IN SEARCH OF HOMELAND: BANTU EXPANSION AS  

PRE-MODERN AFRICAN DIASPORA 

 

By 

 

Blair Rose Zaid 

 

 

 

 

 

 

 

A DISSERTATION 

Submitted to 

Michigan State University 

in partial fulfillment of the requirements 

for the degree of 

 

African American and African Studies— Doctor of Philosophy 

Anthropology—Dual Major 

 

2019 

 

 

 

IN SEARCH OF HOMELAND: BANTU EXPANSION AS PRE-MODERN AFRICAN 

ABSTRACT 

 

DIASPORA 

 

By 

Blair Rose Zaid 

 

The expansion of Bantu speaking people is one of the longest periods of premodern 

human migration. Despite a century of research on their expansion into Central Africa, little is 

known about how Bantu speaking groups successfully traversed the varied landscapes of the 

Congo River Basin especially amid the cultural and environmental shifts that characterize the 1st 

millennium B.C. This study addresses the fundamental question of how and why was Bantu 

Expansion into Central Africa during the Early Iron Age successful? To answer this research 

question, I examined published and unpublished archaeological accounts of sites located across 

Central Africa from western Cameroon to southwestern Democratic Republic of the Congo. I 

then categorized the presences and absences of the archaeological material from each site to map 

when and where artifacts emerged during the Early Iron Age, ca 1000 B.C. to 200 A.D. Then I 

developed a framework for examining premodern African Diasporas which prioritizes massive, 

long term dispersal and the interregional connections such dispersals produce. Through 

analyzing an interregional network of shared cultural knowledge throughout the Equatorial 

Forest between 1000 B.C. and 200 A.D. I illustrate that Early Iron Age Bantu speakers were 

actively constructing connections to their homeland via the trade, technological, settlement, and 

aesthetic choices. The similarities and dissimilarities reveal an intentional pattern of interregional 

connection, previously thought to be discontinuous and disjointed. This research ultimately 

reveals that the patterning of behaviors strengthened their common local and regional identities 

and thus gave way for continuous reconnection and cultivation of the Bantu homeland. 

 

 

Chapter One introduces the Bantu Phenomenon, the major models that illustrate cultural 

change, and the research questions that shape this dissertation. Chapter Two situates the research 

within major theories of cultural change in pre-historic Africa and outlines the pre-modern 

African Diaspora framework to explore mechanisms of interregional change given the duration 

and magnitude of the expansion of Bantu speakers during the period. Chapter Three explains the 

use of the existing published record of archaeological materials as the primary source of 

information for the dissertation. Chapter Four then discusses the development of the Early Iron 

Age package in the Bamenda Grasslands and the Northern Plateau of south western Cameroon. 

Chapter Five reports on the archaeological materials recovered from the Forest interior and 

Lower Congo serving as illustrations of the dispersion of the Early Iron Age package throughout 

the Congo River Basin. Chapter 6 presents the analysis of when and where behaviors arrived 

across these sites and the inter and intraregional patterns that emerge between the cases. Chapter 

7 concludes the dissertation with a re-construction of Bantu expansion as a premodern African 

Diaspora. This re-construction reveals the connections between the Northern Plateau, namely 

Obobogo and the rest of the basin as indicators for the establishment of the Bantu Diaspora that 

would persist for the next two millennia. Lastly, I suggest that future research would benefit 

from paying attention to both the continuities and discontinuities of assemblages at the regional 

level and how those connections change across space and time. This research contributes to the 

scholarly engagement with the origins of the Bantu Phenomenon and how it endured across the 

variety of environmental, cultural, and technological conditions of the Central African Early Iron 

Age. 

 

 

 

 
 
 
 
 
 
 
 
 
 
 
 
 

 

 
 
 
 
 
 
 

 

This dissertation is dedicated to the memory of my grandfather James William Starnes and  

my mother Darlene Elaine Starnes. 

Love Always, Blair Rose

iv 

ACKNOWLEDGEMENTS 

 

 

 

First, I would like to express deep appreciation to those who’ve directly impacted the 

execution and completion of this dissertation project; the African American and African Studies 

Program, the Department of Anthropology, the African Studies Center, and the Graduate School 

at Michigan State University for the dissertation research support. To my co-chairs Dr. Lynne 

Goldstein, Dr. Yomaira Figueroa, and lead advisor Dr. Pierre de Maret for never losing faith in 

my ability to complete this process and contribute to archaeology and the study of the African 

Diaspora. To my uncle and aunt Mr. and Mrs. Andre and Kimberlyn Starnes and my dear, dear 

friends Dr. Michael and Mrs. Sylvia Galaty whose travel and writing support was paramount to 

the completion of this project. To my field research assistant Ms. Charlie Dobson whose faith, 

conviction, and compassion was indispensable in Belgium. To the archaeologists and staff at the 

Royal Museum of Central Africa in Tervuren, Belgium who provided real-time feedback, 

support, and guidance during data collection. I also want to acknowledge my committee 

members from all phases of the process, for their contributions along the way. 

 

Second, I would like to recognize those who’ve been instrumental in the dual degree 

process in general: The College of Arts and Letters, The College of Social Science, the Center 

for Latin American and Caribbean Studies, the MSU Alliance for Graduate Education in the 

Professoriate, and Ms. Mavinidi Mitogo for the funding, mentoring, guidance, housing, and 

support over the course of my undergraduate, master’s and doctoral degrees at Michigan State 

University.  

 

To my base; Drs Amy Michael, Emily Riley, Adrianne Daggett, and Charlotte Cable, 

whose own pursuit as women academics and archaeologists kept me striving and focused 

v 

throughout this process. To Queayna Lynā Battle, Jasmine Denise Oke, I-Sha Nicole Reeves, 

who kept me grounded and lifted me up for the last 15 years. To Pastors Sean and Tayana 

Holland and the Epicenter of Worship church family, the Kraus-Petterson and Carr families, and 

a host of community supporters who provided meals, encouragement, and prayers for a strong 

finish. 

 

Also, my family who have literally taken trains, plains, and camel rides, to support this 

project. To my grandmother Mrs. Carolyn B. Starnes, my dad, Pharius Pacouloute, and my 

siblings, Alicia Starnes, Carolyn Murphy, and Stephen Starnes for having my back along this 

path; my nephews Gabriel Starnes-Wright and William Starnes-Dortch for keeping my heart full 

and my mind sharp; to my Aunt Laurie and Uncle James Hargrave, Aunt Linda Murphy and the 

many aunts, uncles, and cousins who’ve kept me motivated and kept my spirits high. To Shanti 

and my baby girl Kiluna Rosali Zaid, thank you for giving me the space to work, keeping me 

focused on where I came from, and reminding me of the infinite possibilities of things to come. 

To my mother Darlene Elaine Starnes for her endless encouragement, wisdom, and food until her 

last intercontinental call of support to Morocco days before her passing in 2017, I thank you. 

 

I end with giving all praises to God, for the focus and strength to complete this journey. 

My heart is full of gratefulness for allowing me to walk in my purpose and providing the ever-

growing community of support along the way. With much gratitude, Blair Rose Zaid. 

 

 

vi 

PREFACE 

 

 

 

In 2002, my mother Darlene Elaine Starnes and I arrived at Michigan State University for 

the freshman orientation program. She encouraged me to explore the world and learn new things 

but never forget my hometown New Jersey roots. I would eventually travel to the Caribbean, 

Europe, and Africa always considering how my family, community, and background prepared 

me for these new experiences. Through wondering about how my past continuously effected my 

present, I began to consider how the past may likewise still influence modern times.  Whether I 

explored historic coffee plantations in Cuba or rock art paintings in the Drakensburg Mountains 

of Lesotho, I continuously wondered about how these cultures that existed hundreds and maybe 

thousands of years ago impact our day to day lives. 

 

This wonder led me to pursue a dual major in African American and African Studies 

(AAAS) and Anthropology at MSU. The former program provided a clear perspective and 

framework for understanding a group of people from their particular social, historical, and racial 

vantage point and the latter offered archaeology, a science that examines the past and a focus on 

culture, or the ways human groups behave and act as a group. As such, my project began by 

exploring the African presence in the Americas to understand the ways culture changes within 

the modern African Diaspora. However, those projects would eventually evolve to consider how 

and when patterns of cultural change identified within the African Diaspora first began to 

emerge. I wondered if they were strictly products of modernity bought on by the conditions of 

trans-oceanic trade of enslave people and its aftermaths or where these patterns of cultural 

vii 

change products of some shared Africaness that existed from the beginning of time? Or were 

there other possibilities scholar simple had yet to consider? 

 

As my questions evolved, I moved my research focus to central Africa, and specifically 

the Kongo Kingdom to better understand mechanisms of culture change that may have been 

present prior to 15th century. Given the Kongo’s extensive historic and cultural literature, I 

assumed their archaeological research would be just as robust. However, when I contacted Pierre 

de Maret in 2012, little did I know that they were beginning on one of the few archaeological 

projects on the historic period in Central Africa. It was only after learning KiKongo and working 

on the KongoKingdom field project with the University of Ghent that I realized that, perhaps, the 

questions of the impact of the ancient world, may indeed need to start with the ancient world. 

 

It was at this point that I began the specific study presented here: understanding Bantu 

Expansion as premodern African Diaspora. I attempted an equally AAAS and archaeology 

project to identify what the African Diaspora would look like in the premodern era and why it is 

important to examine the Bantu Phenomenon as such. This project thus sought to traverse 

disciplinary boundaries to identify potential pathways for interdisciplinary work between 

AAAS/Black Studies and archaeology. This project thus represents a starting point in the 

reexamination of Bantu Expansion and the conceptual ground work for framing the Bantu 

Diaspora. My hope is that it reinvigorates interest in the distant past and how it continues to 

shape today.  

 

 

 

 

viii 

 

 

 

 

TABLE OF CONTENTS 

 
 

 

LIST OF TABLES ......................................................................................................................... xi 
 
LIST OF FIGURES ...................................................................................................................... xii 
 
CHAPTER 1: INTRODUCTION ................................................................................................... 1 
I. The Bantu Phenomenon........................................................................................................... 2 
II. Understanding the Bantu Phenomenon .................................................................................. 3 
III. The problem, research question, and research design ........................................................... 8 
IV. Dissertation outline ............................................................................................................. 11 

 
CHAPTER 2: LITERATURE REVIEW ...................................................................................... 13 
I. Introduction ........................................................................................................................... 13 
II. Frontier-metropole hypotheses ............................................................................................. 13 
III. Intra-frontier hypotheses ..................................................................................................... 15 
IV. Environmental hypotheses .................................................................................................. 18 
V. Frontier cultural and identity construction frameworks ....................................................... 18 
VI. Summary of literature review ............................................................................................. 20 

 
CHAPTER 3: METHODS ............................................................................................................ 21 
I. Introduction ........................................................................................................................... 21 
II. Data collection...................................................................................................................... 21 
III. Description of research materials........................................................................................ 25 
IV. Data extraction and analysis ............................................................................................... 28 
V. Research challenges and limitations .................................................................................... 32 
VI. Summary of research methods............................................................................................ 33 

 
CHAPTER 4: CASE 1: THE BAMENDA GRASSLANDS AND CASE 2: THE NORTHERN 
PLATEAU .................................................................................................................................... 35  
    I. Introduction ........................................................................................................................... 35 
II. Case 1: The Bamenda Grasslands ........................................................................................ 36 
III. Case 1 Summary ................................................................................................................. 41 
IV. Case 2: The Northern Plateau ............................................................................................. 45 
V. Case 2 Summary .................................................................................................................. 50 
VI. Chapter Summary ............................................................................................................... 55 

 
CHAPTER 5: CASE 3: THE FOREST INTERIOR AND CASE 4: THE LOWER CONGO .... 56 
I. Introduction ........................................................................................................................... 56 
II. Case 3: The Forest Interior ................................................................................................... 57 
III. Case 3 Summary ................................................................................................................. 63 
IV. Case 4: The Lower Congo .................................................................................................. 70 
V. Case 4 Summary .................................................................................................................. 74 

ix 

VI. Chapter Summary ............................................................................................................... 77 

 
CHAPTER 6: ANALYSIS AND DISCUSSION ......................................................................... 78 
I. Data analysis .......................................................................................................................... 78 
II. Summary of analysis ............................................................................................................ 82 
III. Discussion ........................................................................................................................... 84 

 
CHAPTER 7: CONCLUSIONS AND SUGGESTIONS FOR FUTURE RESEARCH .............. 87 
I. Conclusions ........................................................................................................................... 87 
II. Bantu Expansion as Premodern African Diaspora: a culture history revised ...................... 88 
III. Suggestions for future research ........................................................................................... 89 

 
APPENDICES .............................................................................................................................. 92 
APPENDIX A: THE BAMENDA GRASSLANDS TABLE OF DATA SOURCES ............. 93 
APPENDIX B: THE NORTHERN PLATEAU TABLE OF DATA SOURCES .................... 95 
APPENDIX C: FOREST INTERIOR TABLE OF DATA SOURCES ................................... 96 
APPENDIX D: THE LOWER CONGO TABLE OF DATA SOURCES ............................... 97 

 
BIBLIOGRAPHY ......................................................................................................................... 98 
 
 

 

x 

LIST OF TABLES 

 

Table 1: Premodern African Diaspora Framework Spreadsheet…………………………………32 

Table 2: Shum Laka Early Iron Age Attributes………………………………………………….43 

Table 3: The Northern Plateau Early Iron Age Attributes……………………………………….54 

Table 4: Imbonga Early Iron Age Attributes…………………………………………………….66 

Table 5: Batalimo-Maluba Early Age Attributes………………………………………………...67 

Table 6: Pikunda-Munda Early Iron Age Attributes……………………………………………..68 

Table 7: Lower Congo Early Iron Age Attributes……………………………………………….76 

Table 8: Timeframe of Early Iron Age Behaviors in Central Africa…………………………….84 

Table 9: Premodern African Diaspora Framework: Bantu Homeland Diasporal Frontiers……..86 

Table 10: The Bamenda Grasslands table of data sources…………………………………….....93 

Table 11: The Northern Plateau table of data sources………………..………………….……....95 

Table 12: The Forest Interior table of data sources…………………….…………………….….96 

Table 13: Lower Congo table of data sources………………………………………………..….97 

 

 

 

 

xi 

 

LIST OF FIGURES 

Figure 1: Map of present-day distribution of Bantu languages…………………………………... 1 

Figure 2: Map of Central African Rainforest…………………………………………………….. 4 

Figure 3: Map of research areas…………………………………………………………………. 25 

Figure 4: Map of the Bamenda Grasslands……………………………………………………… 35 

Figure 5: Map of the Shum Laka cave site, present-day…………………………………………38 

Figure 6: Map of archaeological sites of the Northern Plateau…………………………………. 44 

Figure 7: Map of archaeological sites of the Forest Interior……………………………………. 56 

Figure 8: Map of archaeological sites of Lower Congo………………………………………… 69 

 

 

 

xii 

CHAPTER 1: INTRODUCTION  

 

Figure 1: Map of present-day distribution of Bantu languages  
Letters refer to their grouping in 15 language zones based on their similarities (de Maret 
2013:628). 

1 

I. The Bantu Phenomenon 

 

The Bantu phenomenon, the presence of a closely connected linguistic, cultural, and 

genetic system across western, southern, eastern, and central Africa south of the Sahara Desert, 

has preoccupied historical understandings about the nature of cultural change in Africa for over 

500 years (See Figure 1 for contemporary depiction of the Bantu language family of sub-Saharan 

Africa). One of the earliest modern historical accounts occurred in 1498 aboard Vasco di Gama’s 

ship, when the chronicler noticed that Martim Afonso, a general familiar with the languages 

spoken along Africa’s Atlantic coast could effectively communicate with several language 

communities on the Indian Ocean coastlines as well (Damião 1566:100). This initial interest in 

the commonality of the languages that spanned a continent continued well into the 19th century 

when the term Bantu was first coined by Wilhelm Bleek (de Maret 2013:627). The Bantu 

Phenomenon was also a point of interest during the early phases of the development of Black or 

African American Studies as noted in the 1940s when W.E.B. Du Bois stated, “Finally there are 

the Bantu, who are congeries of peoples, belonging predominantly to Central and South Africa 

and occupying the southern two-thirds of black Africa. The Bantu are defined on purely 

linguistic criteria” (Du Bois 2007:60). Since Du Bois’s observations in the mid-20th century the 

Bantu phenomenon has grown to encompass 300 million speakers and 500 distinct languages 

spread throughout the globe. What first appeared to be a distinctively linguistic phenomenon 

soon morphed into a cultural, ethnic, and even racial category (de Maret 2013:627). This shift 

towards understanding the origins of cultural aspects of the Bantu phenomenon is the focus of 

this dissertation. Through examining the Bantu phenomenon as the result of adaptations to the 

emergence of the Bantu Diaspora, I have identified key mechanisms for the geographic spread 

and maintenance of Bantu culture.  

2 

II. Understanding the Bantu Phenomenon 

Linguistic models: 

 

Since the Bantu Phenomenon was first observed as a linguistic system the prevailing 

frameworks for understanding are based on linguistic models of cultural change and connection. 

One of the earliest attempts at categorizing the languages was from Henry Johnston in 1919 who 

stated, “…it might be said that the whole of this southern third of Africa contained bus one 

indigenous speech family, the Bantu…” (Johnson 1919:15). This encouraged the perception that 

the Bantu “speech family” emerged from a single mother tongue (Vansina 1979). Later by the 

beginning of the 21st century Jan Vansina had identified several distinctive characteristics among 

the various Bantu speaking communities such as when the encountered non-Bantu speakers, the 

agricultural and technological skills each group possessed and the extent of exchange and 

intermarriage with Bantu and non-Bantu speakers, particularly in Angola, just south of the 

Congo River Basin (Vansina 2004). Vansina’s work illustrates the shifts towards a complicated 

linguistic system, that did not move across space and time at a unified pace. The definitive 

recognition of Bantu speakers as complex heterogenous cultural groups reignited the debates 

about their origins, their contemporary experiences and the cultural, social, and political 

processes, events, and circumstances that occurred in between. This has influenced the current 

interdisciplinary strategies that have sought to combine environmental, economic, 

archaeological, and cultural frameworks with linguistic models to produce a more nuanced 

understanding the phenomenon (Bostoen 2015).  

3 

Figure 2: Map of Central African Rainforest 
Stars represent research sites. (Butler 2013). 

 

Environmental models:  

 

 

 

Central Africa is defined by the geographic zones associated with the Congo River Basin 

and the Equatorial Forest stretching eastward from and along the Atlantic Coast (Lanfranchi and 

Clist 1991, see Figure 2). These contemporary countries include: northern Angola, southern 

Cameroon, Equatorial Guinea, Gabon, western Democratic Republic of the Congo (DRC), 

Republic of the Congo (Congo-Brazzaville), São Tomé and Principe, and the southern border of 

Central African Republic (de Maret 2013). The defining geological feature is the Congo River 

Basin, a large depression that averages about 400 meters above sea-level, that begins at the East 

African rift Valley and extends to the Atlantic Ocean (See Figure 2) (Eggert 1993:291). The 

Congo River Basin is made up primarily of the Congo Rain Forest, or the Equatorial Forest, 

which makes up 8% of the continents surface area as it stretches from the coast of present-day 

Liberia to coastal Angola (Eggert 1993:291).  

 

The Equatorial Forest developed over the last 40,000 years with significant climatic 

4 

events every five to ten thousand years (Clist 1989:62). The climatic events have created 

significant variation in the density, structure, and extent of the forest, particularly around 2000 

years ago which produced both forest environments- dense primary and secondary forests that 

are humid and have an above average rainfall and savannas- non-forest, open areas that are arid 

more than two months out of the year (Gond 2013:2; Maley and Brenac 1998:145). As a result, 

the region is made up both savanna and forest environments which continuously shift- creating 

new open, less dense areas as access to and support of new plant and tree species. Both the forest 

and savannas vary in density, duration, and humidity levels to create distinctive zones, 

particularly along the edges of the Equatorial Forest or the outer hilltops of the Congo River 

Basin. Given the vast arrange of climates, geographies, and environments of the region, several 

models of cultural change have focused on the effects of the environment on human behavior 

across time, forest pathways models and scarcity frameworks.  

 

The forest opening models argue that the fluctuating forest and savanna boundaries 

created access points for migration into and through the dense rain forest (Bostoen et al 

2015:357). Therefore, the peopling of the region is attributed to the shifts in the forest which 

required adaptation to new environments and allowed access from groups the north and west.  

These new dryer savanna conditions likely caused the slow but steady introduction of macro-

lithics, a few axes, and early crop cultivation of pearl millet as early as 2400-2200B.P (Neumann 

et al 2012:54; Lavachery 2001:244; Lupo 2014:59). The dry climate removed some forests, 

particularly along optimal travel routes like the SRI which aided in the southern expansion of 

Bantu speaking peoples. Likewise, the shifts in the climate introduced new crops such as wild 

palm that were added to the forest-based hunter-gatherer diet as the forest started to decline 

(Lavachery 2001). The forest openings are then understood to produce the environmental 

5 

conditions that support movement through an otherwise less hospitable region. This theory 

suggest that Bantu speakers were ultimately passive in their expansion into the forest as they 

were mostly prompted by the shifts in the environment. 

 

Scarcity frameworks begin with the assumption that Central African Rainforest was 

generally inhospitable and as resources became scarce, communities had to move further south 

(Lupo 2014:59-62).  Like the forest opening framework, this framework suggests that climate 

changes caused unavoidable shifts in the available resources as well as access to them and Bantu 

speakers would have to develop new settlement and subsistence patterns in other parts of the 

region. However archaeological data throughout the region suggests access to an abundance of 

resources and trade networks, thus reducing the probability that communities would move when 

resources became scarce. 

Economic models:  

 

The shifts in the climate and environments lead to shifts in their ability to adapt to new 

surroundings and access to resources. These circumstances increased the need for trade and 

exchange both inside and outside of the Congo River Basin. Trade and exchange become the 

major sources of innovation and thus cultural change for Bantu speakers. This trade could be for 

subsistence goods such as fish, ideas such as metallurgy and agriculture, and knowledge such as 

new subsistence strategies from the autochthonous populations. Further, along the lines of the 

scarcity framework, Bantu speakers who lived in open savannas relied on the forest products 

such as meat and may have secured those goods from the people who lived in the forest. Further, 

there is evidence of intermarriage which may have shifted social and political relationships 

between migrants and indigenous people (Lupo 2014). These exchanges produced an 

6 

interrelationship between these groups which supported the continuous migration of Bantu 

speakers into the forest (Lupo 2014:610). 

Archaeological contributions: 

 

Given the linguistic, genetic, historic, and even environmental antiquity of the Bantu 

Phenomenon, archaeology has been particularly suited to offer materially based evidence for 

various aspects of the emergence, dispersion, and continuity of Bantu cultural practices. 

Archaeological research suggests that Bantu populations entered the Congo Basin from the 

Nigerian/Cameroon boarder in a series of continuous migrations down along the coast and into 

the equatorial rainforest over a period of 2000 years (Blench 2006; Needham 1984). The 

distribution and pace of changes in microlithic foraging technology suggests that the migrations 

occurred during climatic events that caused the rainforest to recede and savannahs to expand 

allowing for better access to resources (Blench 2006:127; Bostoen et al 2015:358). 

Archaeological evidence also suggests that Bantu people exhibited a mixture of subsistence 

strategies such as cultivation, hunting and gathering, as well as various forms of trade (Blench 

2006; Lupo 2014; Bostoen 2015). This has allowed for advanced theories on changes in food 

production strategies in the region.  For example, although there is evidence of proto food 

cultivation during the Later Stone Age, food production sees a distinct increase during the Early 

Iron Age (Neumann 2012; Oslisly 2013). Also, aborio-culture remains a stable feature of both 

pre-agricultural and agricultural groups during Iron Age site in the region (Neumann et al 2012). 

