Environment, Development, Crisis, and Crusade:

feelingmomInternet and Web Development

Dec 7, 2013 (3 years and 4 months ago)


World Developmenr, Vol. 23, No. 6, pp. 1037-1051, 1995
Copyright 0 1995 Elsevier Science Ltd
Printed in Great Britain. All rights reserved
0305-750x/95 $9.50 + 0.00
Environment, Development, Crisis, and Crusade:
Ukambani, Kenya, 1890-l 990
Clark University Worcestel; Massachusetts, U.S.A.
Summary. - For over a century Ukambani, the home of the Akamba people, has been the object of intense
scrutiny and repeated interventions by international and national experts. Outsider narratives have por-
trayed the region as a crucible for a series of crises, including human and livestock epidemics, overgraz-
ing, soil erosion, low productivity, underdevelopment, fuelwood shortage, biodiversity loss, and
threatened wildlife. Akamba farmers and herders recount a very different story in which land alienation.
land hunger, and limits on mobility of people and their herds have restructured the ecological and spatial
order of their homeland, to the benefit of some and the detriment of many. The history of crisis construc-
tion and resolution by outsiders, juxtaposed with the diverse experience of people within the region sug-
gests that simple solutions to single problems may actually create new crisis, in Ukambani and elsewhere.
Throughout the last hundred years the dry savannas
of East Africa-and Kenya in particular - have been
portrayed alternately as a paradise lost (Myers,
1979; Western, 1988) or as a crucible of crisis on the
continent (Timberlake, 1988; World Bank, 1989). The
paradise accounts abound with images of untamed
wilderness and noble hunters, while the crises are por-
trayed in images of poverty, pestilence, hunger, and a
ravaged wasteland. It is hard to imagine that these
recurring, but contrasting narratives apply to the same
place. Nonetheless, they have themselves altered the
shape of history, the landscape, and regional identity
in the savannas of Kenya.
Travellers journals in the mid- 1800s spoke of east-
em Kenyas fertile lands, vast herds of livestock and
wildlife, and prosperous settlements of heathly people
(Krapf, 1860). These reports drew eager investors, big
game hunters, and adventurers. By contrast, accounts
from the 1890s picture societies and ecosystems in a
state of collapse (Hardinge, 1899; Hobley, 19 10; Tate,
1904). These narratives attracted military and relief
administrators, new waves of missionaries, and oppor-
tunistic investors to an imagined chaos in need of
order, charity, and economic progress (Ainsworth,
In the 20th century, a series of crises in the
Kenyan environment and economy have been identi-
fied by experts within the colonial, national, and
international government, academic, and nongovem-
* Over the last IO years, rural residents, local officials, field
extension agents, and researchers in Ukambani have con-
tributed to field work and analyses that have informed this
paper. Especially helpful were Luis Malaret, Ahson Field-
Juma, Calestous Juma, Richard Mwendandu. Ana Nzisa,
Muhammud Jama, Alex Dianga, Lawrence Kyongo,
Veronica Ndunge. Japheth Kyengo, Bernard Muchiri
Wanjohi, David Mutiso Kiilu, M. Musyoki, Alice Mwau.
Ana Ndungwa, Charles Gichuki, Betty Wamalwa, and Isaac
Jondiko. We also wish to recognize many prior research
collaborators, including Patrick Maundu Munyao, Annette
Hoek, Kamoji Wachira, Rosemary Mutiso, Remko Vonk,
and other colleagues and residents too numerous to name.
We are indebted as well to Phil Porter, B.L. Turner II.
Kate Showers, Francis Lelo, Simon Batterbury. Jeanne
Kasperson, Jean Hay, and Michelle Reidel for comments on
earlier versions of the paper and especially to Emery Roe for
careful critique and editorial assistance.
Field research and writing (1988-93) was suppotted by
the World Wildlife Fund, the United Nations Unrversity. the
National Science Foundation, the United States Agency
for International Development, the Ford Foundation, the
Rockefeller Foundation, the International Development
Research Centre, and the International Council for Research
in Agroforestry
A longer version of this paper was presented at the annual
meeting of the African Studies Assoctatton. Boston.
Massachusetts, USA. December 1993 and has been
published under the title A Hundred Years of Crisis
Envnonment and Development Narrattves in Ukambani.
Kenya, as Working Paper No. 189 within Boston
University African Studies Centers Working Papers tn
African Studies series. In addition. portions of tht\ arttclc
will appear m Rochelenu PI ctl forthcoming.
mental organization (NGO) communities. Histori-
cally, the specific crisis identified has varied consider-
ably. Crises have included famine, disease, soil
erosion, desertification, fuelwood and energy short-
age, overpopulation, poverty and unemployment,
deforestation, and, most recently, the decline of biodi-
versity. The name of each crisis has carried with it the
seeds of each solution. Unfortunately, measures pro-
posed to ameliorate any given crisis have generally
failed to address the entire problem, and they have
often assisted in the construction of new crises in their
wake (Watts, 1989).
The variety of crisis narratives suggests something
other than a single recurring crisis endemic to the
region. Rather, these accounts reflect a series of spe-
cific problems, defined by outsiders and extrapolated
to regional characteristics and conditions at different
historical moments. Below, we argue that each of the
crises attributed to the Kenyan landscape and society
can be viewed as successive internal impacts of
processes which have their origin - in large part -
outside the region. Contradictions and conflicts
emerging in the First World and on the global scale
have been continually exported to Kenya, where
they have merged with regional social and ecological
systems to give the appearance of a series of local, uni-
dimensional crises. Our contention is that the series of
crises experienced, documented, and named in the
Ukambani region of Kenya have revolved around an
ongoing and multi-faceted encounter with the global
restructuring of economies, ecologies, and cultures.
We provide our own Crisis Narrative here, the
heart of which is our critique of the way in which
development practitioners have historically accepted
their eras prevailing definition of a unidimensional,
single-scale crisis. By focusing on the representation
and articulation of crises we are not implying that
there is not a real crisis behind each official crisis.
Indeed, there are several problems behind each
named crisis. That is why the selection and reifica-
tion of one crisis to define the problems encountered
by people in a given place within a given era has such
significant ramifications for policy interventions in the
region. It is when the causes and effects of a problem
are complex, uncertain, and are constituted at a num-
ber of scales that it becomes particularly important to
uncover the stories and assumptions - both explicit
and implicit -which have been used in the naming of
the crisis (Roe, 1992, 1993).
Throughout the case study below, we discuss cri-
sis construction in two interdependent senses. On the
one hand, crises are constructed as representational
discourses. lndividual and collective actors attempt to
exert social power by telling one particular story of
crisis. For instance, one might choose to identify a
food shortage crisis during time of drought as a crisis
land degradation
resulting from past
agricultural practices. Alternately, the food crisis
could be constructed as one of global political
economy: changes in the world economic system and
in a regions integration into that system have led
farmers to abandon certain crops that had formerly
provided a highly reliable and drought-resistant food
This representation of crisis is, however, insepar-
able from the construction of actual crisis conditions.
The representational construction of one particular
crisis as opposed to another will lead to particular poli-
cies (and specific acts of resistance) which will con-
tribute to the material construction of the next crisis.
In the above example, an attempt to rectify the con-
structed crisis of past agricultural practices may, for
instance, lead to policies designed to modernize
agricultural production. These policies could, in turn,
further construct the (as yet unnamed) crisis of
abandonment of drought-resistant food crops.
This papers own Crisis Narrative resonates with
the growing critique of the latest development trend:
sustainable development (Peet and Watts, 1993;
Watts, 1993). Early critics of sustainable development
demanded that attention be directed to the question of
what was to be sustained and for whom (Martinez-
Alier, 1990; Redclift, 1987; Thrupp, 1989). Close on
the heels of this debate over equity and the greening of
development as usual has come a broader critique of
the development discourse (Escobar, 1992; Marglin
and Marglin, 1990; Sachs, 1992, 1993).
Our Crisis Narrative - with its implicit critique
of mainstream conceptions of scale and causation -
also builds upon the insights of political ecology.
Political ecologists meld analytical and methodologi-
cal traditions from cultural ecology, international
development, environmental politics, and political
economy in an effort to examine concurrently the
local-scale changes in agricultural and social systems
alongside the dynamics of global political economy
(Blaikie and Brookfield, 1987; Croll and Parkin, 1992;
Hecht and Cockburn, 1990; Schmink and Wood,
1992; Watts, 1993).
Additionally, we critique the way in which devel-
opment experts -
as they go about defining a par-
ticular crisis - unwittingly accept and reproduce
social conceptions of identity and culture. Here we
rely upon feminist researchers who question the
homogenization of distinct groups and their diverse
experiences within simple, one-dimensional analyses
of economy, environment, and community. A number
of feminist authors treat gender, class, race, and cul-
ture as axes of identity and difference that are socially
constructed and embedded in relations of power
(Fraser, 1989; Harding, 1986; hooks, 1990; Merchant,
1989; Stamp, 1989). These axes affect the course of
social, economic, and ecological change and invari-
ably influence peoples experience of those changes.
