Illuminated signs: style and meaning in the beadwork of the Xhosa- and Zulu-speaking peoples.

The beadwork of the Xhosa- and Zulu-speaking peoples holds painterly appeals. Color and composition combine--usually without the obviousness of figuration. No big genitals, no halos; neither Madonnas nor elephants. Sculpt an elephant, it remains attached to its signified: that leviathan. But a blue rectangle or red brushstroke has no lexical translation. It takes a braver sensitivity to admire what escapes us than what we recognize, and a subtler one to glean meanings.

Today, in "the West"--a label that globalization is dissolving--the rhythm of pattern remains less art than craft, still of a lower class. "Decorative." Elsewhere, though, such repetition sounds the music of the spheres. It echoes the spiritual realm like a meditational chant, reflects it like a mandala. It is religious art.

Just as Modern artists appropriated "primitive" sculpture to break the mold of Western art, so too they pillaged non-Western performance arts and two-dimensional traditions, evacuating their sacredness. Though a string of Western painters, from Paul Klee to Sean Scully, say, have drawn "inspiration" from non-Western graphic arts, their sources remain inadequately appreciated in their own right today. To appraise them anew, perhaps we must borrow the eyes of those who have borrowed from them; pause a moment to consider abstraction in the abstract.

In Western art, abstraction has been a tendency toward a purificatory and heroic emptiness of figuration, or of reference. As John Rajchman observes in Constructions, this leaning masks alternative readings of abstraction, such as that of philosopher Gilles Deleuze, who

   identifies an abstraction quite different
   from the self-purifying kind--that
   of those "abstract machines"
   that push art forms beyond and beside
   themselves, causing their very
   languages, as though possessed
   with the force of other things, to
   start stuttering "and ... and ... and ..."
   He connects this stuttering abstract
   "and" not with dying or heroic self-extinction
   but with a strange anorganic
   vitality able to see in "dead"
   moments other new ways of proceeding.
   And this sort of vitally, this
   sort of abstraction, he thinks, is
   something of which we may still be
   capable, something still with us and
   before us.

   (Rajchman 1998:59-60)

Perhaps such re-examination of abstraction by Western artists and theorists returns their cultural debt as a prism through which we can peer afresh. They glimpse a non-Modernist abstraction that is about addition and plenitude, not reduction. This is an abstraction that interrogates signifying languages, just as Wittgenstein used the visual trick of the "rabbit-duck" drawing (Fig. 1) to point to "the gap between perception and what a mind might make of it.... There was always a change lurking, a bit of the unknown surfacing, a duck in the rabbit.... Language could be shown to fail them all" (Nesbit 2000:272). An abstraction that pushes art forms beyond themselves, "as though possessed with the force of other things." The beadwork and murals of many South African peoples appear to be of this order.

[FIGURE 1 OMITTED]

This article sketches some of the parameters within which the patterns of beadwork among the Xhosa- and Zulu-speaking peoples, though apparently abstract, in fact conveyed meaning, provided "readings." Beadwork traditions were semiotic systems, structured through differences. They should be understood as inherently dynamic, like language itself and semiotic systems in general. It is therefore impossible within the scope of this essay to be comprehensive, to fully describe every system and track its changes over the last century or so. Rather, the outline given here sketches illustrative case studies in the hope that it will encourage further in-depth studies, particularly those based on intensive fieldwork.

The Significatory Functions of Beadwork

First, like flags or languages, beadwork traditions signaled a sense of belonging to a people, to a place, and to a chain of tradition. They were signs of identity that flagged ethnic and or regional roots--not that these identities were fixed and rigid nor that people were not uprooted often within their histories. By diagnosing different historical moments, however, we can illustrate how art styles proclaimed origins. Pictorial surveys of South African peoples, such as Barbara Tyrrell's (1968), undertaken from the late 1940s onward, or the photographs of Jean Morris (1959-1990s), show dramatic differences in dress structured according to people and place. The abstract elements of beadwork patterns play a key role in flagging difference--like the tartan kilts of Scottish clans.

Second, because tradition linked the living to their ancestors, beadwork was spiritual art. Absent the ancestor figures and masks that focused belief for other African peoples, this religious function of beadwork was vital. The cultures of Xhosa- and Zulu-speakers were essentially iconoclastic. The Xhosa peoples produced virtually no figurative sculpture. Among the Zulu, even as late as the 1920s, it was said that "any person making an image of any living thing is committing an impropriety" (Dube 1928:43). Visual art for these Nguni peoples was located mainly in abstract forms (1) and in beadwork. Recently, utilitarian objects used by Zulu-speakers have gained recognition, often under the umbrella of "abstract forms." They conform to the Western stereotype of African art as male-carved wooden sculpture, often polished up and darkened for a market that prefers these signifiers in its art from the "Dark Continent." Beadwork, full of color, proclaims another Africa, one of light, abstraction, and women's work. As will be suggested here, it is the primary sacred art form among the Xhosa, and is also of key importance among Zulu-speakers. It is therefore high time that beadwork is recognized as a primary medium both in Nguni art and within the broader context of African art. (2)

Third, beadwork ensembles allowed social identities to be read. Differences in dress within particular beadwork conventions mapped social typologies: a person's gender, age grade, and marital status, and sometimes his or her social rank and role and spiritual state (Fig. 2). Fourth, beadwork was, and still is, a vehicle of self-expression, reflecting the individual styles of both its creator and wearer. Like linguistic systems, it is open to individual inventions and borrowings that expand the language, and redundancies that contract it.

[FIGURE 2 OMITTED]

Fifth, many beadwork colors, patterns, and motifs conveyed symbolic references--an ability to indicate concepts that mirrors the function of language. Such symbolism even enabled the construction of complex narrative messages whose semantics were intelligible within a limited territory, functioning like a dialect. From at least the 1950s onward, some beadwork included literal messages spelled out in Western alphabetic signs. These render obvious the significatory function of beadwork, but the meanings of the written statements were often unclear, even within the territory of their creation. This ambiguity underscores Ferdinand de Saussure's axiom of semiotics that the relationship between the signifier and the signified is arbitrary.

Since these various functions combine, multiple simultaneous readings are possible. A plenitude of possible meanings arises, a kind of stuttering of the abstract signs so that they proclaim, "and, and, and." This is the opposite of the increasing standardization of the languages of Xhosa and Zulu, which occurred through missionary activity and later state controls.

The Xhosa Peoples and the Work of Joan Broster

To speak of either the "Xhosa" or "Zulu" is to elide complex identities. These two great language families share, however, common Nguni origins, which history has obscured. (3)

The Xhosa, also known as the South Nguni or Cape Nguni, are united by dialects of the Xhosa language and other cultural features, but they consist of several distinct peoples. Listed roughly from south to north on the southeast coast of South Africa (Fig. 3), the principal groups are the Xhosa per se, Thembu, Mfengu, Bomvana, Mpondo, Mpondomise, Xesibe, and Bhaca. (4) The distinct beadwork traditions of several of these groups are reflected clearly in a collection of synchronous costumes assembled in the 1950s and 1960s by Joan Broster, nee Clarke, whose grandfather pioneered a trading business in the Engcobo district of the Transkei in 1875. Four Clarke generations lived among the Thembu before Broster, as a young bride, moved to the village of Qebe in 1952 to run a family store (Broster 1967:4). There she studied Thembu traditions, developed a passion for beadwork, and collected and annotated local costumes that demonstrate how minutely beadwork mapped social identity within this Xhosa-speaking community. During the fourteen years she lived at Qebe, Broster used her contacts with the network of white traders to collect costumes from other Xhosa-speaking groups (Broster, personal communication, 1997), which exemplify the flag-like mapping of cultural identities through costume (Fig 4).

