African Archaeology and the Ancestors



© 2009

My message in this memorial lecture derives directly from the ideas of Bassey Andah, who nearly two decades ago wrote about what must now be one of the most prominent issues in African archaeology, or for that matter, world archaeology todaythe need to better understand and communicate with the ancestors:

Africans need to be aware not just of the fact that the past exists in the present “but how and where and why precisely their pasts exist[s] in their presents [and] what value such presence holds for their future meaningful existence. We very much need to acquire a synchronous sense of tune which encourages as well as enables us to see ourselves co-existing peacefully, creatively, harmoniously and truthfully in all spheres of life, at all stages and times with our ancestors and our descendants, and this, with all of our neighbourspast, present and future” (Andah 1990:3).

A dilemma arises out of Andah’s vision for many Western-trained archaeologists, who may feel compelled to ask, how do we engage ancestors when simultaneously we are enveloped within Western science with all of its expectations for objectivity? It is informative that Andah, as an archaeologist trained in the West, saw no contradictions between connections with the ancestors and the practice of archaeology. And, it is also instructive that as a deeply religious Christian Andah’s perspective towards ancestors was very much a pluralistic one that accepted ancestral presence and guidance as complementary to a life committed to Christian practice. The absence of contradictions in his thinking in these matters causes us to pause and to consider more seriously how archaeology and ancestral agency might be re-thought in archaeological practice and theory today.

As we explore the variety of tensions that arise for some between the practice of a Western-derived scientific archaeology and a respect for and link to the ancestors, it is instructive to foreground this discussion with the lessons to be taken from Chinua Achebe’s (2007) short story, “Dead Man’s Path”. Achebe tells the story of an earnest young school teacher, newly appointed as headmaster of what he depicts as a “backward “mission school, the Ndume Central School. The new headmaster Michael Obi wants it to be the most perfect, beautiful, and successful school in the mission system. Enlisting the help of his young wife, he transforms the school compound into an English garden, replete with flowers and hedges around the school buildings. When he comes to learn about a little-used path through the school compound that connects the village shrine with the place of burial, he constructs a fence to keep villagers from using the path through the school. Sure of his mission, Obi informs a concerned village priest that the fence will stay up, even though the priest informs him that “The whole life of the village depends on it. Our dead relatives depart by it and our ancestors visit us by it” (Achebe 2007:249) The priest also warns that if Obi interdicts the path, then he will cut “the path of children coming in to be born.” Obi ridicules the elder, informing him that the purpose of the school is to eradicate such beliefs and to “teach our children to laugh at such ideas” (Achebe 2007:249). Two days later a young woman in the village dies in childbirth, and a diviner prescribes heavy sacrifices to propitiate the spirits insulted by the fence. Mr. Obi awakes the next morning to find his school compound in ruins, the flowers trampled, the hedges torn up, buildings demolished. That very day the white school inspector arrives and writes him up as having poor relations with the local villagers because of his misguided zeal.

Archaeology in Africa must avoid erecting fences like those privileged by Mr. Obi. Distancing people from their ancestors separates archaeology from the past and the present, preventing an interconnection between both. Bassey Andah saw the dangers of distancing the ancestors. He saw that it leads to cultural blunders that foster misunderstandings between cultures, widespread erasure of local histories, and incomplete or seriously incorrect interpretations of the archaeological record. I now want to turn to an examination of archaeology to gain some insights into how archaeology has treated ancestors and what we might see as positive new trends in an archaeology that recognizes, respects, and not only accounts for the ancestors, but also consults and works with ancestors, trends that would have satisfied Bassey Andah’s concern that distancing be reduced and bridges to the ancestors be built.

In an essay (Schmidt 1995) published as a companion piece to an essay by Bassey Andah (1995) in Making Alternative Histories (Schmidt and Patterson 1995), I explore some of the implications of Andah’s views about ancestors, which among his other work made him one of the boldest thinkers and profound critics of African archaeology (Schmidt and Patterson 1995). Postcolonial archaeology in Africa, as envisioned by Bassey Andah, is an archaeology that unveils connections to “significant ancestors”, and that contributes to building contemporary identities freed from colonialist ways of seeing and thinking. The link that Andah insists must be constructed to the ancestors is liberating, leading to the open recognition much like the recognition that ancestors play a roll in contemporary life “that the past is found in the present, and that this awareness is significant for a “future meaningful existence” (Andah 1990:3). Andah’s position on identity in this case is teleological and pragmatiche seeks to build an “enlarged future”, one sustained by regeneration, renewal, and return–a cyclical journey wherein “Africans return or journey back to our African homes, natural, social, and spiritual, of our yesterdays [so] that our present will accede to merge with our past, and to emerge from the past in an enlarged future” (Andah 1990: 3).  Andah’s message resonates strongly with many African and Africanist scholars, including Abiola Irelele (1991), who also has emphasized revaluation, regeneration, and return. For Andah, this journey back depends on the strength of “spiritual” bridge building ”connecting with the ancestors” to move through history in ways sensitive to local values and sensibilities.

Juxtaposed to this humanistic side of Bassey’s thinking was his insistence that the historiography of Africa be revolutionized by using scientific principles. This assertion however swiftly creates tensions, tensions that arise out of Western Cartesian thought, which constructs the scientific world as separate from the spiritual world. Andah’s belief that there is a natural union between spiritual bridges and scientific searchesthe primary metaphors that he usesmakes us reflect momentarily to examine and evaluate the language Andah uses in his advocacy of this kind of archaeology as well as the messages he wanted to communication to Nigerians and the world beyond. He employs powerful tropes [figures of speech], the language of metaphor and metonymyto elicit images of ancestors past, such as ritual grounds in the past, spiritual journeys, and identity with the ancestorsall drawn from the deep wellspring of African life, sensibility, and history. In this perspective he is trying to recast the archaeological discourse, so that it fits African mental constructs.

Andah insists that African and Africanist archaeologist of Africa pay attention to one of the most difficult domains of human existencethe ancestral/spiritual worldand then seek to explain the spiritual world in such a way that it marks respect and acceptance of non-Western ways of communication in which ancestors play a constant, daily role in lives. Many archaeological “scientists” are either not willing or reluctant to take this critical step, a step that surrenders control over scientific measurement and accepts another dimension of human existence normally outside the purview of scientific archaeology. Yet, as archaeologists and anthropologists we may unveil and make more accessible emic examples of spiritual “agency, in which spirits, gods, and ancestors frequently have an active, or even dominant role in directing the course of human life and history (Deitler and Herbich 2007).