Finally, we turn to a growing number of social
science researchers who have centered their studies
on the narrative and rhetorical devices used to frame
an argument or present a situation. These scholars
emphasize how policy makers and academics, as well
as individuals in their everyday lives, imply defini-
tions and employ categories to define a specific prob-
lem and thereby to suggest a particular solution
while precluding others (Dalby, 1992; Der Derian and
Shapiro, 1989; Nelson, Megill and McCloskey, 1987;
Roe, 1992).
A common thread in all of these critical streams of
literature is a return to specific places and particular
histories to understand the broader political outlines
that bind local ecologies, economies, communities,
and cultures into regional and global systems. In keep-
ing with this perspective, much of the critique pre-
sented below derives from the experience and insights
of people in several rural communities in Ukambani
the home of the Akamba, in Kenya - shared over
10 years of consultation and collaboration. The story
developed here is a view through the prism of separate
realities within the IOO-year history of environment
and development policy in Ukambani.
In the case study which follows, we demonstrate
how policies formed on the basis of single perspec-
tives have repeatedly distorted resource use and allo-
cation in a complex, multifaceted landscape with a
diversity of actors. While the historical argument
about colonial, commercial, and national government
intentions rests on choices about the sources of fact
and judgment in a sea of conflicting accounts, the end
result is always clear. The many stories told by a diver-
sity of Akamba people constitute an implicit critique
of past and present theories, policies, and practices of
economic development and environmental conserva-
tion. The measures taken to alleviate one groups
crisis or to foment another groups progress have
often brought other groups to the brink of disaster and
beyond. We conclude by suggesting that the sustain-
able development paradigm, which dominates con-
temporary development theory and practice, needs to
be subjected to this analysis. Sustainable develop-
ment has much to learn from the recurring cycles of
crisis designation in Ukambani and the successive
imprints of single geographies on separate realities.
Ukambani (Figure 1) provides an ideal study area
for tracing the history of crisis labeling. Close enough
to Nairobi to have continually attracted the attention
of colonial and Kenyan governmental authorities, the
region has also served as a semi-arid frontier area for
agricultural expansion (Bernard et al., 1989; Mbithi
Figure I. Ukambani, Kequ.
and Barnes, 1975; Munro, 1975). As a high-profile
frontier area, Ukambani has for a century been at the
center of environment/development crisis labeling
and intervention in Kenya. Its environmental condi-
tion has been subject to scrutiny by colonial officials,
natural and social scientists, government ministries,
and its own inhabitants.
The region has frequently been defined as a site of
environmental degradation and rural poverty (Deacon
and Darkoh, 1987; Mbithi and Barnes, 1975). Rural
residents continue to report frequent crop failures and
food relief has become a permanent feature of rural
life (Mbogoh, 1991; Porter, forthcoming). These
problems have been variously and contradictorily
attributed to overpopulation and environmental degra-
dation, to colonization and development, and to
insufficient development. Most recently, one part of
the region (Machakos District) has also been cited as
an example of successful land use intensification and
dramatic environmental recovery (Tiffen, Mortimore
and Gichuki, 1994).
Ukambani, composed of Machakos and Kitui
Districts,? is a predominantly semi-arid region, falling
from an elevation of 2,lOOm in the west to 440m in the
east. Only the highest altitude regions, primarily in
Machakos District, have sufficient reliable rainfall to
support regular agricultural harvests without irrigation
(Bernard, Campbell and Thorn, 1989). Soils through-
out the region are generally of low fertility, and many
are highly erodible (Barber, Thomas and Moore,
1981). The dominant vegetation in the area is dry bush
with trees, and, in the higher areas, savanna with
scattered trees (Ominde, 1968). The hills were once
forested, but by the beginning of the colonial period
most of the desirable agricultural land had been
cleared (Owako, 1971; Silberfein, 1989) leaving
patches and corridors of forest along ranges, rivers,
ravines, and hilltops, as well as dry forest stands,
Savannah woodlands, and scattered trees in large
expanses of grazing land.
More than 95% of the MachakoslKitui population
is Akamba (Tiffen, 1991), and approximately 85% of
all Kikamba speakers reside in Ukambani (Porter,
forthcoming). The combined area of the two districts
is approximately 45,000 km*, although one-fifth of
Kitui, some 6,300 km2, lies within Tsavo National
Park and is legally unavailable for use by the Akamba.
The 1989 census indicates that the current population
of Machakos and Kitui Districts is approximately two
million people. The vast majority (90%) of the popu-
lation lives in rural villages, primarily in the well-
watered upland areas of Machakos. People rely on a
combination of subsistence and commercial agricul-
ture (an estimated 50% of income), wage labor (24%
of income), and off-farm enterprises (17% of income)
(Ondiege, 1992).
Prior to the colonial era and the beginning of
outsider crisis narratives, Akamba farmers grew
sorghum, millet, maize, beans, sweet potatoes,
bananas, squash, sugarcane, and other crops in addi-
tion to keeping livestock (cattle, goats, and sheep). A
mix of private and common property rights, integrated
crop-livestock systems, spatially separated holdings,
flexible patterns of settlement and mobility, and
mutual reciprocity arrangements served to limit vul-
nerability and provided mechanisms for coping with
drought (Bernard and Thorn, 1981; Lambert, 1947;
Wailer, 1985). The system was apparently successful,
e.g., Krapf (1860) described Ukambani as a fertile
land inhabited by a prosperous people.
(a) 100 years of environmental crisis
Over the past century, land use and cropping sys-
tems in Ukambani have changed rapidly in response
to new markets, new property relations, population
growth, and large-scale migration and resettlement.
State policy has been instrumental in shaping these
responses. Most often, policy changes were made ina
response to a succession of perceived crisis on the
Akamba Reserves: famine and disease in the 1890s;
severe land degradation in the 1920s and 1930s; low
production rates and political instability in the
1940s and 1950s; low productivity, poverty, and land
hunger in the 1960s and 1970s; and fuelwood (energy)
shortages, land degradation, and threats to wildlife
and biodiversity in the 1980s (Table 1).3
Yet these crises have themselves been constructed
out of the particular intersection of processes and con-
ditions operating within a broader context, converging
upon the social and environmental landscape of
Ukambani. Global scale or First World crises and
social concerns have led colonial and government
officials and development experts to identify spe-
cific crises endemic to Ukambani. But many of the
conditions noted by these outsiders have, in fact, been
brought about by the policies and interventions of
their predecessors. The history of Ukambani is as
much a history of crisis construction (in both the
material and discursive senses of the word) as it is one
of an actual succession of crises. Each particular crisis
narrative has engendered a specific response, but since
each response has been designed to remedy only a
small part of a much larger problem, the responses at
best have been ineffective and at worst have con-
tributed to the construction of new crises.
1890 - WWI: Human and livestock epidemics
During the early colonial period, human and cattle
disease was the dominant crisis identified by adminis-
trators and travellers in Ukambani. The 1890s saw the
construction of the Uganda Railroad, which likely
contributed to the spread of rinderpest among cattle
(Hardinge, 1899) as well as the introduction of small-
pox by foreigners (Wamalwa, 1989). These forces
combined with a drought in the late 1890s to bring
about the great famine of 1897-1901 (Ambler, 1988).
Since cattle served as the main drought insurance, the
results of combined drought and cattle disease were
devastating, particularly in Kitui, where official
figures suggest that upward of 50% of the people
perished (Hardinge, 1899).
Viewing the devastation around them, colonial offi-
cials and observers, while recognizing the epidemics,
blamed Akamba cattle-rearing practices as well as
their primitive standard of living for the magnitude
of the disaster. Writing around the turn of the century,
colonial observers identified several reasons for the
Akambas supposedly low standard of living, accus-
ing them of weakness of character, irrational attach-
ment to their cattle, and inefficient and destructive
cultivation practices (Hardinge, 1899; Tate, 1904).
Noting the famine resulting from these social and
agricultural practices, colonial authorities defined a
public health crisis among the Akamba.
This public health crisis paralleled a concurrent
crisis construction by Progressive-era reformers in the
United States and Great Britain who were associating
high population densities in urban slums with poor
household maintenance practices, poor hygiene, and
the spread of disease (Dyhouse, 1978; Meckel, 1990).
