and curators are generally happy to help.]

Over the next few weeks I will be writing a short series of posts based on this visit, beginning today with a look at two  robes from Sierra Leone or, more probably, Liberia. One of these has been frequently published and the other is very obscure.

My attention was drawn to this simple and rather stained looking robe (British Museum number AF,WA.10) by the early accession number and the brief description on the image page “Embroidered garment made of cloth (grass).” In old descriptions and travellers accounts of West African textiles cloths described as woven of “grass” are generally actually woven from raffia (the dried inner leaves of a type of palm tree. While cloths woven from raffia on the upright single heddle loom in West Africa are reasonably widespread, strip woven raffia cloths are extremely rare (primarily, I think, because raffia thread could only provide short lengths that had to be tied together.)

There is no detailed record surviving of the origin or accession date of the robe but it appears to be part of the collection left to the museum by Henry Christy (1810-1865.) Disappointingly the first thing that became apparent when we handled and examined this robe was that despite the rather harsh scratchy texture, it was in fact woven from hand spun cotton not raffia. Raffia is not spun so on close examination the fibres are flat rather than round as here. [The main description on the BM page has now been corrected following our examination.]

So what accounts for the pale beige colour and harsh texture ? The robe appears to have been soaked in some kind of plant based dye (most probably after robe was tailored but before it was embroidered.)  Although I am not aware of any documentation of this practice in the southern part of Sierra Leone or Liberia, it is still in use in the making of “war shirts” called hu ronko among the Limba people in the north of the country and in neighbouring Guinea. In their book Blues et ocres de Guinée Anne-Chantal Gravellini and Annie Ringuedé describe the use of the bark of the tree terminalia ivorensis along with the kola nut Cola nitida to dye cloths and tunics a variety of ochre shades.

Also notable was the regular placement of thicker wefts at intervals along each strip of cloth. The embroidery, although limited in area and elaboration, is quite complex in design. Below is a detail from the back. Imported wool, cotton and silk is used.

The pocket has the folded over corner and oblique placing that were found on many robes from the Guinea Coast region and help to distinguish them from the better known robes of Mali and Nigeria.

The sides of the robe are completely open but it can be seen that they were once sewn up at the lower part. This sleeveless structure would put this robe within the group that the Lambs (Sierra Leone Weaving by Venice and Alastair Lamb)  report are called kusaibi, while the second robe that we looked at, with a more complex tailored design is called in Manding duriki ba.

Below we show front and back views from the BM site (BM accession number Af1934,0307.218). This robe was part of the Beving collection accessioned in 1934 but can be assumed to be from the nineteenth century.

Published by the Lambs with photographs that make it look very yellow this remarkable robe is in fact a very similar pale ochre colour to the first one we looked at, although it is softer to handle.

In this robe also there is a decorative effect in the cloth used achieved by adding thicker threads in the weft, albeit here in blocks rather than the single threads used in Af.WA.10. This is a quite unusual technique in West African strip weaving and may in itself point towards an origin in the same region. In his book Le Boubou –c’est chic (Basel, 2000), Bernhard Gardi attributes the very small number of robes of this type to Liberia. A blue ground robe with rather similar embroidery to this was collected in 1932 from a Mano chief in a village called Blaui in northern Liberia.

Click on the photos to enlarge. In the post above all detail photos are by Duncan Clarke and the full views are from the linked pages on the British Museum site.

“Costume for a King”–An important Sierra Leone or Liberian robe at the Pitt-Rivers Museum.

A couple of years back a research project at the Pitt-Rivers Museum, Oxford, revealed that a previously undocumented West African robe in their collection was in fact among the founding objects assembled by General Pitt-Rivers in the 1870s, and more remarkably, that the same robe appeared in an article in the Illustrated London News on 28 November 1846.

The robe was among a group of objects collected by a Captain Henry Denham during a naval survey of the West African coast in 1845-6. It belongs among the extremely small number of chiefs’ robes of the type that Bernhard Gardi in his book Le Boubou – C’est Chic (Basel, 2000) ‘boubou Manding’ from Sierra Leone and Liberia.

For full details of this robe in the Pitt-Rivers collection click here and for a notice about the research here.

“Mbuna”: collecting textiles in Mali– excerpt from a memoir by Rachel Hoffman.

The chief of Mbuna with an arkilla munnga hanging, 1980s, Mali. Photo ©Rachel Hoffman.

Rachel Hoffman, formerly of the Fowler Museum, UCLA, is currently working on a memoir “Toubab” of her years in Mali, where she played a key role in assembling a major collection of textiles for the National Museum in Bamako. In a chapter “Mbuna” published recently in the online magazine 1966Journal.org Rachel describes a tense incident on the edge of the Sahara – from page 78 here.

“Weaver’s warp” © Rachel Hoffman.

Rachel kindly provided the following context and biography as well as the photos that accompany this post.

“During the late 1980s, the UCLA Fowler Museum joined in collaboration with the Musee National du Mali to collect weavings and associated technologies, oral histories, photographs and film of and from Mali's Inland Delta region of the Niger River. Because of the severe droughts that had plagued the Sahel for 25 years prior, the enormously rich heritage of textile production was undergoing, at best, a vast transformation, and, at worst, a disappearance. The museum directors thought that preserving examples of as many genres as possible was important.