The archaeological, archaeo-botanical and linguistic data show the distribution of the most 

common features of early food producers; pits containing ceramics, charred plant material and 

increasingly, metal objects parallel the linguistic maps of the Bantu phenomenon (Neumann et 

al. 2012). These overlapping conclusions indicate that the expansion of Bantu speaking people 

7 

correspond with the emergence of Early Iron Age culture in the Basin and thus support the 

assumption that the earliest villagers were most likely Bantu speakers (Neumann 2012; Oslisly 

2013; Lavachery 1998). However, presently there is no strictly archaeological basis for 

identifying these sites as Bantu (Egger 1993:223). For the purpose of this dissertation and to 

remain consistent with present terminology, the parallels in the environmental, linguistic, 

genetic, and archaeological data are sufficient to view Early Iron Age sites as Bantu and thus 

illustrations of Bantu culture.  

 

These models ultimately contribute to the shift of understanding the Bantu Phenomenon 

in purely linguistic terms to one that includes enduring Bantu cultural productions, stability, and 

change. Here a broad definition of culture is most appropriate because there are significant gaps 

in our understanding of the nature of Early Iron Age behaviors, identities and activities. Here I 

adopt the broad definition proposed by Stuart Hall, a definition that merely requires a shared 

history and ancestry… “ which provide us with ‘one people’ with stable, unchanging, and 

continuous frames of references and meaning, beneath the shifting divisions and vicissitudes of 

our actual history,” (Hall 1990:223). Drawing on the expansive similarities and differences that 

develop among African Diaspora populations, Hall used this broad definition to establish a 

baseline of characteristics that unify an experience, which I believe is appropriate here.  

Nevertheless, these models set the stage for the prevailing ideas about Bantu expansion, for 

example that it was a colonizing efforts on the part of Bantu speakers, or it required power over 

the autochthons populations, or on the other hand, Bantu speakers were at the whim of nature 

and history and responding to changes as best they could.     

III. The problem, research question, and research design 

 

Given the last 100 years of systematic work on linguistic, environmental, historic, 

8 

cultural, and genetic aspects of the Bantu Phenomenon, some characteristics are now considered 

factual. The movement of Bantu populations is one of the most significant archaeological events 

in Central Africa due to the sheer magnitude of migrants, rapidity of cultural change, and lasting 

impacts on human behavior in the region (Bostoen et al 2015:355). The existing literature on 

Bantu expansion, interaction, and exchange rightly suggests that the expansion of Bantu 

speaking groups required the development of a variety of adaptive strategies including sustaining 

interactions across multiple regions as they are more familiar with these new environments 

(Mercader and Brooks 2001). Genetic evidence has most recently concluded that this expansion 

of Bantu speakers into sub-Saharan Africa was the result of population movements, rather than 

the result of an interregional trade language shared across various populations (Rowold et al 

2016; Li et al 2014:5-8). Lastly, this population movement occurred during the same period in 

which villages, metallurgy, and early food production techniques were appearing across the 

regions (Bostoen 2015; Neumann et al. 2012). 

 

Nonetheless, the issue of how these changes led to the development and endurance of the 

Bantu Phenomenon has yet to be fully addressed. The above models provide context for why 

people would move but not why such movement would cause such monumental changes. For 

example, with the scarcity frame work, if resources are scarce and the forest it inhospitable, why 

then would communities move south, further into the forest, rather than west and north? Or with 

the environmental factors, the Equatorial Forest is continuously influx, with pluvial and 

interpluvial periods consistently changing the nature of the forest (Eggert 1993:293).  If that is 

the case, then why haven’t there been numerous culture phenomena throughout central African 

prehistory? These questions seek to illustrate that while these models help to understand the 

context, they are insufficient for understanding the Bantu phenomenon.  

9 

 

Therefore, I developed the following research question to address the gaps in our 

understanding about the social and cultural change brought on by the movement of Bantu 

speakers into Central Africa: How and why was Bantu Expansion into Central Africa during the 

Early Iron Age successful? The ‘success’ of the Bantu expansion, or the ability of Bantu 

speaking groups to spread people, material objects, and immaterial cultural behaviors throughout 

sub-Saharan Africa, is well documented in anthropological, historical, and biological sciences 

(de Luna 2014). However, as will be discussed in the next chapter, scholarship on pre-historic 

African cultural change tends to focus on the external stimuli such as the environment or an 

external community. Further, such scholarship prioritizes the conservative nature of culture and 

leave little room for understanding the significant cultural and technological advances that make 

up the Early Iron Age package.   

 

Following Colin Palmer’s claim that Bantu Expansion represents the second major 

dispersion of African people, I identified key elements of the archaeology of premodern African 

Diasporas to assess the observable elements of the Bantu Diaspora for indications of its success 

(Palmer 2000). First, I examined four key regions of the Congo River Basin for evidence of 

connections between the Bantu homeland and the diaspora frontiers: The Bamenda Grasslands, 

The Northern Plateau, the Forest Interior, and Lower Congo. Next, I examined these regions for 

archaeological evidence of Early Iron Age trade, subsistence, settlement and technological 

change to assess the extent of intra-African dispersion and the construction of relationships with 

each other and a homeland.  This process can be summarized in the following way; first identify 

the material record, analyze it for regional patterns of dispersion and connection, describe how 

such patterns aided in the success of the phenomenon. I identified patterns of similarity within 

the Early Iron Age package but also regional distinctions that appear to be more intentional than 

10 

previous literature acknowledge. I conclude that dissimilarity of specific artifact classes, for 

example ceramic form and shape, does not inherently mean a lack of interaction between groups. 

Further, I found that an examination of the patterns of both similarity and dissimilarity across 

artifact classes can reveal new insights into the construction of “homeland” by various diasporal 

frontiers. I ultimately argue that the success of the Bantu Phenomenon is due to the construction 

and maintenance of the Bantu Diaspora, a premodern African Diaspora. The significance in this 

research relies in the ability of this premodern African Diaspora framework to offer new insights 

into the interregional connections that affected Bantu speakers.   

IV. Dissertation outline 

 

This dissertation is divided into seven chapters. Chapter 1, the Introduction defines the 

Bantu Phenomenon and research problem, question, and design. Chapter 2 reviews the literature 

on cultural change in African prehistory and African Diaspora studies to construct a framework 

for premodern African Diasporas. Chapter 3 details my methods for understanding this 

phenomenon particularly the focus on the collection and analysis of the archaeological literature 

on the Central African Early Iron Age period. This chapter also constructs the basic tenants of 

archaeology of premodern African diasporas and possible forms of evidence to look for in the 

material record. Chapter 4 examines Case 1 The Bamenda Grasslands and Case 2 the Northern 

Plateau, both located in south western Cameroon along the Nigeria/Cameroonian boarder, the 

linguistic homeland of Bantu speakers. Chapter 5 presents case 4: The Forest Interior and Case 5: 

Lower Congo as these regions are in other parts of the forest and thus represent different 

strategies, connections, and developments of Early Iron Age culture. Chapter 6 offers an analysis 

of the four cases and a discussion of the findings in relation to the framework presented in 

Chapter 2. Lastly, Chapter 7 presents the Bantu Expansion as a premodern African Diaspora 

11 

based on the research. The dissertation ends with suggestions for future research that can 

contribute to a better understanding of the Bantu Phenomenon as well as suggestions for 

identifying other premodern African Diasporas. 

 

 

12 

CHAPTER 2: LITERATURE REVIEW  

 

I. Introduction  

 

While the how question has yet to be addressed within Central African Bantu Expansion 

specifically, the whys of cultural change in premodern or prehistoric Africa have been 

thoroughly examined across bodies of literatures. What was unique about the types of 

interactions, cultural processes, and effects on culture history for people on the frontier? What 

contribution can a frontier analysis make in the era of globalization, global authority of nation 

states, and near permanence of constant movement and interaction of people around the world? 

These questions provided the answer for the field: use the fields emphasis on politics, culture, 

and interaction to analyze these contemporary, ever changing frontier context (Helms and 

Loveland 1976). Helms and Loveland (1976), argues that through an analysis of regional and 

local context we can better understand the construction of global frontiers. The authors suggest 

that the Caribbean and Central America are a social, political, and cultural frontier. They explain 

that the multi ethnic power dynamics of this region can be understood as a framework for 

construction global frontier processes. 

II. Frontier-metropole hypotheses 

 

Frontier studies most popularly begins with the work of Fredrick Jackson Turner, a 

significant American historian from the 19th century. Turner’s (1894) essay is understood to be 

the earliest proposition for a basic framework for identifying and analyzing the social processes 

occurring and arising from a frontier. For Turner, the frontier is characterized as the boundary 

between savage and civilized, a world unhinged and free, a place of settlement (Turner 1894). 

The nineteenth century nostalgia of the U.S. American West becomes the impetus and definition 

13 

of frontier studies. Turner did not incorporate nuanced notions of power, colonialism, nor Native 

American agency when understanding the dynamics of frontier life in the US. Over the last 

century, scholars have emphasized the ethnocentric and narrow use of the Native American and 

European settler interaction as the basis for understanding frontier spaces worldwide. 

Nevertheless, Turner’s work, despite over a century of critique, still structures scholarship on 

cultural change within a frontier context. 

 

The defined boundaries of nation states create a different dynamic as groups who live on 

the borders are typically very clear about political authority in various jurisdictions. Ochonu 

(2015) examines role of violence in the economic and religious cultural changes in the lower 

Benue hinterlands of Nigeria (Ochonu 2015:204). In agreement with Parker and Rodseth (2006), 

Ochonu demonstrates the varying scales of power and influence from multiple sources within the 

frontier formation processes. Nineteenth century India provides another example of political 

impetus for the formation and continuation of a frontier context.  Although British governance 

and punishment was largely structured by the British imperialist in India, the frontiersmen 

established a strict frontier boundary empowered by the Frontier Crimes Regulation (Hopkins 

2015:371-378). This structuring and legitimizing of frontier politics on the British frontier 

demonstrate that use of administration and authority to organize and strengthen social life. In an 

urban frontier settlement of the Ottoman Empire, Mastan (2014) demonstrates how 

“identification agents”, or outsiders, established unity among outsiders based on their social roles 

formed in the imperial center (Mastan 2014:16). Mastan, following in the long history of frontier 

studies, argues that by encountering indigenous groups, the outsiders created new cultural and 

social behaviors. Further, the frontier context established different patterns of behavior that made 

cultural productions diverge from those that occurred in central spaces (Mastan 2014:16). 

14 

 

Lastly, Igor Kopytoff's edited volume, The African Frontier (1987), is the most widely 

recognized source for understanding the nature of cultural interaction and emergence of complex 

societies outside of European interaction mechanisms. The frontier in Africa was unique because 

as Kopytoff purports, because ethnically ambiguous marginal societies on the outskirts of large 

established polities, or frontiers in Western literature, were the model of cultural change on the 

continent (Kopytoff 1989:5). Deveneaux’s (1978) detailed overview of the history of frontier 

studies in Africa, demonstrates that the construction of an African frontier cannot be structured 

around the western philosophical traditions of Aristotle, Hegel, and Marx. Through a 

biographical sketch of a nineteenth century East African leader, Bennett (1971) examines the 

expansion and contraction of regional politics in East Africa. This volume demonstrates how 

local politics shaped Afro-Arab regional dynamics.  Hess (1978) contributes another perspective 

on African-Arab-European interactions, through his analysis of the spheres of influence around 

the Mediterranean Sea. De Kiewiet (1942) assesses regional processes in the frontier zones of 

colonial South Africa, arguing that geographic segregation resulted in cultural and political 

frontiers of British, African, and Boer communities. These three texts demonstrate how analysis 

of local and regional context can help evaluate the entanglement of global relationships and the 

cultural and political boundaries that continue to shape social and political interaction today 

(Hess 1978:3-4). 

III. Intra-frontier hypotheses 

 

Social scientist inspired by the Boasian culture historic critique began to develop an 

interest into the cultural processes of non-western people. Analysis of the frontier is no longer 

framed by the political context alone; rather the social and cultural process that occur provide the 

most substantive insight into cultural change over time in the frontier setting. Bohannan and 

15 

Plog’s (1987) chapter attempts to move the concept of frontier beyond even a geographic 

boundary to describe “any kind of interaction across cultural boundaries” (Bohannan and Plog 

1987:9). This posture seeks to understand the types of cultural processes that occurred within 

multi ethnic spaces, regardless if these processes occur within a bounded geographic zone. The 

frontier began to be seen as more complex than simply a site of cultural domination of one group 

over the other. Further, the rate of acculturation was also more complex and often the process 

itself took on ethnic distinctions. 

 

 

Barth (1969) uses ethnographic evidence from Africa, Asia, and Europe to compare 

dynamics on the frontier and those in cross cultural contact zones. Specifically, Barth proclaims 

that contrary to his social science contemporaries, ethnic boundaries are often strengthened in 

events of multi-ethnic contact. He states, “ethnic distinctions do not depend on an absence of 

social interaction and acceptance but are quite to the contrary often the very foundations on 

which embracing social systems are built” (Barth 1969:10). Stein (1980) complicates this 

narrative by emphasizing that social and political power is not always established by 

authoritarian rule by the larger, core metropolis over the periphery frontier context. In 

conjunction with Feierman cited below, this text demonstrates that the borderlands too control 

resources and socio-religious power which can be levied against the core authority. Feierman 

(1974) also demonstrates that the frontier was a source of power and initiated processes of 

cultural change. He examines the Shambaa, a regional political economic frontier in eighteenth 

and nineteenth century Tanzania. He demonstrates that market activity was multi directional and 

that the activities and needs of the frontier directly affected the needs and cultural developments 

in the more powerful surrounding cities (Feierman 1974:129-131). This text helps demonstrate 

that frontiers do not exist at the whim of the powerful centers. Rather, center political economics 

16 

rely heavily on the social, cultural, and trade relationships developed in the hinterlands. The 

regional perspective also demonstrated that the frontier existed beyond the sites of rural isolation. 

The identification of urban frontiers, whose authority can lie outside of the core political areas as 

well as the frontier zone itself, highlights the significance of socio-political context among its 

neighbors and hos it shapes life on the frontier (Vlassenroot 2013:3171). 

 

 

In Africa specifically, the social, political, and cultural formation on the frontier also 

affected interactions with the areas surrounding the frontier, not just homelands. Ogundiran 

(2014) demonstrates the significance of frontier-frontier interaction through an historic 

archaeological approach to the Osogbo of nineteenth century southwest Nigeria (Ogundiran 

2014:6-7). The Osogbo frontier existed as a part of a regional network of frontiers. Political 

authority on the Osogbo frontier developed through interplay with the surrounding frontier 

networks, rather than an internal conservative dynamic founded on homeland behaviors as 

theorized by Kopytoff (Ogundiran 21:2014). In making the local global, Ogundiran (2009) 

investigates “interstitial frontiers” in the Oyo Empire of Nigeria, or non-metropole frontiers as a 

method to understand indigenous mechanisms of cultural production an authority in the region 

and uses these local realities to better understand the Atlantic trade (2009:356). Usman (2011) 

performs a similar task through in investigation of the peopling of the frontiers of northern 

Yorubaland in north central Nigeria. Usman’s focus on population change invokes ethnogenesis, 

particularly during the time of mass movement of people during the Atlantic era.  Peacock 

(2009) offers a collection of essays about the variety of frontiers surrounding the Ottoman 

Empire that expanded from central to eastern Europe and Iraq to Sudan between the fifteenth and 

twentieth centuries. This comparative approach to the multiple ways in which a specific Empire 

utilized frontier spaces demonstrates mechanisms for social interaction at a large scale. 

17 

IV. Environmental hypotheses 

 

The frontier zone offers a unique perspective on how human behavior shifts and adapts in 

relationship to the environment because of the inherent nature of culture change on the frontier. 

Frontiers are often defined by isolation that emerges from natural conditions such as water, 

forests, and terrains which often create forced boundaries between societies. Osei-Tutu (2006) 

examines the shifting cultural influences between two societies with contrasting landscapes one 

along the Akuapen ridge and those in the Shai hills of Ghana. Osei-Tutu uses the ceramic 

assemblages to demonstrate that cultural exchanges occurred between these societies, 

particularly in the sixteenth and seventeenth centuries, however there was no evidence for 

sociopolitical influences (Osei-TuTu 2006:103). The landscape here, the ridge, acted as a 

physical boundary that constructed complementary needs for the two groups. Similarly, 

Rodseth's article addresses how the Himalayas in the formation of Hindu and Buddhist ethnic 

enclaves at the beginning of the first millennium (2006:90). The Hindu and the Buddhist both 

represent a homeland, core spaces, therefore this work illuminates a core to core model of 

interaction. However, the Himalayas create a natural boundary between these two societies 

because of the difficult terrain and isolation from the homeland networks. The author argues that 

the Himalayan borderland between Hindu military quest and Buddhist missionary expansions 

created a hilltop and valley systems that shifted through time and directly impacted ethnogenesis 

in the region. 

V. Frontier cultural and identity construction frameworks 

 

Frontier studies is also used to explain of identity formation processes. Parker and 

Rodseth (2006) identify three elements that impact identity construction on the frontier: one or 

more parties emerging from a “core”, a core-frontier dialectic, and social interaction between the 

18 

products of this core-frontier dialectic and a new, unrelated group (Parker and Rodseth 2006:4). 

Further, Whitten demonstrates that African descendants in the Americas should be understood as 

frontiersmen because of their marginality to mainstream politics and economics. Whitten’s 

argument for frontier identities also challenged the tenants of the frontier in that the construction 

of a frontier zone or frontier interaction need not be based on geographic isolation but power 

dynamics within cross cultural settings (Whitten 1974). 

 

While the dominance of one political authority over another is readily apparent on 

modern frontiers, cultural change and the resultant identity (re)formation processes develop 

unique expressions of power on the frontier. For instance, the Ekie case in the Goba region in the 

Democratic Republic of the Congo demonstrates a continuous reconfiguration of political 

authority but consistent cultural ideological expressions of power at the regional level (Lancaster 

1987: 106-107). Packard’s chapter adds the element of multiple origins and narratives of first-

comer and late-comer, where outside and inside memberships continuously change through time 

(Packard 1987: 150). Cultural identity formation processes therefore are continuously 

restructured as newcomers become indigenous. Such origin dynamics re-situate political 

authority as each new group claims power over previous groups and thus creates a socio-political 

climate that (re)connects authority among them (Packard 1987:130).  These cases demonstrate 

both stable and fluctuating circumstances that structure identity formation processes on the 

frontier. 

 

Lastly, the multi-ethnic context is inherent to the frontier but ebbs and flows as cultural 

change is determined by one group’s social authority, cultural or political, over another. 

Isaacman and Isaacman (1975) investigate how European traders and farmers in Mozambique 

became subsumed into indigenous cultures over a period of 200 years. The authors use Philip 

19 

Curtin’s concept of transfrontiersmen to demonstrate that within this setting, cultural change 

flowed in many directions, not just toward European, nor towards those with most political or 

military power. Finally, Smith (2006) takes an even more direct approach in his study of 

intermarriage and politics, which uses Bourdieu’s concept of habitus to demonstrate how 

intermarriage reconstituted ethnic identities to construct political power between Nubia and 

Egypt (Smith 2006: 230-231). 

VI. Summary of literature review 

 

This chapter presented the prevailing theories on cultural change in premodern Africa. 

The initial theories suggested that the cultural similarities of Africa could be attributed to the 

power dynamics of centers over frontiers and the resultant conservative reconstructions of home 

culture in new spaces. Intra-frontier theories were also used to understand the interaction that 

occurred between the peripheries. Cultural change was also attributed to environmental factors 

where culture change emerged as premodern people attempted to respond to shifting 

environments. In both cases, cultural change is the result of external forces, be them political or 

natural, and how communities respond to them. However, this project employs the African 

Diaspora framework to assess the internal mechanisms of change that may be present in the 

Central African Early Iron Age. The following chapter illustrates how this framework 

incorporates the range of climatic, technological, and cultural traits to formulate a cohesive 

narrative of cultural change.  

 

 

20 

CHAPTER 3: METHODS 

 

I. Introduction 

 

The goal of this dissertation is to understand how Bantu speaking people moved through 

Central Africa in such a short period; radically changing food economies, technologies, and 

cultural productions for themselves and those they encountered. I sought to understand these 

processes through answering the following question: How and why was Bantu expansion into 

Central Africa during the Early Iron Age successful? This research prioritized archaeological 

data to identify patterns in how Bantu speakers maintained cultural behaviors or established new 

ones given the variation in the equatorial forest during the Early Iron Age. In order to evaluate a 

broad set of archaeological sites within the research period, I relied on published and 

unpublished literature from Central African archaeology as the basis of information from which 

to identify these patterns.  Ultimately, this chapter explains where, how, and why I collected the 

data to answer my research question. I also discuss my analytical approach and why I prioritized 

certain aspects of the data and presents the justification for my analytical framework.  

II. Data collection 

 

The published literature on the central African material culture related to this project is 

for the majority split between 4 museum spaces: The National Museum of Yaoundé (Cameroon), 

the National Museum of Congo (DRC), The University of Tübingen (Germany), and the Royal 

Museum of Central Africa (RMCA) (Belgium). These institutions hold archaeological and 

cultural material dating at least to the 1980s and potentially would each serve as a compelling 

resource for understanding the interest of this project. However, the former three institutions 

typically have the materials from very specific sites and excavations. Further, as I examined 

21 

during my pre-research at the National Museum of Congo in 2015 and the Bibliotheque 

Nationale de la Republique Democratique du Congo in 2011, these institutions were in the 

process of cataloguing and organizing their collections. The Bibliotheque Nationale had very 

minimal resources pertaining to pre-colonial history and culture in the region. At the National 

Museum of Congo, the archaeological materials, neither accessioned or properly catalogued and 

were not in a state for systematic research. Although the wealth of materials at the Museum of 

Congo cannot be understated, necessity to process, organize the materials to get to a point of 

research was beyond the scope of a dissertation project. 

 

Ultimately, I chose the RMCA for my data collection as this museum holds most of the 

Central African archaeology and archival material. The RCMA is the largest museum of Africa 

and certainly Central Africa in the world. It is a major funder and supporter of the Central 

African archaeology projects, since the 1950s but certainly from the 1980s through the present. 

The museum maintains a consortium of archaeologists, curators, and archivist working on the 

region expanding and maintaining collections ranging from upper paleolithic to historic and 

contemporary times. Their collections, often on loan from Central African institutions, contain 

archaeological materials, photographs, field records, and videos. As a library-based dissertation 

project exploring patterns in material culture across multiple sites and regions, this project relied 

heavily on published literature found in the archives and collections at the RMCA. 

 

Data collection at the at Cultural Anthropology and History division of the RMCA 

occurred from June 6 through July 21, 2017. I identified the published and unpublished sources 

considered data for the project from June 1st through August 1, 2017. I spent a total of 6 weeks 

in the Royal Museum of Central Africa (RMCA) in Tervuren, Belgium and the remaining weeks 

using the library resources for Michigan State University (MSU). As a library-based dissertation 

22 

project exploring patterns in material culture across multiple sites and regions, this project relied 

heavily on published literature found in the archives and collections at the RMCA. My first few 

weeks at the RMCA library consisted of photographing and digitizing relevant monographs, 

edited volumes, photo slides, and reports from their archives. I then digitized relevant Ph.D. 

dissertation and master’s thesis chapters from these sites. With the help of Nadine 

Devleeschouwer, the archaeology reference librarian, I ultimately collected over 4GB of articles, 

reports, and dissertations. Upon my return to East Lansing, Michigan I continued my search of 

the digital databases such as included J-Stor, Google Scholar, and African Archaeolog.net using 

the MSU Library system from July 21 through August 1st 2017. I searched generic search terms 

such as “Bantu migration,” “Central African archaeology,” “Cameroon Archaeology,” “African 

prehistory,” “Early Iron Age” as well as specific terms such as “Shum Laka;” “de Maret;” “plant 

cultivation in Gabon” among many others.    

 

I explicitly sought out materials that discussed archaeological data for the following 

archaeological regions; The Bamenda Grassfields, The Northern Plateau, The Forest Interior, and 

Lower Congo dating from 1000 B.C. to 200 A.D or the Early Iron Age. Each of the sites I 

identified for the project had publications on their initial archaeological findings, subsequent 

analysis, and additional insights spanning over at least twenty years. Lastly, these sites had 

subsequent fieldwork during separate periods of the historiography of central African 

archaeology, therefore they represent different modes of collecting and analyzing material 

overtime. These similarities and differences allowed me to minimize differences in fieldwork 

strategies to determine the composition of the artifact assemblages.  I did not include articles that 

were explicitly theoretical or prioritized linguistic, human genetic, or ethnographic data over 

descriptions of archaeological materials. I also prioritized original publications of new 

23 

archaeological data for each of the site over publications that reiterated pervious findings. I 

included regional syntheses only when they included original archaeological data from the 

relevant locations.  