Colonial and urban reform movements identified sim-
ilar problems and solutions, each relying optimisti-
cally on state regulation of household behavior to cope
with the problems which ensued with the advent of
modernity (Chambers, 1988). In both Ukambani and
working-class England, womens household practices
were judged against the standard of the Victorian
women and were frequently found wanting. As work-
ing-class English women received training in mother-
YCUS External phenomena
Internalized definition
of crisis
Official response
1890 - WWI
Rivalry among
colonial powers
Equation of
poverty, high
densities, and
civilization with
1920s &
Equation of poor
practices with
soil erosion
1940s &
War-time and post-WWII
First World resource needs
1960s &
Com~t~tion among
newly independent
states in world
commodity markets
Concern about
declining soil quality
Concern about declining
global energy resources
Concern about declining
global gene pool,
endangered species, and
endangered ecosystems
Identification of need
for settlers to have
privileged access to
land and resources and
control of commodity
Identification of
irrational, disease-
prone Akamba cattle-
rearing practices on
crowded reserves
Identification of
destructive Akamba
agricultural practices
Identification of underproduction
on Akamba reserves; recognition of
land hunger as a source of politica
identification of low
productivity among
Akamba farmers
identification of destructive
Akamba agricultural practices
as cause of river and dam
sedimentation problems
Identification of the other energy
crisis and Akamba overcutting
of trees for fuelwood
Identification of the threat posed
to national parks and the wildlife
therein by area residents
Disruption of Akamba
land tenure system
Further land alienation
and segregation of
Akamba economy
Forced terracing and
production limits
Forced cash crop
production; enclosun: and
privatization of land
Further encouragement of
cash cropping and land
concentration among
efficient producers
Promotion of intercropping
and terracing techniques
Promotion of monocrop or
alley cropping
miracle-tree plantings
Increased efforts to
segregate citizenry from
parkland; separate pursuit
of conse~ation and
agriculture policies in
separate spaces
ing skills and home economics (Dwork, 1987; Ross,
1990). colonial observers criticized the Akamba for
their alleged sexual immorality and noted critically
that the women do all the work in the fields and are
also hewers of wood and drawers of water (Tate,
1904, p. 136).
While the naming of the epidemics which plagued
Ukambani during the 1890s reflected European expe-
riences and prejudices, the epidemics themselves
derived from contact with European people and live-
stock, whose very presence was a response to events
unfolding in Europe. The British were primarily
responding to conflicts on the global stage, as inter-
European political and economic competition intensi-
fied, culminating in the 1884 Congress of Berlin and
the scramble for Africa. The Uganda Railroad was
built in part to solidify British control of East Asia.
and the authorities encouraged European settlers in an
effort to make the railroad pay for itself (Bates. 1987;
Bradshaw, 1990). The settlers, in turn, faced a land
shortage, as the Akamba tenure system depended on
expanses of open, communally held pastureland. The
settlers responded with an extensive land seizure and
enclosure program. By 1920, the Akamba had lost
effective access to about two-thirds of the land they
had formerly controlled (Porter, forthcommg), includ-
ing their most fertile lands and half of all their pasture.
Along with some of their best grazing land. they lost
the freedom to migrate seasonally and periodically in
search of water, pasture, and cropland (Munro, 1975;
Spencer, 1983; Wisner, 1977).
For the Akamba, the crisis of the era was one of
land alienation, as settlers disrupted their tenure sys-
tem and took away their land. At the same time, the
state barred the Akamba from the export-oriented
agricultural sector in an effort to protect the white
settlers monopoly. These policies contributed to a
continuing crisis of cattle disease; agropastoralists
were left with few options other thah to preserve
underfed and sickly cattle, their major assets, in
overcrowded reserves where disease spread easily
(Spencer, 1983). These colonial land tenure policies
also forced the Akamba into sedentary settlements and
continuous cultivation on relatively small areas of
poor quality land. This process of sedentarization and
concentration, in turn, sowed the seeds of future
crises, including those of land degradation, overpop-
ulation, and urban migration.
1920s and 1930s: Soil erosion in the other
dust bowl 
By the second decade of the 20th century, European
expansion in Machakos had ended, while population
on the Native Reserves increased, and the economy
became more commercial (Munro, 1975). Land alien-
ation and associated land scarcity had stimulated
settlement in the plains, but relatively infertile land,
tsetse, lack of water, and the 1924-25 famine forced
an exodus by the Akamba back to the hills, resulting
in the virtual depopulation of parts of the plains and
increased crowding in the hills (Owako, 1971;
Wisner. 1977). While the natural rate of population
increase in Machakos was approximately 2.5 to 3.0%
per year for 1918-39, officials estimated dramatic
increases in some of the highland areas receiving
return migrants from the drylands (Munro, 1975).
In the densely populated Machakos Hills, land
alienation, land scarcity, and official policy resulted,
by the early 193Os, in continuous cultivation of land
and enclosure of permanent farms with fences or sisal
hedges (Munro, 1975). This entailed decreased fallow
and more continuous cropping, which, combined with
the concurrent replacement of sorghum and millet
with maize, resulted in soil exhaustion and reduced
yields. The remaining commonage was under severe
pressure as a source of forage and wood, particularly
because many farmers began cash cropping, plowing
up their grazing land and putting increased pressure on
the commons. Land sales, tenancy. and, for the first
time, landlessness, became features of Akamba life
(Silberfein, 1989). Social stratification based on land
holdings emerged and land disputes became common
(Munro, 1975; Mutiso, 1975).5
As land tenure and land use systems underwent
transformation. so did dominant economic activities
withm Akamha society. Scarcity of pastureland, along
with other financial and administrative pressures,
resulted in an overall decrease in Akamba cattle
wealth and a drastic decline, for most households, in
livestock holdings. By the beginning of WWII, signi-
ficant numbers of Akamba were peasant cultivators
instead of herders, and crops had become a significant
income source (Munro, 1975). Alternative income
sources, such as brick and charcoal production
became important (Silberfein, 1989). Increasing num-
bers of men engaged in migrant wage labor (OLeary,
1984), changing household relations and agricultural
labor availability.
During this period, colonial officials and settlers
were increasingly aware of land degradation on the
Native Reserves (Dregne, 1990). Rather than linking
this phenomenon with the successive forces of land
alienation, population concentration, and disruption of
land tenure and land use systems, colonial officials
instead constructed the soil erosion crisis around
Akamba cattle-rearing and agricultural practices.
While continuing the official policies to repress
Akamba livestock production and quarantine their
cattle (Spencer, 1983), European settlers waged a
successful campaign to portray African agriculture as
a kind of contagion, spreading insidioulsy across the
landscape and infecting productive European lands.
In 1929, the Hall Commission report raised erosion in
Machakos to the status of a major hazard in Kenya
(Myrick, 1975; Spencer, 1983). A witness before the
Kenya Land Commission opined, The African
people have never established a symbiotic relationship
with the land. They are, in the strict scientific
sense, parasites on the land, all of them (Nature,
1938, p. 397, as cited in Jacks and Whyte, 1939,
p. 270).(
The poor condition of the reserves and settler polit-
ical fears coalesced in 1935 with both the global anti-
erosion movement and the reaction to the Dust
Bowl in the United States. This concern over a
soil erosion crisis was exacerbated by the new pro-
fessionalism of colonial agricultural officers
(McCracken, 1982), many of whom were trained in
Trinidad by a soil conservationist whose zeal was
honed in the Dust Bowl. Having identified Akamba
agricultural and pastoral practices responsible for the
soil erosion crisis, colonial authorities responded
with an interventionalist program based upon destock-
ing and reconditioning of farms and rangelands.
After Akamba protests foiled a major destocking
campaign (Tignor,
1971), agricultural officials
focused on reconditioning, actively encouraging the
enclosure and seeding of grazing lands as well as the
enclosure of homestead lands. By the end of 1939,
over 400,000 acres had been enclosed. For most
Akamba farmers during this period, any merit intrin-
sic to conservation paled before its use as an instru-
ment of colonial control and, thus, a focus of
anti-colonial resistance (Beinart, 1984; Blaikie,
1985). Conservation officers also treated conservation
as a *special project separate from everyday farming
and herding. While this attitude was in keeping with
the administrators tendency to identify a one-dimen-
sional crisis emerging from a discrete set of Akamba
social practices, it was foreign to the daily experience
of farmers and was therefore doomed to fail
(Wamalwa, 1989). By 1940, the reconditioning pro-
gram had all but halted, due to both Akamba resistance
and to the call-up of agricultural officers to aid the
war effort.
(iii) 1940s and 1950s: The crisis of
With the advent of WWII, colonial officials and
rural people alike confronted wartime quotas for grain
exports, followed by severe food shortages and
famines in 1946 and 1951-52. Through applications of
technology, officials in Nairobi and London sought to
transform the rural landscape of Ukambani into a
coherent and specialized production unit for grain
crops during the war, and for cash crops in the postwar
period. The crisis was now one of underpr~uction,
and it too had its origins in the preceding crisis
Initially, the European settlers were to intensify,
specialize, and commercialize agriculture with
Africans as wage laborers. Large-scale maize plantings
were reported to have encouraged further soil deple-
tion (Spencer, 1980) and manipulation of the maize
market during the war caused a minor famine. Wartime
benefits to the Akamba reserves were limited to mili-
tary remittances, as thousands of men volunteered or
were coerced to serve (Easterbrook, 1975; Lelo, 1994;
Munro, 1975). After the war, the government assisted
demobilized British officers to acquire land in Kenya.
This increased the pressure to clear out squatters,
many of them Kenyan veterans and their families
(Lelo, 1994) thus contributing to the Land and
Freedom Movement (MauMau rebellion) of the
early 1950s (Bates, 1987).