One of my goals is to explore how archaeology, as it is practiced in Africa, treats ancestors: Are the powers of ancestors in daily life sufficiently understood by those practicing archaeology in Africaboth Africans and Africaniststhat archaeologists incorporate ancestral powers and presence into the narratives that are written about the past? Are archaeologists in Africa respectful of ancestral needs and sensibilities when engaged in investigations that touch on ancestral ground, shrines, and sacred space? How have archaeologists interpreted the ever present power and influence of ancestors in domestic life, in technological life, and in life after death? Is it necessary to be African when proffering interpretations about ancestral engagements that run from the deep past to the present? These and a host of related questions arise out of Andah’s (1995) treatment of ancestors, a treatment prompted by his deep concern over the widespread depredation of ancestral landscapes in southeastern Nigeria under the neo-colonial arrangements between international oil companies and the Nigerian state.

How the Ancestors are Treated by African Archaeology

I turn now to several examples to illustrate how archaeologists working in Africa today treat ancestors. I have selected several studies that readily illustrate the themes and issues that I want to bring into focus, but hasten to note that there are a number of other innovate approaches not reviewed here (e.g., Agorsah 2009; Insoll 2007, 2008; Insoll et al. forthcoming; Stahl 2008), including the important work of Akin Ogudiran (2007, 2009) who makes significant contributions to understanding the role of bricolage in ritual process. One of the more complete and sympathetic treatments of ancestors may be found in Neil Norman’s (2009a) research into the long duree of vodun in Dahomey, a project in which he employs ethnohistoric evidence along with archaeology to develop a deep time history of vodun. Drawing on other scholars such as Suzanne Blier (1991, 1995), Norman emphasizes that vodun practices depend on innovation and assimilation of new components, particularly material components that make up the ritual repertoire, a mixed media where many different parts form the wholea creative process of combination and reinvention in which different materials such as glass, ceramics, metals are bound together (cf. Drewal 1996; Rush 1999). This bricolage, wherein different elements of local and foreign materials as well as symbolic structures are bound together, Norman believes can be found in sacred spaces in which ancestors played a key role hundreds of years ago.

Norman uses more recent vodun shrines as general analogs in the process of tracing out how earlier shrines may have been fabricated, drawing on the vodun belief that there is a “mutual independence between visible world of the living and invisible world of the spirits” (Norman 2009b:4). He goes on to explain that communication between the two domains depends on sacrifice, possession, divination, and prayer, with an ever present awareness that vodun spirits and ancestors have the power to do both good and evil as well as to assist or punish. Most important for our discussion here is that the ancestors depend on their sustenance from those who serve them. Mostly this entails “feeding” the ancestors, enjoining them to intervene favorably in family ceremonies among many other activities; vodun initiates might also ask for vodun agencywhere people require vodun to keep sustenance flowing and happily and reciprocally provide offerings (pers. comm. Neil Norman, October 6, 2009). Feeding and sacrifice conducted by the living for the ancestors goes beyond food and blood to include material items that attract the attention of the ancestors.

As part of his research into the relationship of outlying communities with the Hueda (Ouida) palace at Savi in Benin, Norman focuses on the era of approximately 1650 to 1727, at which time troops from Dahomey burned the palace and surrounding communities (Fig. 1).

Fig. 1

Ethnohistoric records tell us that religious practices during that era were international and inter-ethnic, drawing accumulatively on many different sources. Drawing on recent ritual practices that employ local ceramics, Norman observes that it is the consecration of mundane ceramics by ritual specialists that transforms these and other objects into “charged sacred spaces.” Archaeology enters into a dialogue with the ancestors when these spaces are encountered during archaeological survey and excavation. The primary questions that emerge for an archaeologist working on any landscape, including those surrounding Savi, are: how may such sacred spaces be recognized and how are they to be treated? The second of these questions poses rather intractable issues. First, recognition of ancient or very old sacred spaces may be limited by the availability of pertinent analogs in contemporary society; if there are demonstrable continuities between past and present, then it is possible to build understandings about the possible locations of sacred spaces, such as under house floors, in house walls, and in boundary ditches. Secondly, once sacrificial and other offerings are exposed, their treatment and care is a highly sensitive issue that archaeologists want to complement by informed and thorough interpretation of their significance. Norman’s work does precisely this.

During excavations in the greater Savi region, artifacts and structures were encountered that suggested the presence of sacred spaces. At locus 2 near Savi, Norman excavated a small structure located between a boundary ditch and the interior of the house compound (2009a, 2009b). Within this space a number of items including a trade bead, imported pipe fragments, iron fasteners, and perforated local ceramics were located. Most important was the recovery of a ceramic vessel similar in form to a modern golizen, a ritual vessel that was cached in a wall at the time of its construction. Based on its proximal relationships and size, Norman interprets this as similar to a Dohemean “spirit house” or djeho (after Blier 1991). Within the same excavation unit were two additional finds of great interest: a shallow bowl containing a fragmentary human cranium and a second cranium with mandible intact and encased in a ceramic storage vessel. Given the location of these finds and historical accounts about such remains being used to memorialize ancestors or the celebration of military victories, Norman privileges the interpretation, correctly I believe, that these are not ancestors but rather the skulls of enemies taken in battle (cf. Law 1989). While such an interpretation is underwritten by later historical accounts, the context of the fragmentary skull requires additional attention. It was set inside a shallow bowl, decorated by interior incised decoration in the form of an “X” surrounded by a field of cross-hatching (Fig. 2). Slipping with a red slurry further evokes blood offerings in this shrine, with the important observation that the symbolic depiction contained within the bowl resembles the Dohomean weke or cosmogram applied to bas reliefs, drums, and thrones (Blier 1991). Its significance also lies in its representation of the four directions and the four physical states (air, fire, water, earth) the vodun inhabit.

Fig. 2      Fig. 3

The excavations outside of Savi have helped to define a number of varied religious practices likely associated with earlier forms of vodun, including a range of sacred objects such as trade beads and ritual vessels in association with oil palm nuts in rooms surrounding a central courtyard as well as a figurine near the steep slope of an exterior ditch. Also important were remains documented in another room of the same complex where pipe fragments and iron on a floor marked a lower pit with ritual items. These varied ritual treatments speak to interventions with a wide range of local deities as well as ancestors.

Perhaps the most poignant evidence for the role of ancestors in daily life was found at locus 7, a secondary palace complex. In a collapsed structure, excavations revealed foreign items such as lead shot, a gunflint, and many local and imported pipe fragments associated with a large jar in a prepared pit beneath the living floor (Fig. 3). Within the jar were 12 teeth and many human cranium fragments, an arrangement congruent with historical observations of burials within houses in the region; the presence of ceramic fragments inscribed with an “X” cosmogram further underwrites an association with vodun practices. These cemetery rooms, argues Norman, are where “defied members of the family are installed, taking up permanent residence in the structure.” These intimate interior spaces and the worship rooms that they surround reaffirm the presence of the ancestors, places where the dead occupied prominent though out-of-sight influence. The Hueda world, then, was one in which a plethora of shrines between the main compound and the surrounding ditch were used to protect the interior social spaces from the dangerous forces of the outside liminal world as well as to negotiate with cosmological actors and connect to the ancestors. The shrines in the interior sleeping and storage spaces however were reserved for the interment of ancestorwherein the houses “were built quite literally on the ancestors” (Norman 2009a:214).