 

Lastly, I included materials in any language, relying on my beginner French language 

training and google translate as needed. I also, necessarily, only included complete sources 

available through these methods during the period of research. These basic criteria; published 

sources that provide explicit accounts of archaeological material collected from these sites, 

significantly limited the number of sources I identified as data for this research. This exclusivity 

allowed me to focus specifically on the archaeological materials of the region rather than the 

theoretical linkages proposed between fields. This process revealed 46 sources that I consider as 

data (see APPENDICES A-D for a full table of sources). 

 

 

24 

Figure 3: Map of research areas 
1) The Bamenda Grasslands; 2) The Northern Plateau; 3) The Forest Interior; 4) The Lower 
Congo 

III. Description of research materials  

 

I conducted a meta-analysis of the grey literature on Bantu expansion in Central Africa to 

develop a timeline of when material based cultural behaviors emerged in the region (see 

Appendix A for a full list of these materials). This grey literature included dissertations, 

published field accounts and radio carbon date updates, bibliographies, and articles. The sources 

I used contained four main attributes; 1) has Central Africa or its borders as the geographic 

25 

focus; 2) reports directly on archaeological fieldwork conducted on the phenomenon of Bantu 

expansion, either as primary or secondary data; 3) contains site specific information; 4) is 

accessible during the period of dissertation research (June-August 2017). The following is a 

breakdown of the sources I obtained for the research. The materials are broken down by the 

region as those were the primary categories for connecting the sites to Early Iron Age Bantu 

expansion (See Figure 3): 

The Bamenda Grasslands: 

 

The Bamenda Grasslands materials overwhelmingly focused on the Shum Laka cave site 

located southwest of present-day Bamenda, Cameroon. These sources span from 1987 through 

2002 and represent the artifacts and analysis of the Early Iron Age/Stone to Metal Age time-

period. I identified seventeen total resources: ten English sources and seven French sources. The 

French sources are comprised of one Ph.D. dissertation, one M.A. Thesis, and five articles from 

journals specializing in anthropology and environmental studies. The English sources are mostly 

book chapters and articles from 1992 through 2002 and represents expansions of the work 

presented in the dissertations as well as work relating to specific artifact classes such as the 

burials, lithic, and ceramic materials. 

The Northern Plateau: 

 

The Obobogo materials include descriptions of Obobogo, as well as the contemporaneous 

Nkang, Ndidan, and the Okolo sites along the hilltop ridges of in central Cameroon. I identified 

eight sources: four English and four French sources. Three of these sources are Ph.D. 

dissertations and one is a master’s thesis. Since the dissertations are field based, they offer the 

most concrete articulations of artifact assemblages in the region. They often reference other sites 

26 

which help to demonstrate relationships among this group of sites. The articles represent the first 

introduction of the site and then 20 years of analysis of the sites.  

The Forest Interior:  

 

The Forest Interior represents the location where the Congo and Ubangi Rivers split 

within the rainforest. I reviewed eight sources from this region; four in English, two in French, 

and two in German. As a series of sites inside the Equatorial Forest, the Imbonga archaeological 

project represents a diversion from the work done on the coast and southern stretches of the 

forest. The materials include three book chapters and five articles published between 1984 to 

2014. These materials describe early 1980s fieldwork in the Mbandaka regions along the 

northeast border corner of present-day Democratic Republic of the Congo. Of note, in late 2016 

Eggert and Dirk published the Campo: archaeological research at the mouth of the Ntem River 

(south Cameroon), the first monograph analyzing the work at the sites in and around Imbonga. 

This volume is a much-needed contribution for the field but was not available during my 

research. As such I did not included it in this dissertation project.  

The Lower Congo: 

 

There is a total of fourteen sources for the Lower Congo group. Six of the sources are in 

English and eight are in French. There are five reports, six articles and three master’s theses. 

These sources date between 1959 and 2016. The Ngovo group consist of sites in Lower Congo, 

the western region of DRC just north of Angola. These sites are located south of the Equatorial 

Forest and illustrate significant changes in the ceramic, lithic, and burial traditions of the central 

African Early Iron Age. Although this site represents the later period of the Early Iron Age, it 

also represents the earliest pottery classifications of the time-period for the region. As such, these 

materials illustrate the earliest methods of data collection and analysis of archaeological material 

27 

for the region.  

 

This list of sources represents the available descriptions of the Central African Early Iron 

Age material culture. As with all social science and humanities work where there are frequently 

interviews are left undone, events left un attended, and units left unexcavated. As such, this 

systematic collection of the resources may have excluded valuable materials that fit within my 

research parameters. Nonetheless, the ones I have included have been referenced over the last 

thirty years as foundational text for understanding the material culture of the region (Eggert 

1993: de Maret 2013; Vansina 1990).  Therefore, the present analysis of the local, regional, and 

interregional changes of the Early Iron Age reflects nearly one third of the archaeological 

publications for the region and therefore adequately represents the material record to meet the 

goals of this research project.  

IV. Data extraction and analysis 

 

Once I identified all of the Early Iron Age materials from these sites and listed them 

based on publication, site, and region, I then moved to the analysis phase of research: 

understanding the spatial and geographic distribution of these materials as part of the Bantu 

Diaspora. The assumption, as stated above, is that Bantu populations sustained cultural behaviors 

which aided in their rapid expansion through the region through maximizing geographic and 

indigenous interactions (González- Rubial 2014; Rodseth 2006). To answer my research 

question, I developed a sequence of when Bantu settlements emerged and determined the extent 

to which these sites were similar to sites located within the homeland region. I then assess 

whether these variations were due to environmental conditions, interaction with other groups, or 

connections with the homeland. After organizing my sources based on site designations, I 

created excel spreadsheets to categorize the following categories for each excavation period for 

28 

each site: primary funding and support organizations, primary archaeologists, governing bodies 

associated with each excavation mentioned within each piece of literature. I then identified the 

specific excavation time periods, strategies, and context of material excavated from each field 

session. I then noted the listed excavation techniques and the laboratory and analytical methods 

employed. These first steps helped me to control for the variation of field techniques across the 

various site location and excavation seasons.  

 

Next, I identified the different artifacts mentioned with their various relevant published 

attributes. For example, if there was a description of a burial, I noted the associated radio carbon 

dates, burial materials, location vis a vie the site and other artifacts, as well as its size when such 

descriptions were available. I made notations for the published levels, whether in stratigraphic or 

survey context. I included notations for each kind of artifact mentioned (e.g. polished stone, 

hearth, ceramic fragments, microlitihics, specific floral remains such as Musa residue, specific 

animal remains as well as evidence for the use and or production of iron. These spreadsheets 

allowed me to create a baseline to identify which sites had similar attributes at similar times. 

This also allowed me to identify the main artifacts present at sites during this period and when 

they appeared or disappeared at each of the sites. My systematic identification of where, when, 

how, and what artifacts were found within these sites revealed the following factors that shape 

Early Iron Age sites and activities at the local and regional level: the site location, use periods, 

immediate environment, floral cultivation, pottery production, lithic production, faunal use, and 

metal use. Once this baseline of when what appeared where was complete, I began asking 

questions of how they may have arrived there. The how question would indicate whether such 

items were traded or locally produced to allow to analyze these sites as a premodern African 

Diaspora.   

29 

 

In order to conduct such an analysis, I first examined existing frameworks for global, 

modern, and premodern Diasporas for possible indicators that would be relevant to the Central 

African Early Iron Age context. William Safran, Robin Cohen, and Ruth Simms Hamilton are 

representative of the variety of criteria that constitute a diaspora at these levels. These authors 

share an emphasis on the dispersion (whether forced or voluntary) from a center; retention of 

collective memory; conflict in the new area; and continued connection to the homeland/center.  

William Safran’s work on the premodern Jewish Diaspora, serves as the prototypical diaspora for 

much of the field, provides additional criteria for distinguishing diasporas from other forms of 

movement such as migration (Safran 2005). To summarize Safran’s criteria also include a desire 

to maintain cultural distinction from the host community, collective commitment to the 

restoration of the homeland and maintenance of cultural heritage and symbols of the homeland 

(Safran 2004:37). Robin Cohen adds categorization of diasporas, such as victim, trade, labor, and 

imperial diasporas, as ways to understand that different classifications will produce varying 

degrees of diaspora consciousness and identity (Cohen 2008:14). Cohen also proposes some 

additional common features such as: migration to two or more foreign regions, idealization of the 

collective homeland and collective commitment to it, frequent development of return movement, 

and a strong ethnic group consciousness sustained over time (2008:17). The addition of special- 

foreign regions- and time- sustained over time- make Cohen’s criteria especially appropriate for 

this present project.  Ruth Simms Hamilton’s focus on the criteria that emerged within the 

modern African Diaspora emphasizes the geo-social mobility of dispersed people meaning that 

they may move continuously through time, the importance of homeland connections, and the 

social inequality experienced within the new settings (Hamilton 2008).  

 

This brief overview of diaspora frameworks presents a few common themes that may be 

30 

present in the Central African Early Iron Age as well as some elements that were most likely not 

present: the forced dispersal, the migration to two or more places over time, the construction of a 

homeland and maintenance of its connections; and experience of social inequality in the host 

land. On the latter point, given the endurance of Bantu speaking people and the linguistic and 

archaeological evidence on their interaction with autochthon people, it is unlikely that they were 

persecuted in their new communities. However, when one considers the climatic changes that 

were occurring, it is not implausible that the people of the Nigeria-Cameroon boarders were 

experiencing some pressure to leave and migrate into other regions. It is also clear that this group 

migrated to several different regions over a range of time and continued to move up until the 

present day.  

 

This dissertation, however, is particularly concerned with the criteria of an intentional 

construction of a homeland and connection to it. Through combining frameworks for what 

constitutes a diaspora and what is known about the Central African Early Iron Age as stated 

above, I have identified the following observable characteristics that may be present in the 

archaeological record and which when considered together could indicated a premodern 

diaspora: 

1) Required dispersion  

2) Intentional connections to the homeland  

3) Constitute a clear cultural, ethnic, or racial group 

4) Inter and intra-regional dispersion beginning in Africa 

5) Adaptations to new surroundings  

31 

 

These patterns are observable in the archaeological record and incorporate what is known 

about the Early Iron Age on the continent. For example, a shift in the floral and faunal materials 

can indicate a need for dispersion as well as intentional connections to the homeland. 

Observance of cultural traits at more than one location can be evidence of migration into more 

than one foreign location and when they may have arrived. Lastly, the unique ability to examine 

how these processes through time allow archaeologists a glimpse into how homeland 

relationships and connections may have adapted through time. Table 1. provides an example of 

the potential of this analytical framework. 

Intraregional Location

The Bamenda Grasslands

Archaeology of Premodern African Diaspora
Evidence of 
Dispersion

Evidence of 
Connections

Consitutes 
Homeland

Constitutes 
Diaspora frontier

The Northern Plateau
The Forest Interior
Lower Congo
Table 1: Premodern African Diaspora Framework Spreadsheet 

 

V. Research challenges and limitations  

 

Due to the nature of archaeological research in Central Africa, sampling issues were 

unavoidable within this research of which little remedy could be achieved. First, during the 

1980s and 1990s archaeological research in Central Africa was primarily motivated by 

government or private construction projects which tended to focus on urban areas outside of the 

rainforest.  This resulted in an overrepresentation of salvage projects or projects that develop out 

of the need for contemporary construction projects, along the forest borders or forest-savanna 

mosaic regions, or on hilltops in contemporarily populated areas. Therefore, the sites that are 

excavated and published tend to follow construction trends and thus concentrate around major 

32 

cities leaving a large swath of the interior forest yet to be excavated and or published. As such, 

my data follows these trends, articulating a better understanding of the edge of the forest and 

leaving out most interior parts.  

 

Further, since the projects tended to follow external construction and development 

projects, there was little support for continued analysis of certain sites and thus very few attempts 

to construct interpretations on the nature of cultural change across the region. Even when 

projects had the luxury of long-term field research lasting between 5 and 10 years, the analysis 

tended to identify a set of behaviors that existed within the space without attention to how or 

why those behaviors changed over time. This analytical limitation is precisely why such cultural 

histories are rare for Central African prehistory. By combining several projects across various 

regions, this dissertation attempts to lessen the impact of these limitations to produce an analysis 

of savanna, forest-savanna, and forest settlements and how they changed over time.  

VI. Summary of research methods 

 

This research sought to apply a premodern African Diaspora framework to understand the 

dispersion and connections that occurred during the earliest phases of the Bantu Diaspora into 

Central Africa during the Early Iron Age. To achieve this goal, I identified archaeological sites 

that exhibited the Early Iron Age package that is typically attributed to Bantu speakers. I then 

analyzed published and unpublished literature for descriptions of the material evidence of 

dispersion, such as the geo-temporal development of the sites, and evidence of connections- intra 

and interregional trade or intentional local or regional distinction. I collected the majority of my 

source material from the Royal Central African Museum in Tervuren Belgium and Michigan 

State University. I ultimately identified the Bamenda Grasslands, the Northern Plateau, the 

Forest Interior, and the Lower Congo regions and I cataloged the presence and absence of 

33 

different pottery types, EIA lithics, metallurgy, floral and faunal remains, as well as the site’s 

location, local environment, and use period. These data points allowed me to identify when and 

where specific behaviors and activities emerged within the region. These designations allowed 

me to establish patterns of dispersal and concentration at the local level, indicating patterns that 

occurred within the homeland as well as those at the interregional level. The patterns that 

emerged within and between these spaces ultimately demonstrate connections that lasted through 

time and space thus producing the framework for identifying the premodern diaspora of Bantu 

speakers.  

 

34 

CHAPTER 4: CASE 1: THE BAMENDA GRASSLANDS AND CASE 2: THE 
NORTHERN PLATEAU  
 
 

I. Introduction  

 

This chapter presents case studies 1: The Bamenda Grasslands and 2: The Northern 

Plateau as they illustrate the earliest cultivation of Early Iron Age in the Congo River Basin. 

Case 1 focuses on the archaeological work conducted at Shum Laka, considered the linguistic 

homeland of Bantu speaking people. Case 2 examines the Northern Plateau and the fully 

developed Early Iron Age package. 

 

Figure 4: Map of the Bamenda Grasslands  

Star represents Shum Laka Cave site (Moyerson et 
al 1996:41). 

35 

II. Case 1: The Bamenda Grasslands 

 

The Bamenda Grassland region is a mosaic of highlands and lowlands with both 

woodlands and savanna environments (See Figure 4) (Maley and Brenac 1998:145). These 

Grasslands are fed by the Gulf of Guinea which has supported the forests area since the 

Pleistocene era (Lavachery 2001:240). This location was unusually lush for this part of Africa 

(Lavachery 2001:214).  While currently, the rock shelter is nestled in a savanna, the majority of 

the artifacts found resemble a forest environment during occupation leaving de Maret to 

determine that “not a single savanna element was found” in the preliminary report of the first 

large scale excavations of the site (de Maret et al 1993; de Maret 1995:14). Around 6000 B.P. 

there was a shift in the forest densities which eventually gave rise to the Bamenda grasslands. 

This environmental split gave rise to a cultural split that would leave the Bamenda region 

relatively isolated along its northern and western borders (Lavachery 2001: 241). There was also 

a roughly 2000 year gap between the Late Stone Age and the beginning of the Early Iron Age 

periods for the site observed in the stratigraphy of the rock shelter; 2 layers consisting of ochre 

ash with phases dating before 6000 B.P; and three layers of grey ash occurring since 4000B.P. 

representing the second wave of Bantu migration and the focus of this study (Lavachery 

2001:219; de Maret 1996: 265). 

 

The Shum Laka archaeology site is located in middle of the Bamenda Grassfields that 

would come to divide the western and eastern branches of Bantu culture. Initially, these branches 

shared lithic and ceramic traditions going back to the beginning of the Holocene, but the climatic 

shift produced a barrier which would split these cultural groups for the next 1000 years 

(Lavachery 2001:240). The site is located within a rock shelter situated behind a waterfall on a 

volcano 14km southwest of the city of Bamenda (Lavachery and Cornelissen 2000:153). When 

36 

occupied during its prehistoric times, Shum Laka was roughly 1,200 to 3200 meters squared and 

used for a variety of purposes discussed more below (de Maret, Clist, and Van Neer 1987; de 

Maret et al 1993; Lavachery and Cornelissen 2000:153).  

History of archaeology in the region: 

 

The Grasslands had long been understood as the linguistic homeland of Bantu speakers 

and intermittent archaeological surveys were conducted as early as 1925 (de Maret et al 

1987:560). The systematic survey of the Bamenda Grasslands began in 1977 at the behest of the 

continental shift towards studying Bantu origins and expansion. Abeke, Mbi Crater, Matupi, and 

Njinikejem were also surveyed during these early prospections and contributed to the culture 

history of the region despite not having substantive fieldwork in the following years. Abeke 

contained a clear occupation period while Njinikejem contained what looked like the only open-

air refuse pit in the Grasslands, possibly indicating a nearby settlement (Asombang 1988).  

Nonetheless, the thickness and density of Shum Laka’s collections were a stark indication of log-

term occupation and thus could potentially questions about transitions and cultural shifts linked 

to Bantu origins and expansion in the region (de Maret 1987).  

 

In the winter of 1980 the Pierre de Maret conducted the first prospection targeted around 

Shum Laka specifically and observed, “Plusieurs sites mégalithiques et dix abris sous roche…” 

(translation: Several megalithic sites and ten rock shelters…) in the region (de Maret et al 1987: 

560). Archaeologists quickly began working at Shum Laka and in 1982, Cameroonian 

archaeologist Raymond Asombang excavated 48 square meters within the shelter to reveal lithic 

traditions extending 10,000 years (Asombang and de Maret 1992). The next wave of extensive 

work occurred in the early 1990s under the Wide Bantu Homeland Project. These series 

consisted of several excavation periods from 1991 through 1994 which would ultimately uncover 

37 

over 80 square meters of occupation and almost half a million artifacts e (Lavachery 2001: 215). 

This work was primarily funded by collaborations between the University of Yaoundé, the Royal 

Museum of Central Africa, the University of Brussels, the Belgian National Fund for Scientific 

Research, and the L.S.B. Leakey Foundation producing at least two generations of dissertations, 

master’s thesis, articles, reports, and book chapters. At the end of the 1994 field season these 

excavations ultimately recovered one of the most complete records of human activity in the 

region.  

Figure 5: Map of the Shum Laka cave site, present-day 

 

Shum Laka archaeological site: 

 

This section discusses the Shum Laka artifacts that are dated between 1500 B.C. and 200 

A.D. or the Early Iron Age (See Figure 5). This section highlights when these artifacts first 

38 

appear and how and when they are replaced over time. My goal here is to illustrate how and 

when new behaviors where introduced and which ones endured the social and environmental 

changes of the Early Iron Age. 

Lithics- Hundreds of thousands of lithic artifacts were recovered from the site within the grey 

ash layer associated with the EIA and they far outnumber any other artifact class in the cave. 

(Lavachery and Cornelissen 2000: 154). These materials were locally available from the volcanic 

and non-volcanic mountain ranges though no formal testing has been completed (de Maret 

1995:2). A few interesting features to note are the appearance of macrolithic basalt tools at the 

bottom of the layer with a steady decline through time; a sharp decline in the number of hammer 

stones that indicate on-site flaking, and a decline in the basalt industry that dominate the LSA 

period, and a clearly delineated activity area with accumulation building up in a semi-circle 

along the cave walls (Lavachery 2001:233-4; Lavachery and Cornelissen 2000:159). 

Ceramics and Metal- By contrast the quantity of ceramic artifacts was quite small, numbering 

just over 1200 fragments, one-third of which could be refitted (Lavachery 2000:160). The 

pottery can be divided into three phases “between 2500—2000 B.C.,400 B.C.—A.D. 1200 and 

A.D. 1500—modern,” based on radio carbon dates from the gray ash layer which can indicate 

either different occupation periods or the intensive erosion of the cave (Lavachery and 

Cornelissen 2000:160). They consisted of pot fragments that were low fired, rouletted, stamped, 

incised motifs (Asombang 1992). The nearby site of Njinikejem also located in the Grassfields, 

there is a refuse pit that contains pottery that typologically connect to ceramics dates to the 

earliest gray ash layer at Shum Laka and given its lack of roulette impressions predates 2500B.P. 

Thus, possibly corresponding to a sustained aesthetic, technological, and raw material practice 

that lasted through the early dispersals just north and east of the Grasslands near present-day 

39 

Yaoundé (Bostoen et al 2015:362). Metal objects are curiously absent given the long history of 

metal production in neighboring Nigerian communities such as the Nok which further illustrates 

the increasing isolation of the Bamenda Bantu speakers. The earliest metal objects date to around 

400 B.C. and consist of one iron bracelet and an iron finger ring, dating around 400 B.C. 

(Asombang 1988; Lavachery 1996:270; Lavachery 2001:242). It would be another 300 years 

before concrete evidence of a full-scale metallurgic tradition would enter the cave as evidenced 

the carved incised and grooved pottery motifs observed in the upper ash layer (Lavachery 

1996:270).  

Faunal and Floral remains- There is a significant increase in the sheer volume of animal remains, 

even when erosion and post-depositional perturbances are accounted for. The kind of animals, 

mostly from the forest stayed mostly the same throughout the ash layer and included mollusks, 

reptiles, bats, rodents, giant forest hogs (Hylochoerus menertzhageni), civets (Viverra civetta), 

monkeys, apes, and bushbucks (de Maret et al 1993: 2; Lavachery 2001:231). There is also a 

marked increase in the presence of Achatina shells beginning in the Lower Horizon with a 

tripling in the Middle Horizon (Lavachery 2001:22). Plant remains on the other hand were a 

mixture of forest trees and savanna grasses and shrubs (Lavachery 2001:221). Most notable there 

was a sharp increase in the use of wild banana (Ensete giletii) and Canarium Schweinfurthii an 

edible tree both of which are savanna plants.  There is also evidence that the latter plant was 

possibly protected and sometimes even planted based on its frequency increase during the 

Middle Horizon of the grey ash layer (Lavachery 2001:237). Lastly, wild palm (Elaeis 

guineensis), most likely the first plant to be cultivated by humans does not appear until the end of 

the middle horizon, well into the first millennium AD. 

40 

Human Remains- There were two funerary phases observed in the rock shelter. The first dating 

to roughly 5000 B.C and includes a 30-year-old anatomically male person in a contracted 

position and a “bundle of only long bones, belonging to a single individual” and a double burial 

of a 4-year-old child and a 15-year-old sub-adult with a nutcracker grave good (de Maret 

1996a:275). The second funerary period begins around 1400-1000 B.C. and corresponds to the 

earliest gray ash layer of the rock shelter (Lavachery and Cornelissen 2000:161). There were 

fourteen burials recovered from the 1991-1994 excavation of 82 m2 leaving archaeologists to 

estimate a possibility of 200 inhumations from this period (Lavachery 2001:243). One of the 

young individuals had an arrow still lodge in their hip during excavations and others had cranial 

and other bone fractures. These burials intentional and occurred in discrete tombs of mostly sub-

adults with some grave goods included (de Maret 1996). Most of them included multiple 

individuals and at least two tombs contained additional human remains such as charred and ochre 

covered bones indicating potential secondary burials during inhumation activities (de Maret 

1996:277).  

Miscellaneous features- There is evidence of one hearth from roughly 2000 B.C. and shell beads 

that appear to be even older (de Maret et al 1995:2). To date, there is no evidence of additional 

shelters or structures relating to the Early Iron Age observed in the cave. 

III. Case 1 Summary 

 

The isolation brought on by the shifting of the forests, from roughly 1500 to 400 B.C at 

Shum Laka caused a 500-year period of stagnation in terms of the development of Early Iron 

Age culture. This is evident in the gradual shift in lithics, taking almost a millennium to develop 

a macrolithic toolkit and the lack of substantial evidence of nearby village sites. By the 

beginning of the Early Iron Age then, the people of the Grasslands had developed a markedly 

41 

distinct lithic technology, regional macrolithic technology which may have resulted from the 

need to cut the new trees of the emerging forest-savanna environment (Lavachery 2001:243). 