With less land than ever before, more crops had to
be produced, and Africans had to produce them. In a
further attempt to rectify this perceived crisis of under-
production - as well as to quell the growing political
unrest - the colonial government in 1954 imple-
mented the Swynnerton Plan, a land reform program
that continues to shape the evolving landscape in
Ukambani and the rest of Kenya. The Swynnerton Plan
was designed to reform land tenure, consolidate frag-
mented holdings. and issue freehold title to encourage
the intensification and development of African com-
mercial agriculture (Bradshaw, 1990). Through land
privatization, consolidation, and registration, the plan
was to increase agricultural investment, employment,
and productivity and to raise rural incomes (Okoth-
Ogendo, 199 1: Wangari, 199 1). The plan was predi-
cated on an assumption that successful African
farmers would be able to acquire more land and bad
or poor farmers less, creating a landed and a landless
class (Swynne~on, 1955, p. 10, cited in Wanga~,
Meanwhile, even though colonial officials blamed
overgrazing on the displaced and on those most depen-
dent on extensive herding, the most overstocked
areas of Ukambani at the start of the 1960s remained
the densely populated commercial farming areas in the
hills {Porter, fo~hcoming). Wealthy landholders often
accumulated large herds and grazed them on whatever
land was available, that is no mans land under the
new property regime. The overstocking problem attrib-
uted to the tragedy of the commons seems, quite the
contrary, to have followed the alienation and privati-
zation of the commons and the concentration of private
The response of farming and agropastoral commu-
nities to declining yields, the redefinition of property,
and the continuing concern over land degradation was
not uniform. It was negotiated in a middle region
between the dramatic resistance against large-scale
reconditioning schemes and the apparent success of
terracing programs on private cropland. The wide-
spread adoption of terracing - rather than acquies-
cence to conservation programs8 - reflected farmers
evaluation of specific technologies within their own
changing production systems. The promotion of maize
in place of drought resistant sorghum and millet had
created an increased vulnerability to drought, which
some farmers partially offset by improved water stor-
age in terraced croplands.
By the time Kenya won Independence, the people of
Ukambani had experienced a restructuring of liveli-
hood as well as of landscape. They had also witnessed
the transfo~ation of the property regime that gov-
erned the ecological and spatial order of their home-
land (Bernard and Thorn, 1981). Frequent crop loss
due todrought, like sedentary life and land registration,
had become facts of life for most residents of the region
(Porter, 1965, forthcoming).
(iv) 196cts and 1970s: The crisis of
Following Independence in 1961, the Kenyan gov-
ernment came to define the major crisis facing the new
country as one of underdevelopment, poverty, and
hunger. National plans were oriented toward eco-
nomic development,
with environmental concerns
considered a legacy of the colonial past. In Ukambani
and other rural areas, underdevelopment was defined
specifically as a problem of poor agricultural produc-
tivity, especially with reference to exportable cash
crops. National agencies largely continued the logic of
the Swynnerton Plan (Migot-Adholla, 1984; Wangari,
1991), a policy which created a market-oriented class
of African farmers within the commercial farming
export sector and was credited with tripling agricul-
tural output during 1955-64 (Bradshaw, 1990).
This recourse to the Swynnerton Plan also fostered
land concentration (Shipton, 1988) and social stratifi-
cation,Y as foreseen by the author of the policy. It
plunged many Akamba residents into further poverty,
hunger, and unemployment (Mbithi and Wisner,
1973). Common grazing, gathering, and forest areas
essential to poor smallholders dependent on off-farm
resources were often the first lands claimed for
growing agricultural estates (Mbogoh, 1991). Those
without capital, savings, or investments other than
livestock were often unable to maintain medium-sized
or even small farms under such conditions, and they
sold out or abandoned their holdings (Hunt, 1984).
The simultaneous creation of a successful large-
holder class and a landless and near-landless class
caused unforeseen environmental problems. Land
hunger was often displaced to more fragile areas,
rather than diverted into pursuit of local wage labor,
given the increased freedom to relocate after
Independence. Many of the people who found them-
selves pushed off the best lands in their home areas
went to work in the cities, the army, and the police
force, and a large number went to drier frontier areas
in both Machakos and Kitui in search of new settle-
ment opportunities (Mbithi and Barnes, 1975). Thus
national policy intended to rectify the crisis of under-
production actually encouraged a massive movement
of landless and failed farmers into both state-spon-
sored and independent settlements in the drylands.
The attempts of these displaced farmers to establish
permanent farms in environmentally fragile frontier
areas paved the way for the construction of the
environmental crisis which resurfaced in the 1980s.
(v) 1980s: Multiple crises of environment
and development
In keeping with the worldwide turn toward envi-
ronmental concerns over the past two decades,
Ukambanis problems have increasingly been con-
structed as a regional environmental crisis. Over time,
the exact nature of this environmental crisis has
changed, with the emphasis moving from land degra-
dation and soil erosion at the beginning of the era, to
deforestation and energy shortage, and finally to
endangered biodiversity.
First, a crisis of soil erosion and watershed degra-
dation was identified. With the environmentalist era,
the Kenyan authorities - and indeed many farmers as
they cleared fragile frontier areas - resurrected the
old concern of the 1930s: soil erosion. The govern-
ment and international advisers again identified over-
population and poor land-use practices among the
Akamba as the key culprits (Louis Berger, 1985).
Now, however, the government was preoccupied less
with declining soil productivity on farms than with the
sedimentation of downstream dams at hydroelectric
installations. In contrast, local concerns about erosion
centered on soil productivity and access to water.
Nonetheless, in this region, the two constructions of
the soil erosion crisis pointed to a common solution.
Intercropping and terracing techniques promoted by
the government and in many cases eagerly adopted
by farmers succeeded, slowing down both the rate of
sedimentation at dam-sites and the decline in on-farm
soil quality and moisture retention (Gichuki, 1991;
Thomas, Barber and Moore, 198 1).
These successes notwithstanding, the association
of soil erosion with the agricultural practices of small-
holder farmers-while consistent with crisis narratives
carried over from the 1930s - may have led conser-
vationists to bypass more effective remedies for the
problem. The overwhelming majority of the sedimen-
tation measured in Machakos streams in the 1970s
originated not from cropped fields but from grazing
land (Moore, 1979) and rural roads (Dunne, 1979;
EcoSystems Ltd., 1986), both of which are tied more
to the interests of commercial producers than to small-
holders. Furthermore, erosion from croplands at other
sites has been most severe in forest and pasture newly
cleared and converted to cropland (La], 1976; Roche-
leau, 1984) as is widely practiced by displaced small-
holders in Ukambani. The crisis was thus one of land
concentration, agricultural commercialization, and land
hunger rather than one of bad agricultural practices.
Second, international researchers, nongovemmen-
tal organizations, and national agencies identified
deforestation as a major problem facing Ukambani
and most of Africa. The energy crisis of the 1970s
and the ensuing global concern about the other
energy crisis in Africa translated into concern over
the use of trees as an energy source, whether con-
sumed as firewood in rural households or as charcoal
in urban homes (Barnes, Ensminger and OKeefe,
1984; Mungala and Openshaw, 1984). In response,
international and national development organizations
sponsored the widespread planting of fast-growing
tree species in croplands and in small blocks (energy
plantations). The tree species and planting arrange-
ments were chosen to maximize woodfuel production
to the virtual exclusion of other tree products.
The Akamba evaluated reforestation around a more
complex understanding of the value of trees. They per-
ceived fuelwood largely as a by-product of trees and
shrubs serving other higher purposes (timber, poles,
fodder, food, medicine, boundary markers, fencing,
water supply protection), or looked to weedy species
growing in fencerows and grazing lands as a fuelwood
source (Hayes, 1986; Vonk, 1983). The specific loca-
tion and species of new trees were as important to the
Akamba as their size and growth rate. Independently,
and with the help of the more flexible sources of tech-
nical assistance, many farmers began to integrate
multipurpose trees into a diversity of landscape
niches (Rocheleau, 1991; Tiffen, Mortimore and
Gichuki, 1994).
While many researchers and administrators - fix-
ated as they were on Kenyas energy needs -
assumed that growth in local fuelwood use and urban
charcoal markets was causing deforestation, field
research in some cases has indicated otherwise.
Dewees (1989) has suggested that land clearing for
agriculture was most often the main engine of de-
forestation, with charcoal as a by-product.r For the
Akamba, the deforestation crisis was constructed out
of continuing land hunger, and a complex combina-
tion of population growth, agricultural commercial-
ization, forest privatization, land concentration, and a
high urban demand for charcoal. These convergent
factors gave farmers multiple incentives to clear land
in dry forest frontier areas for farming and for profit
from charcoal sales. Yet, simultaneous with depletion
of the woodstocks in the surrounding savanna and dry
forest, farmers in the drier parts of the region were also
planting new trees for fruit, timber, and fencing on
their own farms, a phenomenon not captured by the
broad-brush summaries of declining forest acreage
(see also Fairhead and Leach, this issue).
