Symbiosis and transformation in Kenya's Meru District


Material Information

Symbiosis and transformation in Kenya's Meru District
Physical Description:
viii, 220 leaves : ; 29 cm.
Goldsmith, Paul, 1952-
Publication Date:


Subjects / Keywords:
Anthropology thesis, Ph. D
Dissertations, Academic -- Anthropology -- UF
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1995.
Includes bibliographical references (leaves 201-219).
Additional Physical Form:
Also available online.
Statement of Responsibility:
by Paul Goldsmith.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 022563270
oclc - 34424470
System ID:

Full Text







,UN:.V.r..:, -Y OF ', L.-. B I ,




This dissertation represents the culmination of many individual efforts. It was
funded by the Social Science Research Council. The Rockefeller Foundation provided
supplementary assistance for the collection of soil data, which although it does not
appear here, also contributed to the overall study by financing an additional round of
revisits to farms in the survey. I also received invaluable material and moral support
from members of my family and friends at several critical points on the way towards
completing the final product.
The chairman of my committee, Professor Ronald Cohen, provided steady and
untiring intellectual stimulation and guidance. To this I would add the example of
uncompromised theoretical leadership of Professor Marvin Harris, the conceptual
acumen of Professor Rene Lemarchand, and the uniquely sincere application of
intellectual ideas to African phenomena Professor Goran Hyden represents. Professor
Art Hansen also provided a supportive and practical angle on the overall enterprise, and
Professor R. Hunt Davis and Professor Peter Schmidt deserve important credit for their
institutional and personal support.
I benefited from an informal Meru think tank on Igembe development while in the
field. Its members included James Laiboni, Joel Kamincha, Samweli Meemee, Elias
Maitheta, Stanley Karunga, Gitonga M'Limung'i, Ngolwa M'Limung'i, Mutabari
M'Limung'i, and countless other local intellectuals operating from different
occupational niches in Meru and other parts of Kenya. Bashir Abdi, Peter Mbabu and
Yusuf Ahmed of Meru Mutindwa provided helpful background on different issues
ranging from local trees to Meru-Somali relations in Isiolo and the Northern Frontier
District. Osman Malta M'Mutunga of Majengo was an invaluable source of information
for the contemporary miraa trade; other Meru elders including Yusuf Mucheke and
Shabaan Hassan Manua contributed information on the historical development of the
miraa trade.
The District Officer for Igembe Division, Mr. Wandiama, warrants special mention
as the kind of unsung leader on the grassroots-administrative interface that rarely
receive the attention they deserve. Professor H.W.O.Okoth Ogendo provided
invaluable institutional support, and accepted me as a research associate at Nairobi
University's Institute for Population Study and Research. Mr. Mani of the National
Council for Science and Technology and Mr. Kibuna of the Office of the President
assisted me with official clearance without which the field study would have remained
stillborn. Professor David Gordon, of USAID's REDSO office in Nairobi helped with

input on my questionnaires, which were based on International Insitute for Tropical
Agriculture survey instruments Professor Abe Goldman was kind enough to share with
The librarians and staff at the Kenya National Archives, USIA library, and at the
International Center for Agroforestry Research all opened their facilities to my use and
made me feel welcome. Dr. Anthony Kinyua of Nairobi University's Center for
Nuclear Science Techniques provided a steady institutional base and feedback over the
duration of the study. I should also mention the divisional agricultural staff in Maua for
their cooperation in the field. The views and encouragement of my colleague, Dr.
Bashir Jama, also warrant acknowledgement, as does the exposure to tropical
agriculture that I received from Professors Hugh Popenoe, Robert Conrad, Ken Buhr,
and P.K. Nair at the University of Florida's Institute for Food and Agricultural
I thank my mother, Ms. Joan Smardon, my wife Safiya Usama, and my children for
their invaluable support, tolerance, and understanding in various ways. I am also
grateful to Peter Sokol and Jill Dygert for their practical assistance during the final
period of the work in progress. Ahmed Samatar of Cold Spring Harbor Laboratory,
Robert Comes, and Dr. Dorothy Edwards of the Washington University School of
Occupational Therapy provided invaluable computer support during the final stages of
write-up. Carol Lauriault's editorial and qualitative inputs were critical in seeing the
project's fruition from its initial conception to putting the document in its final form.
Peter O'Malley provided stimulation from New Orleans. Andrew Goldsmith, George
Thirikwa, Ali "Mwarabu" Mohammed, David Tomlinson, and Peter Brinkwart kept the
wheels rolling. Finally, like so many others who have tread this path before me, I
thank Professor Gwen Carter for her support and inspiration.






The Meru Peoples 3
Anthropological Research in the New Canaan 7
Feedback from the Field 10
Defining Development 13
The Cultural Endowment Problem 17


The Empirical Problem:
Population, Agricultural Systems, and Polity 20
The Evolution of the State 22
Problem Solving and Paradigms 28
Anthropological Research Traditions 29
Analyzing Social Systems 32
The Culture Problem 35
Culture and Cultural Endowment 37
Culture, Feedback, and Social Learning 39
Symbiosis 40
Coevolutionary Processess 43


Sustainability and Cultural Adaptation 48
Kuria: The Meru Definition of Sustainability 49
Precolonial Dynamics 50
Maasai-Meru Relations 53
The Nyambene Meru 55
The Meru Cultural Endowment 57
The Njuri Ncheke 58
Age-Set Organization 61
Clans, Kinship, and the Gichiaro Social Risk-Spreading
Mechanism 64
The Aathi Cultural Ideology 66
The Njuri in Transition 70
Njuri Ncheke-British Administration Cooperation 72

The Nyambene Farming System 75
Modern Dynamics of Nyambene Meru Cultural Institutions 78

Precolonial Trade Networks 86
The Decline of Pastoralist Power 87
The Impact of Ivory on Regional Trade Networks 88
Late Nineteenth Century Environmental Catastrophes 89
The British Influence 91
The Rise of Miraa as a Commercial Commodity 93
Social Aspects of Miraa Consumption 95
The Socioeconomic Context of Miraa Production 97
The Nyambene Range and The Meru Farming System 99
The Development of the Miraa Trade 103
Igembe Control of the Urban Trade 107
The International Miraa Trade 109
Miraa And Rural Development 114
Cash-Crops in Comparative Perspective 115


The Resilience of the Igembe System 120
Agronomic Aspects of Miraa Production 123
The Land Variables in Nyambene Small farm Production 128
Crops and Cultivars 133
Indigenous Agroforestry 135
Technological Innovations 137
Livestock 138
Constraints and Inputs 141
Production and Income 145


The Meru Lowlands in Perspective 150
Kenya's Agricultural Policy
Commercialization and Ecozone Symbiosis 152
Sectoral Goals for Kenyan Agriculture 153
Ecozone Symbiosis 161
Local Linkages 162
Markets and the Dynamics of Technological Change in the
Nyambenes 164
Production Markets 165
Crop Differentials 168
Market Influences: The Case of Sorghum 172
Consumer and Input Markets 173
Consumption Trends and Household Expenditures 178
Social Dynamics: Cultural Endowment and Social Learning 180
Formal and Informal Group Membership 182
African Developmental Processes Revisited 184


African Developmental Processes Revisited 187
The Resilience of African Agriculture 189
The State and Local Systems 193
Fission and Fusion 193
Fusionary Forces: Culture Revisited 195
The Urban Ecozone and Generational Dynamics Revisited 197

Biographical Sketch 220

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Paul Goldsmith

April 1994

Chairman: Ronald Cohen
Major Department: Department of Anthropology

This study addresses processes of agricultural systems development in Meru
District. The study demonstrates how local African societies coevolve in a manner
where populations adapt, specialize, merge, and diverge in various ways that allows
new forms of social and economic organization to emerge over time. This
coevolutionary dynamic links the Igembe Meru to their past while projecting their
developmental processes into the future. Meru environmental history reflects how
variation across ecological zones influences a number of social micro-adaptations. The
Meru agricultural system evolved through continuous cultural symbiosis of hunter-
gather. pastoral, and agricultural groups. Ethnicity followed group eco-niche
specialization, favoring exchange across ecological zones and the rise of regional trade
networks. The Nyambene Meru system featured in this study served as the hub of an
extensive trade network spanning the region between Mt. Kenya and the southern Lake
Turkana area, continues to demonstrate a high level of biodiversity, adaptive responses
to modem market phenomena, and cultural institutions that embody their indigenous,
cross-generational concept of social sustainability within the contemporary context of
capitalist-led change. Local systems adapting incrementally contribute in turn to the
processes of transformation on the national level, just as macro-level developments
create possibilities on the local level.



I am the Lord Almighty, and I have led you out of Egypt to the land of Ca-naan.
Holy Bible, Book of Numbers.

This study is based upon field research1 designed to investigate the commercialization of
agricultural production as the main determinant of small farm agricultural change in the
Nyambene Range, a line of hills extending in a northeastern direction from the foothills of
Mt. Kenya. The study is both an agro-economic anthropological study of how
commercialization in high potential areas influences the adjacent, semi-arid lowlands, and
an exploration of how cultural factors influence and contribute to agrarian systems. The
study demonstrates how local African societies coevolve in a manner where populations
adapt, specialize, merge, and diverge in various ways that allows new forms of social and
economic organization to emerge over time. This coevolutionary dynamic links the Igembe
Meru to their past while projecting their developmental processes into the future, and
defines the multi-sectoral nature of indigenous economy.
The Nyambene region has a rather fantastic appearance. When viewed from afar the
hills and caldera form a geometry of soft curves, angles, and conical shapes. Tropical
alpine forest caps the peaks and upper hillsides of the range. The plains below are
punctuated with lunaresque massifs dominated by the Gibraltar-like silhouette of Shaba
imposed in front of the hills of Samburu country in the distance. The local farms situated
in-between the forest and the plains resemble small jungles, a dense riot of vegetation with
the long crooked boles of mwenjela (Cussonia holstii) trees projecting above the canopies
like fringed stovepipes. The eye takes in a truly large swath of central and northern Kenya
from vantage points high in the range, and the configuration and nature of the hills
themselves appear different when viewed from various angles.
The 200,000 Igembe people, a section of the greater Meru community, inhabit the core
of this area. The Igembe rely predominantly on agriculture, supplemented by cattle and
small livestock that are pastured in the lowlands. Their farming system, based on the
region's biological diversity largely disregards technological monoculture. Access to land

1This research was funded by a doctoral research fellowship provided by the Social Science Research
Council, made possible by funds provided by Hewlett Packard.

is regulated by clans which control strips of land extending from the well-watered hills to
the dry plains. Otherwise, the criss-crossed configuration of the Meru polity and social
structure does not easily fit conventional anthropological categories of group social
evolution (c. f. Johnson and Earle 1987). The lives of the Igembe Meru are still intertwined
with important cultural institutions that have long since lapsed in other parts of Meru and
Kenya; age-sets and warrior initiation rituals, gichiaro fictive kinship relations, and local
councils of the pan-Meru parliament, the Njuri Ncheke continue to influence the daily lives
of the Igembe. The commercialization of agricultural production has not overturned the
internal order as much as it has reinforced it through cultural feedback of cross-generational
continuity and adaptation.
All of this "otherness" survives into the 1990s in the presence of homogenizing forces
of the capitalist world system, including intense commercialization of local small holder
agriculture. Indeed, the influx of "cash money," support Igembe income levels matched in
few other agricultural areas of East Africa. Unlike other small farm areas in Kenya
characterized by the lack of it, money is at least partially responsible for the unusual state of
affairs in the Igembe-inhabited core of the Nyamrnbene Range. A new dedication to "cash
money" has not displaced the role of food as a domestic currency, and as a result the
Igembe have converted the lowland areas formerly reserved for grazing to field crops. This
replaces some of the food production in the highlands lowered as their indigenous
agroforestry plots mature in a hundred year cycle. At the same time they have acquired
more cattle than they possessed before, which enhances the organic input recycling
properties of their mixed agro-silvo-pastoral agricultural system.
The local Meru munch and exchange small bunches of miraa (Catha edulis), a tree
domesticated for the mildly stimulatory qualities of its young twigs, as they go about their
work. Over the last century miraa has evolved from an accessory to traditional social
ceremony and ritual to a commercial crop that serves as the engine of the local economy.
Miraa consumption, and the increased social interaction and information exchange that
accompanies it, is an interesting subject in its own right. Miraa, as the primary commercial
crop in Igembe areas, is an important element within the local economy. Because it is also
an important social component in Nyambene society, it creates socioeconomic and
environmental feedbacks into the developmental process that are somewhat more
complicated than, for example, those associated with coffee, tea, or cotton.
Environmental differentials in the East African highlands engendered ecologically based
specialization and exchange among and within groups. Local markets in the Nyambenes
are situated mainly on the creases of ecological zones, and historically served as a magnet
attracting a number of agricultural, agro-pastoral, and nomadic pastoralists. Both this and

farmer-herdsmen conflict contributed to the Meru precolonial cultural symbiosis. The
expansion of the miraa trade has carried these cultural dynamics into the contemporary era.
Igembe traders, bottled up in the Nyambene region by the Pax Britannica, breakdown of
African regional exchange during the pre-world War H phase of British colonialism, and
the Mau Mau emergency, have through their miraa marketing networks come into contact
with a number of new social and economic landscapes: the majengo neighborhoods in
Nairobi, Mombasa, and other highland towns, the arid expanses of the Northern Frontier
District between Meru, Marsabit, and Moyale on the border of Ethiopia, and the Somali
towns of Northeastern Province. In effect, a variety of social landscapes, the Western
education system, and a steady steam of new information and ideas from urban areas feed
new forms of symbiosis. Symbiosis, in this case study, refers to the coevolutionary
outcomes resulting from the dynamic interactions, confluences, and recombinations of
technical, cultural, and institutional components of Kenya' s ethnic, organizational, and
economic pluralism. The case study strives to show through the Meru material that the
heterogeneous qualities of local adaptations in Kenya are in many ways an important legacy
that, under proper conditions, can contribute to and fuel the country's multi-sectoral
Once in the field, it was soon apparent that commercialization and change within the
Igembe Meru economy, perhaps more so than many local production systems in Kenya and
Africa, cannot be understood without examining the interactions and complementary
relationship between culture and agriculture. The system is therefore referred to at various
points in this study as the Nyambene cultural and agricultural system. This makes the
Nyambene case study more complex, but also, hopefully, interesting and potentially
valuable for comparison with the typical account of small farm commercialization in Africa.
Linguists often repeat the methodological maxim, "no contrast, no information." Perhaps
the most useful aspect of this study is that it does provide a contrasting trajectory of social
development, agricultural innovation, and cultural adaptation when compared to other
examples of African agrarian change involving the production of other world market
commodities and marketing through their organizational structures.
A brief map of where the following pages take one may assist the reader in following
the scheme of this work in progress. This initial chapter is a thematic exposition of the
study's context: the broad theoretical issues, the cultural, historical, and environmental
backdrop of the study, and references to how we go about the study of African societies.
The second chapter provides a more specific treatment of the empirical and conceptual
problems addressed, and the theoretical framework for their analysis. Chapters three and
four provide several perspectives on the historical, cultural, and economic development of

the system up to the present. The micro data on the Nyambene farming system and the
conclusions on ecozone symbiosis do not appear until chapters six and seven. Finally, in
chapter seven, I conclude by placing the case study within the larger macro context of
socioeconomic change and transformation in Kenya.

The Meru Peoples

The Meru are a Bantu speaking people consisting of eight groups united by common
history, culture, and social institutions. The Meru sometimes refer to the fecund country
on the northeastern fringe of Kenya' s central highlands where they live as the New
Canaan, after the Biblical promised land. There are a number of parallels between the flight
of the Israelites from Egypt and the story of how the Meru entered their own promised land
some three hundred years ago. The proto-Meru fled Mbowa, their coastal homeland, to
escape the oppression of the "Nguo Ntune," or "antu meria," red-clothed, light colored
overlords (probably the Portuguese). The Nguo Ntune attempted to enslave them by
posing three riddles, the elders solved the first two, but after they were confounded by the
third, the Mbowans plotted their escape. The Mbowa story makes them to this day wary
of divulging information to light skinned outsiders.
Three groups of clans resulted from the original flight. The white clans escaped during
the day, the black clans fled under the cover of night, and the red clans slipped away at
sunrise. These clans trekked across the dry hinterland as one group led by a heroic leader
named Komenjue. They crossed the Tana River by felling a Mugumu tree (Ficus
natalensis) large enough to span the river. Not long after this the proto-Meru clans
divided into several different groups, and moved up the foothills of Mt. Kenya and the
Nyambenes. The Meru "tribes" retained a large degree of their post-Mbowa unity by
emphasizing their common origin, culture, and language. This is reinforced through
institutions such as the Njuri Ncheke parliament and the religious authority of the Mugwe--
a person of the highest moral character chosen in each section to set an example of virtue,
advise leaders, and sanction the social rituals of each group. Through time, the Meru
prospered, growing just as much through the incorporation of other groups as through
natural fertility increase, into a community numbering over one million people, inhabiting
the district that bears their name.
The Meru district boundaries frame an irregular and ecologically varied topography. Its
main physical features include the eastern slopes of Kirimaara, or Mt. Kenya, a rocky
corridor between its foothills and the Tana River, and the Nyambene Range, a line of
geologically younger, volcanic hills rising on the northeastern flank of Africa' s second

highest mountain. The land of the Tigania, the Igembe' s cultural "cousins," spans the
foothills of Mt. Kenya and the Nyambenes.
The Nyambene region's ecosystem spans a number of different ecozones and ecotones,
and sustains an interestingly hybrid variation on society and economy. From the 8,247
foot peak of Itiene, the highest point of the Range, to the crater of Ntonyiri, the Nyambenes
Range is a variegated, fertile, well-watered extension of the highlands of central Kenya
descending into the austere badlands of northern Kenya. The hills narrow and gradually
drop in altitude before ending in a line of spectacular volcanic caldera that taper off into an
arid and waterless plain. The fringe of rangeland bordering the northern face of
Nyambenes, stretching from Isiolo to Garba Tula is one of the driest blocks of land in
Kenya. The southeastern slopes adjoin the more humid lowland plain that becomes Meru
National Park before shading into the undifferentiated semi-arid expanse that stretches
across the Northern Frontier District (NFD) into Somalia.
From these northern plains, the Nyambenes appear as a long ridge of rounded hills
whose outer slopes are as dry as the floor of the surrounding semi-desert. Their flanks are
covered with acacias, the wiry shrubs of the Barleria family, and other dry land plants,
and several unpalatable species of Pennisetum grass. A traveler viewing the range from the
north vantage point can be surprised to find an unusually green and fertile land upon
ascending and passing into the interior of the range. From the south, in contrast, the range
appears as an extension of the fertile central Kenya highlands.
The Igembe are the northern-most section of the Meru peoples. Their land bridges these
two distinct regions of Kenya: the agricultural highlands, or "upcountry," in Kenyan
parlance, and the "NFD," a term dating back to the time when the different pastoral
populations inhabiting this expanse were administratively separated from the rest of Kenya
in the colony's Northern Frontier District. Igembe Division and the more northerly parts of
Meru District in general display the legacy of both NFD-style benign neglect and the
external influences contributing to change in the highlands. Historically, the Igembe
straddled the NFD-highland divide, and developed a symbiotic, agro-pastoral cultural
heritage. They continue to evolve through linkages to the agricultural highlands and the
pastoral lowlands, although the urban experience accumulating over the past three decades
has extended the Igembe landscape to a third distinct environment.
The Nyambene farmers in this area produce some of the world's best tea, relatively less
of Meru's highly rated coffee, the finest miraa produced anywhere, and a cornucopia of
food crops. Small farms, incorporating numerous domestically valuable shrubs and trees
blend into the lush vegetation of the valleys and hillsides. The Igembe Meru cultivate field
crops on the fringes of their sophisticated agro-forests, multi-storied, jungle-like plots of

trees, yam vines, food crops, and medicinal herbs often developed over the course of
generations. Ruminants were the main livestock component of their production system.
Exceptionally tasty and tender goats, zero-grazed on a traditional diet of miraa and
mwenjela, are a specialty of the local cuisine, fetch very high prices in local markets, and
provide a culturally-patterned pathway for the adoption of new breeds of milch cows that
are now also zero-grazed on local farms.
Previously the Igembe pastoral tradition mainly centered on grazing livestock in the
extensive lower zones stretching from the base of Mt. Kenya to the I'ombe crater, an
important magadi soda lake falling on the extreme northern corner of the district. This
magadi lake is a major intersection attracting the Meru and a number of pastoral groups
who bring their livestock there for the lake's salt and minerals.2 For the Meru, I' ombe is
more than just a source of salt. The lake shares with the Nyambene and Mt. Kenya forests
an important religious position as one of the places where God dwells.
The combined spatial-cultural legacy of Nyambene society kept European influences in
check, and the Igembe entered the post colonial period sustained by a history and
environmental adaptation that supported their independent way of life. Styles of dress
depict their capacity to absorb new elements while maintaining their own traditions.
Straddlers by nature, many Igembe wear several different hats: the cowboy hats of
"godfather" capitalists, the woven caps and sombreros favored by farmers, the open
turbans preferred by herders, the Islamic-style skullcaps urban entrepreneurs often sport,
and the berets of intellectual proletarians. Many Igembe men alternate between town and
country in a high speed cycle of circular migration, especially during that period of the
lifecycle corresponding to warriorhood in the traditional age set system.
Except for the small class of urban-based professionals, the Igembe eschew paid
employment, prefering self-employment or the other opportunities that arise as a
consequence of their travels to near and distant corners of Kenya via the miraa trade. As
miraa traders they travel to dangerous destinations and endure difficult conditions in order
to establish working capital for future investments in their local place of birth. Unlike
urban migrants from other areas in Kenya, they rarely settle permanently outside their home
Back home, Igembe men organize farm and business activities, walking about the
countryside chewing miraa, exchanging the respected reddish-green twigs with friends they
meet on the path. After work they retire to local canteens and bars where they imbibe a
variety of local and bottled alcoholic drinks with gusto. Surrounded by warrior herdsman

2pasotral groups who frequented I'ombe included the Maasai, Samburu, Boran, Turkana, and Somali.

and shifta bandits, their fertile homeland is located in a rough neighborhood. This
ambience reinforces their own cultural tradition of institutionalized warriorhood. In today's
world they have parlayed this culturally directed aggressiveness into a variation on modem
entrepreneurship, retaining with their Tigania neighbors elements of a warrior society now
transposed, at least partially, into the arena of economic competition.
Otherwise an ebullient and rambunctiously outgoing people, the Nyambene Meru are
regarded by other Kenyans as hot-headed and temperamental. The panga, or machete, and
not the hoe, is the main farm implement used in Nyambene farming systems. Meru
ferocity and hair-trigger tempers in the presence of pangas lead to a reputedly high
incidence of interpersonal violence, prompting one local researcher studying incidents of
spontaneous panga-slashing to designate Nyambene society as a "subculture of violence."
The label, although unsubstantiated by actual comparative statistical data from other
districts, stuck.
Journalistic interpretations of Meru culture molded the understanding of real-life events.
One year after local press ran a feature story on the Nyambene "subculture of violence,"
The death of nineteen school girls during a student strike-cum-riot at St. Kizito Secondary
School in August of 1991, confirmed the public perception of the Nyambene's as Kenya's
most unpredictably violent region.3 Urbanized Meru in Nairobi concealed their ethnic
identity for quite some time afterwards. Still, a Nyambene man once told me that he and
his friends like the ethnic image of unpredictable warriors. The sterotype confers fear and
respect in Kenya's urban areas. Indeed other Meru peoples who look down on their
"Tigania" brethren at home, cultivate a common association with their Nyambene "hillbilly"
kinsmen in the cities for the same reason.