[FIGURE 4 OMITTED]

Equally useful are Broster's books. She recounts her experiences rather anecdotally in Red Blanket Valley (1967) and more analytically in The Tembu: Their Beadwork, Songs and Dances (1976), which views Thembu society through its rigid age grades, each with distinctive dress, songs, and dances. Though not scholarly, these volumes provide the most detailed study to date of the social significance of beadwork within a Xhosa-speaking society. In 1967 Broster spent several months among the Gcaleka, the senior chieftaincy of the Xhosa per se. (5) In 1968 and 1969, with noted ethnographic photographer Alice Mertens, she documented ceremonies among the various peoples of the former Transkei, published in African Elegance (1973). During the 1970s Broster, with Herbert Boum, embarked on a study of diviners among the Xhosa peoples, published in Amagqirha: Religion, Magic and Medicine in Transkei (1981). Her later books assist in comparison across different Xhosa peoples, as do similar volumes by others (Tyrell 1968; Morris & West 1976; Morris & Levitas 1984), but they lack the detail of her books on the Thembu.

"Red People" and "White People"

At Qebe, Broster lived among the amaQwati, a refugee Xesibe group that had settled among the Thembu in the early 1800s and adopted Thembu customs and dress. (6) The Qwati were staunch traditionalists who "steadfastly refused all offers of a school and a church and still adhere[d] to their old pagan life" (1967:6), signified through their wearing of red ochre, "beloved by the ancestral spirits" (1976:3). Broster noted, "Beautiful tribal dress, beadwork, songs and dances are the media employed in ancestral worship but in everyday life those who adhere to this faith are recognised by the red ochre or clay which they apply to body, blankets and clothing. This is the color of their faith and they are therefore called 'Red People'" (Broster 1973:2)--or amaQaba, as distinct from Christianized Xhosa-speakers.

Ochred clothing signified religious and cultural continuity. Broster also notes that ochre color "varies from the palest orange to the deepest red-brown. Each tribe [sic] has a particular colour preference and will use no other" (1973:2) (Figs. 5, 6). (7) We know from early accounts that the Xhosa peoples buried or burned the clothes and belongings of their dead, and that periods of mourning were signified by changes in appearance, practices that emphasize the spiritual significance of dress. Early burial practices probably account for the fact that very little Xhosa beadwork survives from the first half of the 1800s--most nineteenth-century beadwork in collections is made from beads that can be identified empirically as manufactured after the 1860s (Peter Francis, personal communication, 2001) (Fig. 7).

[FIGURES 5-7 OMITTED]

Red ochre and white beads inaugurated the social life of a Thembu child, which commenced with a two-day ceremony for the extended family, during which mother and child remained isolated indoors. Before the father went to the cattle kraal to sacrifice a goat in honor of his ancestors, he instructed the mother to qaba the child, to redden its body and face with ochre. The mother prepared herself by drawing a red dot on each cheek and a line from wrist to wrist, via her shoulder and jawline.

   As the father sacrifices the goat he
   calls out, "Mawetu! My ancestors!
   Today I sacrifice to you for the reddening
   of the child." The guests
   reply, "Camagu!" This may be loosely
   interpreted as; "we give thanks
   for blessings from the ancestors."
   The goat is then skinned and the
   right shoulder is roasted and taken
   to the child and its mother in the
   hut.... Early on Sunday morning the
   meat is divided between the men
   and women and cooked separately.
   After all have feasted the father of
   the child addresses the guests. He
   says, "I now offer the beads. May
   the child be healthy and prosper!"
   The guests reply, "Camagu." The
   host approaches the men first and
   then the women. He gives each
   guest two small white beads. Each
   one receives and returns the beads
   with the word "Camagu. " When all
   the beads have been given and
   returned the man takes the beads to
   his wife. She immediately threads a
   simple necklace of knotted beads
   and fastens it on the child. Then
   only, may mother and child meet
   the guests. This ceremony is comparable
   to Christian baptism. The red
   ochre and beads are the first symbols
   of the Qaba Faith. Later the
   child will be taught to worship his
   ancestral spirits in song and dance.
   It is for these reasons that Christian
   Africans will not wear red ochre or
   beads nor will they join in tribal [sic]
   dancing and singing. In the absence
   of any written law, beads and ochre
   carefully convey ideas of custom
   and procedure. Thus they are used
   as regulating agents in the life of the
   tribe [sic] and a form of bead symbolism
   in colour combinations and
   motif has been developed to convey
   messages. Each tribe [sic] has its
   own set of colour combinations and
   patterns for every age group.

   (Broster 1976:4-5)

As this passage suggests, Xhosa peoples regarded white as the primary color of purity and mediation--and traditionalists still do. Only white beads were used as offerings to spirits (8) and, on very rare occasions, to the Creator. (9) White clay (ingceke) is still used to indicate such liminal states as male circumcision, female initiation, and nursing at the breast. Equally persistent is the use of predominantly white beads among diviners, amagqirha, because white is associated with purity and the supernatural clarity and inspiration of the ancestors (Fig. 8). Amagqirha are known as "white people," who must have illumination, inkanyiso, on them to commune with the spirits (Gitywa 1971:124). Important items in a diviner's regalia include a beaded spear that symbolizes the ancestral spirits who inspire the diviner, beaded horns and calabashes for holding medicinal fats and powders, a beaded cow-horn trumpet, and a white-beaded veil (Broster 1976:96-98, 1967:102-7; Gitywa 1971:124). The veil, known as amageza ("beads of madness"), induces trance when swaying before the eyes. Furthermore, for ceremonial occasions, most people wore white clothing accented with beads and black braid, which was often, from as early as the 1930s, stitched on by machine for a small fee (Hunter 1979:102; Costello 1990:7).

[FIGURE 8 OMITTED]

Values

It is probable that glass beads and shell and brass buttons in beadwork were, like red ochre, regarded as immanent media. They were possessed of divine shine (Fig. 9). As in many parts of Africa, smiths were viewed with awe for their magical transformation of earth into shining metal, which was rare in precolonial South Africa. Metal was produced in the north around the first millennium and traded widely. Fourteenth-century plates for drawing wire have been discovered at Ingombe Ilede and Great Zimbabwe (Duncan Miller, personal communication, 2002); by the nineteenth century woven-wire decorations (sometimes combining iron, copper, and brass) were applied to regional weapons, staffs, and gourds. Wire weaving and basketry techniques furnished a precedent for the virtuosity achieved later in beadwork techniques. (In the twentieth century brightly colored, plastic-covered telephone wire provides an interesting extension of both beadwork and woven wire.) Early Xhosa people believed that glass beads washed up on the shore came from ancestors in the sea. Mother-of-pearl buttons probably carried similar associations; certainly, they were seen as extraordinary. The awesome immanence of these materials probably obviated pattern creation in the earliest beadwork. Rather, each gleaming iteration added power within an additive aesthetic. Also, because these materials were currency, each addition was value added, pronouncing the sacrifice of wealth in the service of personal presentation, which in turn saluted the ancestors from whom abundance flowed.

[FIGURE 9 OMITTED]

We know much about the shifting value of beads in trade with the Xhosa. In the mid-1700s one pound of beads worth a few pennies sufficed to buy an ox; by the 1780s two or three pounds were required, and they had to be of a particular type (Shaw & Van Warmelo 1988:868). H. Liechtenstein observed in 1803-6 that two small strings of beads, traded southward from Portuguese sources, were worth one cow and a calf (in Shaw & Van Warmelo 1988:453-54). In the early 1820s, when settlers began to occupy Xhosa territory, Xhosa raiders sometimes accosted English herdboys to steal the mother-of-pearl buttons from their shirts (Tyrell 1968:175). In 1822 the British authorities instituted trading fairs to regulate and regularize what until then had been restricted trade. At the weekly Fort Wiltshire fair (1824-30), beads and buttons were the universal standard of trade with the Xhosa, but the values varied with the caprice of fashion (Smith 1824-25 in Shaw & Van Warmelo 1988:859). According to an observation published in 1826, among the Thembu and Xhosa

   buttons pass for money; the colour
   of the beads is regulated by the prevailing
   fashion, which is as much
   attended to here as at Paris; so that
   many traders at Grahamstown have
   suffered considerable loss by not
   having them of the fashionable
   colour. As the women are exceedingly
   fond of both beads and buttons,
   they put their husbands to
   great expense. The dress of a woman
   in many instances costs twelve or
   twenty oxen.