Further, the abundance of secondary processing materials as well as a wider range of larger game 

indicate a possible significant increase in the EIA population by the middle of the first 

millennium B.C. (Lavachery 2001:231). It is clear that at least some occupants had either 

exposure to or ancient experience with metallurgy and or arboricultural techniques as early as 

1000 B.C as noted earlier the few metal artifacts and possible cultivation of savanna trees. 

Lastly, the increasing isolation of the Grasslands during the initial 500-year period of the Early 

Iron Age lead to the possibility of an increasingly inward awareness of the behaviors, beliefs, 

and practices of the group. These massive changes start to taper off by 400 B.C. leading to a 

period of stability and increasing complexity that last for the next 1000 years. Minimally, this is 

indicative of the establishment of normative cultural practices that endured the complex changes 

brought on by dispersal, agriculture, and metallurgy of the end of the first millennium B.C. 

 

 

 

 

42 

Shum Laka EIA Attributes 

Attribute 

Context 

Description 

Lithics 

gray ash layer 

Ceramic 

gray ash layer 

macrolithic basalt tools, hammer 
stones 
low fired, rouletted, stamped, 
incised motifs 

Metal objects 

Floral remains 

middle horizon, 
gray ash layer  

an iron bracelet and an iron finger 
ring 

lower horizon, gray 
ash layer 

wild banana (Ensete g.), wild tree 
(Canarium Schweinfurthii) 

Faunal remains 

lower horizon, gray 
ash layer 

mollusks, reptiles, bats, rodents, 
giant forest hogs (Hylochoerus 
menertzhageni), civets (Viverra 
civetta), monkeys, apes, 
bushbucks, Achatina shells 

 Table 2: Shum Laka Early Iron Age Attributes 

Timeframe 

1500 B.C. to 
200 A.D. 
2000 or 400 
B.C. to 80 
A.D. 
540 A.D. to 
880 A.D. 

2000 or 400 
B.C. to 80 
A.D. 
2000 or 400 
B.C. to 80 
A.D. 
 

 
 

43 

Figure 6: Map of archaeological sites of the Northern Plateau 
Stars represent locations under study (Mbida Mindzie et al 2001). 

 

44 

IV. Case 2: The Northern Plateau 

 

The Northern Plateau contains the earliest evidence of permanent settlement, food 

production, metal, and the end of the dominance of microlithic, as well as the continuous 

exploitation of both forest and savanna environments (See Figure 6). the sites that make up Case 

2 are nestled within the plateaus that slope in a western direction from the mountainous region of 

the Bamenda Grasslands and are located on the extreme north east corner of the Equatorial 

Forest. This slope created hilltops and valleys that became open air settlements, not cave or rock 

shelter settlements, during the time period (Atangana 1988:25-26). During EIA occupation, plant 

and faunal remains indicate that these sites were in a forest-savanna matrix similar to Shum Laka 

(de Maret 1980:12; Mbida Mindzie et al 2000:157: Gond et al 2013:3). The sites under 

discussion here collectively are Obobogo, Nkang, Okolo, Nkmetou, and Ndindan, a group of 

sites that run north west from Obobogo for about 60 kilometers going up the Nyong-Sanaga 

ridge. These sites are contemporaneous, within close proximity, and have similar in structures 

and assemblages which offer an expanded level analysis of cultural activities for the time period. 

Further, pottery from across the regions had similarities that suggested affinities recognized as 

the Obobogo Group tradition.  

 

Given the site’s proximity to the capital city, these sites have offered continuous 

opportunities for excavations from between 1980 and 1985. The first excavations of Cameroon 

prehistory were done by colonial administrators in the 1930s with the vast majority of work done 

in the Sahelian environment of northern Cameroon., Obobogo specifically was first identified by 

J.B. Jauzé in 1944 where he found potsherds, polished stones and axes (cited by Mbida Mindzie 

1996:31). Southern Cameroon so little activity because of the contemporary belief that the 

Equatorial Forest was generally inhospitable in ancient times. Under the urging of the 1977 Ie 

45 

Colloque international sur l'expansion Bantu (International Symposium on Bantu Expansion) 

1980 Pierre de Maret and J.M. Asombang began prospections and surveys to prepare for 

excavation to specifically explore the Neolithic elements of the site (de Maret 1980:12). These 

initial excavations offered archaeologists the first glimpse of a large refuse pit which inferred a 

permanent settlement. The first excavations were conducted in the summer of 1981 by a group of 

students; Mr. Raymond Asombang from London University, Miss C. Atangana from the 

University of Paris I, Mr. C. Digara from the University of Paris, Mr. M. Elouga from University 

of Yaounde, and Mr. Christophe Mbida Mindzié from the University of Brussels under the 

direction of archaeologists Pierre de Maret of the Royal Museum of Central African, Tevuren, 

Belgium and J.M. Essomba, Minister of Culture from Cameroon (de Maret 1982:11).  

Subsequent excavations took place in 1983 which also revealed the archaeology sites of Nkang 

and Ndindan as well forming a north-western line of sites along the Nyong-Sanaga ridge. I have 

chosen to focus on this cluster of sites as they reveal a continuous stratigraphy that relates to the 

Early Iron Age and illustrate the range of activities in the region. These data were collected from 

reports, articles, master’s theses, and doctoral dissertations completed between 1982 to 2001 

(See Table 1 for complete list of sources). 

Obobogo archaeology site: 

 

 

The first archaeology site is Obobogo which is located in a southern suburb of present-

day Yaoundé, the capital of Cameroon. Its proximity to Shum Laka and the temporal, spatial, 

and cultural connections between the two suggests that Obobogo was the first space of large-

scale dispersion and settlement outside of the Bamenda Grasslands around 400 B.C (Essomba 

1989:53). Thus, the culture history of Obobogo represents the best indication of the development 

of homeland connections for Bantu speaking people. Obobogo is an open-air site defined by 8 

46 

refuse pits that were exposed as a result of construction. The work was carried out in two 

trenches, the first, Trench A, corresponded to Ceramic Stone Age Tradition which included 

grinders, decorated pots, and millstones (Essomba 1989:37).  Bellow Trench A was a Late Stone 

Age occupation which produced quartz microlithic that were like most of the region dated to 

roughly 5000 B.P. (Essomba 1989:37; de Maret 1982:11). Initial radiocarbon dates from the 

1980 excavation suggested that these pits dated to the beginning of the first millennium B.C. (de 

Maret 1982). However, subsequent research in and around the area as well as the ceramic 

typology developed by Phillip Claes suggest a much later date, more than likely to the second 

half of the first millennium (Claes 1985:145). The Early Iron Age at Obobogo extends from 1100 

B.C. until the end of the first millennium B.C. and this Trench B assemblage is presented below 

(Essomba 1989:37).  

Features- Eight pits were excavated in 1980 and they included ceramics, lithics, floral, and iron 

materials mentioned above 

Lithics- The lithics recovered from roughly 40-50 centimeters below surface of Trench B 

consisted of polished stones, adze, polish axes, groove stones and grindstones which dated to 

around 1000 B.C.  

Ceramics- The ceramics included richly decorated potsherds and consisted of a variety of forms 

and styles (Claes 1985: 131). The pottery fragments show variation between pits as well as 

similarity in form and decorative style (Claes 1985:125). Generally, the shapes are spheroidal or 

oval, and a concave neck (Claes 1985:126). The bottoms are flat and they are usually decorated 

with lines, circles, and other waves and polycentric designs (Claes 1985:127). 

Fauna- There were no osteological remains reported at this site (Claes 1985:137). 

47 

Flora- There was an ‘abundance’ of, edible palm tree nuts such as Elaeis guineensis (oil palm) 

and Canarium Schweinfurthii (fruit bearing tree) (de Maret 1982:12). 

Metallurgy - There was undated slag found in Pit 7 and dated to the second century B.C. 

(Essomba 1989:38).  

Ndindan and Oliga archaeological sites: 

 

As stated earlier, the Ndindan archaeology site was found during 1983 regional 

prospections following field work at Obobogo. Ndindan is located roughly 15 kilometers west 

from Yaoundé and about 10 kilometers northwest from the Obobogo archaeology site. This site 

was excavated by C. Mbida Mindzie as part of the regional field work on Early Iron Age 

villages. Based on radiocarbon dating this hilltop village, was occupied between mid-7th century 

B.C. to the end of the first millennium A.D (Mbida Mindzie 1996:630). In 1988, aided by 

additional construction in the area, J.M. Essomba and J. Monteillet discovered the site of Oliga at 

the base of Ndindan hilltop just over a kilometer away (Essomba 1989:44; Mbida Mindzie 

1996:36). Radiocarbon dates from Oliga span the last millennium B.C. into the third century AD, 

though most of them overlap with the occupation dates from Ndindan (Essomba 1989:44-45; 

Mbida Mindzie 1996:629). 

Features: Nindan consisted of 7 shafts that contained stone tools, ceramic fragments, and iron 

slag (Essomba 1989:39). The only feature excavated at Oliga was an iron furnace. 

Lithics- Lithics from Ndindan were primarily made up of local sources such as quartz and silica 

and consisted of millstones, wheels, and strikers typical of the first millennium B.C. of the region 

(Mbida Mindzie 1996:638).  

Ceramics- Ceramics consisted of flat bottom vessels with beveled edges (Essomba 1989:39). 

48 

Fauna- There were no bones or indication of faunal remains found in the pit (Mbida Mindzie 

1996:629). 

Flora: Canarium schweinfurthii and Elaeis guineensis endocarps were present in all the pits and 

the rest of the materials await further study (Mbida Mindzie 1996:629). Pennisetum sp. was also 

embedded in two ceramic sherds found at the site (Kahlheber et al 2014:498). 

Metallurgy - Iron slag was found in one of the pits at Ndindan and dated to 400 B.C. to 400 A.D. 

(Mbida Mindzie 1996:638, 641). The Oliga fieldwork revealed an iron furnace at the eastern 

base of the Ndindan hilltop site that structurally resembled the construction of 6th century B.C. 

Nok furnaces from Nigeria (Essomba 1989:44; Mbida Mindzie 1996:493,629).  

Nkang archaeological site: 

 

The Nkang archaeology site is an open-air site located 70 kilometers northwest of 

present-day Yaoundé (Mbida Mindzie et al 2000: 151). As noted earlier the site was first 

recorded in 1983 during field work related to Obobogo during local road construction which 

exposed the 14 refuse pits found in situ inside an embankment (Mbida Mindzie et al 2000:153). 

The 1983 field work consisted of excavating fully excavating and analyzing 3 of the 14 pits 

ranging from 3 to 15m2 and charcoal dates the site to between 500 and 150 B.C (Mbida Mindzie 

et al 2000:152). No Late Stone Age materials were recovered leaving archaeologist to determine 

that the Early Iron Age lasted from roughly 2580-2170 B.P. (Mbida Mindzie et al 2001:1)  

Features- This site contained 14 shafts that contained pottery and lithics, eleven of those were 

excavated and analyzed (Mbida Mindzie et al 2000:1). Mbida et al describe the pits as “either 

bottle-like, ovoid or cylindrical in shape, and their volumes ranged from 3 to 15 m2 (Mbida 

Mindzie et al 2000:152). The pits were filled with “local soil, broken ceramics and stone tools, 

iron slags, charcoal and faunal remains (Mbida Mindzie et al 2000:153). 

49 

Lithics- Lithic materials were similar to Obobogo and consisted of grinders, grinding stones, 

hammerstones, and polished axes (Mbida Mindzie et al 2000:153).  

Ceramics- Ceramics at Nkang had a collar neck and were oval shaped and appeared to be 

decorated with nut shells and contained had impressions on both the inside and outside (Mbida 

Mindzie et al 2000:153).  

Faunal- Animal remains included fish such as Lates niloticus (perch), some goats, sheep, and 

shells such as mollusks and Achatina shells (land snails) (Mbida Mindzie 1996:650-651, 653; 

Mbida Mindzie et al 2000:159). The Achatina shells had punctured holes in the center (Mbida 

1996:200). 

Flora- In addition to the typical Elaeis guineensis and Canarium Schweinfurthi, found at 

Obobogo and Shum Laka, Nkang also contained evidence of Musa (wild banana) found in Pit 9 

180-300cm deep and dated to between 770-380 B.C. (Mbida Mindzie et al 2001:5). The 

phytoliths were found both on charred residues found on pottery as well as within the sediments 

of the pit (Mbida Mindzie et al 2000:154). 

Metallurgy- Slag and tuyere fragments were recovered from two pits which date to the EIA. 

(Mbida Mindzie 1996:648). 

V. Case 2 Summary 

 

Similar to Shum Laka, this region had Late Stone Age materials that dated between 5000-

4000 years B.P. with a Ceramic Stone Age appearing at the end. In terms of Early Iron Age 

materials, the lithics seemed to be axes, adzes, grinding stones and other tools that were typically 

used for chopping trees, grinding grains, and other activities typically associated with village life. 

 

The region is defined by the Obobogo Group ceramic tradition, which is found across 

Obobogo, Nkang, Ndindan, and Oliga (Claes 1985, 1992). These ceramics are characterized by 

50 

their collared neck and oval shape (Claes 1985; Mbida Mindzie et al 2000:153). The pottery at 

Nkang is slightly distinct in that it is slightly thicker than what is found in other parts of the 

region (Mbida Mindzie et al 2000:153).  This is most likely because Nkang is considered the 

northern limit of the Obobogo tradition (Claes, 1985; de Maret, 1991; Mbida Mindzie et al 

2000:153). The ceramics from Nkang also represented a shared understanding of pottery 

production although some variation was present (Mbida Mindzie 1996: 475).  Although the 

pottery tradition is similar, there is no indication of a pottery trade between the sites but was 

most likely produced at the household or village level across the sites (Mbida Mindzie 

1996:645). 

 

Due to the acidity of the soil in this region faunal remains-specifically bones are scarce.  

The majority of the faunal information then comes from Nkang. Some interesting elements are 

the perch and the Achatina shells. The perch is not local to the area and was more than likely 

obtained through trading (Mbida Mindzie et al 2000:159). Nkang also has sheep and goat bones 

that indicate a modest breeding tradition, based on the low quantity of bones found (Mbida 

Mindzie et al 2000:159). The Achatina shells are interesting because they 1) come from the 

Atlantic Ocean; 2) had holes punctured into them; and 3) have often been linked to trade 

currencies in the region. Mbida Mindzie determined that the shells were most likely traded 

through local micro regional trades from the coast into the region and then used as currency in a 

northern direction along the rivers (Mbida Mindzie 1996:653). The circular holes in the middle 

of the shells may also be attributed to adornment but further analysis is needed (de Maret, 1985b: 

166; Mbida Mindzie 1996:650). 

 

The floral evidence at these sites was well preserved and provide substantive information 

on the dietary, technological and trade practices of this region.  Each of the sites contained Elaeis 

51 

guineensis and Canarium Schweinfurthii indicative of adaptation to both forest-savanna mosaic 

environments. One idea is that the trees found at Nkang were used for construction and for fuel 

for metal production (Mbida Mindzie et al 2000:157). Nkang however had evidence of Musa sp., 

which was domesticated in Indonesia and no wild species have yet been observed in prehistoric 

Africa (Mbida Mindzie et al 2000:157; Mbida Mindzie et al 2001:2). Therefore, evidence of EIA 

cultivation in this region can only be the result of a long-distance trade network stretching to East 

Africa (Mbida Mindzie et al 2000:157). As Mbida Mindzie et al explains, "… Musa is of Asian 

origin, and this study provides the first concrete evidence of contacts across the Indian Ocean a 

millennium earlier than currently accepted" (Mbida Mindzie et al 2001:1). Further, this plant is 

an ideal crop for rainforest agricultural practices and are thought to indicate an increase in the 

population capacity of these villages (Vansina 1990; Mbida Mindzie et al 2001:5). Lastly, this 

plant is significant as it is the first archaeologically identified crop in Central Africa and it 

appears in the first millennium B.C. (Mbida Mindzie et al 2000:157).  

 

While iron or slag fragments were found at all the sites, the most significant find is the 

furnace at Oliga as it represents an intimate knowledge of production rather than the acquisition 

of items through trade. This feature then illustrates that these interconnected villages began using 

iron technology just as they began to disperse into the forest. The peculiar shafts found at 

Obobogo, Nkang, and Ndindan are also found throughout the Congo River Basin. While the uses 

appear to vary from site to site, they are generally interpreted as either animal traps, some sort of 

grain storage system, clay extraction reserves, and most recently grave sites (Mbida Mindzie et al 

2000:152; Meister and Eggert 2008:199). 

 

The general activities of this region consisted mostly of local production of crops, iron, 

and pottery with access to a vast trade networks to the north, south, and east. These villages most 

52 

likely were mainly hunters and used traps and snares as well as a regional trade economy. There 

is also a shift in lithic technology towards more agriculturally focused activities such as crop 

cultivation and production and animal husbandry (Mbida Mindzie et al 2000:160). This cluster of 

late first millennium B.C. hilltop settlements is currently the earliest known settlement outside of 

the Bamenda Grasslands. Obobogo was the largest space of all of the sites and could have been 

the regional market location based on its variety and volume of materials while Oliga could have 

been the iron production site within Ndindan village or its closest neighbor (Mbida Mindzie 

1996).  

 

The settlement of Obobogo and surrounding villages is the earliest complete record of 

Early Iron Age activities- villages, iron production, agriculture, effective use of savanna-forest 

environments- but also the development of extensive trade networks spanning from the northern 

Sahelian areas, down to the southern Atlantic coast and into the East African- Indian trade 

economies. These changes then began appearing in successive ways both in and around the 

forest-savanna regions. The next case study looks at such changes in an interior part of the forest 

to illustrate where and when these practices begin their eastward development. 

 

 

53 

The Northern Plateau EIA Attributes 

Attribute 
 

Context  Description 

Timeframe 

Lithics 

Obobogo   polished stones, adze, polish axes, groove 

1100 B.C.- 100 B.C. 

stones and grindstones 

  

  

Ndindan  millstones, wheels, and strikers 

Nkang 

grinders, grinding stones, hammerstones, 
and polished axes 

750 B.C. - 1000 A.D.  

500 B.C. - 150 B.C.  

Ceramics  Obobogo 

  

  

Ndindan 

Nkang 

spheroidal or oval shapes, a concave neck; 
flat bottoms, decorated with lines, circles, 
and other waves and polycentric designs  
flat bottom vessels with beveled edges  

collar neck with oval shaped; decorated 
with nut shells and impressions on the 
inside and outside 

Metal 

Obobogo 

slag 

Ndindan 

iron slag 

Oliga  

Nkang 

iron furnance 

Slag and tuyere  

1100 B.C.- 100 B.C. 

750 B.C. - 1000 A.D.  

500 B.C. - 150 B.C. 

200 B.C. 

400 B.C. to 400 A.D.  

1000 B.C. - 300 A.D. 

500 - 150 B.C 

  

  

  

Floral 
Remains 

  

  

Obobogo  wild palm (Elaeis guineensis), wild tree 

1100 B.C. - 100 B.C. 

(Canarium Schweinfurthii)  

Ndindan  wild palm and wild tree endocarps 

750 B.C. - 1000 A.D.  

Nkang 

wild palm, wild tree, and wild banana 
(Musa) 

770 B.C. - 380 B.C. 

Faunal 
remains 
  

Obobogo 

none found 

Ndindan 

none found 

  

  

  

Nkang 

perch (Lates niloticus), goats, sheep, 
mollusks and land snails (Achatina) 

500 - 150 B.C 

 

Table 3: The Northern Plateau Early Iron Age Attributes 

 

 

 

 

 

54 

VI. Chapter Summary 

 

This chapter presented the data associated with the initial phase of the Early Iron Age in 

the Bamenda Grasslands and the Northern Plateau regions of the Congo River Basin. The 

Bamenda Grasslands have been the well-established as the linguistic homeland of the Bantu 

speakers (Asombang and de Maret 1992:13; Li et al 2014:2; Lavachery et al 2000:154; Vansina 

1984:132-135). However, the archaeological signatures present an alternative view. There is 

little evidence of a complete Early Iron Age package and their period of isolation during the 

formative years of the emergence of the Early Iron Age suggest that they were most likely not 

the origin of such behaviors. This may be why scholars like Lavachery insist on describing this 

period as the “Stone to Metal Age” because it does not ultimately represent the behaviors that 

would emerge alongside early Bantu speakers (Lavachery 1996).  This case document the some 

of the activities that were present during the beginning of the Early Iron Age. Case 2: The 

Northern Plateau provides the evidence for the initial settlements outside of the Bamenda 

Grasslands and evidence of local, regional, and interregional trade from the Atlantic Ocean to the 

west and Sahara to the north. These village settlements demonstrate that these occupants had iron 

technology, began cultivating wild bananas, and pearl millet, and they became accustomed to 

their forest-savanna mosaic environment: a shift from the isolationism of Shum Laka. The 

Northern Plateau also exhibits the Early Iron Age characteristics that who would eventually 

inhabit central, southern, and eastern regions of the African continent. This cultural complex was 

not just a random, disjointed assortment of practices; rather they had established subsistence, 

technological, and trade networks that indicated a group awareness. The next chapter will 

illustrate when, how, and where these features appeared across the Congo River Basin. 

 

 

55 

CHAPTER 5: 
CASE 3: THE FOREST INTERIOR AND CASE 4: THE LOWER CONGO 

I. Introduction 

 
 

 

This chapter presents case studies 3: The Forest Interior and 4: The Lower Congo as they 

illustrate Early Iron Age activities a significant distance from the Cameroon sites of the last 

chapter. Case 3 focuses on the archaeological work conducted within the forest in and around the 

Congo and Ubangi River split. Case 4 examines the occurrence of these activities along the 

southern extremities of the forest in the Lower Congo region. Together, these cases illustrate the 

evidence of the development of these practices and for trade across the region.  

2 

3 

1 

Figure 12: Map of archaeological sites of the Forest Interior 

1) Imbonga Horizon; 2) Batalimo-Maluba Horizon; 3) Pikunda-Munda Horizon. 

56 

II. Case 3: The Forest Interior 

 

Current theories on Bantu expansion suggests that there were two pathways, one in an 

eastern direction further into the forest and one in a southern direction along the coast. Case 3 

presents evidence for the former. This case focuses on the area north and south of the Congo 

River bend in northern Democratic Republic of the Congo (See Figure 7). Due to its location 

around the Congo River, this location contains to numerous rivers and tributaries that connect the 

Congo River to into Central African Republic to the north and the Republic of Congo to the west 

(Eggert 1993: 295). Despite the climatic changes that occurred around 1500 B.C., the numerous 

rivers and annual rainforest have fueled the growth of the evergreen rainforest that is known 

today (Eggert 1993:294). This part of the inner basin did not experience a forest retreat as seen at 

Obobogo and Shum Laka but maintained a mature, secondary forest mosaic throughout the time 

period under consideration here (Kahlheber et al 2014: 501). It is for these reasons that for at 

least 100 years, colonial administrators, historians, researchers, and archaeologists viewed 

Equatorial Forest as a whole as in hospitable and impenetrable (Eggert 1993:290). These 

perceptions of a static unchanging forest were disrupted during the mid-1980s as environmental 

specialists began to understand the relative fluctuations in the forest density and extension and 

archaeologists began to understand the extensive and ever adapting tool kit of Central African 

people (Mercader and Brooks 2001:197-198). Within that spirit of renewed possibility of human 

activity deep within the forest, Eggert and a team of archaeologist from Germany began 

investigating such activities to understand who these early occupants were, how they survived 

within the forest, and how they interacted with the regions around them. Do date, this work 

remains the most extensive archaeological research conducted within the forest and presents the 

best indications of Early Iron Age occupation.  

57 

 

The initial archaeological survey of the region was led by Manfred Eggert and took place 

in 1977. This archaeology team returned in 1983 as the Rivers Reconnaissance Project and 

mapped the riverways between the Congo River and the Ubangi river which headed north into 

Central African Republic. Funded by Deutsche Forschungsgemeinschaft of Germany, this 

project, "surveyed several hundred villages and other places of potential archaeological interest" 

across an overall distance of 5000 kilometers (Eggert:1987:129; Eggert 1993:295). This project 

was focused on developing local chronologies that would serve as a backbone for future research 

(Eggert 1993:295). The results included an enormous among of ceramic materials which at 

sometimes were the only materials recovered from the surveys and excavations. This led to the 

identification of three ceramic horizons that make up the cultural history of the region- the 

Imbonga, Batalimo-Maluba, and the Pikunda/Munda horizon. As such, Case 3 focuses on the 

Early Iron Age assemblages that correspond with these ceramic traditions.  