Anthropological Research in the New Canaan

The Nyambene area first pricked my interest during a short trip to Meru Park in 1980.
Previously, I had worked among the Swahili speaking peoples of Kenya's north coast.
Over the past several decades miraa consumption has developed into a major Swahili social
institution. I was curious about the trees that have had such a large impact on coastal
society and Muslim neighborhoods in other urban areas. The trip to the Park takes one
through the heartland of the Nyambene miraa production area, but to uneducated eyes
Meru's famed miraa is almost invisible against the backdrop of their multi-cropped, multi-

3The story was splashed across the local press and made CNN. The fact that this national tragedy happened
in "Tigania," as the Nyambene region is called by outsiders, made it understandable in terms of Kenya's
ethnic folk models.

storied agroforestry plots. As a consumption item, except for a few leafy bundles that
hardly represented the banana-leaf wrapped "kilos" one sees in the rest of Kenya, it was
also less apparent than expected. On the way back from the park, we chanced upon a local
farmer who volunteered to guide us to Muringene market, the main center for miraa export,
located right off the main road. This brief visit proved very interesting, beginning with the
rather unusual sign marking the turn-off to Muringene--"Muringene, Soko La Miraa Bila
Ubaguzi," (Muringene, the Miraa Market Without Discrimination).
The miraa bazaar appeared very different compared to other local markets throughout
Kenya. For one thing, it was a temporally concentrated event where most of the exchange,
packaging, and accounting is completed between one p.m. and three p.m. It also seemed
to involve a highly ordered underlying pattern despite the spontaneous and chaotic
impression it conveyed to an outsider. And it obviously involved financial transactions
significantly higher than your average cabbage or maize market
A miraa trader I knew in Nairobi insisted I make another trip to the Nyambenes to visit
his home. I returned the next year in the company of one of my companions from the last
trip, and this time we spent a week touring the vicinity between Kangeta and Maua on foot
and in densely packed matatus, packed Peugeot 404 station wagons crammed with bodies
in the style of the American telephone booth fad. Conversations and walks in Igembe
revealed a fascinating and infrequently investigated cultural-economic complex. The trail
blazed by miraa from Nyambene farms to the coast made for an interesting study, and I
decided to expand my interest in Swahili society to look at the developmental implications
of the production and marketing of this highly used, and sometimes abused, commodity of
Swahili social intercourse.
During another stopover at Muringene during the second visit a local businessman we
met in a small cafe/bar commented to me that "we have money, but we don' t have
development" Urban Lamu, whose relatively unchanged way of life is a magnet for
Western tourists, is often described by Kenyans as also lacking development As with
Lamu, the Nyambene area did not appear undeveloped by Kenyan rural standards. It does,
however, display unique and intriguing features of cultural continuity uncharacteristic of
other areas of highly commercialized small scale agriculture. As a Westerner, I had always
considered "development" to be more than a matter of material prosperity. Heuristically,
the businessman's statement provided me with a research question on local
commercialization. Can the usual forms of sociocultural change associated with
"development' vary independently of money incomes and types of cropping strategies?
On the surface, the urban, internationally oriented, maritime-based economy of the
northern Swahili presents little in common with the small-farm, local economy of the

Nyambene Meru. As in the Swahili settlements of Kenya's north coast, the connection
between indigenous economy and vitality of local culture presented an internally generated
version of African development. Otherwise the urban-rural-seafaring world of the Swahili
and that of Nyambene mountain men couldn't be more different. Under the influence of a
world religion, the Islamic Swahili developed a sophisticated scholarly tradition of religious
study and poetry, and Swahili culture embodies a philosophical and mystical tradition, and
a markedly romantic inclination. Under the influence of nomadic pastoralists, the Meru
have developed institutionalized warriorhood to complement their down-to-earth
predilection for consumption and cultural reproduction supported by an agricultural science
based on an indigenously evolved ideology of trees.
Deeper investigation revealed several theoretical commonalties: the correspondence
between indigenous economy and the retention of indigenous cultural institutions,
production systems based on intimate knowledge of the local environment, and a cultural
ideology supporting an independent and self-sufficient attitude toward the outside world.4
I continued to find the time and resources for annual visits to Igembe division every year
except 1987. I subsequently decided to use the indigenous economy of miraa as the
departure point for a research project encompassing the larger system of Igembe Meru
agriculture and economy. 5
The central economic hypothesis emerged from a 1986 Carter lecture on African
agriculture at the University of Florida delivered by Dr. Michael Lofchie (1988). Citing
Kenya as an example of policy-induced agricultural success, Lofchie expressed a sanguine
picture of Kenya's commercial agriculture as the engine of economic growth. During the
question period I referred to the Meru experience, and questioned whether policies
successful in the high potential agricultural zone could easily be extended to the more
challenging dry zones below them. In the absence of a clear response, I was forced to
ponder for myself the perrenial implications of the enduring issue, "what is development?"6
Despite the progress of commercial agriculture in high potential zones, the problem of
accommodating Kenya's expanding population as it spills over into the lower zones is a
very real concern, and Meru presented an ideal setting to investigate the problem of the

4The initial movement up, as noted earlier, began in the coastal homeland named Mbowa after the Meru
term for the rumen of a goat. The name is symbolic of the Meru's desire to conceal their origin in order to
disguise them from the Nguo Ntune who could potentially pursue them should their true identity be known.
The Bajuni and the Nyambene Meru also share cultural personalities typified in folk models as an outgoing
friendliness that easily gives way to tempestuous reactions to perceived slights. Manda island, directly
across from Lamu, matches Meru descriptions of Mbowa right down to minor details.
5Early data on the Igembe economy became a preliminary article on the production and marketing of miraa
in Meru (Goldsmith 1988).
61n all fairness, the problem posed was beyond the scope of Lofchie's lecture on "China's Lessons for
African Agriculture."

very real concern, and Meru presented an ideal setting to investigate the problem of the
commercialization of agriculture across ecological zones. Demographic growth in the
densely populated Nyambene highlands presents three options for the expanding
population: (1) intensify highland production; (2) migrate to urban areas; and (3) establish
agricultural settlements in the lower ecological zones. The Igembe often do all three of
these things spread over time in the course of their lifecycle.
Surmounting environmental constraints that limit the growth of agriculture and the social
welfare of many Kenyans is, nevertheless, an exceedingly important factor in the current
phase of Kenya's development. Demographic growth, agricultural commercialization,
technological innovation, and governance are major dimensions in this equation. If African
development was simply a question of manipulating these factors, cobbling together
solutions would be easier than it is has proved to be up to this time. The problems
examined in this study, and indeed, the elusive qualities of human society in general, invite
us not to stop too early by limiting our investigations solely to these issues.
When the proto-Meru Thagana clans migrated from the coastal island of Mbowa to the
eastern side of Mt. Kenya, they found the other side of the river more congested than did
the first wave of Bantu clans. Population pressure pushed them up the ecological gradient
of the hills, and different sections of the Meru conquered and/or absorbed elements of the
Agumba or Aathi, hunter-gathers, non-agricultural autocthones accustomed to harvesting
the natural bounty of the forest and savanna since prehistoric times.
The composite culture that coevolved out of this process reflects elements borrowed
through interaction with non-Meru communities complementing the Meru agricultural
adaptations to the different ecologies of the highland environment. The newcomers
integrated several distinct cultural institutions within the unified polity of Meru society,
subsuming a congruent set of production techniques adapted to a pattern of zonal
agriculture that spread production across several ecological zones.
During the 1960s the upward movement of the population slowed as the last belt of
agriculturally viable forest reserve--the tea zone--was cleared and settled. What goes up
must come down. The tea zones soon became saturated, and during the 1970s the flow
reversed as people migrated down to establish farms in areas formerly used for grazing and
temporary cultivation. This is the initial, empirical departure point for this study.

Feedback from the Field

It is not accidental that the research reported here provides a metaphorical
correspondence between the intercropped, multi-storied Igembe farming systems and

monocultural agriculture on one hand, and a multicausal conceptual framework of analysis
compared with the more monocultural emphasis of most Western social science theory. In
effect, the many tiered aspects of my research parallels an epistemological posture guiding
the investigation as a whole. Patterns emerge out of reoccurring relationships among
variables. Sets of these relationships said to explain the patterns are expressed as theories.
The test or plausability of theories derives from how well these hypothetical relationships
metaphorically or symbolically represent and explain the essential characteristics of the
patterned events of empirical phenomena.
Metaphor allows us to express abstract ideas in the language of commonly understood
relationships. Metaphors we use and manipulate to impose a conceptual order on diverse
phenomena often derive from the naturalistic order embedded in our cultural and linguistic
orientations (Lakoff and Johnson 1980, 4-6). Thus astronomy was the metaphysical
model for much atomic physics, just as the Oedipus myth helped Freud conceptualize the
internalization of social norms. Conceptual models are metaphorical abstractions of the
relationships among variables. For example, when we observe modernization, structures
of dependency, and other such phenomena in one setting, we then extrapolate the
conceptual relationship from one historical experience for the purpose of examining and
analyzing other case studies. The test of the model is how well it holds across many cases
and contexts.
In this manner, observed relationships are translated into comparatively useful analytical
tools. The model is operationalized to test the empirical validity of the metaphorical
relationship. The metaphor/theory is revised in the course of theory-building, critique, and
explaining the implications of new data. Conceptual processes therefore represent a two-
way street involving the interaction of empirical investigation and theoretical analysis. New
empirical and conceptual constraints arise and conceptual models are revised or expanded in
response to the new questions and problems (Laudan 1977).
This is a roundabout way of saying that the culture and system of Meru agriculture
provided important clues for understanding the more general dynamics operating within
African society and economy. Past failures to properly understand the dynamics of African
agriculture reflects to some extent the imposition of prevailing conventional models of
theoretical change. Analysis through Western theoretical lenses often produces examples
of what Hyden (1988) has identified as the monocultural tradition. Monocultural
agriculture based on single factor responses to inputs leading to increased outputs has
dominated much Third World agricultural research. According to Hyden, the particular
emphasis placed on monocropping in agriculture, the range of Western economic models--
be they neo-classical or Marxist--and external models of political development are all

representative of this monocultural tradition. The highest expression of theoretical
monoculture is elevating conceptual analysis to the level of quasi-religious paradigms.7
It became necessary, therefore, to adopt a conceptual framework allowing one to freely
intercrop theories, concepts, and even paradigms in a multi-storied configuration analogous
to the Nyambene agroforests created over the course of generations by Meru farmers. Not
to do so is an inefficient use of the social landscape from which these data have emerged.
At its most flexible best, only a loose fit can be achieved between the conceptual order and
most local case studies. There is no need to destroy theoretical models, only to follow up
on the full system-wide implications of material causation and the feedback it engenders.
Through the 1970s the monocultural political orientations of African states divided the
Organization of African Unity into two conflicting radical (i.e., Marxist) and conservative
(i.e. neo-classical) camps. No clear winners emerged among the states representing
different positions on the ideological spectrum (Young 1980). As economic reality set in,
the debates between conservatives and radicals became largely an epi-phenomenon. The
African beneficiaries of the theory-makers persevered within their day-to-day theoretical
models: God is rain...and if it doesn't fall eventually many souls proceed on to the
timeless world inhabited by the ancestors.
The consistent pattern formed by Africa's fundamental post-independence problems
contrasts with the trendiness of theoretical discourse during the same period. The relative
lack of conceptual progress of the recent past recalls Africa's "lost decade" of the 1980s.
The situation is succinctly stated in the final footnote of a refreshing critique of the agents
of development in the countryside by a Zairois anthropologist: "I have tried to interest my
colleagues back at the university in these issues but they are too busy debating esoteric
points of Marxism (Vwanawakazi 1989)."
According to this view from the field, what is needed is not another paradigm, green
revolution, or miracle tree, but greater effort to create more synergistic and symbiotic
theories by linking existing conceptual insights and data with greater sophistication. The
building blocks are there. Until we act to integrate our present knowledge, we are not, to
use John Lonsdale's phrase, "intellectual nomads in search of a paradigm (1983)," but
rather, to extrapolate from the oft-told Oriental parable, blind men in search of an elephant.

7For example, in a research tradition that can assert "...theory is timeless and proceeds in a sort of eternal
present (Kitching 1980,6)." 'theory' becomes the equivalent of some secular God.

Defining Development:

Defining development, like positing theory, is problematic. Consistent with the national
ideology, any research in Kenya must have some developmental implications. Few
definitions of development are as elegant as the economic hypothesis based on the shift of
population employed in agriculture to other sectors of the economy: as other sectors of the
national economy develop, the importance of agriculture declines (Johnston and Kilby
1975). Generally an accurate reflection of diversification, specialization, and growth of
income, this indicator is by definition limited to economic variables. Other developmental
definitions address institutions, political systems, health, gender, and a host of other
categories. The intuitive dissatisfaction with simple characterizations of Lamu and Meru as
lacking "development" speaks not to differences in this class of formal criteria, but some
intangible, missing quality.
In Kenya, and the rest of Africa, human capital has long been a critical element in the
developmental puzzle. The concept of human capital has quantitative and qualitative
dimensions. In the small scale societies of precolonial times population growth coupled
with the group's ability to provide for their health and welfare, was the primary
developmental goal. In Africa fertility is everywhere celebrated in art, song, and customs;
the poorest and most forlorn figure among the African poor is the barren woman (Illife
1987). Population, (i.e., labor) under most African conditions, was the variable limiting
the utilization of available resources and opportunities.
Normal environmental conditions meant a steady attrition of population, and the more
extreme fluctuations of climate or disease vectors could cause precipitous population
decline. The principle of safety in numbers was an equally important element of group
defense from the incursions of traditional enemies and marauding groups displaced by
environmental calamities. For these reasons, development, from the traditional African
perspective was a function of the population growth necessary to generate greater wealth,
improve group capacity to withstand extreme environmental vagaries, resist raiding by rival
groups, and to enrich the social environment that life revolved around.
The African concept of development was founded on people, and the notion of fertility
and population growth gave rise to cultural institutions based upon the safety and welfare
of the group. The material and unseen world were connected and interpenetrating as surely
as one term for God (Ngai) shared by a number of Kenyan ethnic groups is also a term for

rainfall.8 God was the great donor Africans appealed to during precolonial times. Igembe
communal prayers embraced their traditional concept of development:

Leader: Thai, Thai Murungu, Peace, Peace, God,
Congregation: Thai! Peace!

Leader: Utue irio, Grant us food,
Congregation: Thai! Peace!

Leader Utue ana babeingi Grant us many children,
Congregation: Thai! Peace!

Leader: Utue nthaka iri na inya. Give us strong warriors,
Congregation: Thai! Peace!

Leader: Utue ng'ombe, ng'ondu, na mburi. Give us cattle, sheep, and goats,
Congregation: Thai! Peace!

Leader Utue belelie nchai Grant us protection from evil,
Congregation: Thai! Peace!

Leader Ndakwelenka kinya ni mburam, I ask that I may also be blessed with them,
Congregation: Thai! Peace!

Source: Chege 1979.

The Meru concept of development expressed here is anchored in the active pursuit of
material progress within defined moral limits not unlike the Weberian Protestant ethic. To
be wealthy, i.e. rich in food, cattle, and children was desirable, but the Meru recognized
that material goals required blessings and protection from evil as well as raids. Excessive
individual pursuit of material gains at the expense of other individuals or the group could
provoke social and supernatural sanctions. Egalitarian ideology was not so much about
restricting individual differentials in wealth and power as it was about insuring the
individual's responsibility and commitment to the group. The Meru developed social,
political, and religious institutions to govern the developmental process: the Njuri Ncheke
council, spiritual-ritual leadership through the office of the Mugwe, age-set organization,
and gichiaro fictive kinship.
The influence of indigenous cultural orientations invokes questions related to the
ideological dimension of development. For two decades Kenya's population growth rate
climbed to record levels despite the sums of money pumped into Western population
control strategies. In Kenya local support for population growth is often expressed in

8This definition does not equate with the pantheism of natural forces: the Meru have several words for God
(e.g. Murungu, Ngai, Barikiba), and the body of Meru ritual and religious expression is unambiguously
based on the fundamental unity of God.

terms of the defense of cultural values; letters to the editor resisting population control as a
Western conspiracy or assault on African values appeared with great regularity in the local
press during the 1970s and 1980s. Historically, Africans controlled population when
necessary. Under certain conditions the demographic equation necessitated biocultural
adaptations for the control of population growth. The Rendille control population growth
within environmental limits through social customs governing marriage and reproduction
that are closely linked to the reproductive cycle of camels (Spencer 1973). The Chuka
developed complex and effective birth control practices when pressed by enemies on all
sides during the nineteenth century (Mwaniki 1984).
The examples of the Rendille and the Chuka underscore the view of population growth
as a process based upon rational, cost-benefit calculations taking into account the
environmental carrying capacity facing local societies (Harris 1987). But other scholars of
demography with long-term experience in Africa see the need to explore the local
institutional facets and "cultural context "of African population demography as well as
unique characteristics of demographic transition in different societies (Caldwell and
Caldwell 1987; Greelaugh 1990). Support for family planning is gaining new acceptance
in Kenya and Meru, but factors such as the AIDS epidemic may yet prove the validity of
Africa's cultural and ideological commitment to population growth. Although HIV
mortality has yet to approach the level exacted by past epidemics and environmental
catastrophes reckoned by still living Igembe memories, it strikes at the economically most
important segment of the population (Caldwell, Caldwell, and Quigan 1989).
Attitudes towards development reflect asymmetrical relationships with variables of
technology, environmental and economic carrying capacity, and structural factors
determining access to resources. The reasons for this are obvious--the West and the
Westernized have long provided the dominant models of material advancement, the critical
patronage linkages, and westernized education is a prerequisite for utilizing donors
institutional and technological transfers to developing countries. But the association
between technological advance and Westernization obscures progressive home-grown
technologies and cultural practices, and promotes cultural duality.
Westernization often enslaves at the same time as it is liberating. The conflict between
material development and social values as a regular feature of day to day life in Africa has
diminished little since Ocol P'Bitek (1967) described the quandary in The Song of Lawino
and The Song of Ocol. The success of Africa's westernized elite fails as a role model for
African society: Kenya's Anglophylic former attorney general, Charles Njonjo, provides
an example of the deep-seated discomfort which permeated Kenyans' view of his political
ambitions, a factor President Moi exploited when he engineered Njonjo's fall from grace.

The ideology of Nyerere's African Socialism, on the other hand, attempted to reconcile
techno-economic development with the social equity idealized in traditional cultural values.
Blending the two cultural traditions proved problematic, and when the Ujamaa experiment
faltered, indigenous ideological and organizational variables embedded in the Tanzanian
social environment reasserted themselves in its place (Hyden 1980).
Mazrui (1984) poses the question of whether or not technological modernization is
possible without cultural and ideological Westernization. His metaphor is that the climate
for technological development in Africa is favorable, but the sociological soil is barren.
The essay, more an analysis of the ideological arguments of the 1970s than the practical
problems accompanying Africa into the next century, fails to illuminate the relationship
between culture and science beyond stating that there is one ibidd; 299). Indigenous
African capitalism prevails in places where capitalism organized on the Western model is
faltering (McGaffey 1991). Formal economies are collapsing while informal economies
thrive. Scientific socialism failed in Somalia, but Somali socialism provides support
networks for refugees in Nairobi's Eastleigh, London, and other Western cities. The
question is nevertheless relevant, and one that we will explore in greater detail in the course
of analyzing data comprising this study.
The ground-level view of African development shares more of what Hirschmann (1984)
described metaphorically as "sailing against the wind" than the on-top-of-the-mountain-
looking-down concepts of Marxist or neoclassical stages of economic development.
According to Hirshmann, development and change describe a zig-zag course where gains
in one sector are often offset by a degree of slippage in another sector. For example,
education, urbanization, and expansion of the non-agricultural sectors, including the civil
service, contributed to state policy biases favoring urban areas while eroding the policy
support for productivity in the rural sector (Bates 1983). Structural adjustment programs
rectifying the imbalance destabilizes urban areas, creating IMF riots. Boosting producer
commodity prices increases production, and increased production lowers the world market
prices for the same commodities. Technological advances in export crop production ends
up subsidizing the Western consumer (Lipton 1988).
The state trims its sails and struggles to stay on course at the same time the tide of the
international economy pushes the vessel backwards. The journey involves not only sailing
against the wind but crossing largely uncharted waters with unpredictable political and
environmental currents, submerged ethnic reefs, and economic sandbars. It is no wonder
that African governance has been described as more a matter of good seamanship than
navigation (Jackson and Rosberg 1983).

Are all African leaders, and the sample is statistically significant, poor mariners? Or do
they share the existential predicament depicted in the common bar-art portrait of the African
woodcutter: treed by a lion, staring into the eyes of the attacking viper coiled on the branch
he's cutting, and dangling over the gaping jaws of a crocodile. The ubiquitous pictures of
this "no escape" version of Africa's triple heritage (lion, snake, crocodile) underscores how
Africans see their quandary. In the circumstances of heads of state, the response has
typically been simply to hang on as long as possible. Many of Africa's liberators--Jomo
Kenyatta, Sekou Toure, Kwame Nkrumah, Siaka Stevens, Kenneth Kaunda-- ended up
leaving the stage with tainted reputations or in disgrace. The truly heroic African leaders
are those who die young.
The realization of a society's developmental goals is like other evolutionary processes:
based on trial and error, subject to unanticipated forces, and producing unpredictable
outcomes. Adaptation need not be optimization (Richerson and Boyd 1992, 75). The
Meru theory of development is the starting point for investigating system dynamics. The
Igembe cultural-agricultural system reproduces by producing food, livestock, and strong
warriors. Behavior is driven by material goals and guided by cultural principles that
generate both blessings and evil within society. Culture matters because it embodies a
social systems memory, and institutionalizes the society's social learning.
We can learn from social learning. Some attempt should be made in the name of
objectivity to recognize the empirically determined borders of professional knowledge.
Because in this situation we are frequently like the blind men describing an elephant, a
priori approaches to social systems is a cautionary exercise. Scientists, social or other, are
all potentially the sorcerer's apprentice. Humility, in this respect, comes from recognizing
the sum record of African developmental interventions from colonialism to the crisis-ridden
present. Bridging the chasm separating theoretical anthropological analysis and practical
application involves integrating our understanding of how the environment and technical
details of production interact with the other components of cultural systems.

The Cultural Endowment Problem

For the anthropologist, economist, and agricultural scientist alike, symbiotic theories
should be primarily a matter of extending the empirical scope of field research and
theoretical linkages among respective fields. Interdisciplinary syntheses are problems in
their own right.

The development economist9, Vernon Ruttan, and his collaborator in the development
of the innovative model of economic development, Yujiro Hayami, claim development
economists and others are forced to deal with culture at an intuitive level, rather than in
analytical terms. Ruttan advocates going beyond conventional, microeconomic models to
understand technical and institutional change, but finds the real understanding of cultural
endowments in the study of economics inadequate for the scholars and practitioners of
development who share the conviction that 'culture matters'. In the 1988 article, which
takes a problem-solving view of the importance of culture in economic change, Ruttan
looks to the specialists in culture for filling the gaps in a generalized "pattern" model for
induced innovation which places "cultural endowments" in an interrelated category
alongside resource endowments, technology, and institutions. This led Ruttan to
investigate anthropology for contributions that can shed light on the relationship between
cultural endowments, technical, and institutional change.
Ruttan claims that the economist who invests time in searching the ethnographic
literature encounters two major problems: interpretations of culture which usually neglect
the historical, political, and economic context because of the commitment to learning
through fieldwork, and the problem of "many anthropologies" resulting from the
intellectual fragmentation within the discipline. Ruttan reduces these anthropological
differences to the materialist-interpretive division. Ruttan notes that the cultural materialist
approach is congenial to the economistic approach and useful for macroeconomic analysis
seeking to understand the impact of resource endowment and technology on differences in
institutional performance and change. But the overlap and similarity with economics also
restricts its utility because it does not substantially address the cultural endowment
problem, and institutions are analyzed as derivatives of the infrastructure. On the other
hand, much of the interpretive school's interpretational analysis strikes Ruttan as
"excessively personal and idiosyncratic," and he says cognitive and symbolic anthropology
must be analyzed in relation to social organization and institutional change.
The detailed attention given to ideology and processes in the work of Clifford Geertz, in
contrast, earns special praise from Ruttan for filling a gap not addressed by other
disciplines: "interpretive anthropology places cultural differences and change at the center
of its agenda, even if it presently tends towards the idealistic and romantic, it offers the
most potential to make a practical contribution to development economics over the long run
ibidd. 255)." Ruttan's article, by defining a general problem, albeit primarily in terms of

9Development economics came into existence as an applied sub-discipline of economics. Hayami and
Ruttan (1985) apply the model discussed here to international development.

his own model, represents a narrow interpretation of anthropology's relevance to
developmental issues.10 Ruttan nevertheless identifies issues of practical, methodological,
and theoretical importance. He concludes that economists will continue using modest
extensions of microeconomic theory as the strategy to deal with culture "until our
colleagues in the other social sciences are able to provide us with more helpful analytic
tools ibidd; 256)."
That insightful cultural exegesis is possible is evidenced by the fact that it is often done
by observers other than professional anthropologists. 11 An example of the interpretive
anthropological approach is Geertz's (1972) famous article, "Notes On The Balinese
Cockfight," where the author combines the curiosity invoked by an unanticipated personal
experience in the field and thorough scholarship into the contextual setting to tease the
meaning of the cockfight as an important cultural institution in Bali. The cockfight
symbolizes the potential for social conflict in a context where political violence otherwise
appears at variance with its cultural setting. The problem is positing a systematic
methodology that can be used to apply this approach across cases.
Symbolic systems link up with their material settings, and the differences between form
and style, function and symbol are not as great as some would have us think (Richerson,
Bettinger, and Boyd 1993). If it is true, as Geertz asserts, that we cannot understand a
culture the same way a native does (1983), it is only logical to make reference to the
standard social science grid lines--economy, ecology, history, etc.--when mapping
culture.12 Either we do this, or concede the study of cultural domains to the post-
modernists, who eschew these grids in favor of their own distancing rhetoric and guilt
complexes.13 How we go about this is treated in detail in the next chapter.

10Critiques of the Haymai-Ruttan model include the fact that the specific social order determining
institutional arrangements in developing countries cannot be understood in purely economic terms.
1 11t is difficult, for example, to find fault with Pico Iyer's (1989) treatment of Japanese baseball, an essay
that uses a sport to delineate institutionalized cultural differences between Japan and America in an
unpretentious and theoretically unobjectionable (from most anthropological perspectives) manner. Also, I
have observed that "native" analysis of sociocultural behavior gravitates towards materialist determinants,
but without erecting a materialist/cultural dichotomy.
12The accuracy of interpretive analyses is contingent upon the researchers' individual skills, their ability to
identify the important features of a cultural reality and to extract meaning from it, not analytical tools.
Indeed, Harris's major criticism of phenomenology or "Geertzian belle lettres" is that it contains no
methodology apart from the researcher's individual insight, placing it beyond established scientific
principles of accountability (1987). It is curious that, although Geertz believes we (i.e. Westerners) cannot
cognitively understand them, they (i.e. non-Westerners) understand us well enough to fit into and contribute
to modern multi-cultural society in the West.
13 Clifford (1986; 1990) is a principle spokesman for the new ethnography that seeks to locate the
ethnographer as a writer of cultural texts. Marcus and Fisher (1986) are the chief critics of the "sins" of
Western anthropology. Roth (1989), Carrithers (1990) and Keesing (1990) critique the postmodern school
from different stances.


Uncertainty in the political, economic, and sociological spheres arises not so much as
a result of data and measurement problems, but because of an extraordinary
atomization of theories of agrarian change, and a failure to link them to the
environmental social dialectic... (Blaikie 1989, 19).

Many African countries achieved political independence during the 1960s with higher
per capital incomes than the poor, predominantly agrarian, former colonies in Asia,
countries which three decades later are newly industrialized economies with some of the
highest growth rates in the world. In comparison, many African nations experienced
economic progress during the early years after independence only to slip back into an
economic and political malaise. The consequences for most of the population are material
and political conditions worse than those of the colonial period.
Africa remains the only region of the world where the sustained efforts of individuals
nations and the international community appear inconsequential in the face of contrary
environmental forces, fractious polities engendered by the continent's social diversity, and
the vagaries of world economy. The widespread and widely noted turbulence manifesting
throughout Africa presents the backdrop to this case study of agricultural commercialization
in rural Kenya. Field research on the Nyambene Meru cultural-agricultural system
provides the departure point for exploring African evolutionary processes, the systemic
dynamics underpinning social change and group adaptation, and the parameters of
socioeconomic transition in Africa. The following discussion refers to the context
accompanying the spread of small-scale commercial agriculture across ecological zones in
Meru, and the practical developmental issues it subsumes.