   (Hallbeck & Fritsch 1826 in
   Shaw & Van Warmelo 1988:304)

Beads and buttons were second only to the currencies of iron and of cattle, which the Xhosa seldom traded because their society revolved around cattle, with increased supply, the value of beads and buttons dropped, but even in the 1830s, when trade was much freer, beads costing 9 pence to 1 shilling could be traded for 6 pounds sterling (Godlonton 1834 in Shaw & Van Warmelo 1988:863)--a profit of up to 10,000%! In the mid-1800s Xhosa day-labor could be hired for beads worth less than 3 pence, or a dozen brass buttons (Shaw 1860, Backhouse 1844, Boyce 1861, in Shaw & Van Warmelo 1988:864).

Viewed against this historical and economic context, the initial aesthetic impression of a hat worn by a Xhosa chief's wife (Fig. 7) today should probably be multiplied by a factor of 10,000 to begin to understand the original impact of this item. A hat like this was fashionable for women of rank from the beginning of the 1800s, when its beads cost their husbands the equivalent of three oxen (Thomas Baines in Shaw & Van Warmelo 1988:479). It declined in favor in the 1830s, when beads began to flood the market and became less significant as markers of elevated status. Colored kerchiefs and spotted head-cloths became fashionable instead, and by the 1850s it was reported that only royal woman still wore such hats (Shaw & Van Warmelo 1988:552). Nonetheless, European artists continued to represent it after it was no longer worn (Van Wyk 1993a:69, 71-72).

Even when beads were readily obtainable in the 1950s and '60s, Broster reveals their expensiveness. In Qebe the average annual family income was $45, allowing less than $4 per month for corn (the staple), salt, matches, paraffin, local tobacco, and small quantities of tea, sugar, coffee, and flour--apart from cloth and beads. (10) A 200-pound bag of corn cost the buyer $3, on which the Brosters made about 15 cents. Wages on nearby white-owned farms were $1.50 per month for men and 75 cents or less for women, while Broster employed young boys to help her in the garden at 7 cents per day. Young men might return after six months' work in the gold mines with savings of $45, of which they would spend several dollars purchasing beads at 1 cents per 10-inch string (Broster 1967:20), which their girlfriends would use to make items for them. Ochre came in little packets costing 4 cents or 8 cents. Although the Brosters' monthly profit at the store was about $70, they were unable to raise bank credit of $150. Life was not easy for anybody.

Nonetheless, between 1932 and 1955 the world's major bead manufacturer, which then had a virtual monopoly, exported to South Africa about half of all the beads sold to Africa, which consumed more beads than any other continent (Saitowitz 1993:37, fig. 21, p. 44). This statistic clearly suggests that South Africa was the world's greatest producer of beadwork in this era (Fig. 10). Beadwork became a remarkably democratic vehicle for aesthetic expression, available to everyone rather than just to leaders, and wearing beadwork on ceremonial occasions became an essential aspect of both identity and tradition. Though global economic lulls slowed bead imports during the Great Depression and World War II, the flowering of local traditions peaked in the 1950s and 1960s.

[FIGURE 10 OMITTED]

Broster noted that the Thembu were reluctant to sell beadwork. "Again and again I have tried to buy their beadwork, but excluding exceptional circumstances it is not for sale" (1967:18). Nonetheless, Broster collected 4,000 objects during her fourteen years in Qebe. More than half of these were items that the owners sold discreetly because of economic need (Broster, p. 176). She acquired three complete outfits as a result of calamity. In one case, a young woman became pregnant by a youth who refused to marry her. As part of the fine levied upon the man's family, her family demanded the return of the beadwork that she had made for him out of her own farm wages, and which represented twenty percent of the bride-price of five head of cattle that his family would have paid upon marriage (p. 177). Spite and jealousy, she remarks, were the motives behind the sale of a costume belonging to an elderly man, whose mistress of fifteen years demanded its return when he took up with a younger woman. (11) In the third case, although it was an "unheard of act to sell the beadwork of a deceased person," the widow of a young man sold his beadwork because she decided to become a Christian and figured that every evil she could imagine had already befallen her (p. 178).

Symbolism in Xhosa Beadwork

Little is known about the color symbolism of Xhosa beads apart from white. In early times, red beads were associated with Xhosa royalty, and they were offered when an elephant was killed (Shaw & Van Warmelo 1988:624). Different terms applied to opaque and translucent red, and special terms described translucence itself (Costello 1990:19). Although Broster states that every color was symbolic among the Thembu in Qebe, she specifies only that yellow beads symbolized fertility and green represented new life (1967:171, 105). These associations also applied to other Xhosa peoples of the Transkei, and accounts for the inclusion of these colors in fertility dolls (Costello 1990:13). Particular shades of the same color were named differently. For example, royal blue is ulwandle, the ocean (Costello 1990:18), and turquoise is ihobe, a dove, which is the same animal association that color has among the Zulu, but without the symbolic meaning of fidelity (Gitywa 1971:118).

Regarding Thembu motifs and symbols, Broster notes:

   Favourite motifs are stars, trees,
   rivers, diamonds, quadrangles, chevrons,
   circles and parallel lines and
   one or more of these are combined
   to form a pattern. The patterns are
   relatively simple and some are exclusive
   to age groups. As in all their
   beadwork their design may be purely
   decorative or the artist may tell a
   story using the motifs as symbols.
   Thus a chevron may denote a hut
   and a diamond a child but there is
   always individual preference and
   artistic license.

   (1967:105)

The constituent elements of the "V" and the triangle were locally named litsomo and idlawa (Broster, pp. 31-32). A favored zigzag motif represented the local river and was named after it, whereas combinations of zigzags suggested multiple rivers in rugged terrain (pp. 104, 174-75). Trees were denoted by chevrons of various orientation and by diagonal or serrated vertical lines; diamonds often connoted stars (pp. 170-75) (Fig. 11b). Broster mentions but does not illustrate an old design called isadunge, which represents pools of water in a dry river bed (pp. 33-34). Behind such environmental symbols may lie concepts related to the ancestors underwriting fertility and the health of the land, but this supposition cannot be corroborated.

[FIGURE 11 OMITTED]

It is clear, however, that by combining motifs and bead colors that are particular to age grades, specific messages could be conveyed. These Broster illustrates (1967:170, 175) in a selection of square beaded panels attached to women's pins (Fig. 11a) and "keeper of my heart" necklaces, which represent the wife of a senior male (p. 65). The symbols indicated states of relationships and pregnancy, bride-price, number of children, and such personal qualities as diligence. (Tourist beadwork in Durban currently packages items with an explanation of "Zulu symbols" based on Broster's account of these Thembu symbols!)

For ordinary people, contact with the ancestors was closest at ceremonial occasions, when animal sacrifices were made or beer was drunk, since both summoned the ancestors and were closely associated with them. On these occasions a profusion of beads was worn, the heaviness of the glass probably impressing upon the wearer the weight of symbolism contained in the beads. Zolani Mkiva, a noted imbongi, or traditional poet, and the current C.E.O. of the Xhosa Royal Council, describes his beadwork regalia as being like telephone wire that connects him with his ancestors (Lisa Brittan, videotaped interview, 2000).

Beer-drinking gatherings for specific age grades were similarly important occasions. Among the Thembu these galas were each controlled by an executive committee, strictly constituted, and regulated by behavioral rules and fines. They included the Umtshotsho and the Isijadu for teenagers, the Intlombe for young adults of marriageable age, and the Ibasi for married men, their mistresses, and mature single women--widows, divorcees, deserted wives, and unmarried mothers (Broster 1976:71). At all such occasions, tradition--underwritten by the ancestors--was expressed through ceremonial white clothing, extensive beadwork, and traditional songs and dances.