Imbonga Horizon: 

 

The Imbonga type site was first surveyed in 1977 with subsequent excavations in 1983 

along the Momboyo river bank (Eggert 1987:130). This ceramic horizon dates between 400 and 

200 B.C (Eggert 1993:306). The pottery was found across 54 separate sites along the Momboyo 

River and its connecting waterways, 23 of which were fully excavated across a 10-year period 

(Kahlheber et al 2014:483). 

Features- Circular pits containing Imbonga pottery were observed throughout the riverbank 

surveys. These pits were up to 1 meter across and 1.7 meters deep any “densely packed” with 

pottery fragments as well as in tact pots (Eggert 1987: 131). It appears that the broken pots were 

most likely broken in situ as they could be put back together in the laboratory with relative ease 

(Kahlheber et al 2014: 483) 

58 

Lithics- One curiosity of the Imbonga type site and subsequent findings connected to the horizon 

is that there are no lithics observed over the course of 4 seasons of fieldwork (Egger 1987:131). 

Ceramics- Imbonga pottery is known for its flat base, regardless of vessel shape or use, and 

grooved, impressed, incised, or adorned pots, bowls, and bottle and beaker like forms (Eggert 

1987:130). The pottery dates to the second half of the first millennium (Eggert 1993:306). 

Fauna- No faunal remains were described. 

Flora- Soil sieving revealed endocarps of Elaeis guineensis and fruits from the Canarium 

Schweinfurthii (Eggert 1987;132; Eggert 1993:323) 

Boso-Njafo: 

 

Boso-Njafo was first excavated in 1985 as part of the River Reconnaissance Project. This 

site is located on the Lulonga River roughly 124 kilometers from Imbonga (Eggert 1984:248). 

Features- 12 circular pits filled with pottery fragments were found in area of roughly 300m2 and 

three of these pits were analyzed (Wotzka 1995:382). These pits contained multiple intruding 

pits that indicated continuous reuse after the middle third century calibrated B.C. (Eggert 

2014:484).  

Lithics- No lithics were reported from this site.  

Ceramics- Ceramic analysis connects this site to the Imbonga horizon (Kahlheber et al 2014).  

Fauna- No fauna remains were reported from this site. 

Floral- Over a dozen species of plants were identified from two pits at the site. The most 

“common and abundant plant” species is Elaeis guineensis but no Canarium s. was reported 

(Kahlheber et al 2014:493). There was also Pennisetum glaucum (pearl millet) materials found in 

in one of the pits as far as 80-100 centimeters deep, as well as Raphia palm and Aframomum sp. 

59 

(a member of the ginger family).  (Kahlheber et al 2014: 486, 494, 500). See Kahlheber et al 

2014 pages 487-492 for a complete list of plant materials found in the pit.  

Metallurgy- No iron materials were reported from this site.  

Batalimo-Maluba Horizon: 

 

Located on the northern side of the Lobaye River, a tributary of the Ubangi River in 

present day Central African Republic, the Batalimo archaeology site was excavated in 1968 by 

R. de Bayle des Hermens (Eggert 1987:135). In 1985, the River Reconnaissance Project found 

similar materials at Maluba, a separate archaeology site located on the Lua river, an eastern 

tributary of the Ubangi located less than 50 kilometers from Batalimo (Eggert 1993:306). The 

ceramics recovered from this site also contained flat bottoms, were highly decorated and had 

similar neck and rim shapes as the Batalimo tradition but were dated to the last half of the first 

millennium B.C., predating the Batalimo tradition by a few hundred years (Eggert 1993:308). 

Thus, the two traditions form the Batalimo-Maluba Horizon dating from roughly 400 BC to 562 

A.D. (Eggert 1993:311). The discussion below describes the assemblage from the Maluba site 

with mention of the lithics and ceramics recovered from Batalimo.  

Features: Maluba had an interesting feature that consisted of three intersecting pits varying from 

between 40 to 100 centimeters in diameter (Eggert 1987:137). The pits contained various 

quantities of Batalimo pottery, but Pit 3 also contained human remains. The burial consisted of a 

secondary burial of a human skull and fragmented long bones. The bones were organized with 

the skull and two long bones, “an ulna and possibly a radius” in the middle of the 5 other long 

bones arranged in a semi-circular pointed towards the skull (Eggert 1987:139). Pit 3 appeared to 

be younger than Pit 2 but its relationship to Pit 1 is yet to be determined. Only Pits 1 and 2 could 

60 

be dated and they range from around 140 B.C. to roughly 70 B.C. with a controversial fourth 

date of 3rd century B.C. for Pit 1 (Eggert 1987:141). 

Lithics- Hermens discovered several hundred ceramic fragments as well as stone tools; polished 

stone axes, scrappers, flakes, and unmodified waste which dated to the middle of the first century 

A.D, (Eggert 1993:311). There were no lithics reported from Maluba (Eggert 1993:326). 

Ceramics- The pottery in this horizon can be divided into two categories; the first group consists 

of flat-based and richly decorated vessels while the second group consists of undecorated bowls 

with a slightly constricted opening and an accentuated foot (Eggert 1987:134 citing Aumassip 

1975). These ceramics, with characteristically intricate designs and flat bottoms, were also well 

polished and highly refined for typical EIA wares (Eggert 1987:136). The pottery specifically 

from the Maluba excavations were specifically “well-structured globular pots and wide mouthed 

bowls” (Eggert 1933:308). 

Faunal- No faunal remains were reported (Eggert 1993:325). 

Floral- Endocarps of Elaeis guineensis were recovered from unspecified context at Maluba 

(Eggert 1993:324). 

Metallurgy- No iron objects were reported (Eggert 1993:326). 

Pikunda-Munda Horizon: 

 

In 1987 the River Reconnaissance project conducted a 6-month field season on the 

Sangha-Ngoko rivers a northern tributary of the Congo River (Eggert 1992: 10). A few sites 

were found of which two have been adequately tested and are thus included in this present study: 

Pikunda and Maluba. 

Pikunda 

61 

Features- The field team partially excavated a shaft that was 3 meters deep which contained two 

pottery phases and a small iron tool (Eggert 1993:326).  

Lithics- No lithics were reported. 

Ceramics- These ceramics consisted of horizontal bands and wavy lines as well as comb-

stamped decorations, round bases, bottle- like (Eggert 1993:314).  

Faunal- No faunal remains were reported (Eggert 1993:325). 

Floral- No floral remains were reported. 

Metallurgy- One small unidentified iron tool with an apparent spatula like blade at one end was 

found in the lower half of a shaft (Eggert 1993:326). 

Munda 

 

Located on the Likwala-aux Herbes river less than 50 kilometers north of Pikunda, the 

Munda site also defines the ceramic horizon for this region (Eggert 1992:16; Eggert 1993:305). 

Similar to the style and form of the Pikunda pottery, the Munda pottery dates between 353 B.C. 

and 435 A.D. (Eggert 1992: 16). 

Features- This site contained two shafts and a “shallow bowl-like” pit (Eggert 1993:311). One of 

the shafts was packed with pottery and displayed evidence of intense heat and contained a 

“considerable portion of slag” (Eggert 1993:326).  

Lithics- None reported from the Munda site. 

Ceramics- Consisted of “wide-mouthed bowls with approximately parallel sides and a flared 

rim” and included globular pots, bottle-like vessels and round bases (Eggert 1993:311). The 

designs were grooved and incised horizontal lines (Eggert 1993:311). 

Faunal- None reported from the site (Eggert 1993:325). 

62 

Floral- Endocarps of Elaeis guineensis were recovered from unspecified context (Eggert 

1993:324). 

Metallurgy- The shafts showed evidence of “intense heat” and the fill contained “a very 

considerable proportion of iron slag” (Eggert 1993:326). 

III. Case 3 Summary 

 

In general, Early Iron Age interior forest settlements begin in the second half of the first 

millennium B.C. and continue until 400 A.D. Pottery entered the region as early as 400 B.C. and 

iron production soon followed as early as the second century at Munda. Individually, these sites 

represent different environments of EIA development- Imbonga situated south of the Congo 

River is deep within the forest; while Batalimo-Maluba represents the cross currents between the 

savanna regions to the north and the forest regions south of the Congo River and Pikunda-Munda 

represents the Sangha River Interval region, a diverse ecosystem of savannas and secondary 

forests that may have been a throughway for early dispersal from Cameroon and into Central 

African Republic and DRC(White: 1979:30; Bostoen et al 2015:355).  

 

Like Case 2, the features in this region tended to be large pits and shafts that were nearly 

full of ceramics. Though further study is needed to understand all of the pits, they tended to be 

interpreted as refuse pits, with the exception of potential furnace at Munda and the secondary 

burial at Maluba. The only lithics that are reported from the area belong to the Batalimo 

archaeology site, which Eggert suggests may have been a lithic workshop given its abundance of 

unmodified waste (Eggert 1993). Further, the poor preservation conditions of the forest 

prevented the archaeological recovery of faunal and osteological remains.  

 

Hans-Peter Wotzka conducted a ceramic analysis and concluded that there is a potential 

ancestral link, or Co-Tradition, emerged between the development of Imbonga and Obobogo and 

63 

then later Imbonga and Batalimo (Clist 1989:80; Wotzka 1995:286). Imbonga ceramic design 

motifs such as the “so-called impression piovante” seem to connect to Saharan and Sahelian 

regions farther north, while other such as flat bottoms connect to Obobogo (Eggert 1984:282-5; 

Claes 1985:127). Nonetheless, at present no known precursor to this tradition has been identified 

and dated as such. Batalimo-Maluba and the Pikunda-Munda emerge along the Sangah River 

Interval (SRI) a region that is closer to potential Cameroonian and Saharan access points, which 

may indicate a continuous flow of occupants into the region as that area is younger but also 

closer to the homeland area.   

 

Another issue is the lack of “foreign” ceramics among all of the pits except Pikunda 

which has been interpreted as the result of little to no interaction (Eggert 1987:144; Eggert 

1993:322; Wotzka 1995:287). This is based on the fact that none of the ceramic traditions were 

observed within the same context: meaning no Imbonga pottery was found on the Ubangi and 

vice versa. This suggests that either these groups maintained strict ceramic production 

boundaries over time despite their contemporaneity, locality, and evidence of trade into both 

local and non-local savanna environments (Kahlheber et al 2014:503).  

 

Palynological data indicates that the forest was at least a secondary forest if not a primary 

forest during the EIA and thus the occupants did not live in a forest-savanna matrix like those at 

Shum Laka or the Nyong-Sanaga. However, floral remains give us some indication that the 

forest dwellers most likely had access to a forest-savanna mosaics. Imbonga, Boso-Njafo, and 

Munda all contained evidence of Elaeis guineensis which typically needs a sunny environment 

than the forest can provide. Also, the evidence of Pennisetum glaucum (pearl millet) materials 

indicates interaction with a savanna environment because of its hydrophobic properties which 

make the basin is a terrible climate for it (Kahlheber et al 2014:494). According to Kahlheber et 

64 

al (2014) it appears to be domesticated and most likely came from the Sahel region to the north 

or minimally from Cameroon.  It is typically a forest/savanna border crop when it is wild 

(Kahlheber et al 2014: 497). It could have been traded into the region or the occupants could 

have developed the necessary technology to grown it such as elaborate drainage system, though 

more work is needed to know for sure (Kahlheber et al 2014: 493, 502). According to Kahlheber 

et al (2014) pearl millet is not even cultivated in the interior today as a result of the rainfall 

(Kahlheber et al 2014:498). Further, it is significant because it now adds grains to what has 

typically been considered the Equatorial diet, namely yams (Khalheber et al 2014:504). There 

are only three other archaeological sites that have pearl millet which are all located in southern 

Cameroon and were also discovered by German research teams (2014:498). Lastly, the swampy 

river environment also provided several different trees and grasses for use in the forest such as 

Raphia palm (a member of the palm family) and Aframomum sp. (a member of the ginger 

family). Even given the fact that the forest borders have fluctuated through times, these plants 

are not typical of this environment and most likely indicate that the occupants maintained at least 

some level of interaction with savanna dwellers over the time period (Kahlheber et al 2014:503). 

 

In terms of general technology of the region, the curious absence of lithics and the two 

instances of iron use make it difficult to assess the nature of stone to metal transition. With the 

lack of stone at Imbonga and the use of stone at the younger site of Batalimo and iron at the 

seemingly intermediate sites of the Pikunda- Munda transition, I agree with Eggert’s assessment 

that these transitions were not linear or uniform when it comes to the forest. Further, the river 

systems made these regions more accessible, rather than isolated as was thought to be the case 

and the archaeological evidence suggests that such travel was normative during the period 

(Eggert 1993; Kahlheber et al 2014:496).  

65 

Forest Interior: Imbonga Horizon Early Iron Age Attributes 

Attribute 

Context 

Description 

Lithics 
  
Ceramic 

Imbonga sites 
Boso-Njafo 
Imbonga sites 

  

Boso-Njafo 

none 
none 
flat base, grooved, impressed, incised, 
or adorned pots, bowls, and bottle and 
beaker like forms  

flat base, grooved, impressed, incised, 
or adorned pots, bowls, and bottle and 
beaker like forms  

Metal Objects 

Imbonga sites 

none 

  
Floral Remains 
 
 

Boso-Njafo 
Imbonga sites 

none 
wild palm (Elaeis guineensis) 
endocarps  and wild fruit tree 
(Canarium Schweinfurthii) 

  

Boso-Njafo 

wild palm; pearl millet (Pennisetum 
glaucum); raphia palm; ginger 
(Aframomum sp.) 

Duration 

  
  
<500 B.C. 

<500-350 
B.C. 

  

  
<500 B.C. 

<350 B.C. 

Faunal remains 

Imbonga sites 

  

Boso-Njafo 

none 

none 

Table 4: Imbonga Early Iron Age Attributes 

 

  

  

 

 

 

 

66 

Forest Interior: Batalimo-Maluba  Horizon Early Iron Age Attributes 

Attribute 
Lithics 
  

Context 
Malabu 
Batalimo 

Ceramic 

Maluba 

Description 
none 
polished stone axes, scrappers, flakes, and 
unmodified waste  
flat bottomed, highly decorated, distinct 
neck and rim shapes, well structured 
globular pots and wide mouthed bowls 

Timeframe 
  
<150 B.C. 

400 B.C. - 562 
A.D.  

  

Batalimo  Consistent with Maluba ceramics 

<150 B.C. 

Metal Objects  Maluba 
Floral Remains  Maluba 

Faunal remains  Maluba 

none 
wild palm (Elaeis guineensis) endocarps   
 
 
none 

  
400 B.C. - 562 
A.D.  

  

 
Table 5: Batalimo-Maluba Early Age Attributes 
Attributes  
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

67 

Attribute 
Lithics 
  
Ceramic 

Forest Interior: Pikunda-Munda Horizon Early Iron Age Attributes 
Timeframe 
  
  
110 B.C. - 120 A.D. 

Context 
Pikunda 
Maluba 
Pikunda 

Description 
none 
none 
horizontal bands, wavy lines, 
comb-stamped decorations, 
round bases with bottle- like 
forms   
wide-mouthed bowls, parallel 
sides, flared rim; globular pots, 
bottle-like vessels, round bases; 
grooved, incised horizontal 
lines  

  

Munda 

353 B.C. - 435 A.D. 

Metal Objects 

Pinkunda 

unidentified iron tool  

110 B.C. - 120 A.D. 

  

Munda 

shaft, iron slag  

353 B.C. - 435 A.D. 

Floral Remains 

Pikunda 

none 

  

  

Munda 

wild palm (Elaeis guineensis) 
endocarp 

353 B.C. - 435 A.D. 

Faunal remains 

Pikunda 

  

Munda 

none 

none 

Table 6: Pikunda-Munda Early Iron Age Attributes 
 

  

  

 

 

 

 

 

 

68 

Figure 8: Map of archaeological sites of Lower Congo 

Circled sites are discussed below (de Maret 1986:105). 

 

 

 

 

69 

IV. Case 4: The Lower Congo 

 

The Lower Congo region of the Democratic Republic of the Congo make up the 

southwestern range of the Equatorial Forest. This area stretches from the mouth of the Congo 

River in the Democratic Republic of the Congo to its present-day capital of Kinshasa (See Figure 

8) (de Maret 1975:133; de Maret 1986:105). Therefore, a discussion of the cultural practices of 

this area offer a glimpse into how a similar forest/savanna environment affected the spread of 

Early Iron Age culture nearly a millennium after it developed in the Bamenda Grasslands. 

Although this region is relatively young in terms of EIA occupation, it has the longest history of 

systematic archaeology and thus provides information for a greater range of sites and activities 

then other parts of the Equatorial Forest. 

 

The first archaeological investigations of the Lower Congo began as early as the 

nineteenth century with a study by Stainier titled L’age de la pierre au Congo (Stainier 1899). 

Subsequently, periodic, though extensive excavations occurred from the 1930s and 1960s with 

some project focused on establishing stone age chronologies and others focused on historic sites 

such as those related to the Kongo Kingdom (de Maret 1984:40-42). With regard to the Early 

Iron Age archaeology in Lower Congo, the first excavations were carried out by M. Bequaert in 

1939 in Mbanza Ngungu/Thysville, where he identified seven pottery groups for the area and R. 

F. van Morsel who excavated the Malebo Pool around Kinshasa in 1949 (Bequaert 1962:337; de 

Maret 1984:42). Later, G. Mortelmans excavated Ngovo, Dimba, and Ntadi Ntadi caves in 1957 

and 1959 to establish 6 ceramic sequences for the area (de Maret 1986:106). Mortleman’s Group 

VI is relevant here as this group was dated to the EIA whereas the other six groups relate to 

much more recent times into the Later Iron and Historic Ages (de Maret 1986:125, 127). The 

next phase of work began in 1972 and through 1984 under the direction of Pierre de Maret and 

70 

include surveys and excavations of Ngovo, Dimba, Ntadi-Ntadi, and Sakuzi rock shelters and 

open-air sites as well as several others (de Maret 1986). The four mentioned sites are included as 

they are had systematic excavations and are dated to the period in question. 

Ngovo archaeology site  

 

Ngovo is a rock shelter located south-east of present day Mbanza-Ngungu (de Maret 

1986:106). Mortelmans’ excavations revealed slag, lithics, and identified 6 groups of pottery 

which have undergone much revision since then (Mortelmans 1962: 419-420). De Maret 

excavated two 1x2 meter test pits to reveal a few Late Stone Age artifacts and a thick grey ash 

layer roughly 200 meters squared dating to 195 ±45 B.C. calibrated in his initial test excavations 

(de Maret 1986:109).  

Features- There were no features reported from Ngovo. 

Lithics- There was one polished stone axe found in Test Pit A and a few shards of cut stone that 

was associated with materials that dated to 195 ± 45 B.C. and no lithic materials found in Test 

Pit B (de Maret 1977:5; de Maret 1986:109). 

Ceramics- There was an average of 805 ceramic fragments per cubic meter at Ngovo for a total 

of 30 vessels (de Maret 1986:109, 121). The materials are attributed to Mortelmans Group VI 

pottery, which is described as reddish brown, oval shaped beakers with a short, flared collar 

often having multilinear motifs, with dotted bands across the top (Mortlemans 1962:416). These 

ceramics also have flat-bases, or slightly convex with oval bellies (de Maret 1977:6) 

Faunal- The 1972 excavations revealed some Potamocherusporcus (bush-pig) metatarsals (long 

bones in the foots), and some long bones, teeth, jaw bones, vertebra, and rib of a Cephalophus 

sylvicultor (medium sized antelope) (de Maret 1975:113). This season also yielded an Achatina 

71 

sp. shell, rodent, other mollusks and avian bones were also recovered from the cave (de Maret 

1975: 109). There were no faunal remains noted from 1973 excavations (de Maret 1975: 109).  

Floral- Plants were in a “grand nombre” at this site (de Maret 1977:6). Charcoal analysis 

revealed Canarium Schweinfurthii among several other forest or forest/savanna tree species (de 

Maret 1986:113).  

Metallurgy- There were no iron objects found in Ngovo cave (de Maret 1986: 109) 

Dimba archaeology site 

 

Located just 6 kilometers north of Ngovo and 3 km from Mbanza-Ngungu, Dimba cave is 

an “exceptionally large cavern” nearly 1600 meters long, 30 meters wide, and 20 meters high at 

the base of a valley (de Maret 1986: 114). In 1972 and 1973 excavations consisted of seven 3x3 

meter units to reveal a blackish layer roughly 20 centimeters deep and a Late Stone Age 

occupation 300-350 centimeters deep (de Maret 1986: 114). Late Stone Age layer contained a 

polished axe (de Maret 1980: 452). 

Feature- No features have been reported. 

Lithics- Early Iron Age lithics consisted of an oblong a polished stone axe of green dolerite 

weighing 305 grams dating, a punctured mother of pear among other cut stones that dated to 390 

B.C. and 70 A.D. (de Maret 1977:5; de Maret 1986:114). 

Ceramics- Like Ngovo, the ceramics from the test units were mostly Mortlemans’ Group VI 

pottery however the surface contained several unspecified varieties of ceramics and showed 

significant disturbance (de Maret 1986:114). 

Flora: No floral remains have been reported. 

Fauna: Achatina sp. shells were recovered from (Lavachery 1990:105). 

Metallurgy: No iron materials were reported from this site 

72 

Ntadi Ntadi archaeology site 

 

Ntadi Ntadi rock shelter site is about 60 kilometers south east from Ngovo and Dimba 

caves (de Maret 1986:116). The Early Iron Age dates from 205 B.C. to 60 B.C. (de Maret 

1986:116). 

Features: This cave also has painting and engravings which have yet to be analyzed (de Maret 

1986:116). No ground features were identified. 

Lithics- Lithics included mother of pearls, flakes, and a fragment of a grinder, no polished axes 

were found (de Maret 1977:6; de Maret 1986:116). 

Ceramics- There were four decorated rim sherds and six small rim sherds found at this site and 

all but two of them were considered Group VI pottery (de Maret 1977:6; de Maret 1986:116).  

Fauna: Numerous Achatina sp. fragments were recovered from this site (de Maret 1986:116). 

Flora: No flora remains were reported from this site. 

Metallurgy: No iron materials were reported from this site. 

Sakuzi archaeology site: 

 

Lastly Sakuzi is an open-air site located roughly 80 kilometers north west from Ngovo 3 

kilometers south of the Congo River (de Maret 1986:105, 126). This site was first surveyed in 

1984 by Pierre de Maret in an attempt to better understand a separate pottery group named Kay 

Ladio (de Maret 1986:119, 125). Charcoal sampled from the site give an occupation range of 160 

± 55 B.C but an abundance of surface artefacts suggests “long use of the site” (de Maret 1986; 

118;).  

Features- de Maret reported 49 archaeological features including pits, hearths, and iron smelting 

furnaces” of which, 19 were excavated (de Maret 1986:121). Only feature 37 was excavated. It 

was 85-90 centimeter in diameter and 100 cm deep and contained “earth reddened by fire- at a 

73 

depth of 20 centimeters- a cutting edge fragment of a polished schist tool… and the only feature 

to contain fragments of Group VI pottery” (de Maret 1986:121). 

Lithics- A few fragments of worked chalcedony, quartz and quartzite as well as fragment of a 

polished stone tool were recovered from feature 37 (de Maret 1986:128). 

Ceramics- Surface finds included Group VI as well as Kay Ladio traditions, a separate group 

based a few kilometers away which dates to the first century A.D. (de Maret 1986:125). 

Ceramics excavated from feature 37 were consistent with the tempering material, surface-finish, 

and decoration of Group VI pottery (de Maret 1986:121). 

Faunal- No faunal remains were reported from the site. 

Floral- Charcoal samples were recovered as well as calcined fragments of Elaeis sp. (palm-nuts) 

(de Maret 1986:121). 

Metallurgy- This cave consisted of iron smelting furnaces that dated to 160 + 55 B.C. (de Maret 

1986 :121).  