The Empirical Problem: Population, Agricultural Systems, and Polity

Kenya is an agricultural country. After one century of precipitous population growth
and enormous social change, over eight out of ten Kenyans still live in the countryside.
Few individuals are more than one generation removed from the family farm, and rural
values still exert a strong influence on modem society. The country's deepening economic
and social crisis is often blamed on demographic factors: the population nearly doubled

over the past fifteen years reaching 24 million in 1992; growth has slowed from a high of
4.1% to a rate of 3.8% a year, the highest demographic growth rate in known history.
This demographic-environmental equation presents a formidable challenge to the economic
status quo. Competition over land follows from the fact that eighty per cent of the
population live on less than twenty per cent of Kenya's land mass. Another 20% of
marginal agricultural land overlap with the arid and semi-arid regions that occupy over 60%
of the country.
Intensified production in high potential zones and the development of arid and semi-arid
areas (ASAL) is not only necessary to feed the projected population of 34 million Kenyans
by the year 2000, but is also expected to remain the primary source of their employment
and the main source of the export earnings necessary to spur multi-sectoral growth
(Government of Kenya 1989; World Bank 1990b). This general prescription applies to the
agricultural sectors of other African countries, according to the World Bank, where
increased food production is necessary to raise per capital nutrition as well as reverse the
trend of natural resource degradation (World Bank 1989, 89).
Liberal commodity price policies, a supportive state presence in the agricultural sector,
and the capitalistic private sector have produced a rate of economic growth sufficient to
keep pace with population growth into the early 1980s. Critics of Kenya's small holder
agricultural policies (e.g. Hunt 1984; Peterson 1986) were silenced by evidence that small
farm production for markets improves regional equity (Gyimah-Brempong 1988) and
bolsters personal incomes while enhancing household nutritional levels (Young 1988;
Heald 1991).
By the middle of the same decade, however, the effects of population growth, falling
prices for Kenya's export crops (especially coffee), and a bloated and inefficient public
sector began to erode living standards and intensify the competition for scarce resources
ranging from land to social services. Internal and external pressure on the political system
increased at the same time. Political mobilization highlights the mismanagement of public
resources, blatant corruption, and manipulation of the political process by state actors.
Many rural areas suffer the general decline of state institutions, while the private sector
forces, which are expected to take over many state functions according to the logic of
structural adjustment programs, remain largely inert. Social change has left a gap in many
rural areas: many traditional institutions of rural society are vitiated, but new ones have yet
to fully congeal in their place. Consequently, societal response to agrarian crisis such as
drought and famine is limited to individual and household-level coping strategies (Stini

1987, Corbett 1988; Cossins and Upton 1988b).1 Fortunately, in Kenya the state has been
able to act effectively in he face of such crisis, but its ability to engineer solutions to other,
more fundamental problems is questionable due to the erosion of its organization capacity.

The Role of the State

The paramount role of the state as the key to economic progress in many areas of Africa
grew out of the sheer dominance of the African colonial state in relation to other local social
structures. After independence, it was assumed that development would take-off once the
state was redirected towards advancing the cause of indigenous society. Modernization and
radical political economy critiques alike endorsed development policies based on strong
centralized state structures. These ideas proved to be seriously flawed.
Post independence state-society discontinuities disclosed the endemic pathologies of
state structures in Africa. Local responses ranged from active opposition to withdrawal on
the grass roots level back into the subsistence economy (Rothchild and Chazan 1988;
Hyden 1980). The international community turned to private sector policies while the
dysfunctional African governments continued to gamer most of the blame for the
continent's condition. One of Africa's most eminent historians even went so far as to label
the state the 'Black Man's Burden' (Davidson 1992). On the other hand, the current
situation in Somalia indicates that no state at all is even worse.
The state is a distinct form of social organization that grows out of a socioeconomic
system's internal dynamics, or is precipitated by the presence of other states in the region.
Although other organizational structures perform the function of the state in local socieites,
the emergence of the state as an organization is never partial: it is either there or it is not,
and more than other forms of political organization, the state is correlated with
multiculturalism and uni-ethnicity (Cohen 1993).
Contemporary African states came into existence as the result of the expansionary
behaviors of European states. This ancestry produced many problems. Ostensibly, the
state promotes the interests of its constituent societies, but this has often not been the case
in Africa the externally imposed state has failed to develop a viable relationship with civil
society. Many problems on the state level are manifestations of forces incubated within
ethnic groups and localized interests. Although the state may both foster developmental
processes in some instances, and may sabotage them in others, these contradictory

1The Igembe have been able to muster coherent social responses to problems arising in their area. The
institutional aspect of local society receives detailed treatment in chapter 3.

processes nevertheless proceed in a manner that makes their study a matter of empirical
The thrust of Western strategies has been large-scale and capital-technology dominant
strategies whereas the African environment historically favors local cultural adaptation,
niche specialization, and degrees of cultural pluralism more than large organizational
structures. The character and behavior of African states reveal the centrifugal strength of
these small-scale social formations, and these local forces embody the cultural dynamics
that will eventually define the salient features of African capitalism. Adaptive large scale
organization, the state being the primary example, can be expected to emerge out of
complex processes of change, innovation, and adaptation on the local and regional levels.
Thus agricultural change and the contribution of the state to it cannot be studied in isolation
from each other. In any event, the agricultural problem transcends solutions solely located
within state institutions.
Determinants of land and resource use are complex, and a number of factors impose
clear limits on agricultural progress in Kenya's productive highland areas. High potential
agricultural zones are now demographically saturated and easy policy and technology
options have been exhausted. Intensified productivity of highland agriculture can at best
incrementally increase crop yields. Over half of the country is an arid expanse supporting
nomadic pastoralism and wildlife reserves. The remaining twenty per cent is semi-arid land
of low agricultural potential, much of which forms an uneven band of traditional grazing
lands, interspersed with swidden plots of dryland grains, on the margins of the high
potential zones. This land occupies an important swing position as Kenya negotiates the
uncertain territory between a Malthusian future and economic transformation.
Colonial governments and external donor agencies have for decades pursued
technological solutions for the agrarian problem consistent with the view that, by Western
standards, traditional African agricultural production is a technologically primitive, pre-
scientific, and embedded in culturally conservative peasant social formations. The project
approach in the relatively non-commercialized areas of years past has produced a
succession of outright failures, false starts, poorly coordinated donor interventions, and ill-
conceived policies, beginning with the infamous Tanzanian groundnut scheme
(Hogendorm and Scott 1981), repeated during the Sahel crisis of the 1970s, and continuing

2Under capable managers the Kenyan government achieved laudable success promoting production and
social welfare in sectors such as small holder tea and milk production (Leonard 1991). Local research
institutes have made large strides in developing improved cultivars for coffee and maize, and improved
livestock breeds.

to the present.3 Externally-conceived panaceas usually involved the creation of new
institutional structures to oversee implementation.4 The sum record of these interventions,
discounting fiscal cost-benefit analysis, is one of mixed success aggravated by
unanticipated administrative, social, political, and environmental problems (Moris 1981;
Batisse 1988; Lusigi 1988).
Commercialization and privatization now replace to a large degree the state and donor-
assisted projects of the past as the primary force of change in the agricultural sector at this
stage of Kenya's economic transition. Donor and International Financial Institutions
structural adjustment programs are now the guiding policy blueprint for empowering
market and private sector forces. They also set off serious social problems, with
unexpected internal consequences across Africa (Howard and Nafizer 1991, Fearon 1988,
Hopkins 1986). Real income has declined, and cuts in government expenditures have
effected education, medical facilities, social, services and agricultural extension much more
than those inefficient sectors of the state most responsible for the forced acceptance of the
structural adjustment policies in the first place.
In Asia, many other preconditions for economic transition were already in place and
waiting for a catalyst to spur multi-sectoral growth. African socioeconomic conditions are
obviously different, and at best uneven. During the period of large-scale agricultural
expansion, the share of Gross Domestic Product for agriculture, an important economic
indicator of sectoral development (Johnston and Kilby 1975), decreased from 35 per cent
in 1965 to 31 per cent in 1988. In terms of the sectoral development model, African
industrialization is still-born (Mylteka 1989). Although this generalization only partly
applies to Kenya, prospects for industrial growth are constrained by a number of internal
and external factors (Godfrey 1987). High technology drives the world economy, and the
dependence of Third World economies on primary commodities reflects conditions of
growth that discourage significant investment in commercial crop systems since the break-
up of colonial semi-monopolies (Dow 1989, 34).5 Export performance reflects the cycle of

3 Horowitz (1979) critiques Sahelian pastoral and livestock projects. Hogg (1984, 1987) reviews the
problems in small-scale ASAL water and agricultural projects in northern Kenya. Wiggins (1985)
examines different donor approaches, including the British EMI project in Meru, and Lusigi (1987)
discusses the problems of UNESCO's Integrated Project for Arid Lands, also in northern Kenya. Franke and
that Chasin 1980) provide a general critique showing the link between famine and Western development
strategies all of these case studies represent.
4The new field of development administration became the organizational adjunct of disciplines addressing
technical and policy-based innovations for developing countries. From the perspective of administrative
science, however, the whole idea is oxymoronic since administrations mainly seek to promote their own
survival (Shaffer 1969).
5While crop production and primary commodities no longer figure highly in Western capital markets, local
investment in agricultural enterprises by agribusiness and a variety of national and international research
institutes qualify this observation. Three agricultural corporations ranked in the top ten companies quoted

decreasing relative value for primary commodities and stagnant manufacturing (Svedberg
Kenya has passed the point of no return in the transition from a collection of
subsistence-oriented communities to a more integrated and economically diversified
society. If the process of multi-sectoral economic growth and diversification falters,
economic stagnation, shortages and maldistribution of essential commodities, and conflict
over the country's limited resources will rapidly undermine past progress. The agricultural
sector's failure to maintain its past momentum relative to population growth heightens the
danger of system-wide entropy overtaking the historical trend of growth and
The pace and structure of demographic growth is a critical parameter. From an
estimated population below two million at the beginning of this century, Kenya's
population approached 10 million at independence in 1963 and passed the 22 million mark
in 1988. Fifty per cent of the population is 15 years of age or less. Despite the early
warning signs of population to resource imbalance, from 1965 to 1988 the fertility rate for
women of child-bearing age has dipped only slightly, from an average of eight to seven
children: some forty per cent of the population are women in the childbearing category.
Between 1979 and 1989, the 3.3 per cent growth rate for agriculture, the engine of the
national economy, lagged behind population growth that peaked at 4.1 % according to
figures quoted in the Economist Intelligence Unit for Kenya (1989). The reduced, decade-
end population growth rate is estimated as 3.8 per cent--a net per capital production deficit
of 3 per cent for the decade.
Although Kenya's agriculture is considered to be a policy model for other African
countries (Lofchie 1988), and net per capital productivity has increased, Kenyan farmers
face formidable constraints influencing agriculture's ability to satisfy the requirements listed
above. Even after the considerable progress of the past several decades, national food
security is tenuous (in December of 1991 the Government of Kenya announced an
impending shortfall in the country's staple food, maize, despite adequate rains). Food
consumed on-farm partially explains the gap between production and consumption (Cohen
1988; Bates 1990). But more and more Kenyans are leaving the subsistence sector. The
rate of urbanization is above eight per cent. Twenty per cent of the population now live in

at the Nairobi Stock Exchange based on post-tax returns to capital: Brook Bond (#4), British American
Tobacco (#8), and Sasini Tea and Coffee (#9). Standard Chartered Bank Kenya, which has sizeable
agricultural investments ranked #2. and Kenya Breweries, which directly relies on the agricultural sector for
its raw materials ranked #10 (source: The Economic Review Oct. 4-10, 1993: 13-15).

urban areas compared to nine per cent in 1965; 57 per cent of these urbanites reside in
Nairobi (World Bank 1990).
The experience of other world regions offers generalized insights into socioeconomic
transition but no ready answers for Africa's problems, which are themselves partially the
consequence of the post 1500 AD capitalist transition in the West. This is why theoretical
formulations of capitalism have to be redefined relative to Africa's unique historical
circumstances (Callaghy 1988).6 Understanding social transition to capitalism in Africa is
a theoretical challenge due to the sheer diversity of associated phenomena and historical
dynamics specific to time and place that it subsumes. A number of reasons, including the
fact that agrarian capitalism in Kenya is more advanced than other African countries,
designates Kenya as the unofficial test case for agriculturally supported economic transition
in sub-Saharan Africa (Illife 1983, 41-42). The incorporation of ASAL regions, including
the periphery of the Meru highlands featured here, is a test of the test case.
The adaptive qualities of indigenous production systems are synergistic indicators of
adaptive agencies already observable within the transformation of traditional societies. It
follows that African capitalism, insofar as capitalism posits the most economically adaptive
path to development historically, will evolve apace with other sociocultural adaptations
including the all important variables of national and local political organization,
environmental management, and cultural guidance. The Nyambene production system
under consideration here yields substantial data of this order.
The Nyambene case study is especially interesting because, as discussed in Chapter
One, socioeconomic change in the Nyambenes has followed a pathway of indigenous
capitalism that appears as unique as it is successful in comparison to other similar regions
in Kenya. The Igembe system, in regard to specific technological and social components,
is adapting in the direction of the relatively Westernized macro economic environment. It
is exceedingly important to note, however, that on the macro level (in this instance I am
referring to the eastern Africa region) the regional political economy also appears to be
moving in the direction of a Nyambene-type economic and social order as reflected in the
growing economic power of informal economies and indigenous sectors of production
relative to the state and economic sectors dominated by foreign capital, the patterns of
disengagement from the state, and the largely invisible internal order created by the myriads
of inter-connected local ethnic systems and economic organizations embedded within the

6Callaghy addressed Marxist and Weberian theoretical approaches, but the situation where capitalism is both
part of the problem and the generic solution suggests that neoclassical economic theories and regional
variations on capitalism itself require further qualification within the African context.

regional system (Hyden 1983; Chazan 1988; Azarya 1988).7 The ongoing exercise in
privatization and creating what the World Bank labels an "enabling environment" for
economic development favors the growing influence of local, societal phenomena on the
order of the data on Nyambene society and economy presented here.
This study attempts to extract some of the cultural and agricultural dynamics of the
system. I do this by placing the field data collected through surveys, interviews, and
observation in the context of historical patterns linking highland ecological zones and
lowlands. The contemporary forces of commercialization intersect the cultural management
of agricultural resources to define the local developmental trajectory. The five main factors
listed below in table 2.1 have selected for this unique coevolutionary trajectory of
Nyambene commercialization.

Table 2.1: Factors Selecting For Coevolutionary Development in Northern Meru

Environmental Variation
Econiche Specialization
Internal Organization
Markets and Exchange
The Domain of the State and National Economy

The way these factors have operated in this case form the basic structure of the data
presentation which follows. They can be summarized briefly in theoretical terms
Environmental variation conditioned production strategies based on econiche specialization:
hunting and foraging, pastoralism, and agriculture. Local societies, which together form a
system of relations have developed adaptive strategies based on combinations of different
production complexes. The internal organization of these systems represent complex sets
of institutions and cultural orientations supporting specific group production strategies. In
Meru, environmental variation conditioned specialization and exchange. Adaptation and
change are ongoing processes driven to a large degree by markets and other forms of
exchange that extend across the sociocultural context to factors such as cultural
borrowings, intermarraige, and the incorporation of isolated clans or members of ethnic

7Much regional economic activity is only "invisible" insofar as it is not reflected in official statistics and
involves by-passing state control in the form of licensing, marketing boards, and customs duties.

groups into other ethnically defined systems. The state represents a domain of variables
that, in the case of many areas of East Africa, appeared within the regional environment in a
sudden and dramatic manner. During the major portion of this century the colonial state
and later, its indigenized, post-independence extension exerted a direct economic influence
on a number of Kenya's local societies in Kenya, and only an indirect influence on others.
The effect of this contact in Meru has been to allow for the autonomous and therefore
coevolutionary development of this particular region in relation to other areas more directly
affected by government policy and its implementation.

Historical Patterning

The pattern of development in Meru represents the culmination of long term historical
processes set in motion by the primarily environmental variables listed in Table 2.2. These
regionally specific factors patterned the subsequent adaptations and impart directionality to
each local system's developmental pathway, which have become in term subsumed within
the Igembe cultural and agricultural system.

Table 2.2: Africa's Environmental Parameters8

Large land mass spatially separated from other world land masses.
Environmental conditions encouraging high but cyclical levels of biological activity.
Low population relative to land mass.
Physical and spatial barriers to regional communication.
Old soils of limited fertility.
Unpredictable long and short cycles of humidity due to ITCZ fluctuations.
Emphasis on cultural adaptation over forms of technological innovation.

Africa assumed a slow and decentralized developmental profile after the desiccation of
the Sahara sealed most of the continent off from other world regions 20,000 years ago.
Holding other factors constant, the low population relative to Africa's large land mass alone
would explain a slower developmental trajectory in comparison to the quickening of
developmental processes in Europe and many parts of Asia. Physical barriers, ecological
constraints, erratic climactic conditions and periodic environmental disasters also
contributed to uneven cycles of growth in African production and population. Africa's

8 Sources for this list include Curtin 1984, Nicholson 1986, Jones 1987, and Oliver 1992.

population was 30 per cent of the world total in 10,000 B.C.; it declined to 10 per cent in
500 A.D., and slipped to 8 per cent in 1800 (Jones 1981, 155).
Africa's environmental conditions produced the following continental pattern.
Concentrated populations supporting levels of production and the state structure necessary
to surmount environmental barriers to exchange originally developed in mainly Sahel-type
environments, not the highland areas which are the main loci of production and change
today.9 Low population densities in the territorities outside state systems facilitated a de
facto exit option weakening the central control of local rulers. The exit option encouraged
migrations contributing to both the rise and decline of state systems during the last one
thousand years.
Large tracts of the continent fell outside the influence of forces supporting primary and
secondary state formation.10 In these areas development was a slow, gradual process
occurring over decades, centuries, and millennia. The ecological variation of eastern Africa
underpinned a range of environmental adaptations. The rise of Meru society subsumed
econiche specializations that were able to survive alongside the developing hybrid system
until recently.
Ethno-environmental specializations included hunting and gathering represented by
"Aathi" groups like the Agumba and Ndorobo, settled agro-pastoralism supplemented by
small scale irrigation in the settlements of the Mwoko the nomadic pastoralism of the
Maasai, Samburu, and Boran, and the predominantly agricultural specialization of the
Tumbiri and Thagana clans (Fadiman 1976; Mwaniki 1984; Sutton 1990; Lambert 1949).
These legacies resurface in the cultural-agricultural configuration of the Nyambene system.
Precolonial group interactions in Africa reveal the small scale and local nature of social
organization, economic strategies, and complementary specialization. Post-colonial
interactions among groups in the region continue in the vein of similar coeveolutionary
The dramatic socioeconomic changes of the past century have had little impact on this
aspect of indigenous social organization. Clans easily percolated through the territorial
boundaries created by the Pax Britannica, underscoring the mutable and plastic quality of
ethnicity in general (Kasfir 1979). The clan--a group defined by common descent--remains
a primary social unit conditioning access to resources where land is a critical factor of
production. Culturally homogenous clans cluster into tribes, but clan relations with other

9The Ethiopia highlands are an exception to this trend, but even in this case the rise of states usually
entailed exchange relations with the adjacent lowlands.
10Secondary state formation stems from external forces such as circumscription, or defense against other

groups can transcend tribe.11 The proclivity for small organizational units resurfaces in
economic organization.
The Igembe system of Meru described in the following pages represents symbiotic
outcomes generated by interactions and combinations of the forces listed above. The
Igembe have continued to straddle different socioeconomic domains in response to the
influences of factors introduced by state during the past eight decades. The concept of
symbiosis proved to be emblematic of general evolutionary forces influencing system
change and transition as well as a discrete mechanisms of adaptation underpinning the
growth and expansion of the Meru polity. As we shall see, ecozone variation produces a
powerful and interrelated set of forces affecting the development of the northern Meru
region. Fieldwork in Meru spanned the commercialized highland agro-ecological zones
and the drylands on their periphery, and provides us with an example of an instance where
the configuration of agricultural production, change and response to external factors is
generated by locally evolved agricultural technology and organization.
The productivity of the Nyambene Meru system rests upon indigenous technology,
research, and development sustained by internal cultural institutions that have proved to be
responsive to innovation, social learning, and market opportunities over time. The
reproduction of the internal order under commercial forces is, however, dependent upon
critical linkages to the external capitalist order. The nature of this relationship sheds light
on the dynamics of agrarian-led socioeconomic transition in Africa. Like in the case study,
the emergence of a unitary state-society system within Kenya's borders also embodies
linkages among groups and regions leading to an uni-ethnic outcome through processes
similar to those operating in Meru historically. Culture is a critical part of the overall
picture. I will, therefore, attempt to address the theoretical context of how we study and
integrate cultural factors into the analysis of social systems before delineating the specific
conceptual components that guided the research.

Anthropological Research Traditions:

A question emerges immediately about how these significant elements should be related
to one another. Fieldwork demonstrated, and pointed to, their relevance. Anthropological
theory points to ways in which they can be understood and explained. Absorbing the

1 There are cases of one clan spread across two tribes--the Atangwa clan is represented within both the
Kamba and Gikuyu peoples--and of clans that alternate between two distinct cultural units according to
material conditions, as the alternating identity of the Ariaal clan, which is associated with both the Rendille
and Samburu, and as of late, the Turkana, according to changing contexts (Spencer 1973; Hjort 1981).

unparsimonious sequence of detail and complexity present in a small scale system creates
an objective dilemma: a system is properly analyzed only by breaking it down into its
component parts, but it is properly understood only in its entirety. Social scientific
modeling and specification of the relationships among causes and effects in society is
therefore inherently problematic, leading to Achen's observation that "a functionally
specific theory of a realistic social situation may be just barely conceivable in principle, but
it would be massively complex. And if it were achieved, no one would want it (1982,
Hirschmann observes that, in the distinctive intellectual climate of the social sciences,
gradual loss of certainty replaces the cumulative growth of knowledge, "as the slow
mapping out of the extent of our ignorance, which was previously hidden by an initial
certainty, parading as paradigm (1981, 59)." The view presented here is that the quest for
paradigms tends to divert social science disciplines away from their pragmatic orientations,
undermines their societal relevance, and by encouraging theoretical inbreeding, produces an
intellectual rigidity that in the long run sabotages conceptual progress as well as practical
application.12 For this reason, this study is based on a problem-solving orientation
consistent with Gellner's observation that fully refined formulations of initial concepts are
the product of multiple reformulations:

Validity in science is not established in accordance with some simple, orderly criterion, the Natural
Law of cognition so to speak; the constitutional law of science is not the logical product of a
single mind and one design. It is not like those buildings admired by Descartes, which were
designed by one architect, but more like those oriental palaces which grow by accretion. It
resembles the shreds and patches, the untidy amalgam of a growing, plural, ongoing culture, its
law is a customary law which emerges from one contingency and crisis to the next, and not from
some supreme constitutional edict issued by Nature (1974, 7).

From a problem solving perspective, there is little sense in allowing paradigmatic
boundaries imposed on the corpus of social science theory to limit one's use of potentially
useful theoretical concepts to those found within a single intellectual school: indeed,
"individual theories working in isolation generally solve no problem (Laudan 1977,
115).13 Problem-solving is the larger paradigm behind scientific theories. We can talk

12Parsimonious theories produce logical and often elegant models out of the inherently disorderly data of
social science. But Kuhn's (1967) dialectical interpretation of scientific revolutions as the outcome of
paradigm competition acted to distort social science theoretical discourse on the complexities of causality in
sociocultural systems. Competition among theoretical schools fosters the development of theory, but the
rivalry among paradigms, or paradigm-candidates, often assumed an all-or-none aspect in the social sciences.
13The Nyambene Meru farming system, which developed much in the style of such an oriental palace, has
much in common with this view of science. The Igembe have progressed as far as they have largely
through developing syntheses characteristic of what Laudan describes as the problem-solving scientific
model. The distinctiveness of the system reflects the synthesis of borrowings from other local systems.

about theoretical ideas in the social sciences associated with different research traditions
without placing these traditions in zero-sum relationships with each other.
Adam Kuper (1990), writing in his capacity as editor of a leading journal of
anthropological theory, identifies three distinct institutionalized research programs which
constitute the stable core of twentieth century anthropology. He describes the first as the
Boasian study of culture: a relativistic approach concerned with the particulars of culture
and cultural differences, focusing on description and explanation as opposed to
explanation. The second program derives from the Darwinian emphasis on materialistic,
evolutionary processes: models based on general principles related to the nomathetic
biological, ecological, and technological forces at work across different periods of human
history. The third program is the study of social structures identified with traditions of
Weberian and Marxian sociology. Kuper clearly intended to sidestep the non-productive
wrangle that results from characterizing these research traditions as paradigms.
The phenomenological school corresponds to the Boasian program: the culturology of
mental, ideational, cognitive, and otherwise cultural manifestations of social systems. The
best known example of this approach is the work of Clifford Geertz whose semiotic
approach is representative of other work in the tradition referred to as interpretive
anthropology. This school is essentially humanistic, and as such it is not at odds with
other approaches aspiring to scientific criteria (Lett 1988). Its practitioners identify with the
humanities, and often distance themselves from formal scientific formulae.
Cultural materialism as formulated by Marvin Harris (1979) exemplifies theories in what
Kuper labeled the natural science, Darwinian evolutionary research program. Cultural
materialism succeeded the ecological school of Julian Steward and Leslie White, and
incorporates Marxist influences, although it assigns theoretical priority to the
infrastructural--the demographic, technological, economic, and ecological--determinants of
systems development and collapse in place of the Marxist emphasis on structural causality.
Sociobiology, where infrastructural factors of an environmental and techno-economic
nature are secondary to the biological traits selected for in evolutionary successful
populations, also falls into the general evolutionary category.
Kuper's "third 'project," the sociological program, consists of theories emphasizing the
structural or social relations of production as determinants of social formations that range
from early British structural-functional anthropology to the Marxian tradition, including
neo-Marxist dependency formulations, Wallerstein's world system analysis, and French
structural Marxism with its emphasis on modes of production.
Diesing (1991, 139-140) also classifies the conceptual plurality of social sciences in
three categories defined by their respective goals. The tradition of logical empiricism

represents approaches aimed at deriving causal or statistical laws that enable us to control
something. Hermeneutics is deployed for improving communication and mutual
understanding, and the applied goal of joint action and living together. The third type of
science, typified by psychoanalysis, targets emancipatory self-knowledge of causes
determining consciousness and behavior. It uses both causal laws and interpretive
technique to probe hidden operations.
All of these objectives, relevant to our analysis in their own fashion, encourage a
synthetic theoretical approach consistent with the problem-solving orientation articulated by
Laudan.14 There is no universal principle demanding the separation of concepts and goals,
and as Laudan observes, scientists can often work within different, and even naturally
inconsistent research traditions (1977, 140). This approach equates well with another
stream in anthropology which chooses or combines elements from different paradigm-
oriented research traditions for their utility. Cohen defines such problem solving
approaches as eclecticism: epistemic creed whose foundation is a stubborn commitment to realism, relevance, and
testable validity. Eclecticism...places the onus on researchers to depict reality, its causal nexi, and
the changing weights given to specific factors in its most empirically valid, inclusive, and relevant
format. The criterion for choosing questions and seeking answers is therefore not commitment to
a paradigm, but rather policy testing and guidance, i.e. the relevance of the explanation in its
applicability to the solution of a real world problem (1988, 26).

The eclectic-empirical tradition emphasizes precision and rigor in place of the
paradigmatic criteria as a precondition for the exegesis of data.15 Social research under this
set of guidelines follows problems and demands accuracy and relevance. Clear, real world
conditions produce complex problems requiring choices about which data are most relevent
to puzzle solving that cannot be limited by paridigmatic limitations. Eclecticism is accused
of disorganized searching for any and all factors that prove interesting to an individual
researcher. But "eclecticism" need not be eclectic: it can also incorporate a systems
framework capable of embracing a variety of theoretical constructs.