During the 1980s a third environmental crisis was
increasingly identified within Kenya: that of biologi-
cal impoverishment (loss of biodiversity). Yet the
accompanying homogenization of the agrarian land-
scape, along with ecodiversity loss and reduced liveii-
hood options, remained an untold story. Once again,
various interest groups constructed the single
named crisis in different ways, around distinct cen-
tral issues of concern, and with different suggested
remedies. For the Kenyan government, biodiversity
was a national economic issue. Tourism had become
Kenyas largest industry, and the government priori-
tized the preservation of high-profile mammals in
national parks such as Tsavo in Kitui, renowned for its
elephant and lion populations. Official policy and
practice sought effective exclusion of neighboring
people from the parks, rather than the integration of
local livelihoods with biodiversity, inside and outside
park boundaries (Lusigi, 198 1).
Meanwhile the international scientific and wildlife
conservation communities paid increasing attention to
the habitats of endangered animals, and they identified
plant species and finally whole ecosystems as worthy
of preservation (Anderson and Grove, 1987). But few
members of this community constructed this crisis of
endangered habitats and ecosystems amid the reality
of agricultural land-use intensification, and even
fewer took up the relation of biodiversity to diversity
of livelihood options and the preservation of culture
in agrarian landscapes (Juma, 1989; Oldfield and
Alcorn, 1991).
The Akamba viewed biodiversity in the context of
subsistence and drought preparedness, as well as reli-
gious and cultural values. A wide variety of flora and
fauna had historically provided them with medicinal
herbs, wild and semi-domesticated foods, and a host
of fibers, dyes, wood products, and fodder sources
within agrarian landscapes. The diversity of flora and
fauna, with their attendant products for human use and
their invisible services to the ecosystem (nutrient fix-
ation and cycling, seed dispersal and plant pollination,
water retention and regulation), depended, in turn, on
the complexity and diversity of the landscape itself.
Farmers tolerated, protected, and even fostered the
growth of a multitude of wild plants, including many
characterized as weeds by colonial, national, and
international agricultural scientists. These plants grew
in interstitial spaces such as fencerows, field bound-
aries, and drainage channels, as well as between row
crops, in fallow croplands, in grazing lands, and in
forest remnants and patches (Rocheleau er al., 1989).
As agricultural intensification and expansion spread
through Ukambani, this complex landscape was
increasingly threatened, and many plant and animal
species surviving in the spaces between disap-
peared. For the Akamba, then, the biodiversity crisis,
like so many others, was constructed within the
broader crisis of ecological and economic diversity
based on agricultural intensification and expansion,
land alienation, land concentration, and - ultimately
integration within the world and national
economies under highly unfavorable terms.
(b) The crisis experienced: The famines of
While the international environmental and devel-
opment communities convened in the nearby capital
of Nairobi to discuss deforestation, soil erosion, van-
ishing wildlife, and threatened energy and water sup-
plies, the people of Machakos and Kitui Districts
experienced one of the most devastating events of the
century, the 1984-85 drought. Although Akamba
farmers knew by mid-1984 that the deepening drought
signalled impending famine, the situation did not gain
national or international crisis status until the subse-
quent nationwide failure of rains and widespread grain
shortage in urban markets (Wangati, 1985). Once rec-
ognized, the famine was designated as a national food
shortage due to natural causes and solvable through
more efficient delivery of food, primarily through
market channels. In fact, both the causes and the con-
sequences of the famine were varied and complex.
interwoven with the other dimensions of crisis which
affected Akamba society.
The drought and famine of 1984-85 began in
Ukambani with the failure of rain and crops in two
consecutive seasons in 1983 and 1984, combined with
the near total depletion of green fodder sources. The
Akamba suffered up to 60% reductions in hvestock
and liquidated many of their hard-won assets in order
to purchase food (Downing, 1990; Downing e/ cl/.,
1989). The loss of draught animals (oxen) hampered
many farmers recovery from the drought and reduced
their ability to cultivate their croplands in subsequent
years (field interviews, 1985-93; Kamau etal., 1989).
Children and the elderly died from starvation in many
areas and severe and chronic malnutrition were wide-
spread in Machakos and Kitui.
The extent and impact of the drought and the expe-
rience of famine varied substantially among regions,
communities, households, and individuals, with
results ranging from death and disability of family
members to windfall profits from livestock and food
trading. Many people in the wetter, upland areas of
Ukambani had recourse to off-farm enterprises, sav-
ings, and investments, as well as to cash remittances
from urban migrants, while the drier, lowland areas
were harder hit. There, farmers had typically offset
crop losses through livestock sales or remittances and
purchased food at normal prices from the cities or
the highlands. In 1984-85, however, livestock prices
were greatly reduced in a nationally glutted market
(Anyango er al., 1989). Due to the national scale of the
drought, there was for a time virtually no food to be
purchased in the dryland communities of Ukambani,
except at vastly inflated prices beyond the means of
most rural people (Anyango ef al., 1989; Downing et
ol., 1989). Hence, in Machakos, 1984 was known as
the famine Nikwa Ngwete (I Shall Die With the
Money in my Hand) (Alice Mwau, Japheth Kyengo,
and Francis Lelo, personal communication).
Eventually, at least 300,000 people in Kitui and as
many in Machakos received food aid from the
Government of Kenya relief program in 1984-85,
roughly 35% of the population in Machakos and 95%
in Kitui (Anyango et al., 1989; Deloitte, Haskins and
Sells, 1985).
There were pronounced differences in the magni-
tude and character of drought impacts on men and
women. In 60% of households women were already
fending for themselves, their children, and their live-
stock (with or without cash remittances) as legal heads
of household or as dejucto farm managers. As their
usual dry season water sources dried up, women and
children in the drier zones spent substantially longer
hours fetching and carrying water. The feminization
of poverty in Ukambani expanded in 1984 to include
the feminization of famine and of famine response.
Women sought the advice of elder men and experi-
mented widely to identify emergency fodder plants -
a knowledge and responsibility previously in mens
domain (Rocheleau, 1991). They fell back on political
and social skills to gain access to food, fodder, water,
and cash from neighbors, relatives, and absentee
husbands and sons. Their returns to labor decreased
sharply in the daily search for water, food, and fuel,
while their purchasing power with remittance income
(when available) also declined due to rising prices of
scarce food.
As Ukambani began to recover in 1985 rural
people variously attributed their successful survival of
the famine to indigenous food, fodder, and medicinal
plants; cash remittances; migration; group work and
group contacts with official and external sources of
relief aid; and mobilization of family, clan, church,
and other networks of mutual support. Government
authorities and development agencies, in contrast,
emphasized the success of market-based relief poli-
cies in the 1984-85 famine, an approach which they
extended when Ukambani faced another famine seven
years later.
The drought of 1991-92 was more localized and
food was more widely available in the markets, but
cash- and asset-starved farmers could purchase it only
at inflated prices three to four times the normal cost.
Due to the 1984-85 drought, as well as land hunger,
most farmers had few animals to sell and many none
at all. In addition to wage labor on plantations, many
families in the dry lowlands of Machakos and Kitui,
from the poorest to the more successful farmers,
produced and sold charcoal at one-fourth the usual
price to earn cash for food purchases (field interviews,
1993). The landless and near-landless, as well as
smallholders with no trees, either purchased trees to
burn or entered into tree sharecropping arrange-
ments.* The inflated price of food and depressed price
of charcoal resulted in a distorted exchange of trees for
food; trees as assets were liquidated at 8-10% of their
usual market value as measured in food purchasing
power (Rocheleau etul., forthcoming). Unforeseen by
the analysts of the 1985 drought response, the market
as a mechanism for food relief in Ukambani carries
with it a strong incentive to deplete standing trees for
charcoal, preempting future use as sources of food,
fodder, timber, fuel and watershed protection.
As for drought recovery in 1991-92, several
national and international development agencies
attempted to apply another lesson of the last famine,
that the nations drylands were overdependent on
maize. They distributed sorghum and millet seed,
much to the consternation of Akamba farmers other-
wise lacking any seed to plant at the onset of the next
rains, When the millet matured there was a milk short-
age, since the traditional millet porridge requires milk
- already a scarce commodity due to large-scale live-
stock reductions in the 1984-85 famine. With no milk
to prepare the millet, poor farmers sought to trade or
sell millet to procure maize. The price of milk soared,
the price of millet plummeted, and once again,
decreased returns to land and labor, in real terms, left
poor farmers (mostly women) paying for the mistakes
of crisis response policy (field interviews, 1993).
The history and landscape of Ukambani have been
contoured in large part by a succession of crisis con-
structions. After the nomination of Ukambani as
paradise lost -
and soon thereafter as a scene of
devastation -efforts to support the colonial economy
led the British to concentrate the Akamba and their
cattle on limited-area reserves, thereby constructing
a recurrent crisis of famine and cattle disease within
the Akamba community. During the 1930s the Dust
Bowl in the United States turned the eyes of young
colonial officers toward soil erosion and overgraz-
ing, both of which figured prominently in land-use
policy in Kenya for the next 20 years. In the 1940s and
1950s the need for grain to support the war effort
fueled a counter-narrative: wartime low agricultural
productivity and postwar political instability (caused
by land hunger) replaced land degradation as the joint
crisis perceived by the colonial administration in the
dry lands of Kenya. Concomitantly, concern shifted
from overgrazing and soil conversion to a greater
focus on poor cultivaticu practices and a concerted
effort to promote cash cropping by African farmers.