The investigations around Savi are important for two clear reasons. First, the material items included in the shrines are a complex bricolage (see Fennel 2007a, 2007b; Schmidt 1996, 2009), a mixture of potency from the local world (e.g. ritual ceramics, oil palm nuts, offerings of meat, etc.), and the inclusion of items from the Atlantic worldimported trade beads, tobacco pipes, wine bottles, nailsan accumulative and varied set of permutations that speak to the mutability of the vodun ritual process and its openness to the inclusion of exotic foreign objects with highly charged values. Second, the veneration of ancestors is an ever-present part of doing archaeology in Benin and other parts of the African continent, requiring a thorough and deep knowledge of the history and ethnography to adequately situate these important findings in contexts that will shed light on the present as well as the future, as Bassey Andah suggested.

Initiating Archaeology Inquiry in African Settings

The second example I examine pertains to how the ancestors guide and direct how archaeologists practicing in Africa conduct archaeological inquiry when they begin to study the history of landscapes. There are a variety of toolkits that can be used to document sites of archaeological value, including sampling schemes that draw on random and systematic samples, the latter perhaps structured according to topographic criteria such as forests, river valleys, and grasslands to name several. Such sampling programs while instrumental for retrieving a statistically reliable understanding of different kinds of archaeological sites over time nonetheless under-represent and insufficiently emphasize key sites associated with ancestral shrines, sites often associated with the deep time of ancestral figures belonging to important social groups. It is now apparent that in Sub-Saharan Africa archaeological survey is richly informed by methods that use oral traditions and oral histories to uncover recent and ancient sacred spaces and places that oftentimes hold highly charged meanings, perhaps stretching over centuries if not longer (Crossland 2001; Schmidt 1978, 2006).

When initiating archaeological research in northwestern Tanzania, I first spent many years learning local histories about important places on the landscape, places that were considered sacred shrines, places important for local identity, or locales that memorialized ancestors. In Buhaya, ancestors are remembered by their graves marked with specific trees and by their names being used to demarcate property linked to ancestral identities. Thus the ancestors act on the landscape to configure it socially and spatially, a relationship also seen among the Luo of western Kenya where “space is also structured by constant reference to the ancestors, and this has very real consequences for the lived experience of social life of present day Luo men and women” (Dietler and Herbich 2007). While such identities with named ancestors may persist for centuries if not millennia (Schmidt 2006), Zoe Crossland has shown that in Madagascar violence and the forced relocation of populations by a hegemonic state can lead to serious erosion of such identities with ancestral land, including the abandonment of memory sitesmostly tombs in the Andrantsay Region of Madagascar.

These features on the landscape, venerated and highly valued by local social and political groups, compel deeper understanding for they often act as daily remindersfunctioning as mnemonic devicesabout the recent and ancient histories of ancestors among many other groups including the Haya of northwestern Tanzania. Because such places were highly valued, they had remained mostly undisturbed and obvious reminders of the significant place that ancestors continued to hold, a constant role in structuring the world even in a culture that by 1970 had become highly Christianized.

As research unfolded over a period of years, I grew aware that among the many sites of great antiquity, one place (including a shrine to a 17th century king, a monumental memorial to the Bacwezi ancestors, and yet another shrine to a Bacwezi ancestor Mugasha) stood out amongst these shrines in Katuruka village in Maruku Kingdom (once part of Greater Kyamutwara Kingdom), just 15 km south of Bukoba (regional and district headquarters) (Fig. 4).

Fig. 4      Fig. 5

Kaiija tree or the “place of the forge” marked a sacred place revered for its antiquity (Fig. 5), its association with an ancient Bacwezi god, and its identity as the locale of the first iron making in the region (Schmidt 1978, 2006, Forthcoming). Ancestral claims to Kaiija tree by the Bayango clan are recognized as privileged (legitimate and tied to the land for centuries) over those of the more recent dynasty, the Hinda. The Bayango are noted for their origination of iron production in the region and closely associated with the Bacwezi ancestors through spirit mediums possessed by the spirit of Wamara. Ancestral spirit mediumship was closely affiliated with Kaiija tree and a neighboring shrine treeKya Rugomoraalso linked to the same iron tower myth attached to Kaiija tree. The Bayango spirit medium overseeing rituals at Kya Rugomora shrine tree, also dedicated to Wamara, represented spiritual authorities and social interests opposed to the royal Hinda royals who had coopted and taken physical control of Kaiija shrine in the 17th century, their occupation initiating a vastly different ancestral identity for Kaiija shrine.

After necessary propitiation of the ancestral spiritsparticularly Wamaraat Kaiija tree shrine and Mugasha’s nearby shrine house, archaeological excavations opened a new era of inquiry into the deep time histories preserved and maintained in the presence of shrine trees and other built features of the cultural landscape. The first excavations at Kaiija tree (Fig. 6) (Schmidt 1978, 1997, 2006) were conducted precisely where elders asserted that King Rugomora Mahe of the 17th century Hinda dynasty had his smiths produce iron to construct an iron towerleading to the documentation of an early iron working forge that dated to the mid-1st millennium BCE.

Fig. 6

This strong affirmation of the oral traditions at Kaiija tree, within the palace compound of King Rugomora, occurred only after extensive research into oral traditions and histories of ancestral lines in the immediate villages surrounding Kaiija tree. Hinda ancestral claims to the site can be affirmed to the mid-17the century, for the commemoration of ancestors at that time may be read in the archeological recordthe performance of multiple ritual activities in the area containing Rugomora Mahe’s “grave” and gashani estate, where the spirit house of the now ancestral king once stood (Fig. 7). Scores of oral narratives focus on a wide range of historical topics, most of them touching on the history of King Rugomora and Hinda appropriation of ancient Bacwezi myth associated with the shrine tree. Most important of these testimonies were those of the Bayango spirit medium who kept rich histories of their ancestors, revealing the relationship, roles, and attitudes held by the Bayango ancestors toward the Hinda royals who displaced them from the Kaiija shrine tree. Were it not for the deeply nuanced histories kept by this servant of the ancestors, then our knowledge of this most ancient of African shrines to iron production would have remained mostly inaccessible.