Status Mapped Through Dress

Broster's work provides detailed evidence of how status could be read from dress in Thembu society. An example of how costume marked the Thembu individual's progress within and between age grades is the change in penis sheaths. Boys aged 13 wore calabash sheaths; aged 15, sheaths were of civet or wild cat fur, which they could decorate with the tail from the pelt at 17. At 18 or 19, they wore fluffy angora sheaths. At around 20, before circumcision, a young man was entitled to a goatskin sheath, a beaded pipe and can for tobacco (like those seen in Fig. 2a), and a jackal-tail headdress. His broad, navy-and-white choker necklace contrasted with the dark turquoise beadwork, highlighted with cerise wool, that indicated younger men and women (Broster 1976:14-17). The costume now included around 135 items. A teenage girl's costume would have almost 200 items, including 50 or more brass armbands and 50 or more rubber leggings adapted from jam-jar seals.

After circumcision a Thembu man attended the Intomble wearing skirts, a turban, and a wide bead collar. His beadwork could include a "waistcoat" (Fig. 12), numerous long necklaces, throatbands, armbands, leggings, and belts. Navy and white beads predominated, contrasted with red and blue. Some yellow and green beads, symbolic of fertility, were included in the costume, especially for bridegrooms and newly married men. Black Empress beads replaced cerise pompoms for additional embellishment. The young man carried a beaded stick and cloth tobacco bags decorated with beaded leather streamers (Broster 1976:32-33). A young woman wore broadly similar dress, with some variations. The costumes at this stage of life included 150-200 items.

[FIGURE 12 OMITTED]

As Thembu men and women aged, their beadwork remained predominantly navy and white, but now contrasted with pink. Male elders generally wore two beaded collars, ten or more throat necklaces, at least three long necklaces reaching almost to the ground (Broster 1976:76, see also 1967:65), and numerous beaded items for the limbs. At the Ibasi, an older man was dressed by his mistress; she was assisted by another woman who handed her, in order, the approximately 100 elements of the costume.

Similarly, among the Xhosa per se, specific types of beadwork were worn by specific age grades and genders. These were described in some detail by V.Z. Gitywa (1971); Xhosa costumes from a collection were illustrated by Dawn Costello (1990), although with some mislabeling; and extensive field photographs among the Xhosa (1956-69) were published by Aubrey Elliot (1970) and Jean Morris (1959 to early 1990s, mostly 1970s).

Together, Broster's costumes and accounts illustrate how beadwork signified synchronically both within a specific Thembu community and across the various Xhosa-speaking groups. It is not always possible to identify individual beaded works precisely in terms of ethnic origin, gender, age group, and other criteria. However, a synchronic, cross-cultural comparison reveals how the beadwork of Xhosa-speakers, especially when compared as complete ensembles, constituted a semiotic system of structured differences through which identity could be minutely read.

"Zulu" Identity

Among Zulu-speakers, the identity question is much more complex than for the Xhosa peoples. The rise of the Zulu Kingdom under Shaka in the 1790s and early 1800s catapulted the minor Zulu chiefdom into the dominant regional power among hundreds of others. Shaka coerced the incorporation or dispersal of neighboring North Nguni chiefdoms, which, though they had a broadly similar culture, were descended from distinct Nguni branches. (12) Cultural difference was both effaced and preserved within the kingdom--and certainly asserted on its borders by those who resisted its power. After the British defeated the Zulu Kingdom in 1879, they installed to prominence chiefs who had been subordinates within it, further altering the distribution of power among regional chiefs. By then the label "Zulu" had seized hold as colonial shorthand for the diverse North Nguni peoples of the region, including those who had never been amalgamated into the kingdom, and subsequent administrations further standardized the Zulu language and entrenched the label, which continues to overarch differences today.

In traditional rural settings, North Nguni societies followed a historical pattern of settlement, aggregating in clans or chiefdoms that could be widely dispersed and interspersed. This means that identity cannot be as neatly mapped as among Xhosa-speakers, which is reflected in the ethnographic survey carried out by Van Warmelo (1935) (Fig. 13). Such surveys were grist for the "homeland" policies of apartheid. Difference, thus defined, helped divide up the land, but white interests remained an overriding consideration: what constituted black "homeland" was largely what was strategic to recognize as such.

[FIGURE 13 OMITTED]

Despite this territorial complexity, clan or regional identity is flagged in several distinct mid-twentieth-century (ca. 1920-1970) Zulu beadwork styles (Fig. 14). These include the styles from the regions of Nongoma, Msinga, Maphumulo, Ndwedwe, and the Drakensberg foothills, where Ngwane, Ngweni, and Hlubi peoples scattered by the rise of the Zulu Kingdom are settled today. Each is identifiable by its palette, often by the conventional form or texture of the beaded object, and sometimes by distinctive motifs although these all vary historically. These diagnostic features, which visual comparison reveals sufficiently for present purposes, combine within a gestalt that seldom leaves any question of origin (Fig. 15). The illustrations here pay special attention to the Maphumulo (Fig. 16) and Ndwedwe regions (Fig. 17), which have been published less often than the others. (Unfortunately, the styles of the South Coast and North Coast of KwaZulu-Natal and the Zulu-speaking Bhaca of southern KZN must remain beyond the scope of this article.)

[FIGURES 14-17 OMITTED]

Social identity, or status, is also reflected in dress among Zulu-speakers. Barbara Tyrrell's 1971 book on the Zulu-speaking Bhaca of the Richmond region provides a perfect example. Tyrrell illustrates and describes sixteen social "types" (pp. 169-84), much like Broster's work. Her meticulous color drawings also record Bhaca style between 1945 and 1970.

Zulu Color Symbolism

The broad outlines of what meanings could be conveyed by North Nguni beadwork styles are best understood against the background of Zulu spiritual concepts and the historical record of communication through bead symbolism in the Zulu Kingdom during the late 1800s. (13) Red, white, and black are the main sacred colors in Zulu symbolism. In combination, as particularly in the clothing of diviners, who also code medicines as white, black, or red (Ngubane 1977), (14) these colors suggest an understanding of the universe and the cycle of life within it. White is associated with the ancestors, concepts of purity, calm, and good intentions, and light and divine enlightenment. In some contexts, black represents darkness, evil, death, and defilement. In other contexts, black is linked to the ancestors and carries positive connotations--it evokes the dark rain clouds necessary for the sustenance of life, for example, and Zulu pots, some sculptural objects, and the leather skirt of marriage are blackened to please the ancestors. Red is the color of blood, menstruation, and fertility, and red ochre is strongly associated with the earth and women and their fertility; white clay is associated with the ancestors (Berglund 1989). (15)

Within the Zulu Kingdom in the late 1800s, an elaborate system of bead language was used, mainly to communicate messages about courtship in love tokens. In 1963 Princess Magogo, a daughter of King Dinizuhi (1868-1913) and the mother of Chief Gatsha Buthelezi, provided a window into this semiotic system (see Grossert 1968, vol. 1:148-64). Raised among the wives of King Cetshwayo (1826-84), she had a detailed knowledge of Zulu history and customs at the royal court. She interpreted the meaning of an elaborate message conveyed by a string of sixty beads consisting of twenty-five different types. In her reading, for example, translucent red beads implied a heart inflamed with love, and pink implied poverty. The meaning of each bead, however, depended on its syntactical placement between neighboring beads. Thus, in one grammatical arrangement a white bead could imply a "white heart," calm and full of love, but in another it questioned why the lover was laughing. The princess pointed out that the reading of such messages depended on in-depth knowledge of Zulu customs, proverbs, and associations related to flora and fauna. She criticized contemporary women who deviated from these conventional rules in their beadwork for not respecting Zulu custom. Apparently, by this time the courtly language of love conveyed in beads had atrophied. The dissolution probably proceeded from about the 1920s onward, as regional beadwork styles developed, because their flag-like palettes contained few colors compared to the twenty-five types of beads that Princess Magogo interpreted.

This courtly language evidently was legible in the early 1900s far beyond the borders of the kingdom. The missionary Franz Mayr, stationed close to Pietermaritzburg, recorded the following reading of the color sequence white-yellow-blue-black-pink-black-yellow-red-white-red-white:

   Here is my letter to you; I know
   you have two oxen (=two neighbouring
   beads) for my father [to
   pay the bride-price to marry me];
   if I were a dove (blue), I would fly
   to you; but darkness (black) prevents
   me; you are still poor (pink);
   and the dark night (black) disturbs
   me; your cattle are only two, work
   to get more; my eyes are red looking
   out for you in vain; but my
   heart is white as the long days go
   on; I have looked out for you, my
   eyes are red.