14Choice of theory is rationally evaluated--normative rationality is admittedly time and culture-dependent--
according to its superior properties vis a vis the designated problem (1977; 130). "Rational choice" among
theories is therefore governed by the "progressive" capacity of theories in terms problem-solving, and being
scientifically reasonable is doing whatever we can to maximize the progress of scientific research traditions
ibidd.; 123).
15By-passing the paradigmatic conflict over ultimate causality reduces the Kuperian research programs to
two main approaches, the materialist tradition and the phenomenological tradition. The former is further
simplified by refraining from assumptions regarding dialectical processes at the core of the Marxist conflict
models which, as Ellen states, reduced to the ecological level can be expressed in terms of a predator-prey
relationship (1982, 246). Incorporating the latter research program, on the other hand, is viewed mainly as
a methodological problem which we mentioned in the first chapter.

Paradigmatic approaches to development and governance held sway in Africa during the
first quarter century of the post-independence period. Not only has the objective condition
of most Africans deteriorated in relative if not absolute terms during this period, but the
social sciences themselves are faced with a major problem of reproduction on the African
continent. 16

Analyzing Social Systems

The nature of the subject matter makes social sciences, unlike the material-based
sciences, wary of the Western positivism traditionally rooted in asserting control over
physical processes. We know more about the pathways social phenomena travel on than
their destinations: social variables combine and articulate in established patterns, but at any
given juncture any one variable or component of a social system can behave in a manner to
make the development of a system over time decidedly unpredictable. Social science, like
the life sciences which also labored under the scientific precedents of mechanistic, iron-clad
laws established in the physical sciences as they came of age (Mayr 1982), has to develop
theories appropriate to the phenomena it embraces in order to reach an understanding of
underlying principles.
This problem is a manifestation of a deeper dichotomy between the material and cultural
that runs through anthropology.17 Every theory has embedded design constraints
demanding trade-offs. The trade-off for elegance and parsimony in theoretical models is
the reduction of empirical reality to abstract constructs. The trade-off for contextual
sensitivity in phenomenological theories is reduced inter-connectiveness, testability, and a
proclivity for cultural relativity. Together these two issues present an essential quandary
for anthropology.
Materialist theory is comparative and evolutionary in nature. The basic analytical
divisions presented in Harris's Cultural Materialism illustrates these features of materialist
theory (1979). The cultural materialism research tradition explains system change through
the principle of infrastructural causality: assigning analytic priority to what Harris labels as
the demo-eco-econo-techno infrastructure. The evolution of cultural differences is shown

16progressive social science is self-reforming, seeking to extend our knowledge and understanding through
improved theories and methodologies rather than revision and reinterpretation of societal data within a
predetermined, paradigmatic mold.
17History proves that the borders between theory and myth, science and scientism are rife with binary and
ethnocentric we/they categories applied to simplistic or non-developmental frameworks (Goody 1977). The
assumed dichotomy then propels us to seek positive and/or negative data to support dualistic hypotheses in
terms of primitive and advanced rational and irrational, or savage and domestic as in the study of the
evolution of thought in general, or as in the case of Levi-Strauss in particular (ibid: 4-8).

to be the result of the operation of infrastructural variables. The separation of sociocultural
phenomena into domains facilitates interdisciplinary perspectives by organizing a full range
of system variables within analytical categories that can be applied to all other social science
research traditions.
Prioritizing infrastructural shifts strip structural relations of their causal primacy,
otherwise cultural materialism (i.e. Harris) and historical materialisms (i.e.Marxists) are
conceptually similar, especially in the case of some French structural Marxist models where
structure is merged with the infrastructure.18 Harris unequivocally broke with other
Marxisms by rejecting the dogma of Hegelian dialectics, replacing it with a formulation of
system reinforcing and system changing feedback loops connecting infrastructure and the
other levels of the system that largely replicates the functions subsumed by dialectical
processes.19 The mode of production school pioneered by French scholars working in
African societies established a holistic base-superstructure framework within the Marxist
research tradition, but failed to escape the basic assumptions of dialectical materialism.20
Some cognitive and interpretive social scientists on the other end of the theoretical
spectrum recoil at Harris's behaviorism and what they see as the reduction of all culture to a
dependent variable of material processes, and the low methodological priority given to
cultural expression in what Harris categorizes as emic mental phenomena, or informants'
reports on their own cognitive, cultural orientations (1979).
Phenomenological theorists work instead displays almost exclusive reference to the
domain of superstructure (i.e. culture) in the course of explicating specific cultural
orientations, deep structures, ideas, and belief systems as keys to a particularistic

18The greatest enmity, as Laudan notes, occurs among proponents of similar theories: "precisely because
some theories represent attempts within the framework of the tradition to improve and correct their
predecessors (1977, 81)."
19The materialist critics of cultural materialism do not object to the analytical division of sociocultural and
economic phenomena, but to the principle of infrastructural determination. Marxist critics argue that
determinacy resides solely in the relations of production (Harris's "structure"), and assert that revising the
core assumption of social change in Marxism--the dialectic process of history--to the level of feedback loops
reduces Harris's theories to mechanistic materialism (e.g. Worsley 1983).
20Harris (1979) praises the general concept of the French Marxists as being consistent with cultural
materialism except for the inclusion of dialectical assumptions. Godelier, however, reverts to the "vulgar
materialism" stereotype of cultural materialism, accusing Harris of demoting social relations to
epiphenomenal adjuncts of a set of adaptive techniques to the natural and biological environment (1986,
49). Godelier's The Mental and The Material (1986), for example, provides an insightful discussion of
cultural-structural/infrastructural dynamics by way of analyzing socially generated information about the
environment and social relations. In the end, however, he imposes the structural relations-cum-historical
contradictions monocausal framework of his research tradition even while honestly presenting useful data
generated by other theoretical approaches.

construction of social reality.21 Explorations into the more symbolic, metaphorical, and
localized side of culture represent an entirely different ethnographic research tradition where
the observation and interpretation is independent of formal scientific analysis. The logical
empiricism of science versus the interpretive, humanistic nature of phenomenology, not
causal theory, is the real issue here. Harris is not guilty of ignoring research domains that
he distances himself from a priori. These are the more ephemeral manifestations of
cognitive and cultural phenomena, which he considers to be beyond the methodological ken
of scientific observation and hypothesis testing. Methodologically, cultural materialist
theory is behavioral, and uses modifications of Pike's linguistic categories of emic and etic
to distinguish between reliably observed behavioral data and cognitive data--to separate
what people say and what they believe from what they do and how they behave.
Long time frames clearly confirm the validity of infrastructural theories but present a
methodological gap for cultural materialism and its direct forerunner, Julian Steward's
cultural ecology:

Given a sufficiently long time span and degree of generalization, technical, economic, and
environmental factors can 'explain' almost all trends of survival and decline, but the nomothetic
rhetoric of similarities, tendencies, and relativities, and broad classification of types of social
formations has somehow to be translated into the language of ethnographic analysis (Ellen 1982,

Culturally-centered studies, in comparison, often lack a comprehensive methodological
framework in the first place. We can, therefore, also note that the language of ethnographic
analysis has yet to be uniformly translated into an inter-connected framework of social
system attributes. Ruttan's (1988) criticism of anthropology centered on this reduction of
human cultural activities and behaviors to an individual ethnographer's reaction and
understanding. We are back to where we started. Human cultural systems are too complex
to be easily grasped within a single theory or paradigm.
Materialist, evolutionary theory is the starting point for looking at how and why cultures
are similar and different. The discord over level and source of causality in the materialist
domain at least partially evaporates if we shift the semantic emphasis on infrastructural
causality to infrastructural selection, or, for that matter, for causality on the structural and
superstructural levels as well. This allows us to discriminate better between the material

21Curiously, however, in an essay titled "Native's Point of View': Anthropological Understanding'," the
principle figure in the cognitive-semiotic school. Geertz, rejects the idea that the fieldworker can personally
participate in or enter into the cognitive world of another culture (1983, 58).

determinants of cultural systems and the cultural feedbacks that they generate.22 As Riedl
(1977) and Kaufmann (1993) note, incremental adaptations produce an organism or system
individually tuned to its specific environment. But this also alters the environment. The
evolutionary process, once underway, generates a mutually causal relationship between
adaptation and feedback.
Variables linked to culture present the major constraints for robust system analysis
encompassing the full spectrum of ethnographic problems anthropologists encounter. The
analytical utility of ethnographic method largely lies in observing the dynamics that
maximize our understanding of the interplay between culture and other forces, and
distinguish the critical differences emerging out of broad tendencies and similarities.
Materialist and ethnographic perspectives employed as complementary conceptual frames
allow us to take full advantage of anthropology's (and social science's) conceptual
endowment. Anthropologists' work oscillates between an unrelenting attraction for the
holistic and general explanations of human phenomena and the investigation of the most
localized, idiosyncratic, and peculiar details manifesting throughout the social universe.
Anthropology is a young discipline, and as Gellner notes, the answer is usually found at
the end, and not at the beginning (1973, 7).

Culture and Cultural Endowment:

Culture, by virtue of its adaptive qualities and synergistic presence in the system, is
synonymous with the self-organizing property of human systems. Culture operates
through a group's internal network of linkages. Cultural components are aligned within
specific social structures as defined by conventional criteria of kinship, age-set systems,
social hierarchies, economic class, and regional, religious, ethnic, and other organizations.
Cultural adaptation, including the role of social learning and decision-making in group
processes, is perhaps even more amenable to analysis through evolutionary principles than
genetic change (Boyd and Richerson 1985). Evolutionary principles can enhance our ideas
of culture as the internal organizing principle of social systems. Cultural responses, more
than biological adaptation which requires long time frames for accumulated mutations to
exert their influence, influenced the human developmental sequence from the simian origins
and prehistoric roots of the earliest societies to the present. This distinguishes cultural

22Feedbacks present the theoretical problem of how different causal variables, articulating together over
time, develop synergetic properties that complicate the material and interpretive analyses of ethnographic

adaptation as the dominant species characteristic enhancing biological survival and genetic
transmission (Ellen 1982, 237).
An important distinction has to be made between selection on environmental and internal
cultural selection within a population or ethnically defined system. Language is often a
criteria for membership in a group, which in turn determines access to land and other
resources. Inclusive criteria defining groups vis a vis other groups is a fundamental
assumption of Marxism and other theoretical variations emphasizing structural
determinancy be it socioeconomic class or ethnicity. Culture, and the behavioral repertoires
it represents, is often colinear with language criteria. As this case study demonstrates,
however, cultural adaptation and inheritance subject to selectionary forces transcends ethnic
categories. One of the goals of this study is to show how the Meru cultural endowment
evolved to reinforce practices and mechanisms supporting sustainable development in the
sense of cross-generational adaptive fitness.
Ruttan's term "cultural endowment" is used here in reference to cultural institutions,
culturally transmitted behaviors, and ideologies arising out of a system's internal order.
Defining a cultural endowment is useful for problem solving, but not in isolation from
other system components. Structure is further qualified by factoring in interstitial and
system-wide organizational phenomena such as patronage nets, patrimonial political order,
factions, societal institutions, interest groups, and decentralized and centralized state
The inter-connectivity of these societal components and how their configurations
respond to the information flows and feedbacks determines the state of the system and its
dynamics. The nature of a system's economic and social connections are thus one key to
cultural dynamics. Methodological stress on the means of communication, as "the
technology of the intellect (Goody 1977, 10)," are a crucial feature of the system's
development. In the Meru case study, much of the intra-system communication involves
informal and non-state loci. I try to represent to some degree the social dynamics involved
here. As we shall see, miraa, in fact, is a basic adjunct to the multi-ethnic, cross-cultural
processes of socialization (Weir 1985; Kennedy 1987). As mentioned earlier, this fact in
itself, would be a thesis of its own. The Igembe have adapted to their economic
environment by producing miraa on a commercial basis, and the consumption of miraa
contributes to the reformulation of social interactions and the processing of information
which is itself the outcome of what Boyd and Richerson (1985) term biased transmission
within a number of local systems.
Most independent variables designated in anthropological literature qualify as system
parameters. Infrastructural variables provide the starting point for the conceptual grid used

for analyzing and locating socioeconomic change over time, but it is the synergetic
combination of material and non-material parameters that produces system dynamics. We
concur with the conceptual order posited in Harris's (1979) cultural materialism regarding
the primacy of infrastructural variables, with the important caveat that evolutionary
development introduces other parameters which may neutralize the primacy of
infrastructure in given situations. In other words, parameter change where change in one
variable effects the others does not a priori refer to infrastructural variables even if they
ultimately derive from shifts in demography, technology, and economy. This is
tantamount to a multi-causal qualification of materialistic theories (including neoclassical
economics, Marxist and neo-Marxist). A holistic approach does not negate monocausal
variables, but it does weaken their causality by linking them to changes in other parameters
and the accompanying feedbacks.
This study builds on the foundation of Johnson and Earle's work by exploring the Meru
variation on the causes, mechanisms, and patterns of evolution (1987, 1). These authors
reinforce the point made by Cohen (1978) that, once it appears, the state becomes a
primary determinant of social evolution. The Meru example is interesting, however,
because the reorganization of society that occurs via the transition to capitalism is occurring
largely in the absence of the state influence on commercialization in other similar areas.
The extent to which internal variables are a direct influence on the production system is
reflected in the farming systems data (Chapters 5 and 6).
The discontinuity between state-initiated economic change and the grass root processes
of social development conceptually complicates proper understanding of both macro level
and micro level dynamics. Small scale formations remain a primary feature of the African
environment, despite the influence of centralizing forces of mainly external origin. The
African situation makes the task of analyzing how the parts of the state-society equation
contribute to socioeconomic transformation more difficult because local systems are
coevolving at the same time strong selective forces are cutting African state systems to a
more manageable size. The situation prevailing in Somalia, following the dissolution of the
state, exemplifies the ability of cultural institutions to reorganize smaller units of social
organization--traditional Somali segmentary lineages in this case. The Somali case is an
extreme example of the fluctuating and "lenticular" qualities described by the relationship
among sub-ethnicity, the state, and supranational organizations (Cohen 1992). As it also
shows, it is not likely that once statehood is present, a region can not easily return to
statelessness. Some form of state-like organization, if not another state, will emerge and a
regime will then step into the gap. The case of Meru involves similar theoretical questions.
The question is not one of state-led versus local development, but a question of how

evolution is adapting states and local systems to each other in current the post-colonial
phase of historical development.

Culture. Feedback, and Social Learning,

We will now list some attributes of culture within the social systems. To reiterate the
materialist concept, culture is generated by other levels of the system, and cultural variables
produce various positive and negative feedback (Harris 1979). There is little argument
that, in this sense, culture reinforces the directionality of infrastructural and structural
pathways. The conclusion reached by (Kaufmann 1993), paralleled in Riedl's (1977)
work, is that after a certain level of system complexity emerges, neither structure or
feedback enjoy adaptive primacy. This hypothesis supports the critical importance of
feedback in the system, and at least part of that feedback is cultural.
Feedback generated by the system reflects some of the internal dynamics generated by
its own development, and can help us understand a system's capacity for adaptive change.
Local feedback influences the larger system's development as local systems or social units
are increasingly integrated into the larger system. This idea receives support from Riedl,
who notes that "the systems conditions which link different levels of complexity to
feedback loops of cause and effect are responsible for the evolution of life (1977, 358)."
The key factor is the group mechanisms for discrimination and classification of the
information coming into a system. The cultural analog appears in Boyd and Richerson's
(1985) concept of biased transmission, referring to knowledge that has been selected for by
the cultural system and replicated across generations. Naturally, cultural information
gathering and social learning are more efficient than individual trial and error. The
importance of culture in creating and transmitting information is another important adaptive
quality that makes it critical for indigenous-led development. In a recent article, Richerson
and Boyd (1992) stress that people pick and choose whom and what to imitate in the first
place, and that social learning is favored over individual learning because it allows
individuals to avoid the costs associated with learning. The strength of biased transmission
depends upon the amount of cultural variation. The variation generated by the
environmental differences underpinning group specializations is reflected in the symbiotic
properties of Igembe Meru society.
On the societal level, they link the increase of non-parental transmission of knowledge
to cultural success and how this aided the spread of low-fertility norms contributing to
demographic transition. This has very interesting implications for the Nyambene case
study where cultural institions increase the weight of "teachers," i.e., individuals from each

generation recognized on the basis of their achievement. One ramification is that Africa's
local systems do not have built-in cultural proclivities supporting reproduction leading to
over-population. Demographic indicators contained within the survey data presented in
Chapter Six indicate that the Nyambene system is not only self-regulating (and that the
Meru do not have to be taught family planning), but also indicate that its culturally
institutionalized age set system deploys labor within different spatial sectors of the
production system (ecological zones) and economic sectors of the production system
(economic nodes) according to an efficient scheme. Another ramification is that,
theoretically, the Nyambene system data argues for selection operating on the level of the
group, in contrast to group evolution as being a cumulative outcome of selective forces on
individuals within the population. The folk-knowledge corollary of this in Meru is
conveyed in the saying "muntu ni antu: a person is people" [the person is a function of the
group to which he or she belongs].
Cultural systems contain highly evolved self-regulatory mechanisms that identify
problems, classify constraints, and evaluate outcomes. A complex system "knows" its
world by virtue of an internal model capable of compressing information and allowing
action: it is driven in part by the dynamic behavior generated by its internal state and in part
by the steady flow of information from its world (Kaufmann 1993). Riedl (1977)
mentions the replication and separation of identical information as a fundamental principle
of life. Culture is the informational software guiding social systems.
Kaufmnann's idea of spontaneous internal organization embodies the idea that culture is
constrained by the nature of a system's internal linkages. Riedl mentions the same idea
when he states that "old systems" are highly interconnected and in principle, unalterable. A
system becomes conservative because new adaptations building on the older ones burden
the system's ability to change (1977, 362). This explains why a highly developed,
complex system with copious information storage capacity can be less responsive to
change, and more prone to entropic forces than smaller, information-poor systems (like
Igembe society) that maintain broad structures for the exchange and dissemination of
information. The conservative caste of non-adaptive social orders explains why history is
punctuated with revolutions and short periods of far-reaching change like the Neolithic shift
to domesticated food and livestock production (Schusky 1989).
By analogy, an adaptive cultural endowment is one that is receptive to and encourages
the integration of social learning. The feedback into the system is also selected for or
against depending upon the internal order of the system. An evaluation of a cultural
system's adaptive properties, examined in this case study in association with the notion of
sustainability, must take these arguments into account.


I referred to symbiosis in the preceding chapter as an important but neglected
mechanism of social evolution in the African environment. Symbiosis helps us to elicit the
pattern and dynamics of the Nyambene system's development. Forces supporting small
and localized adaptations also support symbiotic interactions. Incremental changes, such as
those at the center of the Nyambene system's development, feed back into the macro-level
transition. Together, these forces drive the local and national level systems towards a state
where the multiple interactions among micro and macro variables interact to generate
socioeconomic transformation.
Symbiosis grows out of exchange, conflict, cooperation, diffusion, and systemic
syntheses. This produces synergetic outcomes in which the whole becomes greater than
the sum of its parts; a cultural equation where one plus one is greater than two.23 Riedl
(1977) and Kaufmann (1993) note that adaptive gains are paid for in the long run by
reduced adaptive potential. A reduced scope for internally generated adaptations implies
that the potential advantages of symbiotic adaptations are increased. In our case study, for
example, symbiotic interactions expand the range of ecological niches available to an
agrarian society. Increased complexity promoted the Meru polity's stability, and decreased
its vulnerability to environmental catastrophe heightened in other local societies by the
growing regional population density.
There are numerous manifestations of accommodation and cultural symbioses among
African societies. Symbiosis may result from hostile and predatory relations between
organisms where a threatened population survives by integrating elements of the dominant
population in a shared environment. As noted above, anthropological examples of
symbiosis often emerge from conflict relations such as Bantu-pastoralist interactions
(Kimambo 1974), the Nilotic penetration of intralacustrine kingdoms in Uganda (Oliver
1991), and the Maasai-Meru relations depicted in this case study (Mboroki 1972; Rimita
Perhaps the most developed thesis on this theme is Spencer's (1973) study of Rendille
and Samburu pastoralists who evolved a symbiotic alliance founded on econiche
adaptations based on camel and cattle specialization in adjacent environments. The

23Margulis and Sagan's 1986 book Microcosnmos is a tour de force developing the critical role of symbiotic
processes from the very beginnings of life on earth. The authors extend symbiotic syntheses in the world
of microbes to a microcosmos of evolutionary development subsuming the details of human physiology.
The role of symbiosis, "the merging of organisms into new collectives (pg. 20)," is described as a major
power of change on earth. Symbiotic coevolution couples large-scale and small-scale in a seamless cycle of
mutual causality (Jansch 1980. quoted in Peat and Briggs 1989,164), echoing Riedl's (1977) description of
two way causality.

relevance of symbiosis to the Nyambene system's internal development cannot be
overstated. We attempt to examine the historical and contemporary aspects of cultural and
agricultural system symbiosis in the following pages. As noted, this problem-driven
analysis attempts to provide information on the important role of the cultural endowment in
socioeconomic processes. The specific problem is that due to the saturated nature of the
high potential highlands, the lower zones and the informal sector of the urban economy
share the greater responsibility burden of supporting Kenya's surging population as the
historical flow of population gradually moving up the ecological gradient has reversed into
a flow back down.
National agricultural goals are therefore contingent upon technological change and the
environmental carrying capacity of fragile ecological systems under conditions of high
population growth, low input use, and limited off-farm economic opportunities. Both of
these factors of innovation and conservation appear to be largely beyond the existing
institutional capacity, including government ministries, PVOs, and NGOs. In recognition
of this reality, the role of market forces in the more problematic ASAL areas is expanded to
include what the Kenya National Development Plan 1989-1993 refers to as the potential for
ecozone symbiosis: ASAL strategies include "determining ways and means of effecting
symbiotic exchange of resources and products between the ASAL and high potential areas
(page 134)."
Unfortunately, the mechanisms and socioeconomic loci of "symbiotic exchange" are not
defined in the Development Plan. These are ostensibly informal linkages, new and
traditional forms of eco-niche specialization, migration and new settlement, reinvestment of
agricultural profits from high potential zone commodities in the lower zones, and the
exchange of location-specific information contained in indigenous knowledge systems in
local resource use and management. The research indicated that ecozone symbiosis has
been integral to Meru historical processes.

Coevolutionary Processes

We end our review of evolutionary concepts with the idea of how the linkage of systems
or their components influence development.. Individual systems and organisms that alter
in response to parameter changes also change or deform their landscapes with
consequences for other entities sharing the environment. Two general ecological models
differentiate coevolutionary patterns and processes. In the Red Queen Hypothesis (Van
Valen 1973), species evolve continuously in relationship to each other where a fitness
advantage in one species is compensated by a similar adaptation in another. The second is

known as Evolutionary Stable Strategies, or ESS (Maynard Smith and Price 1973). This
states that adaptive changes, tradeoffs, and responses eventually stabilize within an
unchanging, overall system stasis.
The critical variable differentiating the two patterns is design constraints within the
organism. Adaptations carry a price. For example running faster may require a higher
metabolic rate. But some traits can increase without constraints, as Richerson and Boyd
indicate for much social learning. Constrained adaptations settle into ESS, whereas
unbounded traits become trapped in Red Queen rat races. Free market policies assume an
ESS outcome based on differential comparative advantage, but Third World commodity
production, where an exotic market crop is substituted for a traditional cultivar, displays
characteristics of the Red Queen Hypothesis. As noted by Lipton (1988) for improved
coffee and market cultivars, improvements in production are offset by decreases in prices.
The IFI's Structural Adjustment Programs can also feed the rat race: countries, like Ghana,
boost their cocoa production only to experience decreased prices due to increased supplies.
This is central to the question of African development, which seeks a strategy that can
tune local economies so as to propel nations out of the boom-bust cycle of primary
commodity production. The implication is that much of Kenya's development up to this
point, in fact based on social learning and borrowing of technology, supports the Red
Queen scenario. For example, commercial cultivars easily substituted for local crops (e.g.
coffee, cotton, tobacco) supported local change but not a change in the relationship of the
local system (in this case, Kenya) to the world economy. In reality, the position of Africa
in world trade has slipped significantly over the past decades, and by 1988 Africa only
accounted for 5 % of international commerce (World bank 1989). The ultimate question is,
therefore, not one of boosting on-farm production. Rather, boosting incomes, increasing
diversification and exchange, and providing inputs feeding multi-sectoral growth is the
critical variable.
Exchange and linkages among regions and the state are a critical variable here. Other
studies have shown that sustaining the self-sufficiency of local systems in isolation is
difficult (Hansen 1994). A way out of the coevolutionary conundrum emerges out of the
idea that Africa will benefit by developing the richness of its internal couplings, implying a
return to some of the regional exchange dynamics of the late precolonial period. This, in
addition to existing linkages to the world system, will buffer local economies against the
forces of the Red Queen. This, again, implies that a shift in the couplings of local systems
directly aligned with the world economy markets and labor markets supported by colonial

governments will contribute to regional sustainability over time by increasing and
diversifying their couplings.24
The idea advanced here is qualitatively different from the simple delinking strategies
advocated by dependency theorists, described in detail by Blomstrom and Hettne (1984).
Linkages to the international economy are adaptive when they generate internal growth and
diversification (Collier and Lal 1980), but ultimately maladaptive when they reinforce
monocultural complexes (c.f. Hyden 1988). Historical interactions among local systems
underscore how the outcome of increased couplings is often cultural and economic
In what follows in succeeding chapters, we shall examine the system's development: the
historical evolution of the production system synthesized under Meru and Igembe ethnicity,
the internal order of Igembe society, the specific configuration of Igembe agriculture and
the dynamics of change and the symbiosis between ecozones. Chapter Three provides a
picture of the Igembe cultural endowment by providing a history of ethnic interactions, the
development of cultural institutions, and showing how they are attuned to problem solving
and conflict resolution. Chapter Four parallels some of the information in Chapter Three,
but centers on the system's economic development. Chapter Five presents an outline of the
production system, and Chapter Six analyses how it is changing and replicating itself
across ecological zones. Finally, Chapter Seven extrapolates some of the dynamics of the
local system to analysis of developments on the national level.
In the chapter that follows, the origins and role of the Igembe cultural endowment are
examined. The cultural endowment consists of ideologies and institutions. Ideologies
point to the "biases" in biased transmission as discussed by Richerson and Boyd (1992),
and influence the accumulation of choices that guided the evolution of the local system from
the past to its current form. Cultural institutions are seen operating in terms of (1) the
stable and adapting relation of internal factors within Igembe society as they remain in
continuous relation to their past as it meets their future, and (2) the relationship of Meru to
the outside world.

24Local examples of this phenomena are documented in the final chapter of Gluckman's work on the Lozi
of Zambia, in Cohen and Atieno-Odhiambo (1989), Berman and Lonsdale (1991), and many other studies of
local society in Africa. deJanvry's (1985, 1987) work on Brazil provides a similar perspective for Latin


"A great curiosity in the literature is why cooperation and social learning are so widely discounted
given the rich empirical variety of local institutional arrangements for managing common
issues." (Herring 1990, 64).