Fig. 7      Fig. 8

Recent research conducted by Dores Cruz (2006, 2009) in Gaza province of Mozambique amplifies the observation that attention to the ancestors leads to a deeper understanding of meanings attached to the African landscape (Fig. 8), a point of great concern to Bassey Andah. Cruz, working for the first time in a region that is relatively unknown from an archaeological perspective, finds that “the ritualization of trees participates in a complex system that assures continuity between present and past through the reification of ancestors..”(2009:2). Trees and groves not only “materialize the remembrance of royal ancestors,” but they also importantly signal resistance to foreign invaders (Cruz 2009:3). Like Kaiija and Kya Rugomora trees in northwestern Tanzania, the attachment of historical memories to natural elements such as trees concretizes or materializes the past. The strong link between trees (nature) and ancestors is a dynamic part of today’s landscape, sometimes detailing attributes of the persons associated with specific trees in Mozambique. Often the ancestral narratives in Buhaya also discuss such linkages in great detail, using proximal ties between the natural place and the ancestral personality, ties that bundle together in contiguous relationships huge, natural ficus trees (natural material place), material objects of great antiquity including an iron forge and royal installation pit, mythological texts of the Bacwezi ancestors and gods, and the historical texts of the Hinda dynasty.

The deep time ancestral histories of the Bacwezi provide a narrative frame for later dynastic histories, part of the illusion that the Hinda dynasty and its ancestors share in that deep time history. I have earlier observed that these sacred places become the focus of disputes over the productive economy and political power, that they entail heavy negotiations between social groups over identity and power, something also found by Cruz (2006, 2009); in the Haya case it was Bayango displacement and resistance to the Hinda, and in Mozambique, it was local resistance to the Nguni invaders of the 19th century primarily promulgated by local non-Nguni families (among whom the Mondlane family plays a central role in the Manjacaze area) in a number of palisaded compounds called khokholoene (sing. Khokholo; Cruz 2009:6), where some palisade poles are preserved, interactively contributing to memories about the site that are kept vital by the active practice of ritual pertaining to the ancestors and its sacred setting. The point of interest here is that ancestral agency is instrumental in the performance of rituals, the deposition of material cultures at the shrines, and the configuration of these sacred spacesall in the present as active ingredients in day-to-day life and well being.

While early ancestral shrines in Buhaya are not linked to burials of the ancestors, later burial shrines, such as the Hinda shrine for King Rugomoraa circle of ritually important Mulinzi trees inside of which once sat the Gashani house (for ancestral spirit possession)bear affinities to a sacred grove (inside the khokholo enclosure) growing where the “royal” compound once stood (Cruz 2006, 2009). In the khokholo enclosure there is also a small hut that marks the royal grave, all of which are named by the ancestor’s name, much like Kaiija shrine in Buhaya is also known as part of Rugomora Mahe’s gashani. Cruz notes that specific ancestors aref identified by particular trees, limited to three species, in front of which are partially buried clay pots “libations and offerings”, as occurs at the khokholo of Bahule (Cruz 2009:7) (Fig. 9); this is also a phenomenon seen at the base of Kya Rugomora tree in Buhaya, where an ancient ritual pot (c. 0-400 CE) has been documented to the early first millennium CE (Schmidt 1978).

Fig. 9

Specificity in naming by using ancestral names is found in a number of cultures in the larger region; the Nyakusa, for example, name a huge number of sacred trees for specific historical figures (e.g., Ellison 1996). In Buhaya named ancestral graves are marked with omuramura trees (a species commonly used for compound demarcation) carefully tended for more than a dozen generations in some locales. Precisely in the midst of the ancestral markings for Rugomora Mahe’s burial place, archaeological inquiry revealed not human remains but a ritual pit (Fig. 10D), stratified into different depositional activities unrelated to the agricultural or domestic activities. One of a suite of four such ritual structures in the immediate zone, this ritual pit is the ancestral place long memorialized by the careful maintenance of the “burial” marker.

Fig. 10

The final issue that I want to address in this section about finding, respecting, and explaining ancestors addresses circumstances where archaeological inquiry into the history of the ancestors is not desired by local communities. Many times people voice strong objections to any interference with ancestral space, holding such space to be sacred and inviolable. Cruz encountered exactly this response from her collaborators in Mozambique. Refusing to grant permission for any disturbance and even collection of surface materials has caused Cruz to reevaluate her research plan, adapting to the concerns and needs of the community. Each community will have its own needs, desires, and concerns. When I embarked upon excavation at Kaiija shrine at the Rugomora Mahe site 39 years ago significant culture change had already occurred: there was no spirit medium living on site, the royal compound had not been occupied for years, and local traditional and political authorities strongly sponsored inquiry as did local elders and religious leaders. Had it not been for the significant change in attitudes about ancestors at this site, we ironically would not have been privy to the significant knowledge about ancient African history resulting from archaeological inquiries conducted there.

Technology and the Ancestors

The third example I examine illustrates how the ancestors enter into archaeological inquiry completely outside of the spiritual settings of shrines and ceremonial displays as well as outside the quotidian settings that make up most of daily life. The cultural setting is iron smelting, industrial production that has widely recognized associations with invocations and prayers to the ancestors as well as specific sacrifices proscribed to ensure the blessings of the ancestors (Herbert 1993; Schmidt 1996, 1997, 2009). During an ethnoarchaeological study of Haya smelting practices (to better understand archaeological evidence for smelting in the region West of Lake Victoria), the Haya were careful to proffer prayers before initiating their first smelt, invoking the blessings of ancestral iron smelters. When the Haya smelters embarked upon a series of smelting operations in 1976, however, they experienced a number of difficulties, particularly with collapsing furnaces (Schmidt1997). Consulting diviners, they came to learn that their problems arose from two different sourcesthe bewitching of the furnace by someone with a grievance, and their failure to conduct blood sacrifices to the ancestors.

Much debate ensued, some of which entailed a discussion of whether they as smelters were first Christians or iron smelters (Schmidt 1997). Deciding that their first identity was to be traced to their heritage as iron smelters, they constructed a shrine for their ancestor Irungu, also the Bacwezi god responsible for all the resources of the bush used for iron smelting. After making appropriate offerings of coffee beans accompanied by the blood sacrifice of a chicken (Fig. 11), the smelting continued without incident (cf. Schmidt 1997 for more details). These activities mark a ritual process that was once common to other spiritual and healing contexts, such as propitiation of the ancestors through blood sacrifice before a spirit medium is called upon to consult ancestral spirits. In the context of this discussion, it is important to recognize that the ancestors are very much present and must be consulted before exploitation of iron ore, wood for charcoal, and the operation of iron smelting. Ancestors are an integral part of a technological process that is dependent on ancestral approval and assistance to ensure its success.

