‘ .5. u...
‘ r.

I. :1:
.3313 :0 e

r.
:55...“
. a111,?

. .v .. :4

1..) .21... .35..
l: ls: ‘

1
s, V. :11";
9“. 35.
1:3 57- “.9 Runny-v
. F3.“ 2.

e.

 

,. .
a“)

v
‘F

i

:t

‘1:

¢

3.. .1
.2; E. z it
1:31?! 2i. .5}.

 

s; f
£395.! X
‘2‘ .a 3‘. .1. 5!.
{.33.},1517.‘
in: ‘ Z.
. .c. ..
1.1:.
3.1:. ,v
..‘..vv.:‘ .0.
£21.... 5.. 2‘
,. .

.3. it

 

n...
i; . ‘ .

 

 

 

 

 

 

 

 

 

 

.331Z
r » .~ . c.-

‘v. . I , . . :‘-\t $‘D~

, . ‘ , , . lyrfit‘l.

_... uh. V , ‘ . . fit

. .| ., .ly>.. v
. , . » ‘ .7 .
‘ b. r u :1... . v
. s r I» p
.5 r .933. s 1

. .qufi .
.

 

LIBRARY
Michigan State
mversity

This is to certify that the

thesis entitled

LAND, LABOR, AND GENDER RELATIONS:
THE CHAGGA HOMEGARDEN AGROFORESTRY SYSTEM
OF MT. KILIMANJARO, TANZANIA

presented by

Jacquelyn Beatrice Miller

has been accepted towards fulfillment
of the requirements for

M.A. Sociology

Jegree in

Major pmfl/

 

 

Agnew/gm

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/C|RC/DmeDue.p65-p. 15

LAND, LABOR, AND GENDER RELATIONS:
THE CHAGGA HOMEGARDENS OF MT. KILIMANJARO, TANZANIA

By

Jacquelyn Beatrice Miller

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS
Department of Sociology

2001

ABSTRACT

LAND, LABOR, AND GENDER RELATIONS:
THE CHAGGA HOMEGARDENS OF MT. KILIMANJARO, TANZANIA

By

Jacquelyn Beatrice Miller

The case of Mt. Kilimanjaro extends theories of population, environment, land, and labor
relations. It illustrates how, in the current context, even while land is scarce, labor too can
be in shortage. This context includes economic changes brought on by economic reforms,
increasing needs for cash, declining terms of trade for farmers, pressures toward
diversification of income and non-farm activities, out-migration, a shrinking land base,
declining soil fertility, and degradation of natural resources, both on- and off-farm. The
gendered division of labor, the age- and gender-differentiated character of out-migration,
and the inter-dependence of productive and reproductive activities are central to these
processes, which result in a continual squeeze on farmers’ labor and capital resources.
While the case of Kilimanjaro is unique in terms of particular cultural, historical, and
ecological circumstances, these same processes are occurring in other rural locations.
Theories of land and labor relations in rural economies cannot continue to neglect the
trend towards de-agrarianization and diversification of income, even while agriculture
remains of critical necessity for rural households’ sustenance, nor can they neglect the
centrality of gender relations for understanding differentiation within rural households,
and labor constraints in particular. This study used a farming systems approach and
qualitative methods to explore the dynamics of gender, land, and labor relations in

relation to the Chagga homegarden agroforestry system.

Copyright by
JACQUELYN BEATRICE MILLER
2001

ACKNOWLEDGMENTS

Shukrani nyz’ngi sana kwa washiriki wa utafiti huu na wanakijiji wote wa Lyasomboro
walionikaribisha vizuri mno na walionisaidia kwa muda wao, ujuzz’ wao, na juhudi zao
kunisaidia katika utafiti huu. Ni tumaini langu kwamba taarifa hii itasaidia kidogo
kuongeza ufahamu wa shida mnazopambana nazo. My deepest gratitude goes to the
research participants and others in the village of Lyasomboro who welcomed me warmly
and so generously gave of their time, knowledge, and effort in assisting with this
research. I hope that in some small way this report will help to increase understanding of
the complex problems you are facing. I was affiliated with Sokoine University of
Agriculture as a Research Associate in 1998-99. Many thanks to Prof. Aku O'Kting'ati,
my faculty advisor, as well as Profs. L.L.L. Lulandala, George Kajembe, John Kessy,
Joyce Lyimo-Macha, and Dunstan Shemwetta for their support and assistance with this
research. The Kilimanjaro Regional Natural Resource Department granted me permission
to conduct the research. Special thanks to the Selian Agricultural Research Institute,
particularly Charles Lyarnchai and Margaret Kingamkono, the Himo Tree Nursery
Project, “Babu” Dawson Lyimo, and Jackson Mkony (Lyasomboro Village Secretary) for
facilitating my introduction to Lyasomboro and for thoughtful advice. I am grateful to my
committee members, Craig Harris, Alan Rudy, and Anne Ferguson, for their insight and
unflagging support. Last but not least, thanks to my family for their tremendous patience
and encouragement. This research was assisted by an International Predissertation
Fellowship from the Social Science Research Council and the American Council of