V. Case 4 Summary 

 

This case illustrates the development of iron, extensive use of forest-savanna mosaic 

plants, and regional ceramic tradition in Lower Congo at the end of the first millennium B.C. 

These caves and rock shelters were used for hunting, smithing, and refuse. In terms of the few 

faunal remains, the bush pig and antelope are known to inhabit equatorial forest and closed 

environments while the shell and rodents were determined to be food remains, all grow in forest 

or savannas. The flora remains, specifically the Elaeis sp. and Canarium sp. are consistent with 

the time period and likewise indicate the use of both forest and savanna environments. The 

presence of the Achatina sp. shells, the ubiquitous ceramic tradition within the relatively short 

time frame and the presence of both Kay Ladio and Group VI pottery within one site provide 

74 

support for a local trade network. Lastly, the presence of an iron furnace dating to the second 

century B.C. indicates that full use of this skill was present in the area at this early date. 

 

 

75 

The Lower Congo Early Iron Age Attributes 

Description 
polished stone axe 
polished green dolerite axe, mother of 
pearl items, among other cut stones  
mother of pearl items, flakes, and a 
grinder fragment 
worked chalcedony, quartz and 
quartzite, polished stone tool  

reddish brown, oval shaped beakers 
with a short, flared collar, multilinear 
motifs, with dotted bands across the 
top; flat-bases, or slightly convex with 
oval bellies 
Consistent with Ngovo ceramics 

 

Timeframe 
250 B.C. - 150 B.C. 
390 B.C. - 70 A.D. 

205 B.C. - 60 B.C.  

210 B.C. - 110 B.C. 

250-150 B.C. 

  

  

 
Attribute  Context 
Lithics 
  

Ngovo 
Dimba 

  

  

Ntadi Ntadi 

Sakuzi 

Ceramic 

Ngovo 

Dimba 

Sakuzi 
Sakuzi 

Ngovo 

Dimba 
Sakuzi 

Ngovo 

Dimba 
Ntadi Ntadi 

  

  

  
Metal 

Floral 
Remains 

  
  

Faunal 
remains 

  
  

  

Ntadi Ntadi 

Group VI ceramics 

Group VI and Kay Ladio ceramics 
iron smelting furnaces  

210 B.C. - 110 B.C. 
210 B.C. - 110 B.C. 

wild fruit tree (Canarium 
Schweinfurthii) among several other 
forest or forest/savanna tree species  
none 
wild palm nuts (Elaeis sp.)  

bush pig (Potamocherusporcus), 
antelope (Cephalophus sylvicultor); 
land snails (Achatina sp.) rodent, other 
mollusks and avian bones 
land snails (Achatina sp.)  
land snails (Achatina sp.) 

250-150 B.C. 

  
  

250-150 B.C. 

  
  

  

Sakuzi 

none 

 
Table 7: Lower Congo Early Iron Age Attributes 

 
 

 

76 

VI. Chapter Summary 

 

These cases document the introduction of Early Iron Age activities across the Equatorial 

Forest over the span of 1000 years. Case 3 illustrates that the there was some level of ceramic 

connections between the occupants from Case 3 and 3 but clear formal and stylistic differences 

were apparent. Nonetheless, the occupants in the interior were equally well adapted to local and 

long-distance trade, particularly within the water networks of the Congo River. These networks 

may have provided grains to the forest villages as well as influenced the design motifs of the 

local pottery traditions. The contemporaneity, use of trade, and regional extension of the ceramic 

horizons generally support the mechanisms of interaction despite the clearly distinct ceramic 

horizons. Lastly, Case 4’s focus on the Lower Congo illustrates how and where these practices 

reached the southern border of the forest towards the end of the first millennium B.C. The 

evidence of fully functioning iron production, the regional ceramic distribution indicates, and the 

extensive use of forest and savanna plants and trees suggests local interaction and regional trade 

as well. Together these cases provide the archaeological support for the formation of 

interregional development of Early Iron Age technologies and practices over a 1000-year period. 

The following chapter will provide a discussion of this assessment and the research conclusions. 

 

 

77 

CHAPTER 6: ANALYSIS AND DISCUSSION 

 

I. Data analysis 

 

Case 1: The Bamenda Grasslands represent the earliest settlement of Bantu speakers in 

the Congo River Basin. They were initially connected to and interacting with Nigerian and 

Sahelian cultural practices as far back as 1500 B.C. This is illustrated by the fact that much of the 

similarities in the ceramic and lithic remains retrieved from Shum Laka that date to this period 

are attributed to practices developing from western or northern directions. Whether this period is 

called the Ceramic Stone Age, Neolithic, or Stone to Metal Age there was a period when pottery 

and ground tools were the predominate remains across the region (de Maret and Nsuka 1977). 

Then around 1000 B.C. a period of isolation developed, presumably as the rainforest thickened 

around that time (Lupo 2014:70; Blench 2006:124 and 126; Bostoen et al 2015:355). This period 

of isolation may have been a motivating factor in the development of cultivating Elaies, and 

Canariaum trees, as well as the shift towards open air settlements (Sowunmi 1999:202-203). At 

the end of the end of the forest thickening around 1000 B.C., this population began settling in the 

spotty savanna area near the Sangha valley (Maley and Brenac 1998; Bostoen 2015). These 

shifts from the highlands to the west of the Sangha river make up our only understanding of how 

these activities may have changed after settlement dues to the seeming extension of cultural 

practices (Klieman 2003). 

 

Case 2: The Northern Plateau then illustrates the first attempts of settlement with newly 

developed skills for forest-savanna environment. Here we witness the local production of 

ceramics, either at the village or household level as well as interregional trade with shells from 

the Atlantic coast and lake fish from further north. We also begin to see specialized zones of 

78 

activity, for example the furnace at Oliga suggesting the development of a small community 

focused on iron production alone. This is typical since iron production require intensive labor 

from a variety of people and resources. This is the beginning of a pattern of site specialization I 

have noticed across the region such as ceramic production at Imbonga in the interior and iron 

production at Sakuzi in Lower Congo. Even with the necessary disclaimers on the sample size of 

the sites, the fact that some sites appeared to be workshops for a specific technology is clear 

across the region.  

 

These sites also necessarily varied in the ways they adapted to their local environments. 

As mentioned above, one example is the way sites did not follow the LSA nor CSA tradition of 

relying on pottery and stone and opted for one or the other with the frequent addition of 

metallurgy. Such was the case at Imbonga and Pikunda where no lithics were reported as far 

back as the third century B.C. although the latter had already began using metal. Another 

example is the way they began systematically using Elaeis guineensis (oil palm) and fruits from 

the Canarium Schweinfurthii across all of the sites with floral data. These plants have a variety 

of uses including for example the palm oil was used to cook as a slow burning fuel and the fruits 

were eater (Khalheber et al 2014:497). Further, the are specifically cultivated in a savanna 

climate, thus, in cases where they appear in the interior at Imbonga, they suggest something more 

intentional (Sowunmi 1999:201,208). Lastly, they adapted mechanisms to interact beyond their 

surrounding areas for instance developing a river-based transportation system which provided 

access to resources not found in the forest such as grains and oil palm. 

 

Although the environment seems to be a strong indicator for what practices will develop 

across the sites, the occupants often sought ways to maintain stability despite the environmental 

barriers. This was achieved mostly through trade or diversifying settlement activities within a 

79 

region. For example, in Case 4: Lower Congo the hilltop sites such as Ngovo and Dimba may 

have been designated for specific activities related to polished stones, ceramic fragments, and 

plants such as ritual spaces, while the riverine settlements such as Sakuzi may have served a 

different function given its location near the river and the abundance of features found at the site, 

such as a markets. Further, the ubiquitous pits and shafts that are found throughout the region 

suggest some sort of shared understanding- whether they are interpreted as storages or grave 

sites, the similarities between their shapes, sizes, and contents point to a conservatism towards 

such practices (Meister 2010:247). 

 

In terms of variation among the sites, the forest interior sites appear to be the most 

different from the activities found in the Bamenda Grasslands. Here, the forest environment 

required different activities such as the abrupt development of metallurgy at Pikunda-Munda and 

the reliance on lithics at later sites such as Batalimo. As such, the connections between the sites 

illustrate their variability because social, political, and cultural formation on the frontier also 

affected interactions with the areas surrounding the frontier, not just homelands. Ogundiran 

(2014) demonstrates the significance of frontier-frontier interaction through an historic 

archaeological approach to the Osogbo of nineteenth century southwest Nigeria (Ogundiran 

2014:6-7). The Osogbo frontier existed as a part of a regional network of frontiers. Political 

authority on the Osogbo frontier developed through interplay with the surrounding frontier 

networks, rather than an internal conservative dynamic founded on homeland behaviors as 

theorized by Kopytoff (Ogundiran 2014:21). 

 

Ultimately, the data suggested more intra-regional variation than between the specific 

sites and the homeland. For example, none of the sites had the exact make-up of lithic, ceramic, 

floral, faunal assemblages at any point in time (See Table 8). For example, even sites that seem 

80 

to be parallel in terms of their development sequence such as Ngovo and Dimba, the later had 

evidence of multiple ceramic traditions and thus served different purposes through time. The lack 

of pottery overlap may be a distinctive identity marker as is usually the case, but it doesn’t have 

to mean they didn’t interact or weren’t from similar origins or didn’t see themselves as united in 

other ways. As the archaeological work done in the region has focused on identifying these 

groups rather than understanding how these places have changed over time, what I wished to do 

here was illustrate how these beginning practices provided mechanisms for interregional 

interaction that would thus sustain a development of Bantu identity across the region for the next 

millennium.  

 

The documentation of the evidence of inter-regional interaction was the most significant 

find of this project. The prevailing literature has maintained a hardline of isolationism across this 

time period and region that is not supported by the data (Eggert 1993:323). Every site showed 

some indication of the use of non-local resources be them plants, animals, trees, or iron ore, even 

sites deep within the forest such as Boso Njafo illustrated the acquisition of a grain typically 

grown further west (Khalheber 2014:493). The distinctive 1000-year gap between Late Stone 

Age and Early Iron Age settlements across the region as an indication of new arrivals due to 

migration or dispersion. As noted earlier in the discussion, there is little evidence of the 

development of these technologies which most likely indicated that they were transferred as 

complete practices or technologies through population change.  

 

Archaeologists in the region suggested that the lack of ceramic overlap was indicative of 

isolation rather than connections among these contemporary, mobile groups (de Maret 1986:130; 

Clist 1989:80; Eggert 1993:322). There is, however, concession around the idea of an ancestral 

group- perhaps from Obobogo or Gabon that share the “simplest decorative system from which 

81 

the others could have developed,” and thus given support to the theory of human migration such 

as common traits of the flat bottoms and grooved lines despite the distance and time between 

these spaces (Clist 1989:80). If this was a simple migration from one site to the next, why do the 

sites connect to Obobogo rather than each other. For me, this illustrates an intentional 

differentiation rather than isolation. 

II. Summary of analysis 

 

This research attempted to answer the question on the success of the development of 

Bantu culture across the region. This question essentially asked whether the behaviors associated 

with Bantu speaking people- settlements, metallurgy, and forest-savanna adaptations, developed 

organically and independently across the region or if they are the result of migration. If they 

were the result of organic independent development, perhaps due to the climatic shifts that 

altered the forest around 1500 B.C., there should be a continuity between LSA groups, evidence 

of trial and area of complex technologies such as metallurgy or refined ceramic. However, what 

we are faced with is evidence of fully developed activities such as grain storage and furnace 

construction early on during these occupation periods- suggesting that the occupants had 

previous experience or knowledge with such activities before settling the region. The question 

then becomes, where did they get this information and how was it spread? Those answers are 

revealed through an analysis of how and when these activities emerged and the mechanism 

through which they might have spread.  Here, I have demonstrated that the south-easterly nature 

of the development of these activities and the evidence or local and interregional trade support 

my theory that connections to Bantu cultural practices, whether directly or indirectly through 

trade were a strong part of the success of the development of the Central African Early Iron Age.   

Thus, the success of these adaptations does not solely rely on people’s ability to rapidly adapt to 

82 

change. Rather, the success in the case of Bantu settlement of Central Africa comes from their 

process of dispersal from the Bamenda Grasslands and settlement of the region during the Early 

Iron Age.  

 

This dissertation sought to explore the nature of success of dispersion among Bantu 

speakers in the Central African Early Iron Age. In the process, I revealed shared cultural traits 

that indicated an interregional network of shared cultural knowledge throughout the Equatorial 

Forest between 1000 B.C. and 200 A.D. Through analyzing this network of relationships as a 

premodern African Diaspora, I have come to understand this process as a cultivation of 

homeland. As such, the shared cultural traits strengthened their common Bantu speaking 

identities and thus gave way for continuous reconnection and cultivation of the behaviors and 

patterns that allowed them to adapt across a range of environments over time, thus allowing for 

successful transference of Bantu cultural traits over the next 2 millennia. 

 

83 

Table 8: Timeframe of Early Iron Age Behaviors in Central Africa. This graph 
illustrates when certain attributes entered the regions under consideration.  

 

 

III. Discussion 

 

This discussion integrates the above analysis of when and where which artifacts emerge 

across a region with the premodern African Diaspora framework, I developed in Chapter 3: 

Methods. The most significant finding is the need to examine the Northern Plateau as the 

archaeological homeland of the EIA package and most likely Bantu cultural traditions. At the 

end of the Late Stone Age, the symbiosis of indigenous, western, and northern populations began 

to cultivate culturally distinctive patterns within the Northern Plateau. These people began to 

shift and adapt to the increasingly mixed environments of the region. Further, these adaptations 

became increasingly more complex, meaning specific activities that require more people, skill, or 

delegation to achieve. This is evident in the shift in lithic materials and production techniques, 

more systematic, functional, and consistent ceramic decorations, and the string of villages that 

84 

appear to be interconnected. This section will discuss how these features together construct a 

cohesive cultural practice that formed during this earliest phase of the EIA. As such, I hope to 

illustrate how these features are the dominant practices that will be replicated in the dispersal of 

Bantu speakers over the next 500 years. Further, this time period witnessed a shift in a preference 

for rock shelter and cave spaces to an inclination towards open air sites and the ability to live and 

thrive. It is within this context that the Northern Plateau, not the Bamenda Grasslands should be 

understood as the “diffusion center” of a shared identity for almost 500 years.   

 

The Northern Plateau was a more or less a cosmopolitan center, with extensive trade 

networks, early crop cultivation, and iron specialization within the local region. Further, the 

connections between the Northern Plateau and the Forest Interior suggests a continuous 

connection based on the shared ceramic elements, the presence of similar complex floral remains 

such as pearl millet, and complex local pottery traditions. These coupled with the extensive 

waterway that makeup the Forest Interior provide substantial evidence for inter and intra-

regional trade. 

 

At the other end of the forest, while Lower Congo is considerably younger than the other 

regions, it has a well-developed settlement system and metallurgy tradition by the end of the first 

millennium B.C. This suggests that in contrast to the northern regions, the EIA package most 

likely arrived in tack. Lastly, while the Bamenda Grasslands may be the linguistic homeland, the 

archaeological assemblage ultimately does not support that idea. The long period of isolation 

from other regions coupled with the long period of disuse of the cave site suggests that the Early 

Iron Age materials were brought in from another region, perhaps Obobogo.  

 

 

85 

Evidence of Premodern African Diaspora 

Intraregional 
Location 

Evidence of 
Dispersion 

Evidence of 
Connections 

Evidence of a 
Homeland 

Case 1: The 
Bamenda 
Grasslands 

Case 2: The 
Northern 
Plateau 

Case 3: The 
Forest 
Interior 

Continuous 
connection with 
western groups 
in Nigeria 
during the early 
phases of the 
EIA. 

Circular 
movement 
within the 
region, similar 
ceramic and 
floral, and 
possibly metal 
working 
activities. 

The region 
contains 
elements of the 
complete EIA 
package. 
 
 

Case 4: 
Lower Congo 

EIA package 
emerges almost 
80 years after 
Case 2 

  

Period of isolation 
and then introduction 
of new lithic 
techniques and floral 
remains. 

Complete Early Iron 
Age package and 
long-distance trade 
with the Sahel and 
East Africa. 

Contains inter- and 
intraregional trade 
networks, use of 
goods from multiple 
regions, ceramic 
techniques are 
shared throughout 
the Congo River 
Basin. 

  

  

Shares elements of 
the ceramic tradition, 
use of pearl millet, 
ceramic pit 
structures with Case 
2, contains savana 
goods, evidence of 
specialization. 

Inter- and intra-
regional, fully 
developed iron 
technology during 
the earliest phases of 
occupation. 

Evidence of 
a Diasporal 
frontier 

Activities 
from 400 
B.C. onward 
may be the 
result of 
migration 
from another 
area. 
  

Activities 
emerge later 
than Case 2, 
no forest 
interior site 
contains the 
complete 
EIA 
package.  
Activities 
emerge the 
latest here 
but are fully 
developed 
upon arrival.  

 
Table 9: Premodern African Diaspora Framework: Bantu Homeland Diasporal Frontiers 

 

 

 

86 

CHAPTER 7: 
CONCLUSIONS AND SUGGESTIONS FOR FUTURE RESEARCH 

I. Conclusions 

 

 

This research has produced evidence to support two key features premodern African 

Diasporas: an interregional trade network that facilitated the movement of goods, technologies, 

and people during the Central African Early Iron Age; and the connections and relationships that 

emerged between these groups and their homeland. This research demonstrated that the 

integration of frontier and African Diaspora studies into understanding premodern Africa 

provides essential theoretical and methodological tools. Environmental frontiers, defined by the 

boundaries that emerge from natural conditions such as water, forests, and terrains which often 

create forced boundaries between societies were key variables for when and where specific EIA 

activities developed. The application of African Diaspora’s frame work offers a model for 

understanding the processes and conditions that emerge in the processes of dispersal from a 

homeland. Used together, I have constructed an archaeologically based culture history of Bantu 

expansion into Central Africa. 

 

This project sought to construct a cultural history of the Central African Early Iron Age 

(EIA) roughly 1000 B.C. to 200 A.D. I identified 4 key regions that can provide insights into the 

movement across the regions, and how these communities responded to the environment, the 

patterns in which activities arrived in specific places and times and their abilities to trade both 

locally and across the region. This required obtaining a clear understanding of the state of 

literature on archaeology of the region. This body of work has, for the last 30 years, focused on 

understand how and when people developed EIA technologies such as iron production, settled 

villages, and inter-regional trade. Prevailing cultural histories connect these activities to Bantu 

87 

speaking people who emerged along the Niger-Cameroon border around 4000-3000 years B.P. 

Acquiring this assumption, I sought to understand how Bantu speaking people so successfully 

moved from one end of the Equatorial forest to the other in the span of a few hundred years and 

the practices they may have brought with them. I first applied a frontier perspective to 

understand how people would move through the multiple environments of the forests, sometimes 

dense primary forest, sometimes major rivers such as the Congo River, and sometimes mosaic 

environments that have all three.  

 

The relationships that develop between diaspora communities become a pillar of the 

cultural identities that endure for centuries following the departure from a homeland. These 

relationships, borne of various power dynamics and social, political, and religious connections, 

can be expressed as conscious or unconscious currents embedded within the behaviors of a 

community. This process is powerful within a diaspora because it structures the ways the 

homeland becomes remembered and shared between these communities.  

 

The project revealed that both the inter and intraregional variation provide insight into the 

construction of a diasporal identity, even in the premodern era. The integration of the frontier 

framework with African Diaspora studies allowed for additional comparison of the nature of 

regional and local connections. For example, on in the Forest Interior region, the ceramic and 

lithic traditions are not similar and suggests very little interaction. However, when compared to 

the Northern Plateau, these differences indicate a different relationship.  

II. Bantu Expansion as Premodern African Diaspora: a culture history revised 

 

When Bantu expansion is viewed as a premodern African diaspora the following picture 

emerges: an adverse climatic shift creates new behaviors, necessities, and interactions along the 

Northern Plateau. The existing village is cosmopolitan, with long distance trade routes from 

88 

across the region. The villages respond to the shift by establishing specialized activities such as 

metallurgy, cultivating crops, and incorporating both forest and savanna goods into their 

subsistence practices. After about 200 years, these groups began to spread via water or forest 

openings into other regions of the Congo River Basin. After another century new traditions 

emerge but trade and exchange routes are maintained alongside aesthetic similarities as 

illustrated by the pottery remains in the Forest Interior. Other groups travel back west and return 

to the Bamenda Grasslands and establish settlements and introduce new lithic technologies to the 

area. After another generation or two, these populations reach the southern limit of the Congo 

River Basin and establish village sites with fully functional metallurgy, extensive hilltop and 

rock shelter settlements, and extensive trade networks into the forest. This 500-700-year period 

of movement, interaction, and connection set the basis for the Bantu Diaspora and the cultural 

traits that would eventually make it into southern and eastern Africa.  

III. Suggestions for future research 

 

This research established a framework for identifying cultural change in pre-modern 

Africa. In this process I identified two general areas that are in need of future research: first, 

more archaeological field work focused on identifying Early Iron Age sites and understanding 

how these sites change over time and second, examining African pre-history for other moments 

of dispersion that can be better understood through African Diaspora Studies.  

 

Archaeology in the Congo River Basin remains pretty sparse relative to the geographic 

size and the extent of human occupation. Several of the underlying reasons were articulated in 

Chapter 2: Methods but suffice it to say here that the training and support of students interested 

in Central African in both field and laboratory archaeology would begin to fill in much needed 

gaps in this survey. For example, there are several pits from each region that still need to be 

89 

analyzed and processed. Also, some of these sites consist of a few test excavations and require 

more field work. There is also research needed on the sources used for the lithic, ceramic, plant, 

and metal remains and identifying where they came from in relationship to the sites. We can also 

begin sourcing ceramics through a project such as using Instrumental Neutron Activation 

Analysis (INAA) to identify if the pottery materials were locally sourced or not (Usman 

2012:131). There were also several post holes spread across some of the village sites which 

require additional analysis to reveal population density as well as the general uses or 

paleoecology of the sites. These kinds of inquiries can provide a better understanding of the 

possible transportation routes between the regions and potentially how each region functioned in 

relationship to the whole. 

 

Given the general focus of this project, there are other regions in the Congo River Basin 

that need to be evaluated for how they fit into the construction of Bantu culture such as southern 

Cameroon, Sao Tome and Principe islands, eastern Gabon and south-eastern Democratic 

Republic of the Congo. These regions each offer their own environmental boundaries, coastal 

frontiers for the former and forest interior zones for the latter two which may reveal different 

patterns of change and continuity than what I have illustrated here.  

 

Further, Shum Laka’s distinctively long and successive occupations periods foreshadow 

how Bantu speakers adapted to the environmental and social changes of the Early Iron Age. This 

then provides a snapshot of how other frontier communities responded to similar changes. I 

contend that articulating the features that make up the Northern Plateaus can be compared to 

other sites in the found in throughout the Congo River Basin. In this respect, it allows us to see 

how the homeland changes and adaptations within contemporary contexts and environments on 

the one hand and establish a possible framework for non-Bantu settlements on the other. 

90 

 

Lastly, focused research into the ways African descendants have moved across the globe 

in prehistory need more attention. Questions about East African people in the Indian sub-

continent are as relevant to pre-historic studies as they are to modern conceptualizations of the 

African Diaspora. If we wish to disrupt the image of a static isolated prehistoric world, then 

inquiries into Africa’s presence and interaction across both the continent and the world need our 

immediate attention. Such studies will bring us closer to identifying the essential qualities of 

what it means to be in diaspora. I believe that this current project has created a space for such 

future work. 