Fig. 11

The role of ancestors in iron smelting operations comes into more prominent light when ethnoarchaeologists observe and record ancestral rituals during iron smelting activities. Archaeologists are keenly interested in ritual as a transformative process. When rituals are interwoven into the technological process, they also leave behind material signatures that may possibly appear in the archaeological recordconfronting us with the often ephemeral materiality of such practices as well as ritual behaviors that deposit substantive evidence for concerns over the welfare of ancestors. The Barongo iron smelters who live near the southwestern part of Lake Victoria provide an excellent example how ancestors enter into smelting operations (Fig. 12), much more directly than among the Haya smelters (Schmidt 1996, 2009).

Fig. 12      Fig. 13

Ancestral presence and influence interpenetrate Barongo iron smelting. Ancestors play roles of approval or disapproval at the site where iron ore is mined, some 25 miles away from contemporary settlements. Once in Misema Forest, the head smelter chanted praises to the Bayovu, the ancestors of former iron workers whose work stations dot the landscape (Fig. 13). He asked the spirits for good luck in obtaining iron ore and in obtaining iron without the furnace breaking or failing; ill luck might befall the smelt if ancestral spirits thought that the iron ore had been stolen. While in the area during 1980, the head smelter and chief ritual assistant gathered special roots of the mususemi shrub from this ancestral ground for later ritual use in the furnacea concrete and spiritual link between the ancestors and the furnace.

The construction of the furnace includes a plethora of rituals, including the insertion of specialized medicinal cures for infertility as well as apotropaic drugs meant to interdict evil spirits, including possible angry ancestors. The most potent of these protections against angry ancestors is the final sealing agenta white powder called rukago, which in the healing repertoire of Bazinza healers acts against spirit afflictions (Schmidt 2009) (Fig. 14). The morning of the first smelt the head smelter brought a spear and a pot to the west of the smelting house where he prayed for the spirits of his father and grandfather; he dropped these objects after having established ancestral contact and immediately excavated a small trench on the south side of the smelting house, into which he placed the leaves of the muhingura plantalso a critical component of the ritual medicinal package earlier placed in the innermost reaches of the ritual pit, the “womb” within the furnace (Schmidt 1996).

Fig. 14

Perhaps the most important ritual calling on protection against displeased ancestors is first-smelt ritual that took on several different expressionsthe spitting of beer on the ground and furnace by the head smelter’s wife and one or more smelting assistants. While this performance is drawn from a nuptial fertility rite, beer spitting is also the propitiation of ancestors among many of those smelters who identified themselves as Bazinza. This emic meaning is further amplified by a spirit medium’s kizingo cap that is worn by the head smelter’s wife and by an assistant smelterused only when communicating with ancestral spiritsas the furnace operations begin. Once bellows pumping begins, a number of different ritual treatments may occur. On two occasionsnot predictableblood sacrifice occurred, with the spraying of goat’s blood on the furnace, all equipment, and even the smelters (Schmidt 1996, 2009) (Fig. 15). Though ancestors were not specifically invoked during this procedure, it is recognizablethrough comparison to its Haya counterpartas an integral part of the suite of performances taken to propitiate ancestral spirits during the start up phase of smelting.

Fig. 15

Once smelting is under way, there are many additional treatments involving ancestors, including the use of a ritual tuberkalinzaplaced on the top of the charge (and also used as part of the internal medicinal package deep within the furnace). As the tuber disappeared when the flames burst from below, the smelters chanted “kasinge” or, thank you, to the ancestors. At other times ritual solutions varied according the needs and actions of the smelters when other ancestral concerns come to light. In one such ritual treatmentmuch like the kalinza tuberthe components of a package were displayed (but not explained), including an iron needle (Fig. 23). We know from Bjerke’s (1981) study of Bazinza healing practices that needles are used in exorcism rituals, again affirming the idea that unfriendly spirits must be expunged at multiple stages of the smelt. As the smelt progressed, ancestral agents again guided ritual behavior when, for example, the head smelter stopped bellows pumping so he could insert a tuber called mususemi into the tuyere (Fig. 16). Taken from a shrub, earlier mentioned, on land belonging to their Boyovu ancestors some 25 miles away, the root was stripped of its bark, which was then mixed with salt and placed within the tuyere with the head smelter chanting, “Nakurungira tata”  or, “He has fed you well” (a direct ancestral referencetata is father). Much like the beer spitting ritual, this is a fertility ritual as well as an appeasement of direct ancestor(s) who are being “fed”multiple tropes being enacted simultaneously.

Fig. 16

The vital role of ancestors in iron smelting operations has become obscured however by a rich literature in which iron smelting technology is represented as part of the productive economy widely associated with rituals and proscriptive taboos that construct it as a human reproductive activity, a female furnace that eventually gives birth to multiple fetuses (Childs and Killick 1992; Herbert 1993; Schmidt 1997). The ritual transformation of the iron smelting furnace into a fecund bride is a varied and complicated ritual cycle that has been incompletely represented until recently (Schmidt 2009). Earlier symbolic studies (Childs 1991; Childs and Killick 1993), with the exception of Herbert (1993), privilege the reproductive paradigm and have obscured representations of the complex role of ancestors. There is no denying that human reproduction figures prominently in the earliest symbolic treatments of iron smelting furnaces, but reproduction is only one of three major ritual concerns that come to inform the transformational domain in which iron smelting is embeddedothers being protection by apotropaic objects against evil forces such as sorcery and witchcraft and other devices that assuage, propitiate, and consult ancestors as well as protect one from angry ancestors.

These ritual treatments in iron smelting appear to arise independently through time, based as they are on different social and economic settings. It is critical that archaeologists possess the conceptual tools to distinguish among these different ritual treatments when the material residues are documented during excavations, especially since ancestorssuch a large part of the iron smelting processhave been obscured in the literature and archaeological record.

It is reasonable to believe that concerns with the spiritual world of the ancestors has been long-standing, given the widespread distribution of invocations to ancestors as well as sacrifice and ritual performance on their behalf (Herbert 1993; Schmidt 1997); nonetheless, archaeological evidence for this part of the tripartite ritual complex is relatively sparsea case where the ethnographic evidence far outstrips our archaeological knowledge.  Iron smelters of many societies are luminal brokers negotiating among the spirit world, the world of healing, the world of magic, and the world of material production. As such they leave behind material metaphors, signs of their actionsas they act out ritual tropesto appease and to solicit ancestors (Schmidt 2009). No matter how well known and deeply nuanced, ritual veneration of ancestors has a low archaeological visibilityas opposed to the plethora of ethnographic casesthat makes recognition of the central role of ancestors in the smelting process opaque.