Learned Societies with funds provided by the Ford Foundation.

TABLE OF CONTENTS

List of Figures ............................................................................................................. vii
1. Introduction .............................................................................................................. 1
1.1 National and International Context ........................................................ 1

1.2 Lyasongoro, Kilimanjaro Region ........................................................... 5

l .2.1 Physical Geography ................................................................... 5

1.2.2 Social Geography ..................................................................... 1 1

1.3 Chagga Homegarden Agroforestry System ......................................... 14

2. Literature Review ................................................................................................... 16
2.] Population, Environment, Land, and Labor Models ............................ 23

2.1.1 Thomas Malthus: Population Growth and Environmental
Degradation .......................................................................................... 23

2.1.2 Ester Boserup: Agrarian Change Under Population Pressure..24

2.1.3 Clifford Geertz: Agricultural Involution .................................. 25

2.1.4 The Case of Machakos ............................................................. 25

2.1.5 Models of Intensification ......................................................... 27

2.1 .6 Political Ecology ...................................................................... 28

2.1.7 Depopulation and Environmental Degradation ....................... 29

2.1.8 Jane Collins: Simultaneous Land and Labor Scarcity ............. 30

2.2 Gender, Household, Agriculture, and Environment ............................ 33

3. Research Questions, Design, and Methods ............................................................ 37
3.1 Research Questions .............................................................................. 37

3.2 Research Design and Methods ............................................................. 38

4. Results and Analysis .............................................................................................. 41

4.1 Overview .............................................................................................. 41

4.2 Land Scarcity ....................................................................................... 46

4.2.1 Land Tenure, Transfer, and Fragmentation ............................. 46

4.2.2 Land Scarcity as a Constraint .................................................. 52

4.3 Labor Scarcity ...................................................................................... 56

4.3.1 Challenging the Divides ........................................................... 57

4.3.2 Non-Agricultural Income ......................................................... 60

4.3.3 Migration, Work, and Residency Patterns ............................... 64

4.3.4 Division of Labor and Gender Relations ................................. 70

4.3.5 Social Support and Stratification ............................................. 73

4.3.6 Seasonality ............................................................................... 76

4.4 Homegardens, Environmental Dynamics, and Natural Resources ...... 79

4.4.1 Homegardens ........................................................................... 79

4.4.2 Forestry .................................................................................... 86

5. Summary and Conclusions .................................................................................... 88
5. 1 Theoretical Implications ...................................................................... 88

5.2 Recommendations ................................................................................ 90

5.3 Conclusion ........................................................................................... 92
References .................................................................................................................... 94
End Notes ..................................................................................................................... 99

vi

LIST OF FIGURES

Figure 1. Map of Tanzania (detail: Kilimanjaro Region). ............................................. 6
Figure 2. Map of northwestern Kilimanjaro Region ...................................................... 7
Figure 3. Map of Lyasongoro and Marangu. ................................................................. 8
Figure 4. River in the Kilimanjaro landscape. ............................................................... 9
Figure 5. Lyasongoro institutions. ............................................................................... 12
Figure 6. Chagga homegarden vertical strata. ............................................................. 17
Figure 7. Chagga homegarden spatial arrangement ..................................................... 18

Figure 8. Population-environment models in comparison: population, land/labor ratio,
and land base. ............................................................................................................... 20

Figure 9. Population-environment models in comparison: agricultural technology, capital
inputs, and production per area and per person. .......................................................... 21