This chapter provides material on Igembe cultural institutions in order to give substance to
the importance of cultural adaptation and feedback discussed in the two previous chapters.
We try, in particular, to stress the importance of the Igembe's system's cultural endowment
as it applies to land and resource management and social guidance. This has long been one
of the most under-valued dimensions of African production due to a variety of factors. The
reports of early explorers, missionaries, and adventurers initiated feedback that conditioned
a view of African natives as irresponsible tenants of a wild but Edenesque environment.1
In the absence of other constraints, European ideologies legitimized exploitation by positing
poor environmental management as one of the claims that supported European intervention
and control. This impression survived beyond the colonial period.2
Environmental issues go to the heart of development policies that regard African culture
as rooted in conservative peasant values, if not outright maladaptive practices.
Environmental degradation, one in the cluster of failed governance, implicated African
peasants (and goats) as the agents of environmental destruction. The large-scale famines of
the 1970s and 1980s became environmental media events disseminating the dust bowl
disaster image that mobilized new kinds of eclectic international responses. Massive food

1The European image of Africa as a region where the natives "lived in a malevolent conditions of chaos"
was partially true as travelers saw the results of slavery and other forces of conflict and change (Davidson
1992, 80). Slavery deflated the quickening pace of change in some areas by siphoning off population and
destabilizing regional ethnic relations.
2The paternalism of widely read authors like Elspeth Huxley reinforced the environmentally ignorant
African stereotype. If Africans were proving to out-perform the settler economy, it was because they
exploited techniques and behaviors that placed their economic interests over colonial environmental concerns
(Anderson and Throup 1985, Steinhart 1987). Colonial policies and administrators, however, treated the
environmental mission seriously, especially in regard to the entrenched emphasis on forestry and soil
conservation inherited from British India (Anderson 1992).

relief and delivery became a critical component of Western approaches to development just
as colonial enforced conservation measures were earlier (Anderson and Grove 1987).3

Sustainability and Cultural Adaptation:

Sustainable development became a buzz word during the late 1980s as environmental
issues came to occupy an increasingly prominent position in formal discussion of
development. Sustainability does not appear in the topics covered in Eicher and Baker's
definitive survey of African agriculture published in 1982, but by 1992 environmental
sustainability and related issues were the major subject of the Earth summit, and provided
the underlying concept of the World Bank's 1992 annual World Development Report,
Development and the Environment. The report defines sustainability in its broadest sense:
"Sustainable development is development that lasts," in the sense that current generations
should "meet their needs without compromising the ability of future generations to meet
their own needs." By the early 1990s it was de rigour in many texts to attach the adjective
"sustainable" to standard references to agriculture, economy, and development.
Sustainable agriculture, for example, subsumes twin objectives of production and
conservation. Emphasis on sustainable agriculture mirrors heightened awareness of the
global effects of environmental pollution, the breakdown of local systems, and a
recognition of the inter-connected quality of bio-physical environments paralleling the inter-
dependent nature of the world political economy. This is a world-wide concern, requiring
interdisciplinary scientific endeavor (Wilken 1989). Institutional agricultural research has
overcome its monocultural biases to investigate hypotheses related to intercropping,
vegetative regeneration under heavy grazing regimes, and various other technological
aspects of traditional production strategies. In fact, much of this research merely quantifies
the appropriateness of practices based on local knowledge.
Local systems logically embody ideologies of conservation as well, although research
has concentrated on other dimensions of traditional practices, and seldom make explicit the
indigenous conceptualization of sustainability. The sustainability of a system is gauged by
its ability to adapt and to evolve in relationship to changing system parameters. The idea is
used here in reference to cross-generational adaptive fitness. The Nyambene system is an

3Game conservation, state forestry, terracing, cattle-culling, protecting watersheds, and rotational schemes
were the primary elements of the colonial Africans conservation strategy. Some of these policies directly
fed political opposition in the native reserves after World War II (Beinart 1987).

especially interesting case study because it displays consistent internal dynamics
contributing to a gradual, but largely uninterrupted, developmental trajectory.
Other regional societies were arguably more progressive in their adaptation to the
Western presence that redirected Africa's socioeconomic processes over the last two
centuries. The accumulation of material rewards of this accommodation for the Gikuyu,
Embu, and Mt. Kenya sections of the Meru is not disputed. Their adaptation to the new
order, however, was paid for by replacing some of their internal organization and
institutions with monocultural external linkages. The Nyambene traditionalist exemplar
presents us with a rabbit and hare situation where the Igembe are rapidly overtaking their
rural counterparts during the current period of economic stagnancy, and forging ahead
largely on the strength of their own cultural endowment.
The survival, indeed the vitality, of the Nyambene cultural endowment contrasts with
other ecologically similar and culturally related areas across the Mt. Kenya region.4 The
strength of the indigenous cultural orientation is reinforced by the feedback deriving from
the unique trajectory of capitalist development emerging out of the commercialization of
indigenous Nyambene agriculture (Goldsmith 1988).5 The Meru, and the Igembe in
particular, developed social concepts and agricultural practices generations ago that are only
now coming of age in developmental circles. It is impossible to discuss the Igembe cultural
endowment without reference to their production practices, but the main concern of this
essay is its social support system.
Institutions, customs, traditions, and myths represent the cultural DNA that guides a
local system as it adapts to its environment and responds to crises.6 Cultural institutions
coordinate the different parts of the system, store and evaluate knowledge, organize, and
establish priorities for group and individual decision making. These institutions are
cognitively linked to symbolic systems, ideologies, beliefs, and myths that continue to
exert certain influences on the behavior of a system, even when their institutional structures
lapse. They can, therefore, provide loci for investigating the transmitted cognitive and
structural orientations of a society. The adaptive capacity of traditional systems points to
corresponding cultural endowments or knowledge systems used to classify, identify,
respond to, and solve problems as they arise. The ethnographic and historical record
contains numerous examples of cultural adaptations and techno-environmental practices

4The Nyambenes encompass two Meru sub-groups, the Tigania and Igembe. The Igembe are the main
subject of this discussion.
5The Nyambene path to capitalism appears more the exception than the rule in sub-Saharan Africa, not
discounting notable examples of similar cases like Ghana's early cocoa farmers studied by Hill (1963).
6 The description of institutions etc. as cultural DNA was made by the Nobel Physicist, Murray Gell-
Mann, quoted in Lewin (1992, 15).

underpinning sustainable production systems in Africa (Kjekshus 1977, Allen 1965,
Richards 1985). Situations where cultural institutions survive the selective forces in the
social and physical environment, and continue to guide local society, present an
opportunity to analyze their role in local systems.

Kuria: The Meru Definition of Sustainability:

The thick canopy of trees, even in areas of dense settlement, is a striking feature of the
Meru highland landscape on the east side of Mt. Kenya.7 The Meru recognize the
restorative properties of forests and trees, and observers have often commented on their
overarching social emphasis on resource conservation (Laughton 1944, Lambert 1949,
Bernard 1988). Two sizable forests, the Nyambene and Ngaya, remain intact despite the
intense demand for productive land in the Igembe region central to this discussion.8
Numerous groves of indigenous trees also exist in densely populated areas, and local farms
are small, intensely manicured domestic agro-forests that generate relatively high incomes
and a variety of domestic products. Local farmers also express positive attitudes toward
wildlife despite the problems of crop damage by animals straying from the forests, plains,
and Meru National Park.9
Respect for these natural resources runs deeper than simple utility and use value. The
Meru invest their natural environment with material, curative, aesthetic, ritual, and spiritual
qualities. They use concoctions of trees and herbs to treat people, animals, and crop
diseases. Their sophisticated indigenous agroforestry practices are the result of continuous
observation and experimentation with plants and trees over time. Exchange relations have

7 Homan remarked that while large trees are few in Gikuyu country, survey is difficult and costly in Meru
due to the indigenous tree cover, which makes aerial methods impossible in some areas (1960. 237). In
Igembe the ideology of trees holds a number of species sacred, and even a metaphor for marriage: "a women
is a garden where you plant trees."
8The name Ngaya comes from one of the words in Meru for God which also refers to rain. The Igembe
Meru consider Mt. Kenya, Ngaya, the Nyambene forest, and the I'ombe crater in the far comer of the district
to be primary places of God, although there are numerous other sacred lakes and small hills, forests etc.
where sacrifices are offered.
9Field notes, Oct. 31, 1992. Between Kangeta and Nkinyang'a we came across a colobus monkey crossing
the road. There were people attracted to the sight, but the monkey was not harassed: the monkey behaved as
if he/she was in familiar territory. Ng'olwa said to harm such a creature would be a great crime. "It is
believed," he said, "that such a creature is a messenger between God's world and men, and killing it would
invoke a great curse on the area. Coming across it in the bush is a different situation, but killing is not the
issue in either situation." Colobus monkeys are apparently not a "sacred" species. Meru made traditional
headdress from colobus skins and the contextual setting of this attitude is one where a monkey from the
Nyambene forest regularly visits this densely settled area without causing significant damage to crops. In
contrast the only wild monkey colony in the United States, the offspring of monkeys brought to film a
Tarzan movie, living outside Silver Springs, Florida, have been scheduled for termination despite the fact
that there is no evidence that they have interfered with the human population or farms.

yet to lessen their independent disposition or their cultural commitment to resource
The linkage between strong indigenous institutions and the Meru ethic of conservation
shares a complicated historical dynamic that is particularly interesting in the light of the
concept of sustainability. The main thread in the argument presented here is that the Meru
people perceived development in terms of indigenous applied agricultural science,
governance structures represented by their social institutions, and the conservation of
natural resources upon which Meru society depends. For the Meru all these ideas are
conveyed through the formal "Kuria" greeting used by Meru elders and Njuri Ncheke
Kuria means to raise or nurture, as in raising children or planting crops. In the cultural
context it is both intuitively and explicitly an all-inclusive term for what we call
sustainability. The greeting symbolizes the Njuri Ncheke ideology of a pan-generational
commitment to social development. Kuria subsumes socially defined goals that mandate
the proper management of not only environmental resources, but also, the human resources
forming the Meru polity.11 The Njuri Ncheke intersects Meru age-set structures in the
form of age-set councils on several levels from the clan on up. Age-sets symbolize the
procession of generations across time, and institutionalize kuria-sustainability as a concept
of cross-generational adaptive fitness.
Local sustainability grew out of multi-ethnic interaction as well as internal guidance.
The Meru entered the region as a small group of clans and evolved into the dominant
society in the region. Their historical dynamic presents empirical evidence of symbiotic
processes as a strong evolutionary force: social and technological adaptation accompanied
the integration of several different ethnic populations into a unified production system
utilizing the high ecological variation in the region. Meru often identify their relations with
other groups in terms of conflict, but a closer examination of their system and discussion
with informants reveals a synthetic society born of symbiotic processes. 12 Recognition of
the symbiotic qualities of African social phenomena broadens the conceptual framework to
include the syntheses and adaptive socioeconomic arrangements that arise out of the
conditions that engender both conflict and the potential for cooperation.

1OThe salutation and response for individuals and groups is simply"kuria!"
11A progressive leader is sometimes called mukuri, from the same root.
12Analyses of Africa often focus on group interactions in terms of conflict and power relations. Curtin's
study, Cross Cultural Trade in World History (1984). two recent articles by Hopkins (1988; 1986). and
Wolfs Europe and the People Without History (1985) do present an alternative view where exchange and
interaction receive their due credit along with conflict and war.

Precolonial Dynamics:

The Meru environmental configuration fosters a pattern of ecozone specialization expressed
in the subsistence adaptations of different ethnic groups inhabiting the Mt. Kenya region at
different times. Diverse groups entered the area and colonized different ecological sectors
according to their original production strategies. Contemporary Meru society evolved
under the influences of populations present in the region prior to their arrival. Group
interactions were complex, and varied according to the area and the specific section of the
The proto-Meru were actually the last group to enter the region. The original inhabitants
of the region were Agumba and Ndorobo hunter-gatherer groups inhabiting the montane
forests and savannas. Over two thousand years ago, Cushitic agropastoralists migrated
from the Ethiopian highlands and took up residence in intermediate zones where they kept
small ruminants and cultivated indigenous African grains, millets, eleusine, and sorghum.
Approximately one millennia ago, Highland Nilotes entered the region and added cattle to
the list of domesticated elements of local production systems (Sutton 1980). The people
known locally in Meru as Mwoko, probably pre-Kalenjin remnants of this population, are
associated with the abandoned Sirikwa irrigation works on the escarpment of the Rift
Valley. Groups of Oromo speaking Eastern Cushites reached the area by the 1500s adding
to the cultural melange.
A wave of Bantu agriculturalists, migrating from the northeast ,crossed the Tana River,
also around 1500 A.D., eventually displaced most of the agro-pastoralists groups. They
probably absorbed some of them as they congealed into the forerunners of the Gikuyu,
Embu, Mbere, Tharaka, and Chuka (Mwaniki 1984). These Bantu settlers, known in local
traditions as the Tumnibiri clans, were followed by south-bound plains Nilotes, pastoral
Maasai and Samburu spreading from the Rift Valley corridor to the grasslands fringing the
western flanks of Mt. Kenya and the Nyambene Range. A second wave of Bantu settlers,
identified in traditions as the Thagana clans (Thagana is the Meru name for the Tana
River), fled their coastal homeland, an island they call Mbowa, and followed the Tumbiri
clans into the region. Clans entering the Mt. Kenya vicinity during the second migration
crossed the Tana twenty one generations ago, which would date their appearance in the
region to the late 1600s or the early 1700s. These Thagana clans developed into the Meru
speaking peoples who maintained their initial unity through the Njuri Ncheke council
(Fadiman 1976; Laughton 1944).
Initial movements of the first Bantu clans crossing the Tana assumed a north to south
direction; settlement centered on the middle highland zones, or the maize zones in the

ecozone classification employed in this study. The exception to this pattern are the
agropastoralist Mbere and Tharaka, who relied on livestock and millet production in the
lowlands of eastern Mt. Kenya. The Maasai had moved into the high pasturelands in the
rain shadow on the northern fringes of Mt. Kenya and into the lower southwestern flanks
of the Nyambenes, while their Samburu cousins utilized adjacent rangelands on the
northern slopes on a seasonal basis, effectively sealing-off expansion to the west.
The Tumbiri settlers absorbed most of the earlier Cushitic and Nilotic agro-pastoralists.
The pre-Kalenjin Mwoko are the one population mentioned in Meru traditions who
remained unassimilated after the spread of the Tumbiri clans. A small pocket of the earlier
Highland Nilote-Southern Cushite population, the Mokogodo, also survived in the Mt.
Kenya area. The second wave of Bantu clans encountered higher population densities and
later movements were in a west and northwesterly direction. The proto-Meru Thagana
clans divided into different communities in the lowlands between Mt. Kenya and the
Nyambenes. Migration was replaced by irrendentist expansion once pastoralist movements
sealed off the northwestern corner of Mt. Kenya. The Tumbiri migration had pushed the
Embu and Mbeere to their locations south of Meru District, and the arrival of Thagana clans
served to force the Chuka up, becoming the first group to settle in the higher coffee zone
and the tea zone of Mt. Kenya.
The arrival of the proto-Meru Thagana was a major turning point in the region. The
population was reaching its saturation point relative to the extensive production .ati tegiep
prevailing during that period: hunting, shifting cultivation, and the cattle based nomadic
pastoralism of the Maasai. By the mid 1700s the scarcity of game and increasing
population density led to increased sedentarization. The Meru clans occupied the
northeastern corner of the Mt. Kenya foothills, now divided into several sections or sub-
ethnicities, and moved up the slopes of Mt. Kenya and the Nyambenes. This resulted in
the zonal agricultural system described by Bernard (1972). Ethnic relations quickened as
the region's periodic droughts intensified conflict among the now circumscribed
populations. Agriculture underwent a shift towards intensified agricultural production as
groups colonized the higher zones bordering the band of montane forest on Mt. Kenya and
the Nyambenes.
The Meru polity evolved through processes of integration, conflict, and synthesis. The
Meru fought the Maasai after crossing the Tana. They clashed with Oromo pastoralists on
the plains of Tharaka during the Nyamdure drought of 1725-29 (Mwaniki 1984, 151). The
Igembe variously refer to these pastoralists as the Agira, Akara, and Ukara, and claim they
were related to the Mwoko (Fadiman 1976, 155). The Meru absorbed some of these

Cushitic/Nilotic groups,13 and developed peaceful relations based on blood brotherhood
with others (Lambert 1949, 13). They also established fictive kinship relations with the
Waso Nyiro Borana, the southernmost extension of the greater Borana population of
southern Ethiopia inhabiting the rangelands overlapping the northern borders of present day
Meru District, and with the Rendille, another Cushitic group inhabiting the more distant

Maasai-Meru Relations

The Meru term Uuru, enemy, is reserved exclusively for the Maasai and Samburu.
Meru traditions describe their relations with the Maasai in particular as a protracted state of
hostility and conflict. Even so, herdsmen and farmers maintained symbiotic relations even
amidst local hostilities, and considerable intermarraige took place, especially among the
Tigania (Hjort 1981). The Meru of Kangeta traded with the Samburu at a market on the
border of the maize and millet zone. Dolichos (Lab Lab niger), or hyacinth beans, are a
major crop on the northern flank of the Nyambenes. The papery skin of these beans
detaches upon drying; when the wind carried these flakes across the plains the distant
Samburu knew that the Igembe harvest was ready, and they came to exchange skins, goats
and other products for food.
Even so, much of the last century saw extended hostilities between the northern Meru
and their warrior pastoralist neighbors. The Meru ultimately prevailed, pushing the Maasai
beyond the modem borders of the district, and chasing the Samburu to the other side of the
Waso Nyiro River. The enmity, however, was matched by other forms of interaction,
because of all the groups in the region, the Maasai influence on the northern Meru is
particularly pronounced. (Mboroki 1972). The Nilotes marked influence on north Imenti
and Nyambene society led one British District Commissioner to observe that, while
linguistically Bantu, the northern Meru were in many other aspects culturally Nilotic
(KNA/DC Mru 1, 1927).
Famine and internecine warfare accelerated the decline of the Maasai in the Mt. Kenya
vicinity. When sections of the Maasai polity crumbled, refugees sought refuge with their
kin among the Meru and the northern Gikuyu (Muriuki 1976, 133). The Meru absorbed
significant numbers of these Maasai (and Samburu to a lesser degree), especially in Tigania

13Sutton and Oliver discuss Cushitic and Nilotic cultural syntheses occurring on the southern edge of the
Ethiopean highlands as the prelude to an agropastoralist migration and settlement in the Kenya highlands
around 1000 A.D.
14 The Meru term, gichiaro, referring to local practices of blood brotherhood, is discussed presently.

and north Imenti. Maasai territories were divided into two southern and northern reserves
following the allocation of land in Kenya's central Rift Valley to British settlers. In 1911
the Maasai leader, Lenana, concluded an agreement with the British to move all Maasai
proper to the southern reserve. Apparently a number of Maasai were able to avoid
relocation through their association with the Tigania. Some Tigania joined a portion of
their Maasai relations and trekked to present-day Narok District during the Kiaramu famine
of 1918-1919. Several years later colonial authorities detained those Maasai remaining in
the district and resettled them outside Meru. Only the Samburu remained in the Mt. Kenya
In this manner the Meru polity expanded to cover the succession of ecological zones
from montane forest to semi-arid rangeland that occur across the highland-lowland
interface, the production system assimilating various elements of ethnic, econiche
specialization in the process. Conflict as well as peaceful assimilation fed the growth
process, and contributed a number of the differences distinguishing the various Meru sub-
ethnicities. The assimilation of ethnic and cultural elements represents an extremely
complicated process as this summary of the ethnic and cultural mix that produced the
contemporary Gikuyu, which also holds for most contemporary populations of the Mt.
Kenya region, reveals:

These early occupants of the Mt. Kenya region have left a deep imprint on the Kikuyu and their
cousins. The numerous physical types seen amongst them bear testimony that the Gikuyu
represent a fusion of many different ethnic elements. No doubt some of the words and traits were
borrowed at various stages of contact from the plains Nilotic speakers such as the II Tikirri.
Ndorobo or Athi and the Maasai. However it should be noted that many of them were acquired
from the Eastern Cushitic-speaking peoples in the first instance. It is probable, therefore, that the
Thagicu, or proto Thagicu, might have acquired some of these Eastern Cushitic cultural
characteristics directly from the Eastern Cushitic-speaking elements. The most important of these
traits are circumcision and clitoridectomy as the major initiation rites, the Cushitic prohibition
against eating fish, and presumably the idea of a cycling age-set system. It is likely, though by no
means certain, that these features were borrowed from the Gumba. Furthermore, the Gumba are
reputed to have taught the Gikuyu the art of iron-working, while the Aathi are claimed to have sold
large chunks of their former hunting grounds to some of the Kikuyu. Muriuki 1976, 118.

The complexity of ethnic interactions mirrors the variation in the regional environment.
A chief feature of the Meru environment, as stressed earlier, is a succession of ecological
zones based on altitude, the spatial orientation of highland masses influencing rainfall and
rainshadows, and the location of physical features such as rivers, lakes, forests, and other
natural resources contributing to the multiplicity of ecological niches within the region.
Adaptation to specific conditions within ecological zones maximized the production
potential of these zones, and specialization promoted interaction and exchange among
populations across the ecological gradient.

The Nyambene Meru:

The sections of the Meru proper, as defined by membership in the Njuri Ncheke, (the
Imenti, Tigania, Igembe, Miutine, Igoji, Mwimbe, and Muthambe) are descended from the
Thagana clans. The Chuka and Tharaka, products of the Tumbiri clans migration, are
Meru by administrative definition and cultural-linguistic similarity. They claim a separate
identity for legitimate reasons borne out in the historical record of the district's internal
group relations. The Tigania and the Igembe share the Nyambene region and a reputation
dating back to the early colonial period for being in the vanguard of traditional culture, and
resistant to Western influences (Lambert 1947). The Tharaka of the district's eastern
lowlands, the most isolated of all Meru's sub-groups, enjoy a reputation for cultural
conservatism to the point of aggressive xenophobia.
The two major dispersal points from where both the Tumbiri and Thagana populations
expanded into the rest of the district lie on either side of the Nyambenes. The one
surviving group linked to the Tumbiri clans and the Gikuyu, the Thaicu, remained on the
lower slopes of the northern Nyambenes. Igembe traditions trace their migration as also
coming from the northerly direction. They avoided contact with the Thaicu presence in the
middle zones, and initially settled near the fringes of the Nyambene forest. They expanded
first into the small valleys on the windward side, and then down the middle spine of the
Nyambene range. The Tigania approached the Nyambenes from the other side and settled
in the foothills between the two mountain ranges and on the southern slopes of the
Proximity to the forest served as a refuge during Maasai raids, and explains Igembe oral
traditions of contact, conflict, and incorporation of Agumba elements. Except for the
numerically small Thaicu, there is no evidence of large populations inhabiting the
Nyambene range other than the Agumba forest dwellers and the Mwoko, whom the Igembe
associate with knowledge of water (Lambert 1949). The Agumba were defeated by the
newcomers and ceased to exist as a separate entity. The Mwoko were agro-pastoralists.
As Highland Nilotes they would have practiced circular age-sets organization (instead of
the linear age-set organization associated with the Plains Nilotes), and the reference to
knowledge of water may refer to Sirikwa-type irrigation practices. Mwoko relations with
the Igembe and Tigania were hostile, and the Nyambene Meru ultimately prevailed over
them as well.
Both of these groups left their mark on the Meru sub-groups of the Nyambenes.
Although traditions claim they took over age-set organization from the Maasai (Rimita

1988), Igembe and Tigania age set organization is cyclical while Imenti age-sets are linear.
Nyambene Meru have highly developed ethno-botanical repertoires, and this is especially
evident in Nyambene high-zone agriculture. The Igembe also identify certain families with
a strong agro-pastoral orientation as being of Mwoko origin. It follows that Maasai
influence became strong later, and especially among the Tigania who, like the north Imenti,
permitted a Maasai presence in the midst of their agricultural settlements (Mboroki 1972).
The Igembe reflect a particularly strong synthesis of elements derived from the different
populations they encountered and absorbed. Ecological knowledge, medicine, and magic
from the hunters and foragers, cattle and age-set social structures from the Plains Nilotes,
and circumcision practices from the Cushites blended into their own essentially agricultural
orientation. The socioeconomic synthesis, the location and natural fecundity of the
Nyambenes, and strong political institutions underpinned the rise of local production
system as the hub of the regional trade network (Chanler 1893). The ethnic elements
contributing to this ethno-environmental consolidation are presented in table 3.1.

Table 3.1: Ethnic groups and Ecological Zones

Ecological Band Ethnic Group Subsistence Technology

High Zones AGumba Foraging, hunting
Botanical knowledge
Maasai High range "buuri" and
lowland grazing 15

Middle Zones Southern Cushites agro-pastoralism
Highland Nilotes agro-pastoralism
(Mwoko/Sirikwa) knowledge of water
Tumbiri/Thagana Bantu clans agriculture, iron
new cultivars

Lower Zones Maasai/Samburu cattle
military organization
Ndorobo/Aathi hunting,
knowledge of plants
and wild animals.
Borana/Rendille cattle/camels

Each group practiced a dominant subsistence strategy, and each contributed different
technological features to the system. The ultimate result of the process was the

15 Buuri is the Meru/Maasai term for the high altitude grass lands, for the most part occurring in the
Mt.Kenya rain shadow. The Maasai were apparently using this and parts of the Mt. Kenya forest as grazing
land before the Meru moved into these areas.

consolidation of the system within the borders of Meru ethnicity. Chanler (1896)
established a base camp in Thaicu for his forays into the NFD during his 1892-93
expedition. In reading his accounts there is no mistaking that the Igembe and Thaicu were
two distinct communities, a mutually shared attitude that remains until the present. This is
the one exception in the historical pattern of integration and synthesis in the Nyambenes.
The Nilo-Cushitic Mokogodo also maintained a distinct identity, but removed themselves
from the sphere of Meru influence by migrating to a peripheral area between Isiolo and
The ecologically similar environments in Meru and the Kenya highlands therefore
accomodated sociocultural adaptations constituting a distinct medley of developmental
trajectories.17 Sensitivity to common environmental initial conditions influenced
population growth, migration, and symbiotic processes prior to the inclusion of local
systems within an external political economy. Assuming a central role for the
demographic-environmental parameter, and that "population growth and a chain reaction of
economic and social changes underlie cultural evolution (Johnson and Earle 1987, 5),"
understanding the chain reaction that follows becomes the practical concern for tallying and
interpreting tsociocultural variables to describe the congruencies in developmental
processes. 18

The Meru Cultural Endowment

We will now attempt to explicate the dynamic relationships between land and resources,
cultural institutions, and both the ideology and practice of conservation within the farming
systems of the Nyambene region. The Meru cultural endowment consists of an
environmental sensibility which manifests as a set of agricultural practices and formalized
relationships with other groups. Internal and external relations are guided by local political
councils that meet locally and on the Njuri Ncheke level. Each of these components will be
examined in turn.