Perceptions of crisis immediately after Independence
centered on poverty and hunger. To remedy both, the
Kenyan government and many of its donors proposed
further agricultural intensification and commercializa-
tion, coupled with land tenure reform, as the solution
to the prevailing definition of rural problems.
In more recent times, the gaze of national and inter-
national attention has shifted to regional manifesta-
tions of the global environmenta! crisis. The 1984-85
famine put East Africa at the epicenter of the world
hunger map, alongside the Sahelian countries and the
Horn of Africa. Other dimensions of the environ-
mental crisis have included fuelwood shortage, rapid
population growth, desertification, and depletion of
trees, forests, water, and wildlife. The multifaceted
nature of the currently perceived environmental crisis
has led to the creation of a dualistic sustainable
development strategy. In special spaces, such as
parks that are the habitat of large mammals, the cur-
rent crisis is couched in terms of paradise at risk. In the
surrounding agricultural areas the crisis turns on an
image of devastated landscapes to be reclaimed for
productive agriculture.
The historical synopsis of the successive crises
constructed in one region provides an important mir-
ror to the various crises currently being defined by
groups with a stake in the future of semi-arid lands in
Kenya and elsewhere. Most development theorists
and practitioners still seek to document or find so-
called solutions to one, discrete local or national cri-
sis. Some scholars, for example, note Kenyas high
rate of population growth and conclude that the coun-
try is heading for disaster (Brown, 1987; Mungala
and Openshaw, 1984) while others point to agricul-
tural innovation and intensification in Ukambani
and herald the dawn of environmental recovery
and economic growth (Tiffen, Mortimore and
Gichuki, 1994).
The history of crisis construction and resolution
in Ukambani calls into question scenarios - whether
pessimistic or optimistic - which interpret just one
problem facing Kenyan society or environment and at
just one scale. Such analyses tend to obscure larger
scale processes which continue to operate behind the
scenes and which will likely localize themselves in
another crisis shortly. Yesterdays solutions do con-
tribute to todays problems. For example, the coming
decade may well see the full fruition of the social
stratification set in motion by the Swynnerton Plan.
Employment opportunities in the military and urban
wage labor have contracted dramatically for Akamba
men. Consequently, the unemployed and near-land-
less will have few options but agricultural expansion,
whether in their home communities or frontiers, or
alternatively in poorly paid causal labor on the farms
of others.
We conclude that one must incorporate multiple
past and present stories of places and peoples before
attempting to solve their problems. All too often,
development practitioners enter an area looking for a
certain problem, find it, and then design a solution.
Our historical perspective suggests that such experts
are as likely to be part of the problem as they are a part
of the solution. Indeed, the history of development
practitioners successively implementing quick-fix
solutions for perceived unidimensional crises has
itself been an enduring crisis for Kenyan society and
the Kenyan environment. If, under the broad umbrella
of sustainable development, a single crisis is defined
from outside and a single solution proposed and
imposed, the prospects are slim for resolution of real
conflict between distinct interest groups within and
outside of the areas concerned. The history of simple
crisis narratives begetting solutions which beget new
crises suggests that a continuing review of multiple
histories and possible futures might serve to inform,
critique, and perhaps transform development theory
and practice.
1. See Kjekshus (1977) for an in-depth discussion of how
colonization and development have affected the ecology of
East Africa.
2. In 1992 and 1993, respectively, Machakos and Kitui
were divided. Part of Machakos became the District of
Makueni and part of Kitui became the District of Mwingi.
Throughout this article, references to Machakos and Kitui
are to the larger, pre-1992 districts.
3. This periodization, while differing from some conven-
tional national-level histories, derives from our reading of
oral histories and archival records for Ukambani.
4. Highland locations with extreme population increases
from 1918 to 1939 included Kangundu (with a 285%
increase), Matungulu (179%) Masii (152%), and lveti
(144%) (Munro, 1975).
5. This is not to say that there was no stratification in
Akamba society prior to colonialization. Differences in
cattle wealth and social status, however, were firmly en-
meshed in a web of reciprocal relations which lessened the
impact of inequality, particularly during times of stress
(Bernard and Thorn, 1981; Mutiso, 1975; Porter, 1965;
Waller, 1985).
6. By contrast, some Europeans studying land degradation
in Kenya not long after this period attributed erosion to agri-
cultural practices on settler lands (Harroy, 1949; Jacks and
Whyte, 1939).
7. See Bromley (1989) on the tradegy of enclosure.
8. DeWilde (1967, p. 94) credits technology transfer for the
widespread adoption of terracing.
9. For Kenya as a whole, less than 0.2% of farms controlled
more than 40% of farmland by the end of the 1980s
(Bradshaw, 1990).
IO. These two facets of the deforestation crisis cannot be so
neatly separated, however. Some farmers liquidated the
standing trees on their land as assets which they sold off in
order to subsidize the establishment of new croplands and
home compounds. One farmer in a frontier zone in Machakos
reported clearing over IO ha of dry forest for charcoal in
order to subsidize establishment of two ha of cropland and a
home compound (field interviews, 1991).
1 I. By January 1985, 50% of the children of Eastern and
Central Kenya, which includes Ukambani, were malnour-
ished (14% severely), and in agroecozone four of that region
only 14% of children were healthy and 64% malnourished
(Anyango et al., 1989).
12. Tree sharecropping involves *borrowing trees from
neighbors, making the charcoal, and returning from one-fifth
to one-half of the charcoal sale price to the tree owner.
Ainsworth, J., A description of the Ukamba Province, East
1 I, No. 1 (1984), pp. 52-83.
Africa Protectorate and its progress under British admin-
Bernard, F. E. and D. J. Thorn, Population pressure and
Journal of rhe Manchesrer Geographical
human carrying capacity in selected locations of
Sociep, Vol. 16 (1900). pp. 178-196.
Machakos and Kitui Districts, Journal of Developing
Ambler, C. H., Kenycr Communities in the Age of Areas, Vol. 15 (1981). pp. 381-406.
Imperalian: The Central Region in the Lute Nineteenth
Bernard, F. E., D. J. Campbell and D. J. Thorn, Carrying
Cenfury (New Haven, CT: Yale University Press, 1988).
capacity of the eastern ecological gradient of Kenya,
Anderson, D. and R. H. Grove, Conservarion in Africa:
National Geographic Research, Vol. 5, No. 4 (1989).
People, Policies und Practice (Cambridge: Cambridge
pp. 399-421.
University Press, 1987).
Blaikie, P., Polikd Economy of Soil Erosion (London:
Anyango, G. eta/., Drought vulnerability in Central and Eastern Longman, 1985).
Kenya in T. E. Downing, K. W. Gitu and C. M. Kamau
Blaikie, P. and H. Brooktield, Lund Degrudafion und Socieo
(Fds.), Coping \Lifh Drou#r In Kequ: Narionul and Local (London: Methuen, 1987).
Strctregies (Boulder, CO: Lynne Rienner, 1989) pp. 15 I-168. Bradshaw, Y. W.. Perpetuating underdevelopment in
Barber, R. Cl., D. B. Thomas and T. R. Moore, Studies on Kenya: The link between agriculture, class, and state,
soil erosion and runoff and proposed design procedures African Studies Review, Vol. 33, No. 1 (1990) pp. l-28.
for terraces in the cultivated, semi-arid areas of Machakos Bromley, D., Property relations and economic relations:
District, Kenya,
in R. P. C. Morgan (Ed.), Soil The other land reform, World Development, Vol. 17,
Conse~crfion (New York: Wiley, 1981). pp. 219-237. No. 6 (1989). pp. 867-877.
Barnes. C., J. Ensminger and P. OKeefe, Wood. Energy, and
Brown, L., Analyzing the demographic trap, in L. Brown,
Households: Perspectives OH Runt1 KenJo (Stockholm:
W. Chandler, C. Flavin, J. Jacobson, C. Pollock, S. Postel,
Beijer Institute: Uppsala: Scandinavian Institute of L. Starke and E. Wolf (Eds.), Store of rhe World. 1987:
African Studies, 1984).
A Worldvutch Institute Reporr on Progress Tobvrrrd
Bates, R., The agrarian origins of MauMau, A~riculturul rr Susfrrinuble Sociery (New York: Norton, 1987).
Ifirtor~, Vol. 61 (1987). pp. l-28.
pp. 2&37.
Beinan. W.. Soil erosion, conservationism and ideas
Chambers, J., The Tyranny of Change: Amrricct in the
about development: A Southrm African exploration. Progressive Era 1900-1917 (New York: St. Martins,
I YO(& 1960. Jourrml of Sotrtheru Afbictrrl Srudies. Vol.
Croll, E. J. and D. J. Parkin, Bush Base, Forest Farm:
Culture, Environmenr and Development (London:
Routledge, 1992).
Dalby, S., Security, modernity, ecology: The dilemmas of
post-Cold War security discourse, Alternatives, Vol. 17,
No. I (1992). pp. 95-134.