Figure 10. Population-environment models in comparison: standards of living, carrying

capacity, and key factors. ............................................................................................. 22
Figure 11. Model of simultaneous land and labor scarcity. ......................................... 43
Figure 12. Model of factors affecting simultaneous land and labor scarcity. .............. 44
Figure 13. Privatization and commoditization of salati on common property. ........... 63
Figure 14. Senescence of a mature tree. ...................................................................... 89

vii

1 Introduction

Farmers on Mt. Kilimanjaro, Tanzania, are being squeezed—squeezed literally, as
farm sizes shrink, and also metaphorically as constraints are everywhere tightening. They
are struggling to find the time to accomplish all that needs to be done to maintain their
households and they are suffering through periods of food and cash scarcity, forced to
make trade-offs between necessities. They expect little relief from outside, as the entire
nation is enduring an economic and social crisis—a crisis that might be called acute, if it
were not also chronic.

This study examines the constraints faced by Chagga farmers related to their
distinctive agricultural system, called the Chagga homegarden, or kihamba (plural
vihamba). I argue that while land is scarce on Kilimanjaro, labor is also in shortage,
contrary to theoretical expectations. The paradox of simultaneous labor and land scarcity
on Kilimanjaro can only be explained by understanding that population pressure is
mediated by complex and gendered land and labor relations, as well as by externally-
originating economic forces which result in declining terms of trade for farmers and help
to drive non-farm enterprises and out-migration. These national and international

economic processes are described below, followed by background on the local context.

1. 1 National and International Context

Many of the political-economic changes faced by farmers in Lyasongoro are
rooted in national and international processes, though how they play out may vary in
different regions of Tanzania. In the last two decades, these interrelated processes have

included structural adjustment policies (SAPS), an opening of the economy to capitalist

penetration, declining terms of trade for farmers, and land reform. The former two
processes increase farmers’ dependence on and need for cash, while the third trend makes
this cash less obtainable via agricultural enterprises. Land reform in Tanzania has
occurred in a context of market liberalization, resulting in “pressure towards
individualization of land tenure” (Forster and Maghimbi 1999, xx).

The conditions imposed by the International Monetary Fund (IMF) to implement
economic reforms, begun in 1986, included “devaluation of the currency, imposition of a
wage freeze, abolition of price controls, increase of (agricultural) producer prices,
increase of interest rates, removal of import controls, and restraints on government
expenditure” (Meena 1991, 170). Mbilinyi argues that while large farms benefit from
SAPs, real wages and crop incomes have declined, hurting already disadvantaged sectors
such as small- and middle-scale farmers (Mbilinyi 1997, 354-355). Also, the institution
of cost-sharing in social services such as health and education has strained already scarce
household cash resources and has disproportionately hurt women and children (Meena
1991, 177-181; Mbilinyi 1997, 355; Koda 1995, 142). Forster and Maghimbi (1999,
xviii-xix) contend that even initial assessments of growth in the overall economy were
probably over-optimistic.

In light of these analyses by scholars from Tanzania and abroad, the claims of a
recent World Bank publication are disturbing, to say the least. In diarnetrical opposition
to published reports, the World Bank claims credit for allegedly reducing rural poverty
through impressive growth rates. The report goes on to acknowledge problems of
increased inequity and decline in social sectors such as health and education, but still

maintains:

These problems cannot be solved by increased per capita income alone. But
sustained growth in the economy gives Tanzania some breathing room to begin
addressing these problems in the social sectors. Solving them will add to the
growth potential of the economy and will contribute directly to improved welfare
and an improved distribution of income (World Bank 1999, 5).

The notion of SAPS creating “breathing room” is incompatible with the economic
squeeze faced by farmers on Kilimanjaro, exacerbated by these very policies. The World
Bank’s stubborn perpetuation of refuted claims such as these is thus an example of the
international ideological context contributing to the social and economic crisis in
Tanzania.

The economic and social changes shaped by SAPS are also linked with
environmental impacts in Tanzania. SAPS were linked to increased rates of deforestation
and soil erosion (Sankhayan 1995; Temu and Mamiro 1995; Bagachwa et al. 1995).