Yet opaqueness disappears when we examine the strong evidence for ancestral offerings in the furnaces of SW Tanzania excavated by Bertram Mapunda (1995). In a series of furnace excavations, Mapunda recorded large bundles of medicinal and apotropaic devices (called vizimbi when in Fipa furnaces) intended to rebuff evil forces from both the more recent large lilungu funace of the Fipa, but also older, much shorter furnaces called katukutu furnaces dated to the 15th and 16th centuries (wherein they devices were called nchinji). In the katukutu furnaces the sharpened nchinji sticks were sealed in the majority of cases by partial pots coated with black sticky clay (Fig. 17). Purposely split in half for this purpose, such local treatment of pots holds that they are used to exorcise evil spirits in the presence of termite moundsprecisely where these furnaces were situated (Schmidt and Mapunda 1997).

Fig. 17

The katukutu technology took form in the mid-2nd millennium CE in southwestern Tanzania under conditions of intense social and economic competition and interaction with the Fipa smelters, who later used similar solutions for interdicting evil forces. These conditions meant the rapid engagement of multiple knowledges, of interaction of people from multiple social and economic backgrounds with a plurality of ancestral identities that likely required treatment of multiple lines of ancestors in the ritual bundle lodged within the furnace. Our knowledge of how Haya smelters (Schmidt 1997) spontaneously re-invented apotropaic treatments when threatened by witchcraft in new social and physical environments provides another lens for understanding how such innovations occur in the implementation of ritual tropes that carry material consequences.

Mapunda’s excavations show a wide variety of wooden objects in the furnaces’ ritual pitsvirtually a direct mirroring of a well documented ethnographic case of the Barongo, just to the north. As we have seen, the Barongo use multiple tropic devices in ritual performances to act on their furnaces, ensure protection, obtain a full reproductive cycle, and satisfy full consultation with ancestors. The apotropaic practices that address both evil forces and ancestors came into popularity during the 19th century when the Barongo iron smelters begin to experience social chaos and economic disruption during slave raiding; they hid themselves in the forest and conducted rituals with a full panoply of medicinal and magical devices to appease the ancestors and gain their approval, but also to ward off hostile attemptslater repeated by colonial authorities””to bring harm to both furnaces and smelters, who feared for their lives (Schmidt 1996, 2009). Of great significance are the multiethnic identities of the Barongo, coming from a variety of backgrounds in which no one ancestor or one group of ancestors might be involved, but rather a host of ancestral lines (Schmidt 2009). Such plurality continues to this day. The proliferation of ancestral concerns and the material and spiritual solutions innovated by the Barongo, the ritual action and bricolage that developed to address these matterspresent strong inference for materiality associated with apotropaic devices aimed at evil forces as well as ancestors in other multi-ethnic and stressful environments, such as 15th and 16th century interactions in southwestern Tanzania (Mapunda 1995). This archaeological record shows that the spiritual world of ancestors is not limited to the quotidian spaces of the households or the scared spaces of shrines or ceremonial rooms or other spaces. Rather, ancestors interpenetrate all spheres of life, even the forge and the hot and dangerous work of iron smelting.

Self-Reflection and Final Thoughts about the Future

Before I move from issues of spirituality, ancestors, and archaeology, I believe that it is important to understand how my personal awareness and understanding of ancestral presence and importance developed during my practice of archaeology. During my investigations of ancestral places and histories at the Rugomora Mahe site and its related suite of sites (Schmidt 1978, 2006), I adhered to expectations for blood sacrifice of a goat to propitiate the ancestors. In West Africa, this often takes the form of pouring libations for the ancestors as is also the case in Mozambique (Cruz 2009), though this particular practice is not usual in East and Central Africa where the sacrifice of a goat or sheep is held as adequate homage to the ancestors. During our interactions with the Kaiija tree shrine and related shrines, these sacrifices were conducted multiple times, with open dialogue about ancestral needs with the local people. When I returned in 1980 to Ikimbo village (Kanyangereko section of Maruku kingdom) in Buhaya, I delayed propitiating the ancestors. Failing to find the required goat, having satisfied the same ancestral spirits on earlier occasions, and working under strict time constraints, I lapsed in my responsibilities. This proved a significant error, as I was keenly aware that Kya Rugomora tree was one of the places where a leg of the iron tower built to the heavens by King Rugomora Mahe had been placed. The ancestors certainly deserved equal treatment and respect, given this was an ancestral location for the Bayango, the pre-dynastic rulers.

As we began our excavations to affirm the great antiquity of this important Bacwezi shrine, we obtained sparse results and were discouraged by the absence of evidence that would attest to ritual activities beyond those we had documented in 1970. One day as we prepared for work, I was taking notes crouched next to a banana tree. My excavation team was gathered in a large semi-circle around me as I handed out the day’s assignments. Looking up to answer a question, I found a huge black snake just a half meter away from my face. It had slithered through the circle without being noticed, and had approached merising up and stopping just short of my face. What I remember best are the large shinning yellow eyes in a head that was as big as my fist. I froze, speechless and unable to move. “Snake!!” yelled my nephew, three heart beats later. At that shout, I sprang from my crouch and landed about 3 meters away. The snake, nonplussed by all the commotion, slowly slid away under a hedge. It measured about 5 meters in length and was as thick as a man’s upper arm. “Don’t touch it!”, one of the workmen warned those starting to shout that it had to be killed. “It is Enkoratima, the snake of Rugomora Mahe.”

Part of the ancestral belief system associated with kingship in this region holds that the spirit of the dead king will live on in this particular species of snake, an Enkoratima. Its presence ensures that the king’s spirit may leave the snake to possess the spirit medium when necessary. To kill such a creature is to kill the vessel in which the spirit of a very important ancestor resides. Almost immediately, I recognized that I had experienced a visitation, that this marked spiritual agency, and that the ancestral spirit of King Rugomora Mahe had paid me a visit perhaps to communicate concern over my presence or my failure to propitiate the spirits. A nearby resident, familiar with these beliefs, warned me, “Tomorrow his children will come to visit.” Though shaken by the experience and prepared to accept the idea of a visitation, I was not adequately prepared for the next day, when true to the word of the concerned neighbor, several additional Enkoratimas visited other member of the team, one dropping from the shrine tree into an excavation pit immediately behind one of the crew. Everyone stopped work, and suddenly I realized that we had not built those bridges to the ancestors that so concerned Bassey Andah,

I gave instructions to start back-filling the excavations immediately and to make preparations for departure. Nothing justified further disruption. It was a message of great clarity, without ambiguity. No science, no quest for further affirmation of initial earlier findings could be sustained against such powerful spiritual forces. Humbled and grateful that I had survived this poignant lesson on respect for the ancestors, I often think back to this experience as a telling moment in my life. It was a moment when, like an epiphany, I came to realize that archaeology indeed was inextricably interwoven with the spiritual world, particularly the world of the ancestors. This was accompanied by an awareness that the scientific practice of archaeology was not only dependent upon the agreement and guidance of the ancestors but also a natural part of unveiling the central roles that ancestors played and continue to play in the daily lives of people. As Dietler and Herbich (2007:3) succinctly put it, “the ancestors have real agency”.