 

 

 

91 

 

 

 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

APPENDICES 

92 

APPENDIX A: THE BAMENDA GRASSLANDS TABLE OF DATA SOURCES 

 

Author 

de Maret, Pierre, 
Bernard Clist, and Wim 
Van Neer 

Publication 
Year 
1987 

Asombang and de 
Maret 
Collette 

1992 

1992 

Title 

Résultats des premières 
fouilles dans les abris-sous-
roche de Shum Laka et Abeke 
au Nord-Ouest du Cameroun 
Re-Investigating Shum Laka 

L’Abri Sous Roche de Shum 
Laka 2 (Cameroun): Etude 
Archéologique 
Preliminary Results of the 
1991-1992 Field Season at 
Shum Lake, Northwestern 
Province, Cameroon 

Site 
detailed 
Shum Laka-
Abeke 

Kind 

Article 

Shum Laka  Article 

Shum Laka  Masters 
Thesis 

Shum Laka  Article 

1993 

1995 

de Maret, Pierre, 
Raymond Asombang, 
Els Cornelissen, Phillip 
Lavachery, J. 
Moyersons, and Wim 
Van Neer 
de Maret, Pierre, 
Raymond Asombang, 
Els Cornelissen, Phillip 
Lavachery, and J. 
Moyersons 
Table 10: The Bamenda Grasslands table of data sources. See bibliography for full citations.   

Continuing Research at Shum 
Laka Rock Shelter, Cameroon 
(1993-1994 field season) 

Shum Laka  Article 

 

 

93 

1995 

Fouilles Archéologiques à 
Shum Laka (Cameroun) 

Shum Laka  Article 

French 

1996 

1996 

Shum Laka Rock Shelter 
late Holocene Deposits: 
from Stone to Metal 
(North western Cameroon) 
Shum Laka (Cameroon): 
Human Burials and 
General Perspectives 
30 000 ans d’occupation, 6 
mois de fouilles: Shum 
Laka, un site exceptionnel 
en Afrique centrale.  

Shum Laka  Book 

English 

Chapter 

Shum Laka  Book 

English 

Chapter 

Shum Laka  Article 

French 

1998  De la pierre au métal. 

Shum Laka  Dissertation  French 

Archéologie des dépôts 
holocènes de l'abri sous 
roche de Shum Laka 
(Cameroun) 

2000  Natural and Cultural 

Shum Laka  Article 

English 

de Maret, Pierre 

1996 

Table 10 (cont’d). 

Cornelissen, Els, 
Jan Moeyersons, 
and Pierre de 
Maret 
Lavachery, 
Philippe 

Lavachery, 
Philippe, Els 
Cornelissen, Jan 
Moyersons, and 
Pierre de Maret 
Lavachery, 
Philippe 

Lavachery, 
Philippe and Els 
Cornelissen 

Lavachery  

2001 

 

 

 

 

 

 

 

Spatial Patterning in the 
Late Holocene Deposits of 
Shum Laka Rock Shelter, 
Cameroon 
The Holocene 
archaeological sequence of 
Shum Laka rock shelter 
(Grassfields, western 
Cameroon) 

Shum Laka  Article 

English 

 

94 

APPENDIX B: THE NORTHERN PLATEAU TABLE OF DATA SOURCES 

 

 
 

 

Author 

de Maret, 
Pierre 

Publication 
Year 
1980 

de Maret, 
Pierre 

1982 

Claes, 
Phillipe 

1985 

Atangana, 
Christine 

1988 

Essomba, 
Joseph-
Marie 
Mbida 
Mindzie 

1989 

1996 

2000 

Mbida 
Mindzie et 
al 

2001 

Mbida 
Mindzie et 
al 

Title 

Site detailed  Kind 

Language 

Obobogo 

Report 

French 

Obobogo 

Report 

English 

Obobogo 

Masters 
Thesis 

French 

Obobogo 
Group:Okolo 

Dissertation  French 

Article 

French 

Dissertation  French 

Obobogo 
Group: 
Ndidan 
Obogogo 
Group: 
Nkang 

Obobogo 

Article 

English 

Obobogo 

Article 

English 

Preliminary Report on 
1980 Fieldwork in the 
Grassfields and 
Yaoundé, Cameroun 
Belgian Archaeological 
Project in Cameroon 
(July-August 1981 
fieldwork) 
Contribution à l'étude 
de céramiques 
anciennes des environs 
de Yaoundé 
Archeologie Du 
Cameroun Meridional: 
Etude due Site 
D’Okolo 
Dix Ans de Recherches 
au Cameroun 
Méridional 
L'émergence des 
communautés 
villageoises au 
Cameroun méridional. 
Etude archéologique 
des sites de Nkang et 
de Ndindan 
Evidence for banana 
cultivation and animal 
husbandry during the 
first millennium BC in 
the forest of southern 
Cameroon 
First archaeological 
evidence of banana 
cultivation in Central 
Africa during the third 
millennium before 
present 

Table 11: The Northern Plateau table of data sources. See bibliography for full citations. 

95 

APPENDIX C: FOREST INTERIOR TABLE OF DATA SOURCES 

Author 

Publication 
Year 
Eggert, Manfred  1987 

Eggert, Manfred  1992 

Eggert, Manfred  1993 

Wotzka, Hans 
Peter 

1995 

Kahlheber et al. 

2014 

 
 
 

Title 

Imbonga and Batalimo: 
Ceramic Evidence for 
Early Settlement of the 
Equatorial Rain Forest 
The Central African 
Rain Forest: Historical 
Speculations and 
Archaeological Facts 
Central Africa and the 
Archaeology of the 
Equatorial Rainforest: 
Reflections on Some 
Major Topics 
Studien zur 
Archäologie des 
zentralafrikanischen 
Zaïre-Beckens und ihre 
Stellung im Kontext 
der Bantu-expansion 
Pearl Millet and Other 
Plant Remains from the 
Early Iron Age Site of 
Boso-Njafo (Inner 
Congo Basin, 
Democratic Republic 
of the Congo).  

Kind 

Language 

Article 

English 

Site 
detailed 
Imbonga 
Group: 
Batalimo 

Imbonga  Article 

English 

 
 

Imbonga  Book 

English 

Chapter 

Imbonga  Book 

German 

Chapter 

Imbonga  Article 

English 

Table 12: The Forest Interior table of data sources. See bibliography for full citations. 

96 

APPENDIX D: THE LOWER CONGO TABLE OF DATA SOURCES 

Author 

Bequaert, M.  

Publication 
Year 
1962 

Morelmans, 
Georges 

de Maret, 
Pierre 
de Maret, 
Pierre and F. 
Nsuka 
de Maret, 
Pierre 
de Maret, 
Pierre 
de Maret, 
Pierre 

de Maret, 
Pierre 

1962 

1975 

1977 

1977 

1984 

1986 

1990 

 
 
 

Title 

Fouilles à Thysville du 
Musée Royal du Congo 
Belge en 1938, Actes du 
4ème Congrès Panafricain de 
Préhistoire et de l'Etude du 
Quaternaire 
Archéologie des grottes 
Dimba et Ngovo (Région de 
Thysville, Bas-Congo) 
A Carbon 14 date From 
Zaïre 
History of Bantu metallurgy: 
some linguistic aspects 

La datation des haches polies 
du Bas-Zaïre 
L'archéologie en zone Bantu 
jusqu'en 1984 
The Ngovo Group: an 
Industry with Polished Stone 
Tools and Pottery in Lower 
Zaïre 
Le “Néolithique” et l’âge du 
fer ancient dans le sud-ouest 
de l’Afrique Centrale 
Lower Congo Rock Art 
Revisited 

Site 
detailed 
Ngovo: 
Thysville 

Kind 

Language 

Report  French 

Ngovo 

Report  French 

Ngovo 

Report  English 

Ngovo 

Article  English 

Ngovo 

Report  French 

Ngovo 

Report  French 

Ngovo 

Article  English 

Ngovo 

Article  French 

Article  English 

Heimlich, 
Geoffroy 
Table 13: Lower Congo table of data sources. See bibliography for full citations. 

Ngovo 

2010 

 

 

 

 

 

 

 

97 

 

 

 

 

 

 

 

 

 

 

 
 
 
 
 
 

BIBLIOGRAPHY 

 

98 

BIBLIOGRAPHY 

 

Dissertation, University College of London, 2005. 

and Shum Laka.” Dissertation, University of London, 1988.  

Ashley, Ceri Zaria. “Ceramic Variability and Change: A perspective from Great Lakes Africa.” 
 
 
Asombang, Raymond. “Bamenda in prehistory: the evidence from Fiye, Nkwi, Mbi Crater 
 
 
Asombang, Raymond and Pierre de Maret. 1992. Re-Investigating Shum Laka. Nsi 10/11:13-16. 
 
Atangana, Christine. “Archeologie Du Cameroun Meridional: Etude due Site D’Okolo.” 
 
 
Badenhorst, Shaw. 2009. The Central Cattle Pattern During the Iron Age of Southern Africa: A
Critique of Its Spatial Features. The South African Archaeological Bulletin 64(190):148
 
 
155. 

Dissertation, University of Paris, 1988. 

Barth, Fredrik ed. 1969. Ethnic groups and Boundaries. Boston: Little, Brown and Company. 
 
Bassel, Laura. 2010. Middle Stone Age Sangoan-Lupemban Lithic Assemblages of Africa. In 

West African Archaeology: New Developments, New Perspectives edited by Philip 
Allsworth-Jones 15-28. Oxford: British Archaeological Reports International Series 
2164. 

University Press.  

Bennett, Norman Robert. 1971. Mirambo of Tanzania: 1840?-1994. New York. Oxford
 
 
Bequaert M. 1962, Fouilles à Thysville du Musée Royal du Congo Belge en 1938, Actes du 
4ème Congrès Panafricain de Préhistoire et de l'Etude du Quaternaire, Annales des 
Sciences Humaines, Musée Royal de l'Afrique Centrale, 40: 323-350. 

 
Blench, Roger. 2006. Archaeology, Language, and the African Past. Lanham: Rowman and 

Littlefield Publishers, Inc. 

 
Blyden, Edward Wilmot. 1987. Christianity, Islam, and the Negro Race. London: The 

Charterhouse Press. 

Change. Garden City, N.Y.: Natural History Press.  

Bohannan, Paul and Fred Plog, eds. 1967. Beyond the Frontier: Social Process and Cultural
 
 
Bostoen, Koen et al. 2015. Middle to Late Holocene Paleoclimatic Chante and the Early Bantu 
Expansion in the Rain Forests of Western Central Africa. Current Anthropology 56(3): 
354-384. 

99 

Brower-Stahl, Ann eds. 2005. Introduction: Changing Perspectives on Africa’s Pasts. In 
AfricanArchaeology: A Critical Introduction edited by Ann Brower-Stahl, 1-22. 
 
 
Malden: Blackwell Publishing. 
 
Brower- Stahl, Ann. 2004. Political Economic Mosaics: Archaeology of the Last Two Millennia
 
 
Butler, Rhett. “The Congo Rainforest,” Mongabay.com (blog). January 23, 2016. Accessed 
 

in Tropical Sub-Saharan Africa. Annual Review of Anthropology, 33: 145-172. 

February 22, 2019. https://raingorests.mongabay.com/congo/.  

Research Frontier: The Ecuadorian Case. In Process and Pattern in Culture; Essays in
Honor of Julian Steward, edited by Robert A. Manners, 281-325. Chicago: Aldine
Publishing Company. 

Casagrande, Joseph B. Stephen I Thompson, and Philip D. Young. 1964. Colonization as a
 
 
 
 
Casey, Joanna. 2005. Holocene Occupations of the Forest and Savanna. In African Archaeology: 

A Critical Introduction edited by Ann Brower Stahl, 225-248. Malden: Blackwell 
Publishing. 

 
Claes, Phillipe. “Contribution à l'étude de céramiques anciennes des environs de Yaoundé.” 

Master’s Thesis, Université Libre de Bruxelles, 1985. 

 
Clark, J.D. 1996. Decision-making and variability in the Acheulean. In Aspects of African 

Archaeology, edited by Gilbert Pwiti and Robert Soper, 93-97. Harare: University of 
Zimbabwe Publications. 

 
Clist, Bernard. 1989. Archaeology in Gabon, 1886-1988. African Archaeological Review, 7: 59-
 

95. 

Clist, Bernard. “Des premiers villages aux premiers européens autour de l'estuaire du Gabon : 
quatre millénaires d'interactions entre l'homme et son milieu.” Dissertation, Université 
 
 
Libre de Bruxelles, 2005. 

Thilo Rehren eds 22-28. London: Archetype Publications. 

Clist, Bernard. 2013. Our Iron Smelting c14 dates. In, The World of Iron Jane Humphries and
 
 
Clist, Bernard, Pierre de Maret et al. 2013. The KongoKing Project: 2012 Fieldwork Report from
 

the Lower Congo Province (DRC). Nyame Akuma 79: 60-73. 

Clist, Bernard et al. 2015a. African-European Contacts in the Kongo Kingdom (Sixteenth
 
Eighteenth Centuries): New Archaeological Insights from Ngongo Mbata (Lower 
Congo DRC). International Journal of Historical Archaeology 19: 464-501. 
 

Clist, Bernard, Els Cranshof, Toon De Herdt, Roger Kidebua, Igor Matonda, Alphonse Nkanza 
Lutayi, Blair Zaid, and Koen Bostoen. 2016. Le projet KongoKing: Les prospections et 
 

100 

 
 

fouilles menées en 2015 dans la province du Kongo Central (République Démocratique 
du Congo). Nyame Akuma 84: 128-141.   

Collette, Sophie. “L’Abri Sous Roche de Shum Laka 2 (Cameroun): Etude Archéologique” 
 

Vol 1 and 2. Master’s Thesis. Universite Libre de Bruxelles. 1992-1993.  

Cornelissen, Els, Jan Moeyersons, and Pierre de Maret. 1995. Fouilles Archéologiques à Shum 
 

Laka (Cameroun). N.S.T.  Vol. 13.2/3/4:319-322. 

Curzon of Kedleston, Lord G.M. 1907. Frontiers. Oxford: Oxford University Press. 
 
Damião de Góis.1566. Chronica do felicissimo rei Dom Emanuel. Em Lisboa : Em casa de 
 
 

Françisco Correa, impressor do serenissimo Cardeal Infante, ahos xvij dias do mes de 
Iulho de 1566 - ahos xxv dias do mes de Iulho de 1567. 

David, Nicholas and Ian Robertson. 1996. Competition and Change in Two Traditional African 
 
 

Iron Industries. In The Culture and Technology of African Iron Production, edited by 
Peter R. Schmidt, 128-144. University Press of Florida: Gainesville. 

De Kiewiet, C.W. 1942. A History of South Africa: Social and Economic. London: Oxford
 

University Press.  

de Luna, Kathryn M. 2014. Bantu Expansion. In Oxford Bibliographies in African Studies. 
 
 

http://www.oxfordbibliographies.com/view/document/obo-9780199846733/obo-
9780199846733-0165.xml. Accessed February 2017. 

linguistics in Central African history. Quaternary International, 2016: 1-11. 

de Luna. Kathryn M. 2016. Conceptualizing vegetation in the Bantu Expansion: Reflections on 
 
 
de Maret, Pierre. 1975. A Carbon 14 date From Zaïre. Antiquity, 43(14): 133-137. 
 
de Maret, Pierre and F. Nsuka. 1977. History of Bantu metallurgy: some linguistic aspects. 
 
 
de Maret, Pierre. 1977. La datation des haches polies du Bas-Zaïre. Ngonge, Carnets de sciences 
 
 
de Maret, Pierre. 1980. Preliminary Report on 1980 Fieldwork in the Grassfields and Yaoundé, 
 
 
de Maret, Pierre. 1982. Belgian Archaeological Project in Cameroon (July-August 1981 
 
 
de Maret 1984. L'archéologie en zone Bantu jusqu'en 1984. Muntu, 1: 37-60. 
 

fieldwork). Nyame Akuma, 20: 11-12. 

History in Africa 4: 43–65. 

humaines, 31: 22-27. 

Cameroun. Nyame Akuma, 17: 10-12. 

101 

Lower Zaïre. The African Archaeological Review, 4: 103-133. 

African Archaeology edited by Peter Robertshaw, 109–134. London: Currey. 

Archaeology edited by Ann Brower-Stahl 420-440. Malden, MA: Blackwell Publishing. 

Studies, http://www.oxfordbibliographies.com/view/document/obo-
9780199846733/obo9780199846733-0154.xml (accessed 15-Feb-2016). 

les abris-sous-roche de Shum Laka et Abeke au Nord-Ouest du 
Cameroun. L'Anthropologie, 91(2): 559-584. 

Centrale. In, Paysages quaternaires de l’Afrique centrale atlantique edited by Raymond 
Lanfranchi and Dominique Schwartz, 447-457. Editions de l’Orstom. 

Aspects of African Archaeology, Papers from the 10th PanAfrican Association for 
Prehistory and Related Studies Congress, edited by Gilbert Pwiti and Robert Soper 275-
279. Harare: University of Zimbabwe Publications. 

de Maret, Pierre. 1986. The Ngovo Group: an Industry with Polished Stone Tools and Pottery in 
 
 
de Maret, Pierre. 1990. Le “Néolithique” et l’âge du fer ancient dans le sud-ouest de l’Afrique 
 
 
 
de Maret, Pierre. 1990a. Phases and Facies in the Archaeology of Central Africa. In A History of
 
 
de Maret, Pierre. 1996. Shum Laka (Cameroon): Human Burials and General Perspectives. In 
 
 
 
 
de Maret, Pierre. 2005. From Pottery Groups to Ethnic Groups in Central Africa. In African
 
 
de Maret, Pierre. 2013. Archaeology of Central Africa. In, Oxford Bibliographies in African 
 
 
 
de Maret, Pierre, Bernard Clist, and Wim Van Neer. 1987. Résultats des premières fouilles dans 
 
 
 
de Maret, Pierre, Raymond Asombang, Els Cornelissen, Phillip Lavachery, J. Moyersons, and 
 
 
 
de Maret, Pierre, Raymond Asombang, Els Cornelissen, Phillip Lavachery, and J. Moyersons. 
 
 
 
de Maret, Pierre. 1996. Urban Origins in Central Africa: The Case of Kongo. The Development
 
 

of Urbanism from a Global Perspective edited by Paul J. Sinclair 1-15. Uppsala
University: Uppsala, Sweden.  

Denbow, James. 1990. Congo to Kalahari: Data and Hypotheses about the Political Economy 
of the Early Western Stream of the Bantu Expansion. African Archaeology Review 8: 
 
 
139-175. 

Deveneaux, Gus. 1978. The Frontier in Recent African History. International Journal of African
 

Historical Studies 11 (1): 63-85. 

102 

Wim Van Neer. 1993. Preliminary Results of the 1991-1992 Field Season at Shum Lake, 
Northwestern Province, Cameroon. Nyame Akuma, 39: 13-15. 

1995. Continuing Research at Shum Laka Rock Shelter, Cameroon (1993-1994 field 
season). Nyame Akuma, 43: 2-3. 

Donnan, Hastings and Thomas M. Wilson. 1999. Borders: Frontiers of Identity, Nation and
 

State. Oxford: Berg. 

Drake, St. Clair. 1975. The Black Diaspora in Pan-African Perspective. The Black Scholar 
 

7(1): 2-13. 

Du Bois, W.E.B. 2007. Chapter IV: The Peopling of Africa. In, The World and Africa: An 
 
Inquiry into the Part Which Africa Has Played in World History and Color and De: 
The Oxford W. E. B. Du Bois, edited by Henry Louis Gates Jr, 52-62. Oxford: Oxford 
 
 
University Press.  

Ebron, Paulla A. 1999. Tourists as Pilgrims: Commercial Fashioning of Transatlantic Politics. 
 

American Ethnologist 26(4): 910-932. 

Edozie, Rita Kiki. 2012. The Sixth Zone: The African Diaspora and the African Union’s 
Global Era Pan Africanism. Journal of African American Studies 16(2): 268-299. 
 

Eggert, M.K.H. 1984 (1986), Imbonga und Lingonda: zur fruhesten Besiedlung des 
 
 

Zentralafrikanischers Regionwaldes. Beiträge zur allgemeinen und vergleichenden 
archäologie, Band 6:247-288. 

the Equatorial Rain Forest. The African Archaeological Review, 5: 129-145. 

Eggert, Manfred K.H. 1987. Imbonga and Batalimo: Ceramic Evidence for Early Settlement of 
 
 
Eggert, Manfred K.H. 1992. The Central African Rain Forest: Historical Speculations and 

Archaeological Facts. World Archaeology, 24(1): 1-24. 

 
Eggert, Manfred K.H. 1993. Central Africa and the Archaeology of the Equatorial Rainforest: 

Reflections on Some Major Topics. In The archaeology of Africa: food, metals and 
towns, edited by Thurston Shaw, Paul Sincliar, Bassey Andah, and Alex Okpoko, 289-
329. Routledge, London. 

 
Eggert, Manfred. 2005. The Bantu Problem and African Archaeology. In African  Archaeology: 

A Critical Introduction edited by Ann Brower-Stahl, 301-326. Malden: Blackwell 
Publishing.  

Eggert, Mangred K.H. 2006. Pits, Graves, and Grains: Archaeological and Archaeobotanical 
 

Research in Southern Cameroun. Journal of African Archaeology 4.2: 273-298. 

Eggert, Manfred. 2014. Central Africa and the Archaeology of the equatorial rainforest:
 
 

Reflections on some major topics. In The Archaeology of Africa: Food, Metals, and
Towns, Bassey Andah et al eds. 289-329. London: Routledge. 

Eggert, M.K.H. and Kanimba Misago. 1987a. Recherches archélogiques et ethnographiques dans 
les Régions de I'Equateur (Zaire), de la Cuvette de la Sangha et de la Likouala (Congo). : 
 
 
Rapport: préliminaire. Annales Aequatoria 8:481-486. 

103 

Eggert, M.K.H. and D. Seidensticker. 2016. Campo Archaeological Research at the Mouth of the 
 

Ntem River (South Cameroon). Africa Praehistorica 31: Köln. Forth coming 

Ehret, Christopher. 2015. Bantu History: Big Advance, Although with a Chronological 
 
 

Contradiction. Proceedings of the National Academy of Sciences of the United States of 
America 112(44): 13428-13429. 

Essomba, Joseph-Marie. 1989. Dix Ans de Recherches au Cameroun Méridional. Nsi, 6:33-57. 

Fairley, Nancy J. 1987. Ideology and State Formation: The Ekie of Southern Zaire. In The
AfricanFrontier. The Reproduction of Traditional African Societies edited by Igor
 
 
Kopytoff, 91-100. Bloomington, I.N.: Indiana University Press. 
 
Falola, Toyin. 2013. The African Diaspora: Slavery, Modernity, and Globalization. Rochester: 
 
 
Fancher, Jason M. 2009. An Ethnoarchaeological analysis of Small Prey Bone Assemblages 
 
 

Produced by Forest Foragers of the Central African Republic. PhD. Dissertation 
Washington State University. 

University of Rochester Press. 

Press. 

Feierman, Steven. 1974. The Shambaa Kingdom: A history. Madison: University of Wisconsin
 
 
Fenaux, Sylvie. “Fouilles archéologiques à Shum Laka (Cameroun). Etude préliminaire du 
 
 
Fennell, Christopher C. 2007. Crossroads and Cosmologies: Diaspora and Ethnogenesis in the 
 

matériel osseux humain.” Master’s Thesis, Université Libre de Bruxelles, 1992.  

New World. Gainesville: University of Florida Press. 

edited by Ann Brower Stahl, 187-224. Malden: Blackwell Publishing. 

Feuer, Bryan. 2016. Boundaries, Borders, and Frontiers in Archaeology: A Study of Spatial
 
Relationships. Jefferson, North Carolina: McFarland and Company, Inc. Publishers. 
 
Gifford-Gonzalez, Diane. 2005. Pastoralism and its Consequences. In African Archaeology 
 
 
Gilroy, Paul. 1993. The Black Atlantic: Modernity and Double Consciousness. Cambridge: 
 
 
Gond, Valéry et al. 2013. Vegetation structure and greenness in Central Africa from Modis 
 
 
Gonzalez- Ruibal, Alfredo et al. 2011. Early Iron Age Burials from Equatorial Guinea: The
 
 
Gonzalez Ruibal, Alfredo et al. 2013. An Ancient and Common Tradition: Funerary Rituals
 

and Society in Equatorial Guinea. African Archaeological Review 30 (2): 115-14. 

Sites of Corisco Island. Journal of African Archaeology 9(l): 41-66. 

Harvard University Press. 

multi-temporal data. Philosophical Transactions: Biological Sciences 368(1625): 1-8. 

104 

González- Rubial, Alfredo. 2014. An Archaeology of Resistance: Materiality and Time in an 
 

African Borderland. Plymouth, Rowman and Littlefield, Archaeology in Society Series.  