In Tanzania, rates of soil erosion and deforestation have been affected by

structural adjustment to the extent that it has increased the cost of fertiliser and the

reduced availability of credit with which to buy it; changed the erosive/non-
erosive balance in the cropping pattern and increased timber extraction, through
internal and external market liberalisation; and reduced spending on afforestation

(Bagachwa et al. 1995, 8).

The environmental impacts of structural adjustment reforms have been difficult to
measure, since other factors simultaneously contribute to the same trends. However,
analysts agree that small farmers tend to be hurt more by structural adjustment policies
and by land degradation, while commercial ventures benefit from market advantages in
less sustainable forms of cropping and timber production (Bagachwa et al. 1995, 9).
Research in the Kilimanjaro region specifically has supported the conclusion that SAPS
have contributed to expansion of farming to more sensitive soils (Temu and Mamiro

1995, 51) and to increased cash crop production “either at the expense of food crops or

through the clearing of more forest lands” (Sankhayan 1995, 13).

In addition to and related to economic reforms, this study also occurs in a context
of de-agrarianisation and diversification of rural livelihoods throughout sub-Saharan
Africa. Traditional models of rural development and of population-environment relations
ignore the diversity of income sources and employment that rural Africans have long
utilized (Clay et al. 1996, 13). Indeed, Bryceson suggests that “the question is why
people continue to engage in agricultural activities at all” and her answer includes the
high cost of purchased food, drastic cutbacks in social services, and attachment to
cultural agrarian values (Bryceson 1997, 9). All this is not to say that agriculture is no
longer important.

Those engaged in income diversification retain their identity as farmers since that

is their fallback and one of their most vital economic activities in terms of

household self-provisioning. In other words, they are still subsistence farmers in
the last instance (Bryceson 1997, 10).

However, evidence suggests that rural Africans increasingly depend on non-agricultural
rural employment as well. Unfortunately, the assumption that most rural Africans are
strictly agrarian persists in misguided development policies (Bryceson 1997, 11). The
present study helps to refute this myth by illustrating the existence and importance of
diverse income sources and migration for rural Tanzanians on Kilimanjaro.

These economic, social, and ideological factors and processes occurring at
national and international levels set the context for this study. However, as should
become clearer in the discussion below, they do not form a static backdrop for the social
and ecological changes on Kilimanjaro, but rather interact with local cultural, social, and
ecological factors. For example, Howard and Millard, in a study of childhood
malnutrition, argue that ideologies of development and modernism have taken a

particularly strong hold on Kilimanjaro. A traditional cultural stigma on poverty in

combination with increasingly tight household constraints has exacerbated the reportedly
increasing social and economic polarization of the wealthy and the poor (Howard and
Millard 1997, 12). This study will explore a different set of processes related to changes

in the farming system, holding these larger processes in mind.

1.2 L yasongoro, Kilimanjaro Region

1.2.1 Physical Geography

Mt. Kilimanjaro is in northern Tanzania near the border with Kenya at about 3°S
latitude and 37°E longitude (Figure 1). The village of Lyasongoro is located on the
southeast Slopes of Mt. Kilimanjaro in Marangu Ward, Moshi Rural District, Kilimanjaro
Region (Figure 2). The village is long and narrow, running from 1500-1900 m.a.s.l.
(5000-6200 ft). It follows a ridge occupied by the main road to the Kilimanjaro National
Park (KINAPA) and bordered by two river valleys (Figure 3). The uppermost border of
the village is adjacent to the forest reserve and park gate. The natural springs in the
village and the two rivers bordering it, the Moonja and the Una, are the main sources of
water, not only for Lyasongoro but also for zones lower in altitude, which receive water
from rivers and from traditional irrigation ditches (Figure 4). Lyasongoro residents also
benefit from piped water provided by the national park and obtained from a spring near
the park gate.

Meteorological data from a station at nearby Marangu School recorded annual

rainfalls ranging from about 1450 to 1850mm (57 to 73 in), with an average rainfall of

Tanzania

International bounmry

 

IIO Kin—alum

150W

 

unanimous-n a-u

Figure 1. Map of Tanzania (detail: Kilimanjaro Region).

 

 
   
  
  

Kiiimanfiro

ar

Kfliman ' ro
Game eserve

”Merriam
into matrb an! Airport

 

Figure 2. Map of northwestern Kilimanjaro Region.

 
   

To the peak

g; Forest Reserve

 

Figure 3. Map of Lyasongoroa and Marangu.