We have seen from the Barongo iron smelting example how ancestors are active agents throughout diverse activities ranging from mining ore, to sealing the potent apotropaic and fertility drugs inside the furnace womb, to initiating the first smelt, and to making sure the ancestors are fed during the smelt. These are continuous, active, and engaged performances in which the different and often unpredictable agencies of ancestors demand responsive ritual bricolage, altered tempos, and different materialized actions during iron smelting. In much the same way, the role of ancestors in Luo daily life (Deitler and Herbich 2007) show us that we must be sensitive to and familiar with the roles that ancestors take in guiding, limiting, and insisting on the manner in which behavior and its materialization take shape.

I have shared my personal revelations about how I developed a heightened awareness and respect for ancestors, much in the vein of what Ruth Behar (1996) and others have done before me in anthropologyto unveil new knowledge and understandings of how lives are lived through ancestors. In ways parallel to Bassey Andah, I embrace the idea we are dependent upon the ancestors and that ancestors are active agents in our lives and other lives. They are us. They passed on their DNA, our characters, and our deep time cultural history; they are forever our teachers. Bassey Andah had it right: we can either incorporate the ancestors or we can continue to erase the spiritual world from archaeology, proclaiming that science has the answers and that the future has no past.


My deepest thanks go to the Organizing Committee of the 9th Bassey Andah Memorial Lecture. They graciously provided me with pleasant surroundings and a marvelous and appreciative audience. I am indebted to Prof. Claire Smith, President of the World Archaeological Congress (WAC), and to Prof. C. A. Folorunso, Vice President of WAC and Head of the Department of Archaeology at Ibadan, for their foresight in sponsoring an Archaeologists Without Borders workshop, a new WAC program that was scheduled to overlap with the Memorial Lecture. By bringing archaeologists from distant universities, the AWB workshop (along with the Memorial Lecture) exposed Nigerian archaeologists to methods and theories being practiced elsewhere in Africa. I am particularly grateful for the helpfulness and kindnesses of two colleaguesNeil Norman and Dores Cruzwho made available to me unpublished material and illustrations. My thanks also go to Jonathan Walz and Rachel Iannelli at the University of Florida; their comments were most helpful during revisions. Thanks also to Michael Dietler and Ingrid Herbich for providing a helpful unpublished paper. Finally, I join my colleagues at the University of Ibadan in thanking the Vice Chancellor, Prof. Olufemi Adebisi Bamiro, for his enthusiastic support for publishing this memorial lecture honoring Bassey Andah.

References Cited

Achebe. C. 2007. Dead Man’s Path. In Literature: A Pocket Anthology, ed. R. S. Gwynn, pp. 247-250. New York: Pearson Education, Inc.

Agorsah, K. 2009. The Shrines of Kormantse, Ghana and the African Diaspora. Paper presented at the Society for Historical Archaeology annual meetings in Toronto, January 10.

Andah, B. 1990. Prologue. In Cultural Resource Management: An African Dimension, ed. B. W. Andah. West African Journal of Archaeology 20:2-8.

_____. 1995. Studying African Societies in Cultural Context. In Making Alternative History: The Practice of Archaeology and History in Non-Western Settings, eds. P. R. Schmidt and T. C. Patterson pp. 149-182. Santa Fe: SAR Press.

Behar, R. The Vulnerable Observer: Anthropology that Breaks Your Heart. Boston: Beacon Press.

Bjerke, S. 1981. Religion and Misfortune: The Bacwezi Complex and Other Spirit Cults of the Zinza of Northwestern Tanzania, Oslo: Universitetsforlaget.

Blier, S. P. 1991. King Glele of Dahomè Part Two: Dynasty and Destiny. African Arts. 24(1):44-103.

_____. 1995. Vodun: West African Roots of Vodou. In Sacred Arts of Haitian Vodou, ed. D. J. Cosentino, pp. 60-87. Los Angeles: UCLA Fowler Museum of Cultural History.

Childs, S. T. 1991. Technology, Style, and Iron Smelting Furnaces in Bantu-Speaking Africa. Journal of Anthropological Archaeology 10:332-59.

Childs, S. T. and Killick, D. 1993. Indigenous African Metallurgy: Nature and Culture. Annual Review of Anthropology 22: 31737.

Crossland, Z. 2001. Time and the Ancestors: Landscape Survey in the Andrantsay Region of Madagascar. Antiquity 75(290):825-836.

Cruz, M. D. 2006. Archaeology, Memory and Postcolony in Manjacaze (Gaza Province, Mozambique). Paper presented to the Association of Southern African Professional Archaeologists at the Bi-Annual Conference at Pretoria, April.

_____. 2009. The Nature of Culture: Sites, Ancestors and Trees in the Archaeology of Southern Mozambique. Paper presented at the Society for Historical Archaeology annual meetings, Toronto, January 9.

Dietler, M. and Herbich. I. 2007. The Agency of Ancestors: Reflections on an ethnographic case from Africa. Paper presented at the Society for American Archaeology annual meetings, Austin, Texas, April 26.

Drewal, H. J. 1996. Mami Wata Shrines: Exotica and the Construction of Self. In African Material Culture, eds. M. J. Arnoldi, C. M. Geary, and K. Hardin, pp. 306-333. Bloomington: Indiana University Press.

Ellison, J. 1996. The Future of African Archaeology: Re-Writing History. African Archaeological Review 13(1):5-7.

Fennell, C. 2007a. Crossroads and Cosmologies: Diasporas and Ethnogenesis in the New World. Gainesville: University Press of Florida.

_____. 2007b. BaKongo Identity and Symbolic Expression in the Americas. In Archaeology of Atlantic Africa and the African Diaspora, eds. A. Ogundiran and T. Falola, pp. 199-232. Bloomington: Indiana University Press.

Herbert, E. 1993. Iron, Gender, and Power: Rituals of Transformation in Africa Societies, Bloomington: Indiana University Press.

Insoll, T. 2007. “Natural” or “Human” Spaces? Tallensi Sacred Groves and Shrines and their Potential Implications for Aspects of Northern European Prehistory and Phenomenological Interpretation. Norwegian Archaeological Review 40: 138″“58.

_____. 2008. Negotiating the Archaeology of Destiny: An Exploration of Interpretive Possibilities through Tallensi Shrines. Journal of Social Archaeology 8(3):380-403.

Insoll, T., Kankpeyeng, B. and MacLean. R. Forthcoming. The Archaeology of Shrines among the Tallensi of Northern Ghana: Materiality and Interpretative Relevance. In Shrines in African Societies, eds. A. Dawson and P. Shinnie. Calgary: University of Calgary Press.