Deacon, P. J. and M. B. K. Darkoh, The policies and prac-
tices behind the degradation of Kenyas land resources:
Preliminary review, Journal of Eusrern Africnn Reseurch
and Development, Vol. 17 (1987), pp. 34-5 I.
Deloitte, Haskins and Sells Management Consultants, Finul
Report on the USAID/GOK Food Relief Monitoring und
Evuluation (Nairobi: USAID-Kenya, 1985).
Der Derian, J. and M. Shapiro (Eds.), Internarional/
Intertexrual Relations (Lexington, MA: Lexington Books,
Dewees, Peter A.,
The woodfuel crisis reconsidered:
Observations on the dynamics of abundance and scarcity,
World Developmenr, Vol. 17, No. 8 (1989), pp.
1159-l 172.
DeWilde, J. C., Experiences with Agricultural Developmenr
in Tropical ,4frica (Vol. 2: The Case Studies) (Baltimore,
MD: Johns Hopkins University Press, 1967).
Downing, T. E., Monitoring and responding to famine:
Lessons from the 1984-85 food crisis in Kenya,
Disasrers, Vol. 14, No. 3 (1990). pp. 204-229.
Downing, T. E., K. W. Gitu, C. M. Kamau and J. Barton,
Drought in Kenya, in T. E. Downing, K. W. Gitu and
C. M. Kamau (Eds.), Coping wifh Drought in Keny:
Nurional and Local Straregies (Boulder, CO: Lynne
Rienner, 1989), pp. 3-23.
Dregne. H. E., Erosion and soil productivity in Africa,
Journal of Soil and Water Conservution. Vol. 45, No. 4
(1990), pp. 43 1436.
Dunne. T., Sediment yield and landuse in tropical
catchments, Journal of Hydrology, Vol. 42 (1979).
pp. 281-300.
Dwork, D., War is Good for Bubies und Other Young
Children: A History of the Icfunt und Child Welfare
Movement in England, 1898-19/R (London: Tavistock,
Dyhouse, C., Working-class mothers and infant mortality in
England, 1895-l 9 14, Journul of So&l Hlsrory, Vol. 12
( l978), pp. 248-267.
Easterbrook, D. L.. Kenyan Askari in World War II and
their demobilization with special reference to Machakos
District, in B. Myrick, D. Easterbrook and J. Roelker
(Eds.), Three Aspects of Crisrs in Coloniul Keny
(Syracuse, NY: Maxwell School of Citizenship and Public
Affairs, Syracuse University, 1975), pp. 27-58.
Ecosystems Ltd., Buseline Sumex of Muchukos District:
/9X5 and Lund Use Changes in Muchukos District:
IYXI-1985 (Report No. 4 ,fiv M.I.D.P.). Nairobi:
Ecosystems Ltd., 1986).
Escobar, A., Imaging the post-development era: CritIcal
thought, development, and social movements, Sociul
Text. Vol. 10, Nos. 2/3 ( 1992), pp. 20-56.
Fraser, N., Unruly Prucnces: Polver, Discourse, und Gender
in Conremporury Sociul Theov (Minneapolis, MN:
University of Minnesota Press, 1989).
Gichuki, F., Environmen& Change und Dr&nd Mun-
ugemenr in Mochukos District. Kenya. /930-1990:
Consen~urron Profile (Working Paper No. 56), (London:
ODI, 1991).
Harding, S., The Science Question in Feminism (Ithaca, NY.
Cornell University Press, 1986).
Hardinge, A., Report on the British East Africa Protectorate
for the year 1897-98, Reprinted in British Purliamentury
Papers (Vol. 69: Reports and Correspondence on British
Protectorares in Eust und Cent& Africu, 1890-1899)
(Shannon: Irish University Press, 197 I [originally 18991).
pp. 397-427.
Harroy, J. P., Afrique, terre qui meurt: Lu degrudurion des
so1sAfricuin.v sous linfluence de lu colonisarion, 2e!me ed.
(Brussells: Marcel Hayez, 1949).
Hayes, J. J., Not Enough Wood for the Women: Hobv
Modernizution Limits Access to Resources m the
Domestic Economy of Rural Kenya, Ph.D. Dissertation
(Worcester, MA: Clark University, 1986).
Hecht, S. and A. Cockburn, The Fute of Ihe Forest.
Developers, Destroyers, and Defenders of the Amuzon
(London: Verso, 1990).
Hobley, C. W., Ethnology of A-Kumba und orher East
African Tribes, Reprint (London: Frank Cass. 1971
[originally 1910]).
hooks, b., Yearning; Race, Gender, und Cultrrrul Polirics
(Toronto, Between-the-Lines, 1990).
Hunt, D., The impending Crisis in Kenyu: The Cusefor hnd
Reform (Aldershot: Gower, 1984).
Jacks. G. V. and R. 0. Whyte, Vanishing Lund.7: A World
Survey of Soil Erosion (New York: Doubleday, 1939).
Juma. C., The Gene Hunters: Biotechnology und rhe
Scrumble,for Seeds (Princeton, NJ: Princeton University
Press, 1989).
Kamau, C., G. J. Anyango, M. Gitahi, M. Wainaina and T E.
Downing, Case studies of drought impacts and responses
in Central and Eastern Kenya, in T. E. Downing, K W.
Gitu and C. M. Kamau (Eds.), Coping rclth Droughr m
Kenyu; Nutionul and Locul Struregies (Boulder, CO:
Lynne Rienner, 1989), pp. 2 1 l-230.
Kjekshus, H., Ecology, Control und Economic Developmenr
in Eust Africun HistoT (London: Heinemann. 1977).
Krapf, J. L., Truvels, Reseurches, und Missionury Labors
During un Eighteen Years Residence in Eustern A,fricu,
Reprint (London: Frank Cass, 1968 [originally 18601).
Lal. R., Soil Erosion Problems on un AlJlsol rn We.ttenr
Nigeriu and Their Control (Monogruph No. I) (Ibadan:
IITA, 1976).
Lambert, H. E., Land tenure among the Akamba (Parts I and
II). A,frican Studies (Johunnesburg), Vol. 6, No 3 (1947).
pp. l3l-147;Vo1.6.No.4(1947),pp. 157-175
Lelo, F.. Humunizarion of Wildlife Munugemenr: A Cusr
Study of 01 Deny Subuk N&mu/ Purk. Kenyu, Ph.D.
Dissertation (Worcester, MA: Clark University. 1994).
Louis Berger. Inc., Plunning and Design for Sorl und Wurer-
Consenution in Kitrri DJslrict (Nairobi: Ministry of
Fmance and Plannmg; Washington, DC: USAID. 1985)
Lusigi, W., New approaches to wildhfe conscnatlon In
Kenya. Ambio, Vol. 10, Nos. 2-3 (198 I), pp. 87-92.
Marglin, F. and S. Marglin, Dommunnp Knobv/e&e.
Development. Culrure. Re.ristunr,e (Oxford: Clarendon.
Martinez-Alier. J , Poverty as a cause of environmental
Mimeo (Washington. DC: World Bank.
Mbithi. P. M and C. Barnes, .S/)or1m~1eo~/.\. Sculevwrt
Problems in KenFu (Nairobi: EALB, lY7.5).
Mblthl. P. M. and B. Wiancr, Drought and famlnc m Kenya-
Magnitude and attempted solutions, Jolrnzal ($ East
African Research and Development, Vol. 3 (1973),
pp. 95-143.
Mbogoh, S. G., Crop production, in M. Tiffen (Ed.),
Environmental Change and Drylund Managemenr in
Muchakos Disrricr. Kenya. 1930-I 990: Pro&r&on
Profile (Working Paper No. 5% (London: ODI, 1991).
pp. 1-44.
McCracken, J., Experts and expertise in colonial Malawi,
African Afltiirs, Vol. 81 (1982). pp. 110-l 14.
Meckel, R., Save rhe Bubies: American Public Health
Reform and the Prevention qflnfanl Mortality, 1850-1929
(Baltimore, MD: Johns Hopkins University Press, 1990).
Merchant, C., Ecological Revolurions: Nature. Gender, and
Science in Nev England (Chapel Hill, NC: University of
North Carolina Press, 1989).
Migot-Adholla, S. E., Rural development policy and equal-
ity. in J. D. Barkhan (Ed.), Politics and Public Policy
in Kenya and Tanzania (New York: Praeger, 1984).
pp. 199-232.
Moore, T. R., Land use and erosion in the Machakos Hills,
Annuls of the A.ssociariorr of American Geographers,
Vol. 69, No. 3 (1979). pp. 419431.
Mungala, P. M. and K. Openshaw. Estimation of present
and future demand for woodfuel in Machakos District, in
C. Barnes, J. Ensminger, and P. OKeefe (Eds.), Wood
Energy and Households: Perspecrives OII Rural Kenya
(Stockholm: Beijer Institute; Uppsala: Scandinavian
Institute of African Studies, 1984). pp. 102-123.
Munro, J. F., Colonial Rule and rhe Kamba: Social Change
in the Kenya Highlands (Oxford: Clarendon, 1975).