The Fowler Museum paid for the project and - my great fortune - sent me as liaison. Mali's National Museum provided an anthropologist, a linguist, a photographer/videographer, expertise and a pickup truck. Our team collected two or more of everything we found. The better collection, along with unique pieces, remained in Bamako in the National Museum. The second collection went to the Fowler.”

“Indigo strips from Tereli, Mali” ©Rachel Hoffman

The project's initial season went well enough that another was planned. And another. Ultimately, our team of four people worked five field seasons together in five separate parts of Mali. We collected thousands of textiles, loom pieces, photos, histories, proverbs, and more for both museums as an archive for future generations. Our second season was among Dogon people. Some of the photos posted here are from that trip. Others come from my own doctoral research following the textiles project, where I spent an additional eight months with Dogon sculptors, weavers, and smiths.”

“Woman indigo dyer, Mali” ©Rachel Hoffman.

To my knowledge this project represents the only systematic survey of  textile traditions in a West African country to date and as such it will be of great importance to future scholars. Some of the textiles collected can be seen in Bernhard Gardi’s catalogue Textiles du Mali (National Museum of Mali, Bamako, 2003) and we might hope to see the Fowler Museum’s portion at least online before too long.

 

Cloth of the month: A unique wool cloth from Mali.

On my recent visit to Ghana I was pleased to find this remarkable wool and cotton cloth from Mali, collected by one of my contacts from the chief of a small village in the north west of the country.  Itinerant traders from Mali have been selling cloths in Ghana and neighbouring countries for at least 500 years so the presence of this cloth was not too surprising in itself. Wool blankets from Mali have become an accepted part of the court regalia of many Ghanaian chiefs, used for a variety of purposes including lining ceremonial hammocks and palanquins, covering royal drums, and just as a backdrop for displays of the royal treasures.

The surprising, and to my knowledge, unique aspect of this particular cloth is the design and layout. Made up of 14 strips of approximately 3 1/2 inch width (9cm), squares of indigo blue dyed wool alternate with squares of white handspun cotton, serving as a backdrop for quite complex patterning in natural brown dyed wool executed using a tapestry weaving technique, these strips would normally have been laid out in a much more regular fashion to form a far larger cloth, called an arkilla jenngo.

Arkilla jenngo were, according to Bernhard Gardi, woven in the area west of Tomboutou near Lake Faguibine by Songhay-speaking weavers, primarily working for Tuareg patrons. Interestingly he notes that according to his informants this type of wool blanket was not traded to Ghana. They were used as highly prestigious tent hangings for high status weddings.

Photo above by Harald Widmer, January 1952. The names given to the motifs “reflect the significance that weddings and bride wealth have in social life: we have for example,  ‘cushion’,  ‘bedposts’, ‘money’, ‘board game’, ‘leather fringe’, or ‘calabash’. (Gardi 2009.)

Turning back to our cloth we can now see that the standard strips woven for an arkilla jenngo have been put together in a different and unexpected way, with the regular layout of the traditional form replaced by an apparently random placement creating an “offbeat” staggered overlap and contrast between each strip and an overall dynamic oscillation across the cloth.  How can we tell that this was a deliberate effect and not just the random result of a partial or damaged arkilla being re-purposed by someone unfamiliar with the tradition ?

If we look at the ends of the cloth we see that it was carefully finished with a braided wool cord making up a series of tassels. This was the normal technique in Mali for finishing a smaller type of wool blanket known as a kaasa, and clearly shows that this cloth was assembled by someone working within that tradition. So we can only speculate on why this novel layout was adopted and it remains a mystery.

A Rare Chief’s Robe at the Indianapolis Museum of Art

For serious lovers of African textiles the most exciting exhibit of the current show Majestic African Textiles at the Indianapolis Museum of Art (see earlier post here for details)  will be the chance to view this extremely rare robe. At the time of his important survey catalogue Le Boubou C’est Chic (Editions Christoph Merian/Museum der Kulturen, Basel 2000) Bernhard Gardi noted that only 23 robes in this style were known in collections worldwide. Gardi observes that very little is known about the production and use of this style of robe, which he calls “boubous Manding.” He suggests they were made somewhere in the interior of Liberia and/or Sierra Leone where people of Mande origin descended from migrants from Mali were in the majority.

This superb example in Indianapolis can now be added to the small corpus. It’s accession details are:  “Mende people; Sierra Leone, Liberia. royal robe, early 1900s, cotton, wool. Gift of Mr. and Mrs. Harrison Eiteljorg. 1989.808”  I would suggest it is in fact more likely to be made in the nineteenth century.

All photos are copyright Indianapolis Museum of Art, with thanks to Niloo Paydar. Click on the photos to enlarge.

Chief Kai Lunda of Luawa Country, Upper Mendi, circa 1893. He was chief of an area on the border between Sierra Leone, Liberia, and Guinea. Vintage postcard, authors collection.