16The identity of the Mokogodo confused British administrators who saw them as a branch of the Samburu.
Sutton (1974, 160) associates them with the Yaaku, an ancient Southern Cushite community.
17Contemporary differences among the Gikuyu. Embu, and Meru areas continue to resist facile comparison,
a point underscored in recent studies in each respective area of Mt. Kenya (e.g. Haugerud 1984, McKenzie
1989, Kitching 1980, Chege 1989, Leo 1984, Goldsmith 1988, Glazier 1985).
18 The historic advantage of anthropology as a discipline is "the ability to see humans as biological
organisms and creatures of culture, to see the cultural and the biological as interacting and complementary,
and to see humans in ecosystems as well as social systems (Keesing 1976. 15)." is largely negated by the
internal wrangling over these two positions.

Numerous elements of Meru society reflect the cultural influences of earlier populations
in the region. The Njuri Ncheke, however, was a uniquely Meru governance structure
which recreated and institutionalized the Meru clans original unity across the different
sections' geographic and social borders. The Njuri Ncheke system is organized around
both the territorial unit and the generational strata. Land is linked to the Njuri Ncheke
through its local clan-based councils or njuri. Kiama are comparable councils operating on
each rung of the age set system.19 The kiama councils select their most able leaders,
agambi, spokesmen (kugamba=to be wise, judicious), to sit on the local njuri. The system
replicates itself on successively higher levels, the highest being the pan-Meru regional
The Meru's Gikuyu and Embu agricultural neighbors practiced local government based
on the njuri and kiama councils without developing a formal council structure joining local
units. Age set organization was also universal in the region, although perhaps most highly
developed among the pastoral Nilotic populations. Age-sets, besides providing for defense
of the land by maintaining a standing army of warriors, also regulate internal social control
by specifying a codified set of social obligations within and among age grades at each stage
in the progression through the lifecycle. Clans governed access to land and natural
resources, just as the age-set system provided a standing army to protect (and acquire)
them. Kinship, real and fictive, conditioned exchange relations.
The ethnoscience and cultural ideologies reflecting the diverse populations became
incorporated into the Meru polity over time. Environmental knowledge was developed by
all the groups in the region specific to their respective ecological niches. Bantu clans were
agricultural experts, pastoralists specialized in animal science, and the sophisticated
botanical tradition, evident in agricultural, medicinal, and ritual practices, is arguably the
legacy of the autochthonous hunters and foragers. All of these components combined
synergistically in the Meru system. We will now discuss how these different components
worked in combination to generate sustainable social development.

The Njuri Ncheke:

After crossing the Thagana and encountering the military power of the Maasai and Mwoko,
the Meru prayed to God for a leader. When a young boy displayed the divinely indicated
traits, the Meru selected nine elders to act as his guardian until he came of age. The boy

19 Both njuri and kiama are glossed as "council" in English. Because these terms refer to contexts in many
cases, the local terms will be retained in the text.

became the first Mugwe or Mukiama, and the nine elders formed the Njuri Ncheke,
alternately known as the Njuri ya Kiama, the council where selected representatives from
different group's local njuri met to settle disputes and decide policies for the greater Meru
polity. These representatives, or Agambi (sing. mugambi), met at Nchiiru, now a small
market center in Tigania near the border of Imenti, the largest Meru sub-tribe which
inhabits the slopes of Mt. Kenya. All members of a locality could belong to the kiama,
while only chosen individuals underwent formal internal initiation into the higher grades of
the council based on criteria of public service and virtue.
Lambert, the Meru District Commissioner between 1937 and 1942, saw local structures
as the key to governance in the District, and his description of the njuri system is worth

When a man leaves the warrior grades he becomes eligible for initiation into the Kiama kya
Nkomnango, and in general all elders become members of this body. The elders of each mwiriga
[clan] choose representatives from among their numbers to be initiated into the Njuri ya
Kiamna, the working nucleus of the whole kianwma. These men--the Njuri--are the parliament and
the supreme court of the tribal unit they represent. Thus the legislature and judiciary of the
mwiriga are the njuri elders belonging to that mwiriga, who will join in with the similar body
of the next mwiriga for the settlement of matters affecting the two. A section of the tribe is
governed by Njuri Ncheke elders representing each mwiriga in the section: and the whole tribe
by Njuri elders representing each section (1947, 6).

The second stage of njuri initiation inducted members into the "locked, or closed"
council of the Njuri Mpingire. Initiation into the Njuri Imbere, referring to "the winnowed
ones," was the highest, or "pure council of elders," restricted to an individual "who has
never had sex outside of marriage, a person who has never shed blood of a relative, or in
unprovoked attack, stolen except from another tribe, and above all a man of good health
and good reputation (Rimita 1988, 50-51)." Initiations entailed specific contributions at
each level, so the individual members of the Njuri system were men of material means in
addition to distinguishing themselves by their adherence to ethical standards.
The Mugwe (Mu'we among the Tigania and Igembe sections) was the group ritual
expert and leader who presided over the religious and cosmological affairs of each Meru
section. The Mugwe sat with the Njuri Imbere, the highest ranking council within the
Njuri Ncheke, presided over important social rituals such as the initiation of a new riika,
and was consulted in all the important decisions of the group. Chosen from the grade of
ritual elders but separate from the njuri/age-set institutional structure, the Mugwe
represented in personal deportment the highest social ideals of conduct, personal integrity,
and sexual behavior in Meru society. Social change vitiated the ritual functions of this

purely religious authority, and the office perished with the death of the last Igembe Mugwe
shortly before Kenya's independence in 1963 (Bernardi 1989).
The authority and efficacy of the Njuri Ncheke rested on consensus-based decision
making, its political legitimacy, and the supernatural authority accruing to it as a body of
elders. The council set social and environmental policies, arbitrated disputes, established
new policies during times of crisis, and had authority to assess fines and penalties for
individuals and groups. In extreme cases Nyambene council passed death sentences or
marginalized individuals by sentencing them to periods of exile in the inhospitable northern
grazing area. The council could call upon warriors to enforce law and order and police
markets, and each age-set was taxed in the form of sponsoring rituals and group feasts for
other groups and riika.
Supernatural controls contributed to the authority of the njuri councils within the
community, but the council was not as such a body of practitioners of supernatural arts.
Supernatural phenomena took several forms. Individuals and groups were targets of
different kinds of curses. Violation of socially prescribed behavior could also ipsofacto
place the individual in the state of mugiro, or ritual impurity, ostracizing the individual from
normal social intercourse, and serving as a forewarning of mystical misfortune. The
individual then had to confer with the particular ritual specialist, of which there were
several varieties, to remove the curse. The curse in effect functioned to make crimes or a
social offense public, and seeking to cleanse oneself of mugiro amounted to admission of
guilt (Fadiman 1982). The Mugwe, in contrast to the supernatural specialists, was a seer
who interpreted omens and dreams, dispensed benedictions, and arbitrated public morality
(Bernardi 1989).
Curative methods and protective measures against unseen forces on the metaphysical
level utilized a variety of natural objects within a complex symbolic system linking the
natural world with the social and religious realms. Respect for ancestral and other spirits
associated with specific places and natural landmarks further linked the environment and
society in what amounted to a long-term relationship of responsible custodianship. This
respect/fear relationship influenced the manner in which land transfers were conducted.
The dominant agriculturalists could have easily taken by force forest areas from the
autochthones. Instead, however, use rights were conferred through ritual, exchange, and
in the end, as demographic pressure increased, intermarriage and integration of the forest
dwellers into the agricultural polity (Muriuki 1974, Lambert 1949).
In Igembe the clan names of groups bordering the Nyambene (Aathi) and Ngaya
(Amwathi) forests bear witness to the presence of these hunter-gatherers. Group land
ownership entailed a dimension of spiritual stewardship. Lambert observed that "...the

dead and those who are still unborn are as much members of the group as the living (1949,
114)." When the land was settled, the Meru would have wanted to establish some kind of
kinship relations with the spirits of the land they now occupied. The living acted as the
trustees for the souls of the dead and those not yet born, and the Meru settlers respected the
rights and power of those ancestral spirits of a locality who had exercised dominion over
the land and resources since time immemorial.
Within the Meru cultural ideology or world view, respect for ancestors in effect merges
with the practical utility of land, both as agricultural land and in its pristine state. The Meru
regard the forest as a whole and certain individual trees as sacred. This practical notion of
the sacred is not distinct from the use values of forest products and understanding its
importance as a social commons, as a magnet for rain, a refuge from enemies, and as a
place of God.20 The Njuri Ncheke was the group repository of this secular/supernatural
authority with ultimate jurisdiction over individuals, age-sets, and clans. As a political
body, it appeared to be particularly active in issues and policies related to land.

Age-Set Organization:

Age group systems provide a set of checks and balances among generations that includes
the influence of ancestors and generations not yet of age symbolically represented in the
cycle. Age-set organization conditions a cross-generational concept of social development
and acts as a constant reminder of the obligations of a generation to their own children and
to future generations.
The ethnographic record provides a complex array of variations on age group systems.
Reviewing, classifying, and developing a systematic terminology has preoccupied more
than one scholar (Prins 1953, Bernardi 1982, Stewart 1977). Distilled to its most essential
attribute, age-set systems represent a set of rules embedded in generational social structures
(Stewart 1977, 5). These rules vary according to the system and locality, and govern
different demographic, political, and cultural aspects of the system. Even where it is not
formally embedded in the social structure, the notion of an individual's and a generation's
progression through the life cycle represents a universal characteristic of human society.21
The Nyambene Meru retain an age-set structure whose rules define the recruitment and
formation of an age-set generation, the intervening sequence of progression through age-

20The Meru perform special sacrifices in the forest: they used it to hide from the Maasai (and the British
during Mau Mau), and recognize its hydrological importance.
21Americans, for example, refer to succeeding generations with reference to the great depression, the baby
boom, the Vietnam War, and the Reagan years.

grades, and the transfer of political power (ntuiko) from one generation to the next.22 In
this manner age-set organization provides for the uniform progression of male generations
through the lifecycle. Each age-set, or riika, pass through life sharing common
responsibilities to society, obligations toward each other as individuals, and to other
generational groups. The Nyambene Meru system consists of eight cycling age sets
divided into two moieties.

Table 3.2: Nyambene Riika and Moieties

Mbaine (old) Ntangi
Lubetaa Miriti
Guantai Gichunge
Kiramunya Ithalie
Micubu Ratanya

Age-sets fell into two alternating streams, the mbaine and the ntangi, representing the
initial two age-sets in the Nyambene succession. Three circumcision groups initiated at
five year intervals form an age-set. The first is called Ndunguri, the second Kobia, and the
third Kaberia: both age-sets and circumcision groups repeat in cycles of eight and three
years.23 The initiation period for a riika covers ten years; five years later circumcision of a
new riika begins. Circumcision may deviate from the set pattern according to
circumstances such as famine or war. When this happens, circumcision dates are adjusted
to compensate for deviations.
Meru oral traditions state that they adopted the military organization of the Maasai and
Samburu as a matter of self defense, and out of respect for the social discipline it instilled
(Rimita 1988). The Igembe also comment that their circumcision rites are said to resemble
those of the Abagusii and Kalenjin, pointing to greater Mwoko influence than allowed for
in the historical accounts of both local and outside investigators.24 This would also explain
why Nyambene age-sets are cyclical as are those of the Kalenjin while Maasai age-sets are
of the linear model. Cyclical age-sets encourage a subtle cognitive approach to time as
possessing circular qualities, and the Nyambene Meru sometimes refer to particular age-

22Fadiman (1982) provides a detailed account of Meru age-sets.
23Individuals are identified by their riika and circumcision group, e.g Lubetaa Kobia, Minrit Kaberia. etc.
24 Local Igembe informants accept that the age-set structure may be derived from contact with Maasai, but
linguistic analysis suggests Cushitic influences particularly in connection with their circumcision practices.
The Mbaine/Natangi stream possibly represents generations initiated after changes in the original
circumcision rites and practices, and not the adoption of age-sets per se.

sets in terms of repeating themes in their historical memory. For example, the Lubetaa
generation is associated with periods of building and development, Miriti with breaking
down, and Guantai, who assumed warrior status in 1990, with periods of upheaval.
Each stage in the life cycle involves social responsibilities: warriorhood, economic
activity and commitment to family, governance, and for those who survive to old age,
sharing of wisdom, historical knowledge, and religious authority are represented by the
gerontocracy. The balances built into the system checked generational conflicts.
Warriorhood was a time of freedom and exploits not easily abandoned for the
responsibilities of adulthood. The accumulation of novice warriors exacted pressure on the
ranks of senior warriors to leave the communal warrior barracks (gaaru) to marry.
Younger warriors pressured their elders by actions which invaded the realm of senior
warrior prerogatives. A generation is subordinated to the one above it within a student-
teacher relationship. Abuse of the teacher status is checked by the same relationship
between the generation who are fathers to the "students," and their "teacher" generation.
Table 3.3 is adapted from Fadiman's presentation (1982, 149).25

Table 3.3: Traditional Meru Lifecycle Progression

Age Life Stage Termination Moiety

0-7 Infant/Child Growth of permanent teeth
7-15+ Uncircumcised boy Puberty
17-18+ Circumcision candidate Circumcision
19-29- Warrior Marriage Mbaine
29-40+ Junior elder First son's ntuiko Ntangi
40-55+ Ruling elder Ntuiko transfer Mbaine
55-70+ Ritual elder Ntuiko transfer Ntangi
70+ The Gerontocracy Death/Ancestorship Mbaine

Formerly, when the indigenous political order was the sole repository of power,
generational leaders typically resisted the passing of their position until njuri intervention
insured the process took place. This transitional period, marked by official handing over in
the ntuiko ceremony, is often characterized by high-spirited and sometimes violent

251 have added the alternating moeties to the scheme, and adapted dates in the elder sequence to match the
fifteen and five year interval rule governing ntuiko and circumcision. The rituals accompanying these rites
of passage involved the expenditure of resources, and the fifteen year ideal intervals were adjusted according
to prevailing factors such as famine, etc.

demonstrations of virility. Transfer of power not mediated by cultural institutions, in
contrast, is a major governance problem in contemporary Africa.

Clans. Kinship, and the Gichiaro Social Risk-Spreading Mechanism:

The mwiriga, or clan, provides access to land and provides the local unit's institutionalized
governance. Most of the original administrative sub-divisions in Meru correspond to clan
names, and in the middle and high zone areas these units can be considered equivalent to
original clan territories. Present Meru clans came to gain stewardship over the land through
complex historical interactions between the Thagana clans and the various groups already
present in the region. Original Meru clans are divided into three groups, red, white, and
black according to the time of day when they escaped their coastal homeland at Mbowa.
These divisions correspond to important ritual duties, and only white clans can perform
specific religious rituals (Mahner 1972).
Since the proto-Meru Thagana clans departed Mbowa as a small body of refugees, it is
difficult to explain the spectacular population growth which made them the sixth most
populous ethnic group in Kenya today other than through a significant incorporation of
other populations. Increasing population was desirable in order to provide additional labor
for production and to provide defense. Kinship practices were mutable in order to
accommodate outsider acquisition of land, and there is little to suggest that the Meru
discriminated on the basis of ethnicity.
Beyond ties of kinship, the institution of gchiaro provided a form of blood-brotherhood
or fictive kinship which bound different groups and different sections of different groups
into a complex web of formal alliances.26 The underlying logic of gichiaro emerges when
viewed as a risk spreading strategy. These alliances do not make much sense as a network
of military alliances--the Igembe are joined to the Imenti by gichiaro, but the Imenti are also
joined to the Tigania, who are not linked through gichiaro to the Igembe. Gichiaro
permitted a member of one group to help themselves to food in the fields or home of their
gichiaro relation, but groups not joined by this ritualized bond are fair game for each other.
Within Meru it created avenues of safe passage, but indicates a broader social logic than
Fadiman's discussion (1982) reveals. Although raiding fulfilled the cultural need for a
warrior group to prove itself, it was mainly a function of need in time of shortfall.

26Chanler's (1896) account attests to the extent of gichiaro as a fundamental adjunct to trade. In a number
of situations Chanler submitted to gichiaro rituals to cement trade agreements, and he actively used the
institution from the Nyambenes to Rendille territory to promote the success of his expedition.

Gichiaro formalized symbiotic relations between agricultural and pastoral groups. The
Igembe maintained gichiaro with the Borana, and even with the distant Rendille, who came
from near Lake Turkana to trade. Visitors to Nyambene markets included Gikuyu, Kamba,
Embu, and other Meru from the highlands, and a host of nomadic pastoralists including the
Maasai, Samburu, Somali, and Borana. As the Nyambene Range became the center of
inter-regional exchange gichiaro not only facilitated interaction and exchange among the
host of ethnic groups in the region but came to incorporate outside commercial interests as
well. Coastal and Somali traders and Europeans all sought to establish gichiaro ties as a
practical measure to facilitate trade and safe passage through the Nyambene region, which
by the end of the nineteenth century had become a main staging area for movement into the
NFD and west to Baringo.
Often blood kinship followed fictive kinship. Such relations buffered groups against
environmental disasters, as Maasai-Meru relations illustrate. The migration of Maasai and
Tigania Meru to Narok during the Kiaranmu (named after the solar intensity of the period)
famine of 1917-1918 was mentioned earlier. During the 1984 drought, the worst drought
of this half of the century, Maasai over two generations removed from contact with the
Meru returned to seek assistance from their Meru relatives in Tigania, and received it.
Gichiaro relations could be abrogated for specific reasons, and this appears to be the case
when the Imenti Meru of Katheri expelled and fought with the Maasai with whom they had
lived in shared coexistence for generations (Mboroki 1972).
Intermarriage, maternal kinship relations (an individual could be granted land by their
wife or mother's kin) and gichiaro fictive kinship facilitated the access of different parties to
land. Igembe and Tigania settling in each others respective territory usually do so through
intermarriage. Imenti Meru and other settlers from outside the Nyambene area, in contrast,
undergo gichiaro rituals in order to acquire land in the new settlement schemes in lower
zones of Igembe.
These practices represent traditional Meru mechanisms based on the idea of group
strategies to promote social strength and well-being. Sustaining population growth meant
the efficient and wise use of natural resources in order to preserve the inheritance of future
generations. The inheritance consisted of, more than anything else, land. Most Meru
consider wealth in the form of cattle or money as essentially impermanent, whereas land is
fixed a security worthy of long-term investment. Land straddles the intersection of material
production and the institutional factors of clan, council, and age-set, and the cultural
ideology permeating them.
Land, natural resources, the supernatural-religious realm, and cultural institutions are
interlinked. The Njuri Ncheke sanctioned the environmental policies regulating forest and

rangeland as social commons whose use rested on religious authority. The Meru regard for
certain trees and animals as sacred or possessing unique attributes extended to the positive
material social benefits and cultural sensibilities of proper resource management. This
underlying ideology can be discerned in the deeper ethos of conservation embedded in
everyday discourse and actions in Igembe Division.

The Aathi Cultural Ideology27

Comparison of a number of clues indicates that important elements of Igembe Meru cultural
ideologies influencing resource use and management derive from the pre-agricultural
peoples who roamed the East African landscape since the paleolithic era. Any
environmental history and cultural exegis of the Mt. Kenya region must account for the
influence of the areas original inhabitants, although the interactions among these people and
their various Nilotic and Bantu neighbors represent an exceedingly complex phenomena.
Very few of the hunter gatherers of the Kenya highlands maintained a separate existence
into the modem period. In most cases they were absorbed into the expanding agricultural
communities, but their contribution to the overall cultural complexion of the Meru system
deserves a brief discussion of its own.
The map accompanying the account of Chanler's expedition of 1892-93 provides a
detailed account of the Nyambenes and different ethnic group's positions within the region
prior to colonialism, and establishes the near ubiquitous presence of these bands on the
marginal fringes of other communities. Chanler's expedition registers a picture of
highland-lowland interactions perhaps at the brink of precipitous change. Regional
conditions were to undergo an even greater convulsion after he left the vicinity in 1893.
The conversations Chanler (1896) recorded with a group of surviving Ndorobo
resisting incorporation into pastoral and agricultural communities are a poignant testimony
to the demise of (and the ideological commitment to) a lifestyle harking back to the earliest
phase of human culture in eastern Africa:

I asked the old man why they did not go to the mountains settle down with the people there, and
work, and thus be relieved forever from starvation and famine. He said: No. they were fond of their
mode of life; they knew no other; their fathers had lived the same life before them, and they were
unwilling to trust themselves in the vicinage of any other people. The worst time for them was
during the rains; for then they were unable to use their bows, as the strings frayed and broke.
During the rainy season they literally starved; those of greatest vitality surviving, while the weaker
ones died. Their one pleasure is the intoxication produced by honey-wine (279).

27Aathi. as used here. refers to the collective hunter-gatherer heritage in Meru for reasons explained in the

We can hypothesize that some Aathi phenomena resurface in the religious function of
the forest and natural objects, and the pervasive yet intangible realm of magic and religion.
Aathi indigenous knowledge is implicated in the incorporation of botanical expertise, and
supports the influence of an inherited ethos of conservation as a factor in decision making
within land and resource management in Igembe. The configuration of Mt. Kenya
agricultural systems and some of the mystical attachment to land observed among the Mt.
Kenya peoples bear the mark of these autochthonous peoples, even if their role is
frequently obscured in local accounts28. The proto-Meru likely established symbiotic
exchange relations among scattered groups of the coastal hinterland, and their route to the
highlands may even have followed trails established by Boni hunter-gatherers who
migrated up the Tana on a seasonal basis (Stiles 1981). Their encounters with the Agumba
upon reaching higher zones may have reinforced the hunting predilections of some groups
acquired earlier during their migration.
Permanent residence, supernaturally sanctioned land rights, Aathi botanical and magical
knowledge, and chronic demand for labor explain why Bantu agricultural clans integrated
the Ndorobo while the Maasai and other pastoral groups retained their Ndorobo as clients
even if they were in all respects aspiring to be Maasai as the result of contacts ((Berntsen
1979, 112; Lewis 1960). Fundamentally different cosmological values and issues of social
access to the critical resources of land, cattle, and labor differentiate agricultural and
pastoral systems on levels of ethnic interaction and cultural ideology. Agricultural groups
limited by labor rather than land were prone to assimilate outsiders: pastoralists were prone
to incorporate the same groups as lower in status to utilize their skills while limiting their
social access to livestock.
Like Gikuyu traditions, the Nyambene Meru mention the Agumba population as forest
dwellers: short, light complexioned people whose habitations were pits excavated deep
within the forest (Muriuki 1975, Rimita 1982). The Igembe clashed with and defeated the
small Agumba groups in the Nyambenes. Survivors were absorbed into Bantu clans. The
Agumba of Meru traditions slowly melted away into the forests.29 According to a Mwimbe
informant's version of the autochthones fate conveyed to the Kenya land commission in the
1940s, they...

28Reference to these groups varies in content and transmission. The confusion surrounding Aathi.
Agumba, Ndorobo, and Okiek partially stems from the use of these different names in loose and
interchangeable contexts. In highland oral traditions the non-pastoral pre-Meru inhabitants of the region
appear variously as magicians, witchdoctors, ironsmiths, hunters, healers, middlemen, botanists,
29 In 1900 a British party climbing Mt. Kenya encountered a band of impoverished Ndorobo at the extreme
elevation of 12.000' (Mwaniki 1984. 189).

gradually dwindled away with the cutting of the forest, although on friendly terms with the
invaders, until only a few old men were left. These old men then turned into plantation
birds which inhabit the forest this day in large numbers, and they are to be heard talking
their original language together in the forest at night (Orde-Browne, quoted in Lambert
1949, 44).

Savanna-based Ndorobo groups survived much longer, perhaps partially through their
alliances with pastoralists (Blackburn 1976). Their survival also owes to a strong cultural
ideology to their form of existence despite the option for incorporation into other groups.
Commercialization of the ivory trade allowed for a brief upswing in the fortunes of
Ndorobo as hunters, guides, and middlemen. The trade boom in ivory of the 1800s
probably added several decades to their independent existence. The final incorporation of
the Aathi represents the end of the hunter-forager lifestyle, but not its tradition. The
establishment of European conservation effectively alienated the remaining hunters from the
material resources they subsisted on, and the Ndorobo almost ceased to exist, although
native hunting continued into the next century (Steinhart 1989).30 The commercial
harvesting of ivory soon proved to be environmentally unsustainable, and an ancient
cultural strategy largely vanished with the hunting ban. The local knowledge of these
peoples and other influences nevertheless survived via incorporation into the Meru
production system.
At least some aspects of the Agumba/Ndorobo heritage is subsumed within references
to the Aathi. The Agumba and Ndorobo became Aathi clans as the Igembe settlers
absorbed the forest dwellers of the Nyambene forest zones. By occupying land fringing
forested areas they assimilated in the cultural and symbolic sense, their ancestors as well.
High zone agriculture shows the influence of their botanical and pharmacological
knowledge, supernatural practices, and certain cultural sensibilities regarding nature in
general and the forest in particular. This knowledge and Aathi cultural traits appear in the
Njuri Ncheke code of conservation, and the ideology of trees manifesting in Igembe
agroforestry practices. The high cultural status of stately fig trees (Ficus natalensis and
Ficus Graciolor) among Bantu peoples is well known; a number of other trees are also
considered sacred in Meru, have important ritual uses, or are associated with magic.
Cutting the forest in any way was subject to punishment by ostracization.

30The 300 or so Ndorobo listed in early Meru District reports are the last specific group of hunter-gatherers
mentioned in Meru District: known as Mwethi or Mwimbe Dorobo to the Meru. Imwesi to themselves.
they even took place in the 1934 District Sports Trials as "fine handsome fellows, dressed Maasai fashion
(Lambert 1949,13)."