   (Mayr 1907 in Wickler & Seibt
   1996:33)

In this message, for example, the signification of pink (poverty) and white (symbolizing a "pure heart") corroborates Princess Magogo's reading.

A Beadwork Convention: Mchunu Style

Meanings remarkably consistent with Princess Magogo's applied among the Mchunu people of the Msinga region, despite their adoption between circa 1920 and 1960 of a beadwork style known as Isishunka.

Ishishunka was based on a restrictive seven-color palette (Jolles 1994) and applied to most beadwork items, not just love tokens (Fig. 15a). These meanings were provided for the Natal Museum around 1970 by an informant named Salafina MaMchunu, who was born there in 1921, and were corroborated by Mchunu elders in the 1980s (Wickler & Seibt 1991:309). White represented all that is good (love, spiritual purity, happiness, truth). Black was the opposite (evil, misfortune, sorrow) but could also refer to the black skirt of marriage. Translucent red represented hot-blooded passion (the name for this white-cored bead, sometimes incorrectly called "carnelian," is umgazi, meaning blood). Vaseline yellow represented gossip; pink symbolized poverty, particularly the inability to pay the bride-price for marriage; turquoise represented fidelity; deep green represented youthfulness or pining away like a withering reed (see, e.g., Wickler & Seibt 1996:30). Green is the dominant color in the palette, followed by black. When combined, the two colors suggest readiness for marriage, and in this context green means new growth, young life, while black stands for the black leather skirt of marriage (Jolles 1994:53).

According to Frank Jolles, in full expression, Isishunka employed seven colors in a fixed sequence of seventeen bars that included several stable triads: translucent red-black-translucent red; green; pink-turquoise-pink; black; green-yellow-green; turquoise-black-turquoise, green-white-green (1994:47). The central color in a triad was called isiqaba (pl. iziqaba), meaning "field," with those on the sides being called "boundaries" (iminqamulo, sing. umnqamulo) (Jolles 1993). These terms inscribe the idea of boundaries, territorial markers, and link the beadwork to agriculture, which, together with concomitant ideas about fertility, is otherwise absent from the symbolism described for the Isishunka palette. The existence of a fixed sequence within Isishunku would seem to inhibit its semantic potential, because the same symbolic meanings would be repeated continually. It is feasible, though, that symbolic messages still could be conveyed. A beadwork maker could: a) select only a limited color sequence that best condenses the message, b) repeat or increase the compositional size of key symbolic colors, c) make conspicuous omissions, or d) put the key symbolic meaning in a prominent position within the beadwork article, such as at the beginning (Wickler & Seibt 1991:339). The less rigid the sequencing rules, the greater the possibility for variable messages within this restrictive palette.

Dynamism and Continuity

The Mabaso and Mthembu (also known as Thembu, or Tembu, and not to be confused with the Xhosa-speaking Thembu) clans who were neighbors of the Mchunu in the Msinga region signified their respective identities through different palettes and through patterns composed of graphic shapes rather than fields of colored bars (Jolles 1994:58-59). In the 1930s and 1940s they adopted the Mchunu field pattern, but they maintained their own palettes. The Isilomi style of the Mabaso sequenced navy, turquoise, sap green, white, opaque red, and black; the similar Isiphalafini style omitted the turquoise (Fig. 18). The Mthembu used sap green, yellow, red, black, and turquoise (Fig. 19, second from left). Reciprocally, these adjacent palettes made their way into Mchunu areas. The Mchunu, Mthembu, and Mabaso had long been neighbors, "shared similar fates at the hands of Shaka," and then started occupying the Msinga region in the late 1830s; by the 1930s the distinctions between them were becoming less important (Jolles 1994:51). This allowed the Mchunu field pattern to be adopted regionally, although at first the neighboring clans preserved palettes different from Isishunka. In the northern Msinga region, around Nqutu, beadwork employed the Isishunka colors minus white and Vaseline yellow, and favored larger, heavier beads combined with smaller ones to provide richly textured objects. Another change in the regional palette was the adoption of an opaque red bead to replace the translucent red (white-cored) bead. Mussolini halted its production to hoard the gold required in its manufacture, and supplies probably ran out shortly after World War II (Jolles 1994:54, n. 10).

[FIGURES 18-19 OMITTED]

In the late 1950s some Msinga people moved southeast into lowland areas, where they applied a new palette, called Umzansi (sap green, white, opaque red, and navy), to the Mchunu barred pattern. "Umzansi" means "people from the lowlands," and this style reflected both territorial and generational shifts (Boram-Hayes 2000:152). By about 1960 (Jolles 1994:50) the Umzansi palette formed the basis of a modern style, Isimodeni, which departed from the field pattern by incorporating a wide variety of graphic patterns (Fig. 19, third from left). Isimodeni later abandoned the Umzansi palette by substituting black for navy and / or orange for red, departures from convention that were admiringly termed Isinyolovane (Jolles 1993:43) (Fig. 19, two belts farthest right).

Elders, like Princess Magogo, disparaged Isimodeni (Morris & Preston-Whyte 1994:49), but it became convention--much as slang is first corruption but settles with use. Isimodeni defined a fashion, imfeshini, that united the clans of the region at a time when "the growing impact of the modern State in a time of rapid economic expansion was beginning to erode the traditional social order" (Jolles 1994:51). This was also a period when both the apartheid government and Zulu traditionalists (particularly the Inkatha movement) were manipulating constructions of identity. In other words, the beadwork in the Msinga region shifted away from signaling traditional clan distinctions and toward signifying regional identity and contemporaneity. These changes exemplify the dynamism that "finds a way forward" within a signifying system.

An example of both the divergence and continuity of bead symbolism is the blue-striped white bead known as iNtothoviyane. It is named after a species of multicolored grasshopper associated with love, marriage, and fidelity: after mating the male stays on the female for weeks until she dies (Fig. 15c). In one context Princess Magogo read this bead to mean "Happy are the grasshoppers because they are at least sure to die on each other's backs." In a different syntactical context she interpreted it as "I now heard that I stink to you like a grasshopper" (Grossert 1968 in Wickler & Seibt 1991:321-22). The association of this bead with fidelity was still in evidence among some Zulu communities in the 1940s (Twala 1951), but it was also incorporated into diviners' regalia to represent a particular type of spirit (Wickler & Seibt 1991:322). Among the Khuze of southern KwaZulu-Natal, this grasshopper and concepts of pairing were associated with the color combination of white-red-black-green (Winters 1988:48), which also had become the palette that defined the Nongoma regional style for most of the twentieth century.

Nongoma Colors and Motifs

The Nongoma region is associated with the region of the Zulu royal court and the attendant ancestral Zulu clans, in Nongoma beadwork, the sap-green color added to the primary Zulu symbolic colors (white-red-black) is associated both with the food of cattle, which were the core of Zulu traditional life and a physical link with their ancestors, and the gall of animals sacrificed to honor the ancestors (Ngubane 1977:125). A Nongoma variant that adds or substitutes blue or yellow or both is associated with the region between Ceza and Mahlabatini, the territorial base of the Buthelezi chiefdom, long allied with the Zulu. Chief Gatsha Buthelezi is the cousin of King Zwelethini and the leader of Inkatha, the Zulu nationalist party, whose colors black-green-yellow-red are frequently seen in patterns set on white backgrounds from this area.

These related palettes are used to create dazzling geometric designs. They vary from simple patterns, such as alternating checks, triangles, or diamonds, to complex combined motifs. The diamond shape represents a shield, symbolizing protection (Morris & Preston-Whyte 1994:51). Researchers record that an upward-pointed triangle connotes the male principle, a downward-pointed triangle or "V" represents the female pudenda, two triangles joined in an hourglass shape represent a married man, and two diamonds joined top to bottom represent a mature woman (Fisher 1984; Winters 1988; Wickler & Seibt 1996:25). (15) The triangle may also be associated with the heart because they share the same word in Zulu, inhliziyo, and when combined with the diamond-shaped shield may imply protection for a loved one (Boram-Hayes 2000). Wolfgang Wickler and Uta Seibt suggest that a profusion of small diamonds, triangles, and other motifs celebrate fruitfulness (1996:25).