Goucher, Candice L.1993. African Metallurgy in the Atlantic World. The African 
 

Archaeological Review Papers in Honour of Merrick Posnansky, 11: 197-215. 

the Route and Pace of Human Dispersals. PNAS 112(43): 13296-13301. 

Systems. In, The Archaeology of Frontiers and Boundaries, Green and Perlman eds 3n
 12. Orlando: Academic Press. 

Green, Stanton W. and Stephen M. Perlman. 1985. Frontiers, Boundaries and Open Social 
 
 
 
Grollemund, Rebecca, Simon Branford et al. 2015. Bantu Expansion Shows That Habitat Alters 
 
 
Guyer, Jane I. and Samuel M. Eno Belinga. 1995. Wealth in People as Wealth in Knowledge:
 
Accumulation and Composition in Equatorial Africa. The Journal of African History
 
36(1): 91-120. 
 
Hall, Stuart. 1990. Culture Identity and Diaspora. In, Colonial discource and post-colonial 
 
theory: a reader edited by Patrick Williams and Laura Chrisman, 222-237. London: 
 
Harvester Wheatsheaf.  
 
Harris, Joseph. 1993. Return Movements to West and East Africa: A Comparative Approach. In 
Global Dimensions of the African Diaspora 2nd edition edited by Joseph E. Harris, 51-64. 
 
 
Washington D.C.: Howard University Press.  
 
Hamilton, Ruth Simms. 1997. Toward a Conceptualization of Modern Diasporas: Exploring 
 
 
 
 
 
Hamilton, Ruth Simms. 2007. Rethinking the African Diaspora: Global Dynamics. In, Routes of 
 
 

Contours of African Diaspora Social Identity Formation. In, Contemporary Diasporas: A 
focus on Asian Pacifics, edited by H. Eric Schockman, Eui-Young Yu, and Kay Songs, 1-
48. University of Southern California, The Center for Multiethnic and Transnational 
Studies Occasional Paper Series (Monograph Paper No. 3, Vol. II). 

Passage: Rethinking the African Diaspora edited by Ruth Simms Hamilton 1-40. East 
Lansing: Michigan State University Press.  

Hauser, Mark W. and Douglas V. Armstrong. 2012. The Archaeology of Not Being governed: A
 
counterpoint to a history of settlement of two colonies in the Eastern Caribbean. Journal
of Social Archaeology 12 (3): 310-333. 
 

Hawthorne, Walter. 2003. Planting Rice and Harvesting Slaves: Transformations along the 
 

Guinea-Bissau Coast, 1400-1900 (Social History of Africa). Portsmouth: Heinemann.  

Heimlich, Geoffroy. 2010. Lower Congo Rock Art Revisited. Nyame Akuma No.74: 42-50. 

Helms, Mary W. and Franklin O. Loveland, eds. 1976. Frontier Adaptations in Lower Central
 

America. Philadelphia: Institute for the Study of Human Issues.  

105 

Hess, Andrew C. 1978. The Forgotten Frontier: A History of the Sixteenth Century Ibero 
 

AfricanFrontier. Chicago: University of Chicago Press. 

Hodder, Ian. 1985. Boundaries and Strategies: An Ethnoarchaeological Study. In, The
 
 

Archaeology of Frontiers and Boundaries, Green and Perlman eds, 141-159. Orlando:
Academic Press. 

Hopkins, Benjamin D. (2015). The Frontier Crimes Regulation and Frontier
 

Governmentality. The Journal of Asian Studies 74(2):369-389. 

Isaacman, Allen and Barbara Isaacman. 1975.  The Prazeros as Transfrontiersmen: A Study in
 
 

Social and Cultural Change. International Journal of African Historical Studies 8 (1): 1-
39. 

Jamieson, Ross. 1995. Material Culture and Social Death: African- American Burial Practices. 
 

Historical Archaeology, 29(4): 39-58. 

Johnston, Harry Hamilton. 1919. A Comparative Study of the Bantu and semi-Bantu 
 

Languages. Oxford: Clarendon Press. 

Kahlheber, Stefanie et al. 2014. Pearl Millet and Other Plant Remains from the Early Iron Age 
 
 

Site of Boso-Njafo (Inner Congo Basin, Democratic Republic of the Congo). African 
Archaeological Review, 3: 479-512. 

King, Adam and Maureen S. Meyers. 2002. Exploring the Edges of the Mississippian World.
 
 

Southeastern Archaeology 21 (2): 113-116. (SPECIAL Edited thematic section: 
Frontiers, Backwaters, and Peripheries: Exploring the edges of the Mississippian World. 

Klieman, Kairn A. 2003. “The Pygmies Were Our Compass:” Bantu and Batwa in the History of 
 

West Central Africa, Early Times to c. 1900 C.E.. Portsmouth, NH: Heinemann. 

Kopytoff, Igor. 1987. The Internal African Frontier: The making of African Political Culture. In
 
 

The African Frontier. The Reproduction of Traditional African Societies 3-86, Igor 
Kopytoff ed. Bloomington, I.N.: Indiana University Press.  

Kowalewski, Stephen A. 2008. Regional Settlement Pattern Studies. Journal of Archaeological 
 

Research 16(3): 225-285. 

Laden, Gregory Thomas. 1992 Ethnoarchaeology and Land use Ecology of the Efe (Pygmies) of 
 

the Ituri Rain Forest, Zaire. PhD. Dissertation Harvard University.  

Lancaster, Chet S. 1987. Political Structure and Ethnicity in an Immigrant Society: The Goba of
the Zambezi. In, The African Frontier. The Reproduction of Traditional African Societies
 
 
edited by Igor Kopytoff, 101-120. Bloomington, I.N.: Indiana University Press.  
 

106 

Sépia, 

Metal (North western Cameroon), In, Aspects of African Archaeology, Papers from the 
10th PanAfrican Association for prehistory and Related Studies Congress, edited by 
Gilbert Pwiti and Robert Soper 266-274. Harare, University of Zimbawe Publication. 

Lanfranchi, Raymond, and Bernard Clist, eds. 1991. Aux Origines de l’Afrique Centrale. Paris:
 
 
Lavachery, Philippe. 1996. Shum Laka Rock Shelter late Holocene Deposits: from Stone to 
 
 
 
 
Lavachery, Philippe. “De la pierre au métal. Archéologie des dépôts holocènes de l'abri sous 
 
roche de Shum Laka (Cameroun).” Dissertation, Université Libre de Bruxelles, 1998. 
 
Lavachery, Philippe. 2001. The Holocene archaeological sequence of Shum Laka rock shelter 
 
(Grassfields, western Cameroon). The African Archaeological Review, 18(4): 213-247. 
 
Lavachery, Philippe and Els Cornelissen. 2000. Natural and Cultural Spatial Patterning in the 
 
 

Late Holocene Deposits of Shum Laka Rock Shelter, Cameroon, Journal of Field 
Archaeology, 27 (2): 153-158. 

Lavachery, Philippe, Els Cornelissen, Jan Moyersons, and Pierre de Maret. 1996. 30 000 ans 
 
 

d’occupation, 6 mois de fouilles: Shum Laka, un site exceptionnel en Afrique centrale. 
Anthropologie et Préhistoire, 107:197-211. 

Lewis, Kenneth. 1984. The American Frontier: An Archaeological Study of Settlement Pattern
 

and Process. Orlando: Academic Press. 

Li, Sen, Carina Schlebusch, and Mattias Jakobsson. 2014. Genetic Variation Reveals Large-
Scale Population Expansion and Migration During the Expansion of Bantu-Speaking 
 
Peoples. Proceedings of the Royal Society B 281:20141448. 
 
 
http://dx.doi.org/10.1098/rspb.2014.1448 

Lightfoot, Kent G. and Antoinette Martinez. 1995. Frontiers and Boundaries in Archaeological
 

Perspective. Annual Review of Anthropology 24: 471-492. 

Lilley, Ian. 2004. Diaspora and Identity in Archaeology: Moving Beyond the Black Atlantic. In, 
Companion to Social Archaeology edited by Lynn Meskell and Robert W. Preucel 287- 
 
 
312. Malden, MA: Blackwell Publishers. 
 
Lupo, Karen et al. 2014. On Late Holocene Population Interactions in the Northwestern Congo
 
 
 
 
Kimpianga, Mahaniah.1993. The Presence of Black Americans in Lower Congo from 1878 to 
 
 
 

Basin: When, How, and Why Does the Ethnographic Pattern Begin? In Hunter
Gatherers of the Congo Basin: Cultures, Histories, and Biology of African Pygmies
edited by Barry S. Hewlett 59-84. New Brunswick: Transaction Publishers. 

1921. In Global Dimensions of the African Diaspora 2nd edition edited by Joseph E. 
Harris, 268-282. Washington D.C.: Howard University Press.  

107 

Maley, Jean, and Patrice Brenac. 1998. Vegetation Dynamics, Palaeoenvironments and Climatic 
 
 

Changes in the Forests of West Cameroon during the last 28,000 years BP. Review of 
Palaeobotany and Palynology 99:157–187.  

méridional. Etude archéologique des sites de Nkang et de Ndindan.” Dissertation, 
Université Libre de Bruxelles, 1996. 

Mastan, Irina. 2014. Urban Identification Agents and Historical Discourse in a Frontier City. 
Case Study: Brasov During the 16th and 17th Centuries. Zeszyty Naukowe Uniwersytetu
 
 
Jagiellonskiego. Prace Historyczne 141(1): 15-35,196. 
 
Mbida Mindzie, Christophe. “L'émergence des communautés villageoises au Cameroun 
 
 
 
Mbida Mindzie, Christophe, Wim Van Neer, Hugues Doutrelepont and Luc Vrydaghs. 2000. 
 
 
 
Mbida Mindzie, Christophe, Hughes Doutrelepont, Luc Vrydaghs, Rony L. Swennen, Rudy J. 
 
 
 
 
Meister, Conny and Manfred Eggert. 2008. On the Early Iron Age in Southern Cameroon.
 

Swennen, Hans Beeckman, Edmond de Langhe and Pierre de Maret. 2001. First 
archaeological evidence of banana cultivation in Central Africa during the third 
millennium before present. Vegetation History and Archaeobotany, 10: 1-6. 

Evidence for banana cultivation and animal husbandry during the first millennium BC in 
the forest of southern Cameroon. Journal of Archaeological Sciences, 27: 151-162. 

Journal of African Archaeology 6(2): 183-202. 

Meister, Conny. 2010. Remarks on the Early Iron Age Burial Sites from Southern Cameroon
 

African Archaeological Review 27 (3): 237-249. 

Mercader, Julio and Alison S. Brooks. 2001. Across Forests and Savannas: Later Stone Age
 
 

Assemblages from Ituri and Semliki, Democratic Republic of Congo. Journal of
Anthropological Research 57(2): 197-217. 

Mintz, Sidney and Richard Price. 1992. The Birth of African American Culture: An 
 

Anthropological Perspective. Boston: Beacon Press. 

Monroe, J. Cameron and Akinwumi Ogundiran. 2012. Power and Landscape in Atlantic West
Africa. In Power and Landscape in Atlantic West Africa: Archaeological Perspectives
 
 
J.C. Monroe and A. Ogundiran eds. 1-49. Cambridge: Cambridge University Press. 

Monroe, J. Cameron. 2013. Power and Agency in Precolonial African States. Annual Review
 

Anthropology 42: 17-35.  

Mortelmans, Georges. 1962. Archéologie des grottes Dimba et Ngovo (Région de Thysville, 
 
 
 
 

Bas-Congo), In du IVe congrès panafricain de préhistoire et de l'étude du Quaternaire, 
1959, Léopoldville. Section III. Pré- et protohistoire Actes, edited by Georges 
Mortelmans and Jacques Nenquin, 407-425. Musée Royal de l'Afrique centrale, Annales 
des sciences humaines. 

108 

Archaeology of the Borderlands. Journal of Archaeological Method and Theory. 17
(2):101-131. 

Naum, Magdalena. 2010. Re-emerging Frontiers: Postcolonial Theory and Historical
 
 
 
Needham, D.E., E.K. Mashingaidze, and N. Bhebe. 1984. From Iron Age to Independence: A
 
 
Nenquin, Jacques. 1957. Excavations at Sanga: The Protohistoric Necropolis. Tervuren,
 
 
Neumann, Katharina, Koen Bostoen, Alexa Höhn, Stephanie Kahlheber, Alfred Ngomanda and 
 
Barthelémy Tchiengué, 2012. First Farmers in the Central African Rainforest: A view 
from southern Cameroon. Quaternary International 249:53-62. 
 

history of Central Africa. Burnt Hill: Longman. 

Belgium: Musee Royal De Le’Afrique Centrale. 

Nyamnjoh, Francis B. 2017. Incompleteness: Frontier Africa and the Currency of Conviviality. 
 

Journal of Asian and African Studies 52(3):253-270. 

Ochonu, Moses. 2015. Caliphate Expansion and Sociopolitical Change in Nineteenth-Century
 

Lower Benue Hinterlands. Journal of West African History 1(1): 133-178Ogundiran 2008 

Ogundiran, Akinwumi and Toyin Falola eds. 2007.  Archaeology of Atlantic Africa and the 
 

African Diaspora. Bloomington: Indiana University Press. 

Ogundiran, Akinwumi. 2009. Material Life and Domestic Economy in a Frontier of the Oyo
Empire During the Mid-Atlantic Age. The International Journal of African Historical
 
 
Studies 43(3): 351-385. 

Ogundiran, Akinwumi. 2014. The Making of an Internal Frontier Settlement: Archaeology and
 
 

Historical Process in Osun Grove (Nigeria), Seventeenth to Eighteenth Centuries. African
Archaeological Review 31: 1-24. 

Plains. Institute of African Studies: Research Review Supplement 17: 91-106. 

Osei-Tutu, Brempong. 2006. Frontier Archaeology of the Akuapem Ridge and the Eastern Accra
 
 
Oslisly, Richard. 1996. The Rock Art of Gabon: Techniques, Themes, and Estimation of its Age.
 
 

Aspects of African Archaeology Gilbert Pwiti and Robert Soper eds 361-370. Harare:
University of Zimbabwe Publications. 

Oslisly, Richard, et al. 2013. Climatic and Cultural Changes in the West Congo Basin Forest
 
 

over the Past 5000 Years. Philosophical Transitions, The Royal Society, Biological
Sciences 368: 2-10. 

Owen, Bruce D. 2005. Distant Colonies and Explosive Collapse: The Two States of the 
 

Tiwanaku Diaspora in the Osmore Drainage. Latin American Antiquity 16(1): 45-80. 

109 

Packard, Randall M. 1987. Debating in a Common Idiom: Variant Traditions of Genesis among
 
 
 

the BaShu of Eastern Zaire. In The African Frontier. The Reproduction of Traditional 
African Societies edited by Igor Kopytoff, 148-161. Bloomington, I.N.: Indiana 
University Press.  

Palmer, Colin A. 2000. Defining and Studying the Modern African Diaspora. The Journal of 
 

Negro History 85(1/2): 27-32. 

Parker, Bradley J. 2002. At the edge of Empire: Conceptualizing Assyria’s Anatolian Frontier 
 

ca. 700 BC. Journal of Anthropological Archaeology21 (2002): 370-395. 

Parker, Bradley J. 2006. Toward Understanding of Borderland Processes. American Antiquity
 

71(1):77-100. 

Parker, Bradley J and Lars Rodseth. 2006. Introduction: Theoretical Considerations in the Study
of Frontiers. In Untaming the Frontier in Anthropology, Archaeology, and History 3-22.
 
 
Tucson: University of Arizona Press. 

Patterson, Tiffany Ruby and Robin D. G. Kelley 2000. Unfinished Migrations Reflections on the 
 
 

African Diaspora and the Making of the Modern World. African Studies Review 43(1): 
11-45. 

Patin, Etienne, Marie Lopez, Rebecca Grollemund, et al. 2017. Dispersals and Genetic 
 
 

Adaptation of Bantu-Speaking Populations in Africa and North America. Science, 
356: 543-546.  

Preuss, J. 1990. Premières séries d'artefacts lithiques originaires du bassin intérieur du Zaïre . 
 
 

Paysages quaternaires de l'Afrique centrale atlantique edited by Raymond Lanfranchi, 
Dominique Schwartz, 431-438. Paris: IRD Editions. 

Peacock, Andrew C.S.eds. 2009. The Frontiers of the Ottoman World. Oxford: Oxford
 

University Press. 

Phillipson, David W. 1985. Early Food-Production in Central and Southern Africa. In, From
 
 
 

Hunters to Farmers: The Causes and Consequences of Food Production in Africa Clark,
J. Desmond Clark and Steven A. Brandt eds 272-280. Berkeley: University of California
Press. 

Rodseth, Lars. 2006. The Fragmentary Frontier: Expansion and Ethnogenesis in the Himalayas.
 
 

In Untaming the Frontier in Anthropology, Archaeology, and History 83-110. Tucson:
University of Arizona Press. 

Rosen, Jack, María Teresa Planella, and Rubén Stehberg. 2010. Archaeobotany of Cerro del
 
Inga, Chile, at the Southern Inka Frontier. In, Distant Provinces in the Inka Empire:
Toward a deeper understanding of Inka Imperialism Michael A. Malpass and Sonia
 
 
Alconini eds, 14-13. Iowa City: University of Iowa Press. 

110 

Rösler, Michael and Tobias Wendl eds. 1999. Introduction Frontiers and borderlands. The rise
 
 

and relevance of an anthropological research genre. In, Frontiers and Borderlands:
Anthropological Perspectives 1-30. Frankfort: Peter Lang. 

Rowold, Diane et al. 2016. On the Bantu Expansion. Gene 593(2016): 48-57. 

Interaction. New York. Plenum. 

Schortman, Edward M. and Patricia A. Urban. 1992. Resources Power and Interregional
 
 
Schwartz, Dominique. 1992. Assèchement climatique vers 3000 B.P. et expansion Bantu en
Afrique Centrale atlantique: quelques réflexions. Bulletin de la Société géologique de
 
 
France 163(3): 353–361. 
 
Skinner, Elliot. 1982/1993. The Dialectic Between Diasporas and Homelands. In Global 
Dimensions of the African Diaspora 2nd edition edited by Joseph E. Harris, 11-40. 
 
Washington D.C.: Howard University Press. 
 

Stainier, Xavier. 1899. L'âge de la pierre au Congo. Bruzelles: Vande Weghe.  

Staski, Edward. 1998. Change and Inertia on the Frontier: Archaeology at the Paraje de San
 
 

Diego, Camino Real, in Southern New Mexico. International Journal of Historical
Archaeology 2 (1): 21-44. 

Stein, Burton. 1980. Peasant State and Society in Medieval South India. Delhi: Oxford
 

University Press.  

Steward and Robert F. Murphy, eds. Urbana: University of Illinois Press. 

Steward, Julian. 1977. Evolution and Ecology: Essays on Social Transformations. Jane C. 
 
 
Stewart, Kearsley A. 1993. Iron Age Ceramic Studies in Great Lakes Eastern Africa: A Critical 

and Historiographical Review. African Archaeological Review 11: 21-37. 

 
Shepperson, George. 1982/1993. African Diaspora: Concept and Context. In Global Dimensions 
of the African Diaspora 2nd edition edited by Joseph E. Harris, 41-50. Washington D.C.: 
 
 
Howard University Press. 

Singleton, Theresa A. 1999. An Introduction to African American Archaeology. In I, Too, Am 
 
 

America: Archaeological Studies of African American Life edited by Theresa Singleton, 
1-20. Charlottesville: University Press of Virginia. 

Smith, Stuart Tyson. 2006. To the Supports of Heaven: Political and Ideological Conceptions of 
Frontiers in Ancient Egypt. In Untaming the Frontier in Anthropology, Archaeology, and 
 
History Bradley J. Parker and Lars Rodseth eds, 207-231. Tucson: University of Arizona 
 
 
Press. 

111 

South, Stanley. 1977. Method and Theory in Historical Archaeology. New York: Academic
 

Press. 

Sowunmi, Adebisi. 1999. The Significance of the Oil Palm (Elaeis guineensis Jacq.) in the Late 
 
 

Holocene Environments of West and West Central Africa: A Further Consideration. 
Vegetation History and Archaeobotany 8(3) :199-210. 

Stahl, Ann Brower. 1994. Innovation, Diffusion, and Cultural Contact: The Holocene 
 

Archaeology of Ghana. Journal of World Prehistory 8(1): 51-112. 

Stein, Burton. 1980. Peasant State and Society in Medieval South India. Delhi: Oxford
 

University Press.  

Archaeology of Interregional Interaction: Archaeology Division Distinguished Lecture
AAA Annual Meeting. American Anthropologist 104 (3): 903-916. 

Stein, Gil J. 1998. From Passive Periphery to Active Agents: Emerging Perspectives in the
 
 
 
Straughn, Ian B. 2014. North Africa: Historical Archaeology. Encyclopedia of Global
 

Archaeology edited by Claire Smith. New York: Springer, Electronic Web Resource. 

Turner, Frederick Jackson. 1894. The Significance of the Frontier in American History.
 

American Historical Association, Annual Report for 1893: 199-227. 

Usman, Aribidesi Adisa. 2011. The Yoruba Frontier: A Regional History of Community
 
 

Formation, Experience, and Changes in West Africa. Durham, N.C.: Carolina Academic
Press. 

The Role of Domestic Stock Keeping, Hunting and Fishing. In Droughts, Food and 
Culture: Ecological Change and Food Security in Africa’s Later Prehistory 
edited by Fekri A. Hassa, 251-274. New York: Kluwer Academic/Plenum  Publishers. 

Van Neer, Wim. 2002. Food Security in Western and Central Africa during the Late Holocene: 
 
 
 
 
Van Noten, Francis. 1982. The Bantu Expansion Problem. In The Archaeology of Central Africa
 
 
Vansina, Jan. 1979. Bantu in the Crystal Ball, I. History of Africa Vol 6: 287-333. 
 
Vansina, Jan. 1990.  Paths in the Rainforests: Toward a History of Political tradition in 

Van Noten, Francis, et al eds 65-69. Austria: Akademische Druck. 

Equatorial Africa. Madison: The University of Wisconsin Press.  

Providence/ Oxford: Berg. 

 
Vansina, Jan. 1992. Habitat, Economy, and Society in the Central African Rain Forest. 
 
 
Vansina, Jan 2004. How Societies are Born: Governance in West Central Africa before 1600. 
 

Charlottesville: University of Virginia Press. 

112 

Creek: Altamira Press. 

(DRC). Urban Studies (November) 50 (15): 3168-3184. 

Vlassenroot, Koen. 2013. Borderlands, Identity, and Urban Development: The Case of Goma
 
 
Vogel, Joseph eds. 2000. Ancient African Metallurgy: The Socio-Cultural Context. Walnut
 
 
White, F. 1979. The Guineo-Congolian Region and Its Relationships to Other Phytochoria. 
 
 
 
White, Carol Wayne. 2016. Black Lives and Sacred Humanity: Toward an African American 
 

Bulletin du Jardin botanique National de Belgique/ Bulletin van de Nationale Plantentuin 
van België 49(1/2): 11-55. 

Religious Naturalism. New York: Fordham University Press. 

Whitten, Norman E.  1974. Black Frontiersman: A south American Case.  
 

Cambridge, M.A.: Schenkman. 

und ihre Stellung im Kontext der Bantu-expansion.  Köln: Heinrich-Barth Institut. 

Wotzka, Hans Peter. 1995. Studien zur Archäologie des zentralafrikanischen Zaïre-Beckens 
 
 
Wyse, Akintola J.G. 1993. The Sierra Leone Krios: A Reappraisal from the Perspective of the 
African Diaspora. In Global Dimensions of the African Diaspora 2nd edition edited by 
 
 
Joseph E. Harris, 339-368. Washington D.C.: Howard University Press.  
 
Zeleza, Paul Tiyambe. 2005. Rewriting the African Diaspora: Beyond the Black Atlantic. 
 
 
Zeleza, Paul Tiyambe. 2008. The Challenges of Studying the African Diasporas. African 
 
 
Zeleza, Paul Tiyambe. 2010. African Diasporas: Toward a Global History. African Studies 
 

Sociological Review 12(2): 2008. 

African Affairs 10(414): 35-68. 

Review, 53(1):1-19. 

113