The word Aathi is generic throughout the Bantu-speaking region. The original meaning
of hunter is overlaid with various connotations according to context. In the Nyambenes,
the connotation is often linked with magical powers. According to Fadiman, njuri ya aathi
were associated with fringe societies of hunters and blacksmiths who hived off from the
main body of proto-Meru during the migration from the coast. Fadiman refers to them as
fringe groups of Meru; others use the word to refer to non-agricultural foragers whose
descent may reflect any combination of Agumba, Ndorobo, Okiek, Sirikwa, Waata, or
other such groups. Mwaniki indicates the labels were cultural appellations that embraced a
number of autochthonous and possibly early Cushitic ethnicities (1984, 177). Lambert
(1949) considered the Aathi to represent a cultural stage, while he referred to Agumba and
Ndorobo as specific groups; individuals and groups passed through or reverted to this
stage until the recent present.
Hunting is a low-caste activity, nevertheless, it remained a common survival strategy for
poorer members of the community well into this century (Steinhart 1989, 247). Since
agriculturalists resorted to hunting (and sorcery) during times of poverty and famine, it
follows that Aathi represented an occupational mode, and substituted as a generic ethnic
label coterminous with the lawful inhabitants of forest reserves.31 Bernsten (1979) records
that the Maasai depended upon symbiosis with the Ndorobo to maintain economically their
cultural ideal of pastoral purity, but they too would frequently "go Ndorobo" as
individuals, and in large numbers during times of severe drought. The same was true of
the Gabra who became Waata foragers during drought, and sometimes remained so out of
preference for the hunting-gathering way of life (Robinson 1979, 8). "Aathi" thus
represented a cultural stage, but also a Meru sub-identity derived from the incorporation of
Agumbaand Ndorobo peoples.
Mention of the Njuri ya Katheka Kai crops up in connection with two 'secret' societies,
the njuri ya kaitha and the njuri ya aathi. Members of the Njuri ya Kaitha were
vegetarians, and employed vegetable poisons against their victims, and items of botanical
origin in their remedies and charms. The Njuri ya Aathi were, in contrast, hunters and
utilized medicines derived from animals. Both entities were originally groups who
remained outside regular clan structures and developed their own councils, as distinct from
other Meru who reverted to hunting when conditions necessitated it (Fadiman 1982, 18).
These groups relied upon their hunter and blacksmith reputations for supernatural power,
using the practice of public curses as a form of social control and as a subsistence strategy
as their resource base inexorably came under the control of agricultural communities.

31Agumba. in contrast, is a low-caste, derogatory appellation among the Meru.

These mystical "weapons of the weak (c.f. Scott 1985)" worked to guarantee group
survival during the transitional period between association with the Meru polity and full
The evolution of these groups clearly follows shifting patterns of land use and political
control. With integration these groups developed a new role, biamnia ya mithega, or
councils of magic, as land became converted to agrarian uses. The members now used
their aathi magic to protect the remaining land not cultivated, and later, to protect crops
from intruders. During the periodic famines of the 1800s the application of magical
techniques were turned against the owners of crops to procure food: the magicians even
imprisoned wealthier farmers inside their farms until the owners paid fines in the form of
grain and beer to remove the curse. A still more novel form of kiama emerged after the turn
of the century. The kiama kya mwaa, or party of fools, got its original members from
famine victims. The kiama kya mwaa toured an area performing ribald dances and, once
they succeeded in provoking laughter, they would curse the spectators for lack of respect.
A nominal payment removed the curse. The agriculturalists derided these bodies as 'bianima
of the stomach', but nevertheless feared their supernatural power (Fadiman 1984, 19-24).
The cultural endowment of the Meru can be briefly summed up by reiterating how the
cultural institutions of riika or age-sets, the Njuri Ncheke, and the religious authority of the
Mugwe combined synergistically as a set of legal, and metaphysical checks and balances
among generations. The Mugwe exemplified moral and ritual conduct, and the Njuri
Ncheke guided social development. The system worked well during the sixteen generation
span of Meru precolonial history. The Meru prospered, surmounted environmental
disasters, absorbed other communities, and grew in numbers in the process of realizing
their own version of manifest destiny on the highland-lowland interface.

The Njuri in Transition:

The first administrators in the district perceived the Meru political/legal system and its
institutional structures in terms of superstition and sorcery. In 1924 the District
Commissioner (DC) commented that "the district is still backward, belief in witchcraft
prevalent, especially in Tigania and Igembe (Kna/DC/Mru 1, 1924)." The authorities of
this period conducted an undeclared campaign against the njuri system, and considered two
"secret" societies, the Njuri ya Kaitha and the Njuri ya Aathi, as part of it.
In the 1927 district report the DC stated that the Nyambene indigenous authority was a
system of lower and higher kiama. He noted that these societies were part of the njuri
system, and declared that their members were well represented on local tribunals. The

Kaitha and Aathi societies sorcerers used their powers for extortion: bewitched individuals
would fall sick and die if they failed to pay a fine. Payment was a de facto initiation into
the kiama, disclosure of its secrets invoked another curse. The report conveys a certain
degree of confusion as the "Njuri Impingire" and the "Njuri a Katheka Kai" were all
lumped together as part of the formal kiama system.
Katheka kai specifies a place of witchcraft or evil, and was associated with the Aathi and
Kaitha societies, although the report adds all elders have supernatural powers (Kna/DC/
Mru 1, 1927). A campaign initiated the next year targeted the houses of Kaitha and Aathi
members, and the anti-witchcraft campaign of the next several years was apparently so
successful that any reference to it and the Njuri ya Katheka Kai soon vanished from the
official record. It is interesting to note that the Njuri itself, seen as an unprogressive
institution opposing the authority of the British administration, subsequently came to be
perceived by British administrators as a model of progressive state-society partnership.
The Njuri was especially instrumental in advancing state supported environmental policies
and land reform. The question that naturally arises is, did the Njuri change, or,
conversely, did a change in the perception of the local administration lead to this reversal of
Situations where the Njuri Ncheke allowed itself to be associated with low status groups
parallels similar strategies of sabotage, foot-dragging, and other forms of systematic non-
compliance with the authorities in power that peripheral groups had employed against the
Meru polity. And the Meru leaders sitting in the kiama and njuri councils were clearly
separate from the British-deputized local headman and chiefs. If the former were cast in the
role as opponents to change, the latter were typically opportunistic individuals chosen from
outside the local political hierarchy who exploited their position for self-aggrandizing
accumulation of wealth and patronage resources.
Kamba and Gikuyu chiefs acted to consolidate their position by recruiting private armies
and corruptly accumulating resources allowing them to live well beyond their official
means. Their excesses were tolerated by the administration as "a necessary ingredient of
local administration and the fuel of the administrative engine...without it the whole system
would have collapsed (Tignor 1979, 199)." The creation of local administration began
from "colonial policies, conceived largely in ignorance," which "tilted the balance against
the stabilizing egalitarianism of of gerontocratic democracy (Forbes Monro 1975, 65)."
Administrative chiefs and emerging mission-educated elites clearly set local societies on a
new path of political development where traditional status and individualistic opportunism
eventually inverted the social order (Mutiso 1975).

That this did not happen in Meru, and particularly in the Nyambenes, denotes the
presence of a strong local political order that was capable of neutralizing individuals
empowered by the British. Opposition to the administration through supernatural means
and non-compliance quickly evaporated when perceptive administrators recognized the
social legitimacy of the Njuri system.

Njuri Ncheke-British Administration Cooperation:

In 1933 the first mention of a policy shift was signaled by the District Commissioner,
A.G. Hopkins. Hopkins, speaking of Tigania and Igembe, accused his administration
chiefs of superstitiously bending to the will of the Njuri Ncheke every time the council
exercised their veto over an administration policy they did not support. Hopkins branded
the institution as possessing "an attitude of extreme distrust of government and the use of
their powers secretly to condemn any order or suggestion emanating from government."
Despite this condemnation, Hopkins goes on to suggest that the first step to reduce
suspicion would be to recognize the Njuri Ncheke's important position in Meru society
(KNA/DC/Mru 1, 1933,7-8).
His successor in 1938 and 1939, V.M. McKeag, adopted a more liberal view of native
custom. McKeag took a certain anthropological interest in the Meru system. Although he
suppressed the Lenta dance, which follows circumcision ceremonies, as an orgy of
consumption and fighting, he recognized the role of social ceremonies as social
mechanisms for providing the entire community with important animal protein supplements
to their daily diet of grains and tubers. His annual report for 1938 catalogues in detail the
types of ceremony accompanying initiations, marriages, births, and other socially
significant events warranting celebration and the consumption of livestock by different social
groups. McKeag also acknowledged the Njuri Ncheke council as the source of important
environmental controls through the examples of fines assessed by njuri elders for offenses
such as cultivation in prohibited places, felling prohibited trees, failing to adopt erosion
control practices, and setting unauthorized grass fires.
H.E. Lambert served as D.C before McKeag in 1934 and 1935, returning for another
short stint during part of 1939. He completed a third tour of duty in Meru District in 1941
and 1942. Meru clearly inspired Lambert with the desire to investigate and analyze social
change on the local level; the extensive details preserved in his district reports and
subsequent papers are a valuable source of information. Lambert, like many of the British
officials of his time, took a keen interest in local environment conservation. Unlike many
of his colleagues, however, he also developed a keen appreciation for native agriculture.

Although he shared the British colonial obsession with soil and forestry conservation
(Anderson 1992), he also recognized in the local institutions many common developmental
interests. He successfully enlisted the Njuri in a cooperative campaign to promote local
commercial agriculture and environmental protection.
Formal recognition of the Njuri Ncheke was one of Lambert's initial actions. Chiefs
and headmen paid in cash, he observed, "have no place in the customary scheme of
things," and he saw recognition of the Njuri Ncheke as the first step in a progressive
direction: "it is there and will continue to remain there; no amount of ignoring its existence
will cause its existence to cease or reduce its power as the prime indigenous governing
body (KNA/DC/Mru 2 1934, 37)." During the period spanned by Lambert's intervals as
D.C., a number of agricultural innovations involving new food cultivars, the use of inputs
such as cattle vaccines, and the introduction of coffee, took root in local society. Lambert
also recognized the progressive nature of Meru agriculture and threw his weight behind a
number of local practices.
Lambert was an unusual representative of colonial government insofar as incisive
critique of some European behaviors accompanied his support for indigenous institutions.
He criticized local Europeans for presenting to the natives an image of "big business,
hedonism, and ephemerality," and he singled out some Christian missions for proselytizing
without educating, and ignoring the fundamental Meru spirituality based on the belief in
one God: "If the African would retain his own faith all would be well. But he will not; he
will imitate what he thinks is ours." On the problem of progress, Lambert goes on to say
that, "Government should cooperate with the right missions in restoring--in better form--
the ancient faith of the African. And that better form of Christianity based on the
indigenous creed and retaining as many of its social sanctions as are, or can be made,
compatible with it (KNA/DC/Mru 2/1939, 32-35)."
Lambert's specific criticism of some missions--he singled out the Italian Catholic
missions as setting a particularly slothful and corrupt example for the locals--did not
prevent him from campaigning with a large measure of success against other less
progressive local customs including female circumcision and cultural injunctions against
warrior agricultural labor, which he justifiably saw as removing the most physically-able
segment of the population from the most important productive sector.
The new administration-local coalition ceded to the njuri councils the regulation of
grazing, protection of hills, forests, and other resources. Following this new strategy,
based on cooperation, Lambert observed a reversal in hillside cultivation and lack of
erosion measures, the building of trenches to protect crops from wildlife attacks (which
failed to stem the incursions of wild pigs), and a reduction of tree poaching, which in any

case was attributed to Gikuyu settlers in the district. Reports filed during the pre-World
War II period consistently refer to Meru conservation versus the Gikuyu "get rich quick"
attitude, the threat of Gikuyu timber merchants invading Meru forest reserves, and the
dangers of Gikuyu political campaigning among the Meru population.
The Gikuyu problem was rooted in objective conditions different than those prevailing
in Meru, which undercut the threat of Gikuyu land consciousness infecting the district after
World War II. The growing unrest in Central Province did indirectly influence later
developments in Meru. These probably owed to internal factors: divisions in the polity
resulting from missionary opposition to traditional practices, generational conflicts, and
reactions to colonial antipathy for specific traditional practices, as much as opposition to the
British presence itself. The Njuri, for their part, interpreted the international conflict in
local terms, and provided a measure of material assistance to the administration during the

"Hitler is an njama destroying his own people;
Churchill is a Mugambi of the Njuri of the King."

Support for the anti-colonial cause gathered, as to be expected, in the traditionalist areas of
north Imenti, Tigania, and Igembe, but there were incidents of Meru-Gikuyu conflict as
well.32 Another episode of aathi witchcraft led by a government chief in Tigania broke
out; the accidental spread of contagious bovine pleuro-pneumonia into the cattle population
by the vetinerary department spurred local resentment, but in general the war did not
overly upset the routine of Meru life (KNA/DC/Mru 4, 1945).
World War II and the Mau Mau rebellion set in motion a colony-wide revision of native
agricultural policy meant to dampen rural sector forces supporting the liberation movement
(Anderson 1988; Berman and Lonsdale 1992). Meru coffee planting which had begun in
1937 accelerated after the war and commercial agriculture, more than political demands,
precipitated a revision of land tenure in Meru District. Traditional tenure was based on clan
ownership as in other parts of Africa, and also recognized rights of the individual based on
tree tenure--planting bananas, yams, and miraa in Igembe and Tigania was a sufficient
guarantee of individual custodianship. Land could be conferred to outsiders through
marriage and in other cases, by the approval of the clan elders. According to a 1942 Njuri

32Meru--Gikuyu conflict, including one over women in Naari and an attempt by Gikuyu settlers to
cultivate land next to the Ngaya forest in Igembe requiring Njuri intervention also blunted potential support
for the movement.

ruling, land transferred could also be redeemed by repaying the initial purchase price
(Homan 1960).
The growth of the coffee industry called into question the principle of redeeming plots.
In 1949 the Njuri established additional payments as compensation for coffee plantings and
other related investments. The Njuri was also involved in establishing policies for
settlement schemes in new zones, reallocation of land for the purpose of consolidating
scattered holdings, and the registration of titled freeholdings. The slow pace of Njuri
deliberations on these issues led to the formation of local land boards to adjudicate
individual land claims in each area, but the role of the Njuri as the unquestioned custodian
of group and clan land control insured its survival into the post independence period
(Homan ibid.).
The Meru anticipated independence by building a new structure at Nciiru to house the
Njuri Ncheke. The Kenyatta government, however, saw the Njuri as an alternate local
government and restricted its operations to the local level. The Pan-Meru council was not
able to use the new parliament building, and the importance of the Njuri and age-sets faded
among the Mt. Kenya Meru during the post-independence era. In 1979, the Meru town
council nevertheless devoted one side of the monument commemorating the ten year
anniversary of the Moi, "Nyayo" government to a bas relief of Njuri Ncheke elders.
The Njuri and age-set organization remained strong in Tigania and Igembe where the
continued vitality of the Nyambene cultural endowment was supported by the distinct
complex of commercialized traditional agrarian production.

The Nvambene Farming System:

The commercialization process in Meru was uneven, and reflected inherent cultural and
spatial aspects. The Mt. Kenya side of the district responded more enthusiastically to
missionary efforts, education, and progressive measures supported by the administration.
Small scale coffee production began in 1934 under administration supervision in selected
areas of Meru and Kisii districts and accelerated during the post-war decades. Coffee
carries with it the association of being a progressive crop due to its importance for the
settler economy and its connection with modem sector institutions; still, relatively few
Nyambene farmers adopted it as their primary cash crop even as production rapidly

expanded after independence (Bernard 1972).33 Instead, the production of miraa (Catha
edulis), or khat, led the shift to commercialization in Igembe Division. Coffee has never
been able to compete with miraa in returns to the farmer.
The indigenous Nyambene agroforestry system differs radically from production
systems based on monocultural tea and coffee production. In terms of returns to land and
labor, and reinforcement of conservation practices, the Igembe farming system is among
the most profitable and efficient agricultural systems of its kind in Africa.
Miraa was originally cultivated for ceremonial and social uses. During this century the
farming system responded to commercial demands and the cash needs present in the
modern economy without sacrificing its natural environmentally sustainability in terms of
minimizing chemical inputs and maximizing biodiversity through a variety of local multi-
purpose tree species, and integrating high value commodity production. The harvesting,
packaging, and marketing of miraa is labor intensive and generates substantial off-farm
employment. Miraa profits financed much of the subsequent coffee and tea planting in the
vicinity, and in general sustains the high population of the northern Nyambenes.
An acre of high quality miraa can generate, at current rates, many times over the
financial equivalent of an acre's annual production of of arabica coffee in Meru without
investment in chemical inputs. The same acre of miraa also provides food, fuelwood,
building materials, high quality livestock fodder, and herbal medicines effective for treating
both humans and domestic animals. Multi-story farms minimize erosion and insect
problems, trees pump and recycle nutrients from the subsoil, reduce leaching of nutrients
by lowering soil temperature, and increase soil moisture retention,. Careful management of
the vegetative canopy maximizes sunlight intercept for understory crops. Inputs are limited
to manure and ashes, and a periodic smoking by burning dried cuttings and branches to
limit insect infestation. Insecticide use, increasingly popular during the 1980s, is declining
following a recent Njuri Ncheke ban proclaimed in 1991.
Mature miraa farms are the product of generations, and while a miraa tree will produce
good income five years after planting, it only reaches full maturity after one hundred years.
A young tree is called muthairo; as it matures it becomes the respected and economically
valuable mbaine. The truly old mbaine stand out like the older trees in natural forests.
There are trees that can be accurately dated back over two hundred years in the heart of
traditional cultivation. Specimens of yet older trees are said to pre-date the establishment of

33Despite the economic success of miraa in Igembe. its association with unprogressive traditional
behaviors influenced some farmers to plant coffee as a less lucrative alternative for the modernity the
national institution of coffee represented.

the riika system among the Meru, and stand as tangible representatives of group histories
and cultural continuity.
Small groves of mbaine continue to serve as the meeting place for local njuri. The
miraa trees are so valuable that the Meru have developed a science of arboreal surgery for
old trees attacked by a bacterial disease, Aspirilla mellea.34. Diseased spots on the limbs
are excised without damaging the tree itself. Underground infections require the deep
excavation of the tree to remove infected portions of the root system, after which the soil is
returned and the tree resumes its normal existence.35
The average land holding in Igembe Division is three acres. In the high zones where
average holdings are usually smaller, even a developed farm less than an acre can support a
family. The work generated by the picking-tying-wrapping of miraa, transporting and
selling of miraa, provides employment. The healthy income distribution associated with it
is high in comparison to many other commercial agricultural commodities in Kenya.
Marketing miraa exposes warrior-grade youths to other economic and social influences.
This, and a high level of household self-sufficiency, help the Igembe maintain a robust
rural economy relative to other small farm regions of the Kenya highlands.
Results from on farm surveys record twenty-nine species of soil fertility enhancing trees
(all but four were indigenous varieties). Some on farm trees help retain soil moisture; two
varieties are reported to repel insects, and there were several cases of succulent plants that
are buried in the ground with seedlings to aid their establishment for the slow release of
moisture. The nitrogen-fixing potential of local tree species, synergistic properties of
intercropping patterns, and some of the ingenious techniques farmers employ represent
hypotheses waiting to be tested by formal methods. A priori prejudice against Catha edulis
has deflected professional interest in the properties of the Igembe farming system up to this
point in time, even though Lambert called for research into local systems over five decades

It should be our business to discover the principles underlying native agriculture and to improve on
rather than inculcate new ones of which the long-term effects are unknown to us. We can generally

34 The infection is called ura locallyUra. Ura also attacks tea, bananas, and many other trees, destroying
roots and turning heartwood into dry rot.
35The Igembe also describe the root systems of different trees. Miraa has an extensive but not deep root
system, although the roots remain deep enough as to not interfere with crop roots. Mwenjela, in contrast,
has a very deep root system that pumps water up from sub-soil horizons. Farmers say that the moisture of
the succulent mwejela roots "cool" the soil and help diffuse moisture to other species and crops.
36Miraa is exported abroad in increasingly greater volume, earns the country foreign exchange, and planting
is expanding rapidly in parts of Meru district where it is not cultivated. Institutional research offers limited
returns to the sophisticated, long-term Nyambene miraa production tradition in comparison to the potential
research benefits for other traditional food components of the system (e.g. Lipton 1988).

find something in native practice on which to base our teaching. Manuring, wash stopping, stall
feeding, for instance are not new ideas to the African; all we have to do (and it is plenty) is to
select, develop, and extend the ideas already there (KNA/DC/Mru 2, 1939, 59).

A comparative example underscores the influence of the Meru cultural ethos on
agricultural strategies. The Kenya Woodfuel Development Project (KWDP) study of
household tree planting in Kakamega, Kenya established that tree planting and percentage
of vegetative biomass on farms increases with population density and decrease in farm size.
The KWDP, however, operated on the assumption that farmer tree planting is an
agricultural innovation (Chavangi 1992, 152).37 The project achieved some success in
absolute numbers of seedlings established in certain locales. This is offset by cost/benefit
factors and the question of continued support from the project independent of donors
Considering the two districts' ecological and demographical similarities, Meru small farm
practices are sophisticated and rely upon local ethnoscientific guidance to the degree that
need for external assistance does not even arise.
The indigenous Nyambene agroforestry complex in particular offers important lessons
for the study of environmentally adaptive agricultural development systems in Africa.
Igembe avoided the famine accompanying the Kiaraniu drought of 1918-19 which exacted
ten per cent mortality rates in other parts of the district (KNA/DC/Mru 1, 1919). During
the 1984 drought far fewer families in Igembe (800) required famine relief than nearby
Tigania (12,000), Nithi (5,000), and the two Imenti (4,598 combined) divisions.38 These
figures reflect both lower ecological vulnerability to drought and greater food purchasing
power. But the Igembe agricultural system's chief claim to sustainability is its adaptive
internal tradition of farmer experimentation, combining innovation and the practical testing
of techniques.39

Modern Dynamics of Nyambene Meru Cultural Institutions;

In Igembe the role of warriorhood transformed to entrepreneurhood as the economy
commercialized. Warriorhood in the Nyambenes is now largely a category of economic
activity, and just as individual riika were expected to distinguish themselves in the

37The KWDP supplied mainly exotic species to farmers and reported varying rates of adoption. The trees
were grown for harvest over the short-term, and results '"suggest that the fuelwood situation for
participating households has begun to improve (Chavangi 1992, 167)."
38Families assisted in Igembe were predominantly from Thaicu, and not Igembe proper.
391In response to survey questions which asked farmers to list the source of new techniques and technology,
seventy-nine per cent listed other farmers and twenty-one per cent volunteered the unlisted category of
individual, on farm experimentation. Comparative figures for the extension service and national media were
twenty-eight per cent and sixteen per cent.

traditional milieu, an Igembe generation is still expected to advance in some way the
welfare of the community. Rigby reports a similar conversion among Tanzanian Maasai
warriors who now bear responsibility for herd vetinerary treatment, procuring water, and
transport within livestock marketing networks (1985, 157).
The first Miriti generation presided over the initial commercialization of miraa. The
contemporary Lubetaa riika distinguished themselves by greatly expanding the sphere of
commercial activity. They are also associated with a local enclosure movement linked to
the shift to formal individual tenure within the farming system.40 The decision to market
miraa as a commodity is recognized in Igembe as a revolutionary innovation, and one Miriti
informant remarked to me how his father told him his generation bears the responsibility
for pioneering economic innovations in the manner of the first Miriti generation.41 These
examples convey how local society sees change through the optic of Meru cultural
In the Nyambenes, age-set structures still contribute to individual and group discipline
and enforce a culturally defined concept of development as a social process across time.
Age-sets are still controlled by the internal dynamic where adjacent generations remain
linked in a student-teacher relationship. The activities associated with each stage in the
lifecycle, however, have shifted. Table 3.4 transposes traditional progression into the
modern equivalents for stages in the life cycle.
Although formal age-set structures reinforce a stable internal order, they are often
associated with anti-progressive warrior behaviors. Condemnation of the warrior tradition
and its effect on political institutions and behaviors obscures important overall features of
the age-set system where it survives.42 In an edited volume on the subject, Southall (1977,
175) joins others who assail warriorhood as a bankrupt tradition at the center of Africa's
political and economic problems: "lacking any ideological anchor, it degenerates easily into
fascism...technological atavism and anachronism...certainly in Africa its fruits are poverty,
dependency, and bloodshed." Adaptation to changing circumstances, not warriorhood or
other African cultural traits, is the core issue, applicable to this and other institutions.

40 This is recognized by the fact that the common hedgeplant, euphorbia tirucalli, is now called mulubetaa
in some places.
41The current Miriti riika, although the first generation to attend school in large numbers, are hard-pressed
to duplicate the Lubetaa economic success and see themselves in competition with their immediate elders.
42The sequestering of initiation groups in the Menegai forest of Kenya's Nakuru District in December of
1991 was reported in connection with rumors of ethnic violence in the Rift Valley, although there was no
direct evidence that schoolboy circumcision groups were deployed when the violence did erupt several
months later. Warrior initiation has been severely criticized in Meru as well.

Table 3.4: Modern Generational Progression in Igembe

Traditional Progression Modern Progression Age-Set

Kiigumi pre-circumcision organization Primary education, herding, farming

Nthaaka junior warriorhood Primary/Secondary education Gwantai
Senior warriorhood-Ramare council Miraa harvesting, urban marketing

Kiama kya Nkomango Marriage, Stage 1 farmnning, Miriti
Junior elder Urban-rural straddling

Njuri Ncheke Settlement, stage 2 farming Lubetaa
Senior elders Off-farm occupations/employment

Gerontocracy-Ritual Authority Family and social authority Ratanaa

A major land dispute involving Igembe grazing lands in the Northern Grazing Area
during February of 1991 provides a contemporary illustration of institutionalized cultural
dynamics. A large group of Tigania living near the Igembe-Tigania border occupied twenty
thousand acres of Igembe land between the Turuka crater and Mulaa hill on the dry lower
margins of the Nyambenes. The contested area, the flat plain of Kiani Kya Mwitu, or "the
dying place," was the site of a major Meru-Samburu battle near the end of the last century.
Word of the invasion spread rapidly, and an Igembe age-set army of Miriti "warriors" led
by Lubetaa "junior elders" armed with mainly agricultural weapons--machetes, axes, and
some spears--assembled at Kangeta market and marched to Kya Mwitu where they engaged
the Tigania in skirmishes. Support vehicles also made the trip and were instrumental in
delivering injured combatants to the hospital.
The conflict involved the Karama clan of Tigania and a segment of the Njia clan in
Igembe. Hostilities continued for several days, leading to the involvement of divisional
administration. The District Officer and the divisional head of police came under attack by
Tigania who stoned their vehicles when they tried to arbitrate. This led to the "knee-
capping" of several Tigania by armed policemen, adding to the ranks of the injured and the
five Tigania said to have died in the conflict. The Njuri Ncheke was called in to mediate the
conflict at this point. A prior meeting of the respective bodies for Igembe and Tigania had
met on October 23 without resolving their differences, but submitted a report and the list of
elders in attendance to the District administration. All the administrative locations in the
two divisions were represented by elders.