Some Nongoma pattern motifs echo stepped-diamond motifs in Asian weaving, which combine triangles and diamonds. It is possible that Zulu beadworkers adopted these from members of the Muslim community in colonial Natal, particularly Gujerati immigrants from Pakistan and India whose fabrics employed identical motifs, and grafted onto them their own symbolism (Papini 1994) (Fig 20). Gujerati fabrics adopted this motif from kilim rugs, where the motif is widely used from Turkey to western China, but most consistently in neighboring Afghanistan (Papini 1994:35-36). The dispersion of this distinctive motif to communities as widely separated as the Zulu and Native Americans also could have occurred directly via Central Asian rugs seen in white homes and interiors during the 1800s. Navaho weavers, for example, were instructed to copy kilim designs hung up in trading stores (personal communication, Allen Roberts, 2000, 2002), and before long such designs were regarded as quintessentially Navaho.

[FIGURE 20 OMITTED]

Symbolic colors and patterns thus enrich the symbolism of Zulu peoples' regional palettes. Christian and Western motifs are also incorporated. A notable example is the beadwork of members of the Shembe Church, a Zulu Christian sect founded in about 1910, which incorporates stylized crosses into another remarkably consistent beadwork style that has been extensively documented (see Morris & Preston-Whyte 1994:60-73). The meaning of motifs can vary regionally, however. A cross motif from the Estcourt region, for example, was identified as isiambolosi, meaning "ambulance" (Morris & Preston-Whyte 1994:30).

Slippery Words

The adoption of Western letters (both upper- and lowercase) to spell out messages in Zulu is, like the symbolism described by Princess Magogo, surprisingly ambiguous.

Among the examples provided by Wickler and Seibt (1996:54, 59, 60), for example, the message "ISAKUHLE" can be interpreted either as "It is going fine" or "She comes willingly." Likewise, a woman read the necklace message "ngiLikHU ningObAAnginAKi" to mean "Hold everything inside (Ngilikhuni) because (ngoba) I do not care (anginaki)," which implies that the girl knows that her boyfriend is having affairs with other girls. Another informant, however, translated this from a male point of view: "I am very strong because I am careless," like a bull in a china shop. Vive la difference. Similarly, one reader understood the beaded message "wAXALAU mASAniBOnAmAnYA mBAnEPni IA n" as "Wazala umasanibona manya mbanephila," a slang way of saying "Salani sonibona umanxa sisaphila," which means "Goodbye, we will see you if we are alive." Another reader interpreted the same message as "Wazala, uMa! Sanibona manyamba na ephila," suggesting it read "Congratulations (Wazala), mother (uMa); live, be in good health (phila); they, them (bona); [unknown word--to have twins?] (manyamba). A more accurate reading suggested to me might be one between those offered by Wickler and Seibt: "Congratulations, Mother [on a good thing, implying birth]. We will see you all in the future if they/you are well." (17) Two other fluent Zulu-speakers told me the message could not be interpreted because "it is written and spelt badly" and "not enough information was given to conclude anything" (Khonya Rauri Alcock and Jupana Dladla, personal communication, 2001).

One thing, then another, and, and, and then ... becoming undecidable. Language become abstract, mad; in this case language itself becomes unraveled by written words that are much clearer signifiers than are the constituents of visual signs--color, shape, pattern--in most bead language. How then, finally, to view, to read, the significatory systems in beadwork? As duck, rabbit, or elephant, white elephant, or pink elephant, or all of the above, simultaneously?

For Wickler and Seibt, who are experts in the study of birdsong, the patterns of Zulu beadwork are like songs or chants, sung in different dialects that echo the rhythmic systems of poetry, music, and nature. "Like sound strings in birdsong, here colour strings with encoded messages are used in averbal communication; another striking parallel is the genetic code incorporated in macromolecular strings of nucleotids in DNA or RNA, or of amino acids in proteins" (1991:340). The principle of self-similarity that underlies fractal geometry, crucial to the scientific understanding of nature, is reflected in much two-dimensional South African art, including beadwork and mural arts among the Basotho (Van Wyk 1993b, 1998). So complex are many Basotho patterns that the figure/ground relationships of their elements are unstable. First one element seems to dominate and serve as the seed motif that generates the composition; then the interstitial element takes over and proclaims its dominance, and the pattern then reads entirely differently. If one attempts to hold both readings simultaneously, like the problem of the "rabbit-duck" drawing, the eye dances back and forth, the rationale becomes undecidable. The abstract animates without being animal. Instead of ending up with a duck or a rabbit, one senses a visual metaphor of the mysteries of life and beyond. Eye moves, mind moves, spirit moves. Two dimensions flower into the manifold.

These patterns surely make our eyes dance, but semiotics is slippery, in color, pattern, or word. Meaning is invented and preserved or shifts, or it slips away, to die or return. In semiotic theory the relationship between the signifier and signified is arbitrary, but in figurative, mimetic sculpture the signifier mimics the appearance of the concrete signified, and recognition is easy if the representation is good enough: "elephant." Where the signified is abstract, we find ourselves in much more open terrain and we must sniff the wind for direction. The wind shifts, too; it is its nature, wedded to time, embedded in it. Such things as the wind and time, grand abstractions, are bigger than us, (18) and elephants. They are of the order of spirit, in southern African languages, they share a special category of sacred nouns; in Sesotho, for example, wind, smoke, mist, fire, and lightning and spirit all carry the "me-" or "mo-" sacred prefix, together with the Creator, named Molimo, who is also hailed as Motlhodi, the source of all divinity, presence or being, and of the cosmos (Setiloane 1976:77-78, 266 fn. 11, 269 fn. 3). The Creator is also cosmic Lesedi, light and enlightenment. Philosophically, these sacred nouns point to the "ontological gap" in metaphysics between being and Being. In the slippage of this metaphysical gap they point to something eternally Other, beyond direct experience and human assimilation (Van Wyk 1996:120-21).

Beadwork and mural patterns in southern Africa are the artistic equivalent of these sacred nouns. Even their media are sacred and powerful. Why else offer red beads when the elephant is dead, pray for rain in paint, and dance for the spirits dressed in beads that reflect and refract the light?

[This article was accepted for publication in August 2003.]

My thanks to Adrian Turgel for putting me in touch with Zulu-speaking informants, ,and to Lisa Brittan for assistance in preparing the digital images.

(1.) South African peoples never made figurative masks. Only a few cultures in the northeast carved freestanding figures for initiation contexts. Apart from these relatively rare initiation figures, the aesthetic appreciation of South African sculptural forms, including dolls, requires transcending the stereotypical view of African art, for the sculptural drama is often abstract or very subtly figurative rather than dramatically representational. These abstract qualities beautify such utilitarian objects as staffs, headrests, spoons, and meat platters. But these objects are often less abstract than they seem. Zulu spoons, for example, are subtly figurative: the scoop is rounded with a pointy chin like that of a beautiful woman, whose "head" often leans forward in the decorous pose required by hlonipa customs of respect which women owe their in-laws. Some spoons have phallic terminals and central patterned panels of pyramidal bumps, called amasumpa, which refer to cattle (Klopper 1991). In combination, these references may refer to the union of men and women through the exchange of a bride-price of cattle. Such subtle figuration was sometimes made more explicit, particularly when animal or human figures were integrated into such utilitarian items as staffs, headrests, and utensils. These inclusions distinguished owners of exceptional status and spiritual power until at least the 1920s. With the growth of colonial trade in the later 1800s, however, figurative works were also made throughout southern Africa for sale to Europeans, and European materials and inspirations increasingly entered the local African art traditions. Among other North Nguni peoples, particularly the Ngoni groups who fled the rise of Shaka's empire in the early 1800s and migrated far north to central Africa, what remained abstract among the Zulu became literal: Ngoni headrests, for example, clearly represent the cattle that Zulu headrests symbolize.