Mutiso, G. C. M., Kenya: Polirics. Policy. and So&o
(Kempala: EALB, 1975).
Myers, N.. The Sinking Ark (Oxford: Pergamon Press, 1979).
Myrick, B., Colonial initiatives and Kamba reaction in
Machakos District: The destocking issue, 1930-1938, in
B. Myrick. D. Easterbrook, and J. Roelker (Eds.), Three
Aspects of Crisis in Colonial Kenya (Syracuse, NY:
Maxwell School of Citizenship and Public Affairs,
Syracuse University, 1975). pp. l-26.
Nelson, J.. A. Megill and D. McCloskey, Rhetoric of
inquiry, in J. Nelson. A. Megill and D. McCloskey (I%.),
The Rhetoric of rhe Htrtntrt~ Social Science: LunKuage and
Argument in Scholarship and Public @fairs (Madison,
WI: University of Wisconsin Press, 1987). pp. 3-18.
OLeary, M., The Kirui Akamba: Economic and Social
Change in Semi-Arid Kenya (Nairobi: Heinemann. 1984).
Okoth-Ogendo. H. W. O., Tenants of he Crolvn: Elolurion
af Agrarian La\v and b~s~iturions in Kenya (Nairobi:
ACTS. 1991).
Oldfield, M. L. and J. B. Alcorn. Biodiversin: Culnfre:
Conservarion and Ecodel,elopmenr (Boulder, CO:
Westview, 199 I).
Ominde. S. H., Ltrnd and Po~ukrtiorr Mo1,ement.v in Kerqtr
(London: Heinemann, 1968).
Ondiege. P. 0..
Local coping strategies in Machakos
District, Kenya. in D. R. F. Taylor and F. Mackenzie
(Eds.). Development From Within: Sun,ivtr/ in Rural
Africa (London: Routledge. 1992). pp. 125-147.
Owako. F. N., Machakos land and population problems. in
S H. Ominde (Ed.), Studies in Easr Abican Geog-
rap/l! and De~~elopmenr (London: Heinemann. I97 I ).
pp. 177-192.
Peet. R. and M. Watts. Development theory and environ-
ment in an age of market triumphalism, Economic
Geography, Vol. 69, No. 3 (1993), pp. 227-253.
Porter, P. W., The Kamba, in P. W. Porter (Ed.) Adaptation
in Ecological Context: Studies of East African Societies
(Berkeley and Los Angeles: University of California
Press, forthcoming).
Porter, P. W., Environmental potentials and economic
oppottunities: A background for cultural adaptation,
American Anthropologist, Vol. 67, No. 2 (1965),
pp. 409-420.
Redclift, M. R., Susrainable Development: Exploring the
Conrradictions (London: Methuen, 1987).
Rocheleau, D. E., Gender, ecology, and the science of sur-
vival: Stories and lessons from Kenya, Agriculture and
Human Values, Vol. 8, No. 1 (1991), pp. 156-165.
Rocheleau, D. E., An Ecological Analysis ofSoil and Water
Conservation in Hillslope Farming Systems: Plan Sierra,
Dominican Republic, Ph.D. dissertation (Gainesville, FL:
University of Florida, 1984).
Rocheleau, D. E, P. A. Benjamin, A. Field-Juma and P. E.
Steinberg, with A. Nzisa and P. Ondiege, History,
Ecology, and Environmental Change in Easr Africa: The
Case of Ukambuni (Tokyo: United Nations University
Press, forthcoming).
Rocheleau, D. E., K. Wachira, L. Malaret and B. Wanjohi,
Ethnoecological methods to complement local knowl-
edge and farmer innovations in agroforestry, in A. Pacey,
R. Chambers, and L. A. Thrupp (Eds.), Farmer First:
Furmer Innovation and Agricultural Research (London:
Intermediate Technology, 1989), pp. 14-24.
Roe, E., Global warming as analytic tip, Crirical Review,
Vol. 6 (1993), pp. 411427.
Roe, E., Applied narrative analysis: The tangency of liter-
ary criticism, social science, and policy analysis, Near
Liferay History, Vol. 23 (1992), pp. 555-58 I.
Ross, E., Good and bad mothers: Lady philanthropists and
London housewives before the First World War, in
K. McCarthy (Ed.), Lady Bount[ful Revisired: Women,
Philunrhropy, and Puvver (New Brunswick, NJ: Rutgers
University Press, 1990), pp. 174-198.
Sachs, W. (Ed.), Global Ecology (London: Zed, 1993).
Sachs, W. (Ed.), The Development Dictionary; A Guide to
Knobvledge as Power (London: Zed, 1992).
Schmink, M. and C. Wood, Contested Frontiers in Amaronia
(New York: Columbia University Press, 1992).
Shipton, P.,
The Kenyan land tenure reform: Mis-
understandings in the public creation of private property,
in R. E. Downs and S. P. Reyna (Eds.), Land and Sociey
in Contemporan ,4fricu (Hanover, NH: University Press
of New England, 1988). pp. 91-135
Silberfein, M., Rural Change in Machakos. Kenya; A
Hislorical Geography Perspective (Lanham, MD:
University Press of America, 1989).
Spencer, I. R. G., Pastoralism and colonial policy in Kenya,
in R. 1. Rotberg (Ed.), Imperia/ism,
Colonialism, and Hunger: Eust and Central Africa
(Lexington, MA: Heath, 1983), pp. 113-140
Spencer, I. R. G., Settler dominance, agricultural produc-
tion, and the Second World War in Kenya, Journal of
qfiican History. Vol. 21, No. 4 (1980), pp. 497-514.
Stamp, P.. Technology. Gender, and PoLver in A,frica
(Ottawa: IDRC, 1989).
Swynnerton, R. J. M.. A P/an 10 btrens!fi rhe Developmenf of
African Agriculrure in Keny (Nairobi: Government
Printer, 1955).
Tate, H. R., Notes on the Kikuyu and Kamba tribes of
British East Africa, ./orrmc~[ ofrhe Royal Anthropological
Insfitute, Vol. 34 (1904) pp. 130-148.
Thomas, D. B., R. G. Barber and T. R. Moore, Terracing of
cropland in low rainfall areas of Machakos District,
Kenya, Jorrntul of Agriculturul Engineering Research,
Vol. 25 (1981), pp. 57-63.
Thrupp, L. A., Legitimizing local knowledge: From dis-
placement to empowerment for Third World people,
Agriculture and Human Vulues, Vol. 6, No. 3 (1989).
pp. L-24.
Tiffen, M., Environmental Change and Dryland
Mtrnagement in Muchakos District, Kenya, 1930-I 990:
Population Profile (Working Paper No. 54). (London:
ODI, 1991).
Tiffen, M., M. Mortimore and F. Gichuki, More People, Less
Erosion: Environmenral Recovery in Kenya (Chichester:
Wiley, 1994).
Tignor, R. L., Kamba political protest: The destocking con-
troversy of 1938, Infernationul Journal of Africcm
Hismricol Studies, Vol. 2, No. 2 (1971) pp. 237-251.
Timberlake. L., Africa m Crisis (London: Earthscan, 1988).
Vonk. R. B., Report on a Technology Generafion and
Dia,qno.rtic Exercise in the Kathama Areu, MSc. Thesis
(Wageningen, Netherlands: Wageningen Agricultural
University, 1983).
Wailer, R. D., Ecology, migration, and expansion in East
Africa, African Affilirs, Vol. 84 (1985). pp. 347-370.
Wamalwa, B. N., Indigenous knowledge and natural
resources, in A. Kitiro and C. Juma (Eds.), Coining
Ground: Insritrrrional Innovations in Lrrnd-Use
Management in Kenya (Nairobi: ACTS, 1989). pp. 45-65.
Wangati, E., t$$cr.r of L.and Registration on Small-Scale
Farming in Kenya: The Case qf Mbeere in Embrr District,
Ph.D. dissertation (New York: New School for Social
Research, 1991).
Wangati, F. J., Pattern and impact of drought in East
Africa, in Symposium on Droughr in Africa Proaeeding~
and Summary Report. Ottawa August 12-I-/, 1985
(Ottawa: IDRC, 1985).
Wa!ts, M., Development I : Power, knowledge, discursive
practices, Proqess in Human Geography, Vol. 17, No. 2
(1993). pp. 252-272.
Watts, M., The agrarian crisis in Africa: Debating the
crisis, Progress in Human Geography, Vol. 13, No. I
(1989). pp. I-41.
Western, D., Wildlife in rural landscapes. in D. Western
and M. C. Peal (Eds.), Conservarion for the Tbrertfy-First
Century (New York: Oxford University Press, 1988).
pp. 158-165.
Wisner, B., Man-made famine in eastern Kenya: The inter-
relationship of environment and development, in P.
OKeefe and B. Wisner (Eds.), Landuse and Development
(African Environment Special Repot? No. 5) (London:
IAI, 1977). pp. 194-215.
World Bank, Sub-Suharan Africu: From Crisis to Sustainable
Grotcrh (Washington, DC: World Bank, 1989).