In March the council met again. After prayers, individuals addressed the assembly
delineating the need to reconcile the differences over the boundary. Sworn witnesses then
gave testimony which recounted past deliberations involving njuri elders, British
administrators, and post-independence Kenyan officials. The statements were signed and
thumbprinted. The history of the border, down to the individual trees and rocks marking
it, was surveyed with reference to past rituals and social exchanges as indicated in the
testimony of one elder, Daniel M'Amanja:

And you. M'Mailutha, used to sit at Karwamba, the place where we used to exchange miraa
or tobacco. Facing Luthii, the Antuanduru were sitting at Ndubai [Antuanduru is a Karama
subclan], and those who were sitting at Karibubu were from Kangeta. The day that I
circumcised Lubetaa at Maitai ya Ndubai at Ciomathia, were not your boys cut with mine?
Secondly, when I went to Likiau and killed the kudu and got a horn called Rikiau, was I not
with you?..Now let us refer to Kamungine which was the place for the ceremony.
M'Mailutha indicated a tree/pole called mulikinyai; and when M'Akwalu was requested to do
so also, he crossed and went to show his pole of muburwa with lukerenya in the middle--
from those mulikinyai to muburwa markers there were two hundred steps. It was divided
into two, Karama/Kangeta, and there are also poles [showing the border] at Keruria, Small
Kamuciere to Karama, and Big Kamuciere to Kangeta. Big Ndubai [a distinctive crater] to
Karama, and Small Ndubai to Kangeta. The fourth [i.e. marker] is the culvert at the foot of
Kamuciere hill on the road to Maua (Minutes of the Joint. Tigania/Igembe Njuri Ncheke
meeting of 3/90).

The series of meetings established the border according to criteria mentioned in this
account. Important supporting testimony verified that the Karama clan of Tigania and the
Njia clan of Kangeta held traditional sacrifices at Rumai and Turuka, hills aligned with the
respective Karama and Kangeta sides of the border. Maps were drawn to illustrate
positions of important landmarks.
When the Karama rejected the evidence and held to their claim, the Njuri decided to
administer the Nthenge oath to the parties. The curse incurred by violating this, the most
powerful of Meru oaths, is greatly feared. On the appointed day the Karama elders failed
to appear. The combined Njuri Ncheke of Tigania and Igembe ratified the border and
petitioned the administration to formally survey and mark the boundary. The District
Commissioner placed the official imprimatur on the solution by visiting the sight of the
conflict later that year.
The conflict offers a fascinating insight into the confluence of traditional-modem
institutional dynamics. The original action was incited by the local chief for Karama
location (appointed by the administration) and a local member of the county council. The
police and the local administration failed to achieve a peaceful solution. The Njuri Ncheke
settled the issue where administrative intervention failed, but hostilities under the surface
remain: a Karama university student I spoke to conceded the legitimacy of the old border,

but insisted that Karama people's rights to the land is based on the high incidence of
intermarriage between Karama and Njia clans.
The conflict was actually triggered by the division of Muthara location in Tigania into
two administrative locations. After the division, the original Muthara clans promptly
revoked Karama access to the lowland areas of Muthara location. Where access to local
resources still follows the basic unit of the clan, political leaders manipulate clan and
kinship factors: the government-appointed chief of Karama then instigated the issue of his
peoples right to unused land. Land hunger in Tigania, where the high zone agriculture is
based on coffee production, is much greater than in Igembe where returns to land and labor
are significantly higher. In Muthara and Kianjai locations, cultivation has spread all the
way to the borders of Isiolo, extremely marginal land for cultivation. The Karama invasion
was an attempt to obtain new land for future expansion, the Igembe margins of the
Northern Grazing Area are not yet settled.
The njuri system was losing its authority to use warriors to enforce its decisions by
Lambert's time. Now coucils of elders compete with local government and a number of
other institutions. Elders' power lie in their control of many agricultural resources. Future
conflicts may take on a generational aspect as competition grows over internal clan and
family resources. During the early 1980s, I heard younger males charge that their elders
waste the profits of miraa production on their own social consumption. Similarly, younger
individuals, particularly the educated, are resisting recruitment into the local njuri,
complaining that elders use membership to extort monies for the meat and beer used in their
ceremonies. New generational pressures are reflected in the noticable rise in community
support for local education, and schools now rival the njuri as grass-roots level
organizational nodes.
The Nyambene system, despite its conservative, hillbilly reputation, has demonstrated
an internal capacity for subtle and successful adaptation without losing its own internal
structure. Change generated critical feedback energizing traditional institutions even as
some modern institutions atrophied. Agricultural commercialization blended with the old
order to help maintain the system's adaptive qualities. The Njuri Ncheke environmental
ideology continues to provide important ideological inputs as the agricultural system adapts
to lowland farming. The more important issue, not just for Meru, but for the whole of
Africa, is that there is no sustainability without institutions and mechanisms for resolving
the conflicts that inevitably crop up where there is competition for economic resources.


There is no evidence that the qat habit is dangerously injurious to health. It does cause
constipation, but it has never been suggested that it may seriously damage the body as may
tobacco, which is alleged to cause cancer, or alcohol, which may cause liver complaints. Qat does
not create an addiction, like opium or hashish, in that those who are suddenly deprived of it, do not
suffer physical consequences. Deprivation may cause mental distress, but that is all. Confirmed
qat eaters who are deprived of the leaf, when they visit foreign countries, quickly adapt themselves
to its absence. (Serjeant 1983, 172)

This chapter traces the growth of commercial agriculture in the Nyambenes. It overlaps
with the historical sequence presented in the previous chapter, but focuses on economic
rather than cultural and institutional factors. A description of the growth of miraa
production and marketing appears elsewhere (Goldsmith 1988). For this reason, this
aspect of economic growth is summarized, and augmented with information on new
developments in miraa marketing networks.

Precolonial Trade Networks

The phase of local ethnic consolidation detailed in the previous chapter overlapped with
a period characterized by expanding commercial relations among groups. Specialized
commodity production and exchange based on eco-niche specialization grew into multi-
ethnic regional trade networks linking highlands to lowlands. The position of the
Nyambene Range, straddling the ethno-ecological interface, made local markets the most
important in the northern highland region's precolonial economy. It is not difficult to
conceive the spatial advantages of Nyambene markets in the precolonial regional economy.
The pastoral Galla, Samburu, and Somali are proximate to the perimeter of Meru grazing
areas that extend in an arch from the Nyambene foothills to near the Waso Nyiro River. To
the southeast, a short distance across the dryland cultivation areas of Tharaka, the Tana
River begins its curving path to the coast. The gentle plain on the other side of the Tana
leads to Kitui, the center of the extensive nineteenth century Kamba trade networks. Nyeri,
the traditional capital of the Gikuyu, is reached by following routes extending around
Mount Kenya to the southwest; the rest of Meru and Embu district lie on the southerly
tangent skirting its eastern slopes.

The Igembe Meru ethnic trade network, as indicated by the existence of formal gichiaro,
or fictive kinship relations, included all of these ethnic groups, the Tharaka, and the major
sections of the Mt. Kenya Meru. Local markets in the Mount Kenya region were respected
institutions whose neutrality was observed even during periods of warfare (Muriuki 1974,
Mwaniki 1984). Even though Meru-Maasai/Samburu were considered hostile, specific
markets for trade with these groups also existed. Ethnicity did not in any way present an
obstacle to trade, and gichiaro served as the equivalent to most-favored-nation status in the
international economy.
Although harmonious relations among these groups fostered peaceful cultural
symbiosis, the droughts and famines of the eighteenth and early nineteenth centuries
presaged a growing trend towards conflict as the pressures on land resources increased. A
series of droughts between 1825 and 1840 produced more intense social consequences than
the earlier dry cycles in the region. Drought induced famines of 1825-1829 earned names
like "Mbaranganu-the destroyer" in Kirinyaga and Embu, and 'Kithithiria-the finisher," in
southern Meru. It was followed within a decade by two other famines, the "Kiverio-
refusing to help people" famine, and the "Kavovo-hating without reason" famine. The
"sweet potato" famine ofm 1843, as it was named across the region because people
survived by virtue of these tubers, was perhaps the worst (Mwaniki 1972, 6). Famines,
however, were largely localized, and this also contributed to the development of exchange
in the region as groups relied upon exchange relations to buffer periodic shortfalls.
A variety of factors supporting a surge in population growth and the growing regional
and external trade produced a boom in food and livestock production as "networks
overlapped and intersected in intricate patterns that defy reconstruction (Ambler 1988, 87)."

Decline of Pastoralist Power

The growing strength of agricultural groups corresponded to the eventual decline of
pastoral power. Even where linked by symbiotic arrangements, over time cattle herders
became more vulnerable to drought and more prone to internal conflict as rangelands
receded. During the 1840s raiding had replaced the largely laissez faire social relations
characterizing agriculturalist and pastoralist interactions. The pastoral groups living in
pockets and on the fringes of Meru agricultural communities (except areas of Tigania where
extensive intermarriage had occurred) were expelled. In 1856 Meru warriors from north
Imenti defeated the Maasai and pursued them as far as the Laikipia plateau. During the
latter decades of the century, the Tigania and Igembe Meru similarly drove Maasai and
Samburu groups from the lower reaches of the Nyambene Range to the other side of the

Waso Nyiro River.(Mwaniki 1984, Mboroki 1972). By the 1870s Maasai were raiding
coastal settlements (Salim 1975), and the Laikipiak Maasai, weakened by the raids of other
Maasai sections and proximate agriculturalists, self-destructed in a final spurt of
internecine warfare during the 1880s (Sobania 1974, Hjort 1981).
The exception to pastoral decline was the Somali expansion. The movement of the
Somali into present day Kenya, although obviously reflecting demographic and ecologicval
factors, also sprung from a religious revival in Bardera during the 1820s. Somali clans
crossed the Juba River and severely pressed the Boran of northeastern Kenya. The Somali
mixed livestock strategy based on camels, cattle, and small ruminants proved superior to
the Boran cattle-based economy, allowing them to consolidate territorial gains during the
post 1850s droughts (Merriman 1987). During the 1860s Somali defeated the Boran at El
Wak and Wajir. Somali pressure pushed the Boran to the western edge of present day
Northeastern Province. Within a few decades Somali groups had reached Isiolo, on the
doorstep of the Nyambenes.
The decline of Maasai and Boran cattle-based monoculture contrasted with the resilience
of the Somali mixed ecological adaptation. The Samburu and Rendille, present a third
variation where the groups maintained their territorial position through the complex
environmental and cultural symbiosis described by Spencer (1973).

The Impact of Ivory on Regional Trade Networks

Items manufactured abroad started to enter the highland-lowland trade networks by the
early nineteenth century through Kamba linkages with the coast. The first major world
commodity exported from the Nyambene region was ivory. The commodization of ivory
stimulated commercial forces that penetrated into the most remote and inhospitable comers
of northern Kenya, subtlety altering economic relations among groups. Coastal caravans
emanating out of Zanzibar centered primarily on slaves along the routes transversing
present-day Tanzania into central Africa, and on ivory in much of present day Kenya
(Sharif 1987). Parties in search of ivory entered the region with increasing frequency after
the mid-point of the nineteenth century following the same trails, reducing the Kamba to
secondary players.
The ivory boom briefly created a new role for the remaining hunter-gatherer groups who
were able to augment their meager existence by becoming guides, ivory-hunters, and
middlemen between their warrior herdsmen patrons and the traders roaming the area.
During the years prior to and after the turn of the century, the ivory trade became a

multiracial bazaar in the bush as Swahili traders from the coast, Ndorobo, Somalis,
Kamba, and Europeans combed the landscape in search of the commodity.
Judging from the accounts of Chanler (1896), the Igembe and Tigania of the
Nyambenes and the Rendille of the NFD remained somewhat aloof of this phenomena
insofar as these groups traded from a position of strength. Chanler camped in Thaichu and
had to deal gingerly with the Igembe to procure the food and supplies he needed for his
expedition to Rendille country. When he finally found the Rendille, who come across as a
proud and wealthy group in his narrative, he was forced to bide his time and use
considerable diplomacy before completing a deal for small livestock and donkeys with the
white-clad nomads. Five years later, Arkell-Hardwick (1903) describes how the Igembe
ambushed his party and killed the major Somali trader in the region.1 This, and other
accounts indicate that the Igembe did not have to venture away from their area to trade--for
the most part other groups visited them--and were hostile to perceived threats to their local
Like other African commodity booms, the ivory bonanza was short lived. Europeans
became actively involved as hunters, and during the latter years of the century elephant
hunting financed missionary and commercial penetration of the African interior. Procuring
tusks became difficult as elephant populations declined. Not only the elephants suffered.
Expeditions lived off the meat provided by slaughtering wildlife in large numbers. Later,
the European administration banned hunting and demarcated game reserves. Hunting was
transformed into a ritualized activity for sportsmen and a model for imperial training like
Baden-Powell's scouting movement. Before this occurred the European freebooters
manning the hunting expeditions were used to staff the punitive missions being launched
against recalcitrant African populations. Some Europeans even treated the new diversion of
manhunting the same as game hunting, counting their kills in the same terms they used for
their bags of game (MacKenzie 1987).

Late Nineteenth Century Environmental Catastrophes

A number of environmental disasters punctuated the rise of Western influence in East
Africa. Smallpox, rinderpest, jiggers, locusts, and famine appeared on the landscape at
different intervals. Severe drought decimated livestock populations, reduced the human
population, and vitiated local exchange. Trade networks collapsed, replaced by survival
strategies on an extreme scale ranging from the pawning of women and children, migration

1This attack took place on the southern approach to Igembe country, probably around Nkinyang'a.

on an unprecedented scale producing an onslaught of refugees, and finally, the repudiation
of all social and cultural conventions including long-standing ties of fictive kinship in
cases. Largescale raiding grew with the breakdown of law and order. The social linkages
between different groups became more important than economic relations, as the most
affected groups sought succor from kin and patrons who had food, and refrained from
raiding neighboring groups with whom they shared gichiaro relations (Fadiman 1976).
The great famine of 1897-1901 completed what the rinderpest epidemic of 1891 and the
smallpox epidemics of 1890s began. The famine may have caused up to fifty % mortality
among the Kitui Akamba where it perhaps hit hardest (Lindbloom 1920)--a fantastic level
of human devastation by any standard--and effected virtually every area of Kenya's central
highlands. These catastrophes, most severe for populations in the lowlands, and
immediately followed by the British intrusion, served to permanently alter high zone-low
zone group relations. The native economy reverted to survival strategies as local markets
wilted along with crops in the wake of the famines and epidemics of the last decade of the
nineteenth century.
The calamities acted to undermine the conservative cultural orientation towards trade of
NFD pastoral groups who had been formerly secure in their predominantly subsistence
production. Before the rinderpest epidemic, Somali traders eager to procure ivory, rhino
horns, salt, and slaves had sought these products among the pastoralists occupying the far
north of present-day Kenya. These largely self-sufficient societies disdained trade at the
time as a low-status occupation. After the calamities, when a pair of tusks could be
exchanged for thirty head of cattle, the Borana and Gabbra actively hunted elephants and
rhino for their products, and sought out the Somali in their own territory in order to recoup
their herds (Robinson 1979).
The famine gave rise to some important elements of the new order to come. Powerful
individuals who had the resources to withstand the drought were able to profit from the
desperate circumstances of others, often in a manner that previously had been socially
reprehensible. Those who sought refuge in the mission stations returned to their farms
with seeds and a headstart on the traditionalists who spurned contact, and the flight of
refugees to the first nodes of British administration represent the beginnings of African
urbanization. The crisis attracted the first generation of future mission educated elites,
initiated the Swahilization of upcountry natives settling in the "Majengo" areas of upcountry
towns like Nairobi, Machakos, and Kitui (Ambler 1988), and symbolized the beginning of
social inversion in local society where traditional elites were displaced by actors on the
periphery (Mutiso 1975).

The late century events reformulated the previous pattern of ecozone symbiosis. The
rangelands were already experiencing the social and environmental consequences of the
population brake on the expansion-predicated pastoral mode. Many highland communities
less effected by the drought profited from the helplessness of the lowland refugees where
their kin and near-kin on the lower slopes suffered the depredations of hungry migrants and
raiders who crossed the Tana in large numbers. Consequently, they were able to rapidly
recoup their domestic losses much faster when sufficient rains reappeared in 1899.
The calamities apparently wreaked relatively little change on Meru internal organization.
Meru (particularly the important Nyambene region) largely drops out of the historical
accounts of Ambler and other commentators who describe the disastrous events preceding
the turn of the century. Fadiman's (1976, 1982) work on Meru history is also curiously
silent on this score, which leads us to infer that: (1) the effect of the drought and epidemics
in Meru was less severe than in the southern areas of Mt. Kenya and the adjoining
lowlands; (2) except for a significant incorporation of Maasai by the Tigania section of the
Meru, the district remained insulated to a large degree from the social consequences of the
upheaval. Even Arkell Hardwick, an ivory trader who spent time in the Nyambenes during
1898-99, is completely silent on the topic of regional catastrophic events intervening
between his visit and Chanler's journey five years earlier.
The Nyambene area was nevertheless influenced by the far reaching developments that
emerged out of the famine and gathered strength during the next century. The imposition
of British authority at the turn of the century transcended and froze in a stroke the processes
of migration, interaction, and consolidation beginning several thousand years ago. Prior to
this, a regional political economy as indicated by internal institutions like the Meru Njuri
Ncheke, regular rural markets, formalized devices for inter-group interaction, and cultural
methods for demarcation and control of resources such as land, was rapidly evolving. It
was nevertheless incomplete, and exchange oriented or commercial production as a
dominant aspect of the production system's logic was relatively recent, particularly on the
pastoralist side, as well (Smith 1992, Arkell-Hardwick 1903, Chanler 1896).

The British Influence

The British presence was of an altogether different order than local societies could
comprehend initially. As Lonsdale (1986, 141) succinctly states, small political
communities with long histories were engulfed by conquest in a larger state without
memories. The Pax Britannica hardened ethnic groups into "tribes," those tenured
inhabitants of territories demarcated by formal borders. The British reoriented

administration and exchange towards Nairobi, and economic change traveled from the
center on the new "hub and spokes" alignment. As in other regional economies across
Kenya, the indigenous Mount Kenya trade network declined and stagnated. Previously,
the comparative advantage based on econiche adaptations had not conferred undue
advantage on the higher ecozones vis a vis middle and lower zones. This changed as
colonial administrative centers became the new nodes of exchange, and traditional social
relations across Kenya were revised or eclipsed by the new order (Waller 1989, Cohen and
Atieno-Odhiambo 1989). Ethnic differences and the calamities of the late 19th century
nevertheless created the wrong impression among Europeans arriving on the scene. It has
taken a generation of scholarly work to replace the incorrect impression of tribes as
autonomous and hostile units with the actual condition of cultural and economic symbiosis:

The population of what became Kenya may have amounted to 3 million in 1890, of whom three-
quarters would have lived in the highlands. Three main groups were involved in the conquest.
Inland from the coast the first to be met were the highland Bantu, primarily the Kamba and the
Gikuyu, but also the Embu and Meru of Mt. Kenya. They were peoples, not tribes, potential
nations rather than actual dispersions of related lineages. There were boundaries between them, and
they gave their neighbors different names, but these served to demarcate different environments and
the different cultures that had grown up in their management, and not absolute breaks in political
allegiance and economic self-sufficiency. Trade, marriage, and patronage knew no confines. As
the one purely pastoral people of the highlands, the Maasai could only survive in close connection
with the 'mixed farming' peoples who lived beyond their escarpments. They were the bankers of
the highlands. Their capital, and their currency, was livestock. (Berman and Lonsdale 1992. 19-20).

After the initial confrontations between the British and local societies, the shift to
colonialism did not engender substantial conflict (Oliver 1991, Illife 1982).2 The threat of
lethal famine, local group hostilities, and traditional concepts of the internal order were
being replaced by a new spatial order based on nodes of European capitalism. The new
regional organization provided by the colonial administration, modern transport, and the
establishment of settler agriculture substantially reordered the local landscape. In some
cases, like that of the crowded Gikuyu areas, the development of the settler economy
presented new opportunities for accumulation (Kanogo 1987). In general, like the effects
of the epidemics and natural disasters of the 1890s across the Kenyan landscape, the
penetration of European capitalism was uneven, and affected the internal organization of
different groups in different ways.
The European economy set the Gikuyu reserves on a path of conflict, both internal and
external, which significantly impacted on the colony's pathway to political independence

2 In many areas prophecies about iron snakes snorting fire and people wearing colored clothes like the
wings of butterflies had primed highland populations for changes that were beginning to manifest even as
famine and disease traumatized the local order.

and beyond (Berman and Lonsdale 1992). The dynamics in the Luo areas was different.
Nyanza became linked to the colonial economy as a labor reserve, out-migration produced a
Luo cultural gloss on the different environments spanning the rural reserve and the urban
and coastal areas that formed the expanding Luo landscape (Cohen and Atieno-Odhiambo
1989). Decentralized societies bifurcated into two streams of traditional and Western
development. Nowhere was the pattern exactly the same and understanding the transition
to capitalism involves exploring underlying uneven and incongruous, but linked,
processes of change.
The Nyambene Meru response to colonial rule fills in some of the missing patterns in
the standard picture of colonial rule and capitalist penetration. In particular, it highlights a
pattern of change in which internal organization was not overturned by a new class of
modem elites. It also shows how different commercial crops influence and determine the
process of commercialization and rural development in different ways. In more general
terms, environmental factors condition the choice of export crops grown locally, and
different commodities are system attractors that create feedback and influence the cultural-
agricultural system in different ways. Income distribution, household labor allocation and
consumption, use of inputs, investment strategies, social orientation, relationships to
modern institutions and emergent or deepened cleavages within a local society vary
according to dominant market parameters.
A number of marketed crops are produced in the Nyambene area: coffee, tea, miraa,
cotton, tobacco, and various food crops are all grown for sale. Of all these commodities,
miraa emerged out of the complex of indigenous production as the commercial flagship of
the Nyambene system. Assessing the dynamics supporting the emergence of miraa as a
major commercial commodity is important for understanding the unique properties of the
Meru agricultural system and for understanding what it signifies for small-scale African

The Rise of Miraa as a Commercial Commodity

Miraa (Catha edulis) is a tree indigenous to the highland forests of eastern Africa. It is
traditionally valued for its stimulatory and medicinal properties in parts of Africa and Asia.
In the case of Ethiopia and southern Arabia, where miraa is known as chat and khat
respectively, its use is a longstanding social institution. Catha edulis domestication may
actually predate the domestication of coffee in these regions where it has been cultivated on
a commercial basis for centuries (Weir 1985; Kennedy 1987).

The fresh leaves and tender young shoots of the tree are chewed to induce a state of
increased alertness and enhanced concentration.3Because it is highly perishable, traditional
markets for miraa and khat were located within a short radius of the areas where
domesticated production thrives, and almost always involved alcohol eschewing Muslims.
Still, traditional miraa consumption extended to a remarkably large geographic area prior to
this century. In Meru its use was governed by social controls similar to those associated
with beer. Consumption outside ceremonial occasions was limited to senior elders and the
gerontocracy. These controls lapsed as commercial production drew younger generations
into the miraa industry.
Wild miraa occurs as a tree growing up to seventy feet in indigenous East African
forests (Dale and Greenway 1961). Miraa is a hardy species that can grow in a variety of
environments. Domesticated miraa is usually limited to highland areas between 4,000 and
7,000 feet above sea level. It is cultivated as a tree, large shrub, or closely cropped bush
according to ecological conditions. In most systems it is intercropped with other trees and
crops. In Yemen, it is grown with a variety of fruit trees and grapes. The best miraa was
irrigated by natural springs, a practice that has expanded to include small petrol pumps in
recent times. Yemeni khat is also monocropped on terraces carved out of steep hills, and
at higher altitudes it is cultivated as a bush cut back to the ground every several years, much
in the manner of tea. In Ethiopia's Hararghe region, it is planted among field crops.
In Kenya the popularity of miraa beyond the geographically restricted areas where it is
grown is a relatively recent phenomenon following in the wake of urbanization and the
development of transport infrastructure. The popularity of miraa among Kenya's Muslim
communities was the principle stimuli for commercial production after the precolonial era,
but today its use has outstripped earlier religious, ethnic, and social boundaries. Kenya's
consumption has doubled several times since independence, and its use outside the areas
where traditional social controls limited the potential for abuse often generates heated
debate, particularly within Kenya's Muslim community (Troughear 1982; Goldsmith
1985). The contentious aspects of miraa consumption has not hindered the expansion of
production. In fact, miraa production is increasing at a faster rate than most other market
commodities in Kenya for a number of reasons. The popularity of miraa, like other
socially consumed commodities, has increased as urbanization and economic change within
the region accelerates.

3 The active organic compounds, cathine and cathinone, are unstable phytochemical precursors of
norpseudoephedrine (Kalix and Braeden (1985). Pseudoepherine is a common ingredient used in commercial
cold and asthma preparations.

Social Aspects of Miraa Consumption

Pre-eminently a social commodity, miraa consumption presents an interesting variation
on the social impacts created by coffee, sugar, alcohol, and exotic spices in response to
Western socioeconomic change (Mintz 1985; Schivelbusch 1993). Like coffee, tea, sugar,
and the non-medicinal use of hard alcohol, miraa consumption outside of traditional
settings is linked to other important social and economic changes.
The history of miraa consumption closely parallels that of coffee. Like Somalis, Arabs,
and Swahilis, the Meru tell the same legend explaining the discovery of miraa by a
herdsman, who found his goats happily munching on branches of miraa after an
uncharacteristic overnight absence (Getahun and Krikorian 1973, 354-356). The happy
goat story is also invoked to explain the discovery of coffee (Jacob, 1935; Hattox 1985,
13). It is altogether likely that the Igembe learned the uses of miraa from the hunter-
gatherers' extensive pharmacopeia, domesticated the tree, and incorporated into their social
and ceremonial milieu.
When coffee first became available in Europe, contrary and fanciful reasons were
produced to condemn it as a poison: "a vile and worthless foreign novelty...the fruit of
goats and camels...induced palsy, impotence, and leanness." A Women's Petition Against
Coffee published in London in 1674 blamed coffee for removing men from their homes
and causing impotence. Italian priests appealed to Pope Clement VIII to ban Christian use
of coffee claiming that Satan gave the "hellish black brew" to infidel Moslems as a
substitute for the forbidden wine. The irony here is that coffee consumption was
persecuted in the Islamic world as well: coffee houses in Cairo, Mecca and Constantinople
were charged as being centers for vice and immorality. As late as 1656 the Ottoman
authorities suppressed coffee for political reasons (Roden 1977, 21).4
Some users of the subversive drink praised it as an elixir. It was reserved for royalty
with severe penalties for use by commoners. The rising economic elite then adopted it as a
civilized pastime, a parlour drink in contrast to the indecorous conduct prevailing in the
pubs that catered to the growing working class. Voltaire's sentiment, "I have been
poisoning myself for eighty years and I'm not dead," ultimately carried the day, and coffee
entered the sphere of mass consumption (Hattox 1985). The term "Penny Universities"
came into use when coffee houses made their appearance across the English landscape.
Coffee imbibed in these Penny Universities provided an alternative to the endemic beer and

4Criticism of miraa and khat matches almost word for word these sentiments expressed in connection with