(2.) A subscriber to the listserve H-AfrArts pointed out recently that the omission of beadwork from "Material Differences" (April 10-October 6, 2003), an exhibition at New York's Museum for African Art that purported to examine the issue of materials in African art, was glaring.

(3.) The two classic studies of Nguni origins art" by J.H. Soga (1930) for the South Nguni and A.T. Bryant (1929) for the North Nguni. They are contradictory in certain respects, and the histories they purport to record cannot be verified. (See also note 12.)

(4.) In the late 1790s and early 1800s, the various Xhosa peoples were sandwiched between two burgeoning states, the Zulu Kingdom to the north and the Cape Colony to the south. In the south, the chiefdoms of the Xhosa per se bore the brunt of colonial expansion. In the north, the Mpondo were the first to confront the waves of refugees fleeing Zulu expansion in the 1820s and 1830s, either repelling them or allowing them to settle--as did the Bhaca, Xesibe, and Ntlangwini, who are counted today among the Xhosa peoples. Other refugees, including the Bhele, Zizi, Hlubi, and Ngwane, continued their path southward and were taken in by the Xhosa and Thembu, who called them collectively "Mfengu"--a name that probably derives from their reply when asked who they were: "Siyamfenguza," meaning "We are destitute, we seek service." It follows therefore that several of the peoples integrated into the South Nguni, or Xhosa peoples, were earlier settled among North Nguni groups. Furthermore, several of these groups were splintered, leaving fragments behind during their migrations. The Ngwane provide an interesting illustration, because a substantial Ngwane community remains in the Drakensberg region, where it is typically counted as "Zulu." Similarly, different Bhaca communities identify themselves as either "Zulu" or "Xhosa."

(5.) David Hammond-Tooke notes that though the Gcaleka took precedence in ritual matters among the ten chiefdoms into which the Xhosa per se split over a period of six generations, they had no political control over the others, received no tribute, and had no rights in adjudicating legal matters. It is therefore inaccurate to regard these Xhosa chiefdoms as constituting a polity (contra Peires 1981) that was, like the Zulu Kingdom, expanding during the early colonial era; and colonial officials were mistaken to regard the Gcaleka king as a "paramount chief" (Hammond-Tooke 1993:68-69).

(6.) Broster does not consider whether Qwati beadwork, which she takes as fully assimilated Thembu, contains residual Xesibe dements. However, her collection from these Thembuized people includes a motif that can be termed a "red-hearted lozenge," consisting of a diamond (usually outlined in white on a navy ground) with a red center. This motif, not used in other Thembu beadwork, may have Xesibe origins. Since the amaQwati were displaced, they might at one time have been labeled "Mfengu," although they had been long integrated into the Thembu community by Broster's time.

(7.) Among the southern Xhosa-speakers, imbola (red ochre) was a precious substance reserved for application to the body until it became more available as a commodity in colonial trade after the 1820s, when imported ochre began to be sold. Among the Xhosa peoples to the southwest, red ochre was mandatory for certain states of life, for example when males returned from initiation or before a bride bore her first child (Shaw & Van Warmelo 1988:709). Another indication of the importance of ochre is provided by Hammond-Tooke's 1956 study, The Tribes of King William's Town District (1958), where the primary distinction in dress was between the Christianized and "pagan" (e.g., p. 56). Although the Mfengu, Xhosa, and other peoples of the region reflected considerable integration (e.g., p. 47), including within their own political structures, the dress of Mfengu and Xhosa women was readily distinguishable (p. 125). The study also notes that the constituent Mfengu groups--Hlubi, Bhele, and Zizi--had decided to unify politically (p. 124).

(8.) After the meat of a sacrificed animal had been eaten, each guest was given a piece of medicine called ichakata, which was then returned to the household head for him to make into a necklace. White beads have been substituted for this medicine, and the resultant necklace is of white beads (see Sobahle 1977:272-73).

(9.) Among the Thembu during times of disaster, prayers, songs, and offerings including white beads were made to uMadi, the Creator (Tyrrell & Jurgens 1983:51).

(10.) These value estimates are derived from converting the sterling values at the time to the dollar at today's rate.

(11.) The first costume is now In the collection of the Art Institute of Chicago, the second in the French national collection.

(12.) A.T. Bryant summarizes the origins and migrations of three main Nguni groups, which generated the various peoples who became delineated as South and North Nguni peoples by the rather arbitrary fact of where they ended up on the map. One major Nguni group so merged with Venda/Karanga peoples that they eventually retained only a Sothoized version of the ancient Nguni name of BaKoni. A second major group, the Tekela-Nguni, split into two; one branch merged briefly with the Thonga, after which some (including the Mthethwa) settled in northeastern Zululand while others spread along the coast as far south as floe Transkei. The other Tekela-Nguni branch settled between the coast and the Lebombo Mountains, and accordingly became known as the Mbo. Different Mbo migrations gave rise to the Ndwandwe, Swazi, Bele, Zizi, and Mpondo. The third major Nguni group, the Ntungwa Nguni, which mixed less with other earlier inhabitants of the region, provided the common ancestors of the Zulu per se, Xhosa per se, and Thembu (see Mertens & Schoeman 1975). In terms of this genealogy, the Mpondo, today regarded as South Nguni or Xhosa, are less closely related to the Xhosa than are the Zulu. The relevance of origins is subject, though, to memory--the degree to which heritage and genealogy are reinscribed, reenacted, carried forward, and "owned," and the answers to this question must be sought on a ease-by-case basis within the various manifestations of culture, including dialect, oral history, and custom.

(13.) The question of North Nguni beadwork styles before the defeat of the Zulu Kingdom in 1879, discussed by Sandra Klopper (2000) and Carol Boram-Hayes (2000), is beyond the scope of this brief essay. Suffice it to say that strict royal restrictions over beadwork within the Zulu Kingdom gradually slipped as bead supplies increased. North Nguni peoples living beyond the kingdom had easier and earlier access to beads, but we have insufficient data to establish in any detail how such differences were marked within beadwork. It is clear, though, that the people who settled in colonial Natal, particularly to the southwest, close to Pondoland, created beadwork that was markedly different from that in the Zulu kingdom. Among other features, this headwork frequently employs a white background and several translucent beads, including clear ones that have a silver appearance.

(14.) Winters's fieldwork was conducted among the Nyuswa.

(15.) A useful source for understanding this symbolism is Axel-Ivar Berglund, from whom many of these associations are drawn. For example, regarding white, see Berglund 1989: 144, 146, 167, 177, 184; for red, pp. 160.-61; for black as representing dark rain clouds, p. 51.

(16.) Hilgard Schoeman (Mertens & Schoeman 1975:7) provides a contrary interpretation that diamonds are associated with women while downward-pointed triangles are male symbols--and Rhoda Levinsohn (1979:61) follows his interpretation for the same symbols in pottery.

(17.) The central ambiguity hinges around the word "manyamba." It is not a recognized term for "twins" in Zulu; even if it were slang or obscure dialect, the second phrase becomes ungrammatical: "Hello twins if they are well." It would be grammatical, though, to interpret the phrase as suggesting that "we will see you all in the future." In this case "manyamba" confirms an implied future tense. "Manyamba" could be a proper name, Mrs. Nyamba, which would personalize the message. Another alternative is that "manyamba" is a misspelling of "manyanya," meaning ancestors, and the phrase means "We see you the ancestors if they are well" (Mark Alcock, personal communication, 2001). In the context of the originating community, the ambiguity might not exist, because locals would understand one or other interpretation to have been intended. It is not infrequent, however, that beaded messages, whether textual or conveyed through color, were so ambiguous that the maker would need to be asked to explain them.

(18.) The grammatical "bigger than we" sounds clumsy compared to the colloquial "bigger than us--an illustration of the semiotic axiom that spoken language, "parole," perpetually breaks and redefines the rules of codified language, "langue."