Irele, A. 1991. The African Scholar: Is black Africa Entering the Dark Ages of Scholarship? Transition 51:56-69.

Law, R. 1989. The Slave-Trader as Historian: Robert Norris and the History of Dahomey. History in Africa. 16:219-35.

_____. 1991. The Slave Coast of West Africa, 1550-1750: The Impact of the Atlantic Slave Trade on an African Society. Oxford: Clarendon Press.

Mapunda, B. B. (1995). An Archaeological View of the History and Variation of Iron Working in South-western Tanzania, PhD dissertation, University of Florida, Gainesville.

Norman. N. L. 2009a. Powerful Pots, Humbling Holes, and Regional Ritual Processes: Towards an Archaeology of Huedan Vodun, ca. 1650-1727. African Archaeological Review 26(3):187-218.

_____. 2009b.Tracing Sacred Routes: Cosmograms, Cosmologies, and Material Cross-Roads in the African Atlantic World. Paper presented at the Annual Meeting of the Society for Historical Archaeology. Toronto, Canada, January 8-11, 2009.

Ogundrian, A. 2007. Living in the Shadow of the Atlantic World: History and Material Life in a Yoruba-Edo Hinterland, ca. 1600-1750. In Archaeology of Atlantic Africa and the African Diaspora, eds. A. Ogundiran, A. and T. Falola, pp. 77-99. Bloomington: Indiana University Press.

_____. 2009. Inventing Symbols, Constructing Self, Reproducing Community: On the Materiality of Culture in Mid-Atlantic Age Yorubaland. Paper presented at “Excavating the Past: Archaeological Perspectives on Black Atlantic Regional Networks”, a conference honoring Merrick Posnanaky, April 3-4, William Clark Memorial Library, Los Angeles.

Rush, D. 1999. Eternal Potential Chromolithographs in Vodunland. African Arts. 32(4): 60-75, 94-6.

Schmidt, P. R. 1978. Historical Archaeology: A Structural Approach in and African Culture. Westport, CT: Greenwood Press.

_____. 1995. Using Archaeology to Remake History in Africa. In Making Alternative Histories: The Practice of Archaeology and History in Non-Western Settings, eds. P. R. Schmidt and T. C. Patterson, pp. 119-148. Santa Fe: SAR Press.

_____. 1996. Reconfiguring the Barongo: Reproductive Symbolism and Reproduction among a Work Association of Iron Smelters. In The Culture and Technology of African Iron Production, ed. P. R. Schmidt, pp. 74-127.Gainesville: University Press of Florida.

_____. 1997. Iron Technology in East Africa: Symbolism, Science, and Archaeology. Bloomington: Indiana University Press.

_____. 2006. Historical Archaeology in Africa: Representation, Social Memory, and Oral Traditions. Lanham, MD: AltaMira Press.

_____. 2009. Tropes, Materiality, and Ritual Embodiment of African Iron Smelting Furnaces as Human Figures. Journal of Archaeological Method and Theory 16:262-282.

_____. Forthcoming. Social Memory and Trauma in NW Tanzania: Organic, Spontaneous Community Collaboration. Journal of Social Archaeology.

Schmidt, P. R., and B. B. Mapunda. 1997. Reading Ideology in the African Archaeological Record. Journal of Anthropological Archaeology 16:73-102.

Stahl, A. B. 2008. Dogs, pythons, pots, and beads: The Dynamics of Shrines and Sacrificial
Practices in Banda, Ghana, 1400″“1900 CE. In Memory Works: Archaeologies of Material Practices, eds. B. Mills and B. Walker, pp. 159-186. Santa Fe: SAR Press.


Fig. 1.  Historic map of Hueda, with Savi indicated. (Norman 2009a:Fig. 2; after Law 1991:232)

Fig. 2.  Shallow bowl from an excavated shrine, incised with a cosmogram (weke) and holding a human skull burial (Norman 2009a:Fig. 14).

Fig. 3.  Jar with skull burial in a pit beneath a house floor at locus 7, Savi (Norman 2009a:Fig. 20).

Fig. 4.  Map of Tanzania with Buhaya, along the hills and the western shores of Lake Victoria, marked in black.

Fig. 5 Map of Kanyangereko peninsula showing shrines, key sites, and landscape names associated with human reproduction.

Fig. 6.  Kaiija (place of the forge), a huge shrine tree that once memorialized early iron working dating to the first millennium BCE.

Fig. 7.  Map of Rugomora Mahe’s royal palace compound in Katuruka, with incorporated shrines, e.g., the king’s grave and gashani, where the spirit house of the ancestral king was located.

Fig. 8.  Map of Gaza province of Mozambique, where ritual propitiation of the ancestors is a deep practice running from 19th century up to the present day. Map by Dores Cruz.

Fig. 9.  Libations being poured near trees at the khokholo of Bahule in the Gaza Province of Mozambique; blood sacrifice of chickens, also depicted here, occur at these sites. Photo by Dores Cruz.

Fig. 10.  A cluster of highly specialized pits near the gashani of Rugomora Mahe marked coeval ritual activities during the mid-17the century; the pit illustrated in Fig. 10D was excavated below the omuramura trees marking the king’s grave, which in this instance was a ritual, symbolic statement and did not contain human remains.

Fig. 11.  Smelters in Nyungwe village, northeastern Buhaya, in 1976 conduct a ritual sacrifice of a chicken to Irungu, a Bacwezi ancestral spirit whose spirit house is behind the arms of the head smelter on the left.

Fig. 12.  Map of Tanzania; the Barongo iron smelters are found south of Lake Victoria.

Fig. 13.  A Barongo smelting house to the southwest of Lake Victoria in 1979; structures such as these proliferated in the forests south of Lake Victoria during the 19th century.

Fig. 14.  The Barongo head smelter sealing an iron smelting furnace, immediately before applying rukago powder, a white ashy substance applied in the form of a cosmogram, suggesting the more widespread and diverse use of cosmogram symbolism for ancestral interdictions than previously imagined.

Fig. 15. Barongo blood sacrifice coats the furnace, tools, and even participants with goat’s blood; this is part of the propitiation and direct appeasement of ancestors as well as a potent reproductive metaphor (see Schmidt 2009).

Fig. 16. The Barongo head iron smelter inserting a piece of mususemi root from ancestral lands to feed the ancestors, and being mixed with salt, also a fertility ritual (see Schmidt 2009 for more on salt and fertility).

Fig. 17. A profile of a katukutu furnace excavated by Bertram Mapunda (1995) East of Lake Tanganyika; the sharpened nchinji sticks were often sealed by partial pots coated with black sticky clay, treatments intended to exorcise evil (others’ ancestral) spirits near termite hills. Drawing by Bertram Mapunda.