Braided Headbands

Home | Program | Participation |

Keynote general

Braided headdbands

Loop braiding

Kumihimo jewelry

Textile composites

Distribution of mummies found at the Neocropolis site 'A' (after Tello and Mejia 1979) from Anne Paul 'Paracas Ritual Attire' Fig 86


Cross-knit looping



Family A Family B

Family C Family F

Braided Headbands

Rodrick Owen

When referring to Pre-Hispanic Peruvian Cultures certain familiar names come to mind; such as Norte Chico, Chavin, Paracas, Moche, Nazca Tiwanaku (Tiahuanaco) Wari and Chimu. Each of these cultures made their own individual textile mark in Peruvian history which is seen in the weaving and embroidery that has survived.
The braided structures that will be shown in my presentation are from many museums most of them identified as being from Paracas burial sites. However we should look at the preceding cultures as their textiles will have influenced Paracas textiles.

Earlier Cultures

Most civilizations have gone through a process of using vegetable fibres to make simple looped, twined and netted structures. Tapestry and network fragments that have been carbon-dated to 8600-5780BC were found in the Guitarrero caves in the northern highlands of Peru. It is suggested that this ability to make twined structures would have also included the skill to make rope by twisting strands together, and a natural consequence of this would be the manipulation of three and four strands together to make the first braids.

Up until the 1960's the Chavin culture 2500-300BC was identified as the earliest society in the Peruvian timeline, but work by three archaeological researchers Solis, Haas and Creamer has now identified an earlier culture known as Norte Chico. (See bibliography)

It is believed that Norte Chico 3000-1800BC society was comprised of some 30 major population centres situated in the northern coastal area of Peru. The people lived in this pre-ceramic culture with a sophisticated form of government. They built large platform mounds and sunken Plazas. There is evidence that they cultivated plants for both food and textiles fibres. The only textiles found were nets and quipo.

Further fabrics were discovered at a site in the Omas valley on the south coast. The finds consisted of over 300 funerary fabrics, dated to approximately 2000BC. These included plain cloth with feathers attached and bags made from simple looping or knotting. In Peru, on the north coast, at Huaca Prieta in the Chimu Valley where most of the oldest textiles have been found, the presence of early hunter-gathers as long as 20,000 to 23,000 years ago have been attested to by Richard S. MacNeish a Canadian archaeologist.

The Huaca Prieta site yielded some 2,700 textiles excavated by Junius Bird from a refuse midden of a village site. These fragments were cast-offs, made from cotton and bast fibres. When the structures were analysed it was found that 78 per cent of the textiles were twined, 10 per cent were looped and the remaining 12 per cent were made by various nettings. The twined fabrics were patterned, showing snakes, birds, crabs and other designs. These fabrics were dated at 2125 - 950BC.

It is however the Paracas burial sites that revealed a vast array of textile structures, which included those that are generally referred to as turban braids.

PARACAS finds at the Paracas Peninsula 600 - 175BC

Evacuations on the dry sandy Paracas Peninsula revealed some 450 mummy bundles from three sites, known as Cavernas, Arena Blanca and Necropolis. Some of these mummies were as much as 1.2m (4ft) high by 1.7m (5ft 6in) in diameter. The outer wrapping of the mummy consisted of plain cotton cloth about 4m (l3ft) wide and up to 20m (65ft) long. This fabric helped to protect the decorated fabrics that lay inside. Included within each mummy bundle were various artefacts, such as spinning spindles, headbands, fans, slings and other textiles.

Supplying the Weavers: Cotton production

It is worth considering the source of cotton supplies that would be needed to make all the fabrics included in one mummy bundle. As there is virtually no rainfall in this region the rivers were the sole source of water and the people learnt to construct canals to divert and irrigate their fields. This was immensely important for survival not only to grow editable crops, but also to cultivate cotton for textiles.

ÒFor example at the Paracas Neocropolis site the simplest cloths are woven more tightly than those of Huaca Prieta where it was estimated 200 people needed at least 306 km. of yarn for clothing and about 150 km. for fishing nets, ropes and other items of everyday use.

If we then take one mummy bundle from the Neocropolis and assume an average of 8 warp threads and 8 weft threads per square centimetre of plain weave fabric.

That means one square metre would contain 8oo metres of warp and weft threads. For the production of a normal cloth of 20 x 4 metres, of the kind used for the outer wrapping of a mummy bundle, the weavers needed 156.8 km. of cotton yarn, i.e. more than half the population of Huaca Prieta needed for all their clothing, including 'the last shirt'. Such consumption could only be guaranteed by extensive high-yielding cotton fields, and they in turn required efficient irrigation systems.Ó

From Ferdinand Anton, 'Ancient Peruvian Textiles' pages 68&69.