References cited

Bedford, Emma (ed.). 1993. Ezakwantu: Beadwork from the Eastern Cape. Cape Town: South African National Gallery.

Berglund, Axel-Ivar. 1989. Zulu Thought-Patterns and Symbolism. Bloomington, IN: Indiana University Press. 1st pub. 1976.

Boram-Hays, Carol S. 2000. "A History of Zulu Beadwork 1890-1997: Its Types, Forms and Functions." Ph.D. dissertation, Ohio State University.

Broster, Joan A. 1967. Red Blanket Valley. Johannesburg: Hugh Keartland.

Broster, Joan A. 1976. The Tembu: Their Beadwork, Songs, and Dances. Cape Town: Purnell.

Broster, Joan A. and Bourn, Herbert C. 1982. Amagqirha: Religion, Magic and Medicine in Transkei. Goodwood, South Africa: Via Afrika.

Broster, Joan A. and Mertens, Alice. 1973. African Elegance. Cape Town: Purnell.

Bryant, A.T. 1929. Olden Times in Zululand and Natal. London: Longmans.

Conner, M.W. and Pelrine, D. 1983. The Geometric Vision: Arts of the Zulu. Purdue: Purdue Research Foundation.

Costello, Dawn. 1990. Not Only for Its Beauty: Beadwork and Its Cultural Significance among the Xhosa-speaking Peoples. Pretoria: University of South Africa.

Dube, John L. 1928. "The Reintegration of Native Social and Religious Life Through Education," in The Realignment of Native Life on a Christian Basis. South Africa: Lovedale Institute Press.

Elliot, Aubrey. 1970. The Magic World of the Xhosa. New York: Charles Scribner's Sons.

Fisher, Angela. 1984. Africa Adorned. London: William Collins.

Gitywa, V.Z. 1971. The Arts and Crafts of the Xhosa in the Ciskei, Past and Present. Ft. Hare, South Africa: University of Fort Hare.

Grossert, J.W. 1968. Art Education and Zulu Crafts. 2 vols. Pietermaritzburg: Shuter & Shooter.

Hammond-Tooke, David. 1958. The Tribes of King William's Town District. Pretoria: The Government Printer.

Hammond-Tooke, David. 1993. The Roots of Black South Africa. Johannesburg: Jonathan Ball.

Hunter, M. 1979. Reaction to Conquest. Cape Town: David Phillip.

Jastrow, Joseph. 1900. Fact and Fable in Psychology. Boston: Houghton Mifflin.

Jolles, Frank. 1993. "Traditional Zulu Beadwork of the Msinga Area," African Arts 26, 1:42-53.

Jones, Frank. 1994. "Messages in Fixed Colour Sequence? Another Look at Msinga Beadwork," in Oral Tradition and Its Transmission, eds. E. Seinaert, N. Bell and M. Lewis. Durban: University of Natal.

Klopper, Sandra. 2000. "From Adornment to Artefact to Art: Historical Perspectives on South-East African Beadwork," in South East African Beadwork 1850-1910: From Adornment to Artefact to Art, eds. M. Stevenson and M. Graham-Stewart. Vlaeberg, South Africa: Fernwood Press.

Levinsohn, Rhoda. 1979. Basketry: A Renaissance in Southern Africa. Cleveland Heights, OH: Protea Press.

Mertens, Alice and Hilgard S. Schoeman. 1975. The Zulu. Cape Town: Purnell.

Morris, Jean & Preston-Whyte, Eleanor. 1994. Speaking with Beads: Zulu Arts from Southern Africa. London: Thames & Hudson.

Morris, Jean and West, Martin. 1976. Abantu: An Introduction to the Black People of South Africa. Cape Town: Struik.

Morris, Jean and Levitas, Ben. 1984. Tribal Life Today. Cape Town: College Press.

Nesbit, Molly. 2000. Their Common Sense. London: Black Dog Publishing.

Ngubane, Harriet. 1977. Body and Mind in Zulu Medicine. London: Academic Press.

Papini, Robert. 1994. "The Stepped Diamond of Mid-Century Beadfabric: Aesthetic Assimilation beyond the Second Izwekufa," in Oral Tradition and Its Transmission, eds. E. Seinaert, N. Bell and M. Lewis. Durban: University of Natal.

Peires, J.B. 1981. The House of Phalo: A History of the Xhosa People in the Days of Their Independence. Johannesburg: Ravan.

Rajchman, John. 1998. Constructions. Boston: M.I.T. Press.

Saitowitz, Sharma. 1993. "Towards a History of Glass Beads," in Ezakwantu: Beadwork from the Eastern Cape, ed. Emma Bedford. Cape Town: South African National Gallery.

Setiloane, Gabriel M. 1976. The Image of God among the Sotho-Tswana. Cape Town: Balkema.

Shaw, E.M. and N.J. Van Warmelo. 1988. "The Material Culture of the Cape Nguni: Part 4 Personal and General," Annals of the South African Museum 58:4.

Sobahle, W.M.K. 1977. "Xhosa Beadwork from Victoria East and Middledrift Districts," Fort Hare Papers, 6:266-89, University of Fort Hare, South Africa.

Soga, J.H. 1930. The South-eastern Bantu. Johannesburg: University of Witwatersrand Press.

Twala, Regina D. 1951. "Beads as Regulating the Social Life of the Zulu and Swazi," African Studies 10, 3:113-23.

Tyrrell, Barbara. 1968. Tribal Peoples of Southern Africa. Cape Town: Books of Africa.

Tyrrell, Barbara. 1971. Suspicion Is My Name. Cape Town: T.V. Bulpin.

Tyrrell, B. and Jurgens, P. 1983. African Heritage. Johannesburg: Macmillan.

Van Warmelo, N.J. 1935. A Preliminary Survey of the Bantu Tribes of South Africa. Pretoria: Government Printer.

Van Wyk, Gary N. 1993a. "Drawing the Bead on Blacks: Eastern Cape People Painted by Baines, Shot by Pocock," in Ezakwantu: Beadwork from the Eastern Cape, ed. Emma Bedford. Cape Town: South African National Gallery.

Van Wyk, Gary N. 1993b. "Through the Cosmic Flower: Secret Resistance in the Art of Sotho-Tswana Women," in Secrecy: African Art That Conceals and Reveals, ed. M. Nooter. Museum for African Art, New York: Museum for African Art.

Van Wyk, Gary N. 1996. Patterns of Possession: An Art of African Habitation. Ann Arbor, MI: U.M.I.

Van Wyk, Gary N. 1998. African Painted Houses: Basotho Dwellings of Southern Africa. New York: Harry N. Abrams.

Van Wyk, Gary N. 1999. "Abstraction," in 5, ed. J. Andrew. New York: J. Andrew.

Wickler, Wolfgang and Uta Seibt. 1991. "Structural and Semantic Constituents of Mchunu Bead Language: Msinga District, Natal, South Africa," Baessler-Archiv n.f. 39:307-44.

Wickler, Wolfgang and Uta Seibt. 1994. "Colour-coded Zulu Bead-Language and a European Medieval Equivalent," Baessler-Archiv n.f. 42:61-73.

Wickler, Wolfgang and Uta Seibt. 1996. "Zulu Beadwork Messages: Chromographic versus Alphabetic Notation," Baessler-Archiv n.f. 44:23-75.

Winters, Yvonne. 1988. "Contemporary Traditionalist Bhaca and Khuze Beadwork from the Southern Natal/KwaZulu Areas," Southern African Museums Association Bulletin 18, 2:47-51.

Wood, Marilee et at. 1996. Zulu Treasures: Of Kings and Commoner's. Durban and Ulundi: KwaZulu Cultural Museum and the Local History Museums.

Gary van Wyk directs exhibitions at Axis Gallery, New York. Four recent exhibitions featuring beadwork of the Zulu, Xhosa, and Ndebele peoples may be viewed at www.AxisGallery.com. He is currently advisor to the exhibition "A Decade of Democracy: Witnessing South Africa," which opens at the Museum of the National Center of Afro-American Art in Boston in April 2004, and is editor of the accompanying volume.