Weaving and Embroidery: How long did it take?

Anne Paul writes about the length of time it would take to complete the garments found in the two mummy bundles numbers 378 and 310 she was examining.

ÒI estimated the number of square meters of plain weave cloth in the garments and the number of square centimetres of embroidery found in each bundle, and determined the approximate time required to make the woven garments in two high status bundles. 11,293 hours for bundle 378 and 29,493 hours for bundle 310. According to these rough calculations, the time required for the production of textiles in bundles 378 and 310 was approximately 40,786 hours.

When this figure is increased by 15 percent in order to compensate for errors, the total number of hours becomes 46,904. If only 185 women, or 10 percent of the projected female population, worked 20 hours a week for one year, a total of 192,400 hours would have been spent in the production of cloth and elaborate garments.Ó

From Anne Paul, 'Paracas Ritual Attire,' pages 32&33

Headbands v Turbans

'Turban Braid' is a name that is generally used to describe a narrow textile that is wrapped around the head of a mummy at the time of burial. It is suggested that the use of these two words is misleading as it presupposes that all head wrappings are in fact braided and this is not the case. Headbands can also be made by weaving, stitched using cross-knit looping and constructed with netting techniques

Based on the iconography of woven and embroidered images it has been discerned that there are two garments. The turban was a long rectangular fabric that hung down the back to mid-calf level. The headband was wrapped around the head to hold the turban in place.
There are lengths of obliquely interlaced headbands that measure up to 6 metres long which have been found wrapped around the heads of mummies, there are also very small ones measuring no more than 200mm that are not associated with a head wrapping at all, but all are grouped under the title of Turban Braids. It is not known how the word Turban came to be associated with headbands.

The method to make interlaced braids sited by Speiser and Turner is to work downwards toward the weaver from a fixed point with loose ends. This Mary Frame turns on its head by suggesting that they were made working away from the weaver from a fixed point.

To work upwards means that there would have to be a tensioning device and a way to control the warp. Was the solution to use heddles and treat these structures as weaving if this is the case then can these woven structures still be called braids or should they now be called oblique interlaced weaving as would be found in triaxial weaving?

Taking into consideration that the possible length of the warp would be around 9 metres to make 6 metre length using up to 600 ends of yarn, which method we would choose work with and how would we control the tension?


This presentation will be asking more questions than it will be giving answers. We will do so looking at the differences of the braids between and within each family. The images shown below represent the largest number of like braids
It has already been demonstrated that a woven version of the twined braids of Family C can be made on the takadai and it is more than likely that all of these Peruvian braids can be reproduced accurately on the karakumidai.


Anton Ferdinand, Ancient Peruvian Textiles, Thames & Hudson, 1987.

Bird Junius & Bellinger Louisa, Paracas Fabrics and Nazca Needlework, The Textile Museum, Washington DC, 1954 Library of Congress 54-9410.

D'Harcourt, Raoul, Textiles of Ancient Peru and their Techniques, University of Washington, USA 1962.

Paul Anne, (ed) Paracas Art and Architecture, Object and Context in South Coast Peru. University of Iowa Press, 1991. isbn 0-87745-327-6. Chapter 4 Paracas Neocropolis Headbands as System Templates

Paul Anne, Paracas Ritual Attire, Symbols of Authority in Ancient Peru, University of Oklahoma Press, 1990. isbn 0-8061-2230-7.

Sawyer Alan R, Early Nazca Needlework, Laurence King, 1997. isbn 1-85699-088-1

Sorber Frieda with others, Rediscovery of Pre-Columbian Textiles, Fotogravuare Datascan, Brugge, 1994. isbn 90-5276-008-X

Tsunoyama Yukihiro, Textiles of the Andes Catalogue of the Amano Museum Lima, Heian/Dohosha, isbn 0-89346-017-6

Norte Chico references

Ruth Shady Solis Museo de Arqueolog’a, Centro Cultural de la Universidad Nacional Mayor de San Marcos, Avenida Nicol‡s de PiŽrola 1222, Lima 1, Peru.
Jonathan Haas Department of Anthropology, The Field Museum, 1400 South Lakeshore Drive, Chicago, IL 60187, USA.
Winifred Creamer
Department of Anthropology, Northern Illinois University, DeKalb, IL 60115, USA