 

a Lyasongoro was formerly known as Lyasomboro. Some people (especially elders) still
call it by this name.

 

Figure 4. River in the Kilimanjaro landscape.

1625mm (64 in) (Anderson 1982, 11). Rainfall is bimodal, concentrated from March to
May and October to December (O’Kting’ati and Kessy 1991, 72). Temperatures vary
between about 19° - 25°C (66° - 77°F) in the region, with wide variations between day
and night temperatures (Anderson 1982, 8). Lyasongoro is significantly colder than the
middle and lower altitudinal zones of the mountain, resulting in much slower crop
growth; for example, the maize growing season in Lyasongoro lasts over seven months,
compared to three months in the lowlands.

It has been said that Kilimanjaro exhibits a “complexity of geology, topography,
climate, and vegetation which is probably unmatched elsewhere in Tanzania” (Anderson
1982, 1). Specifically, the soils between Marangu and Lyasongoro, known as the
Marangu Series, are characterized as moderately to steeply sloping, very stony, dark
brown sandy loams over basalt or analcitite (Anderson 1982, 72 and map of “Soils: South
and East Kilimanjaro”). The parent materials of soils on Kilimanjaro are mostly basic
igneous rock, which tends to produce soils higher in moisture equivalents, nitrogen
content, and cation exchange capacity than acidic rock, accounting for the relatively
fertile soils on Kilimanjaro (Anderson 1982, 31; O’Kting’ati and Kessy 1991, 72).
Relevant to Lyasongoro, “With the lower temperatures at higher altitudes, production of
organic matter tends to be greater than its decomposition and so the soils tend to be high
in organic matter” (Anderson 1982, 9). This contributes to favorable soil physical
properties, such as porosity, which help to reduce erosion. However, the soil is
considered relatively Shallow, pointing to the need for continued attention to soil erosion
prevention (Anderson 1982, 111). While soil fertility is currently declining on

Kilimanjaro, as will be explained in the analysis below, the plentiful rainfall and

10

relatively fertile soils account for the long history of permanent agriculture in the high-

potential coffee-banana belt of Kilimanjaro.

1.2.2 Social Geography

Population densities on Mt. Kilimanjaro are extraordinarily high for a rural
region. According to the most recent national census conducted in 1988, the average
population density in the immediate vicinity of the mountain was 264 people per km2
arable land (684/mi2), but densities as high as 650-1000 people per km2 have been
reported. Population in the region more than tripled between 1948 and 1988 (Gamassa
1991, 1). In Moshi Rural District, average density in 1988 was reported as 220 people per
km2 (570/mi2), the highest density of the three rural districts of Mt. Kilimanjaro (Gamassa
1991, 5). Probably due to out-migration, Moshi Rural had the lowest population grth
rate of the districts in the region: 1% as compared to the regional average of 2.1%
(Gamassa 1991, 3-4).

In Lyasongoro, farmers reported farm sizes ranging from 'A to 5 acres at the most,
with many in the range of 1-3 acres per household. A study published in 1985 found an
average farm size of 0.495 ha (1.22 acres) and an average household Size of 8.5 people
(O’Kting’ati 1985, ii, 92). The number of laboring adults averaged 4.8 per farm or 9.7 per
hectare (O’Kting’ati 1985, iii). Another study found an average homegarden size of 0.68
ha, ranging from 0.2 to 1.2 ha and an average household size of 9.9 people (Femandes et
al. 1984, 82). Households on Kilimanjaro tend to be markedly dispersed in distribution.
Village centers usually represent locations of stores and institutional buildings, such as
schools, churches, and town offices, not concentrations of households. In Lyasongoro, the

local institutions include the Lyasongoro Primary School (Figure 5), the Lyasongoro

11

4

KK ' ‘
”5333105115: YA (‘KASKAZlNl

IVRi" i};y~.r~

5“” 884 31140

wit '-
2,,”

"'43
809 utility

 

Figure 5. Lyasongoro institutions.

Lutheran Church, the town office, and the Kilimanjaro Forest Reserve and National Park,
known locally as “KINAPA.” Neighboring villages house secondary schools and
Catholic churches. The major occupations available to non-migratory residents of the
village include farming, store-keeping, guiding or portering for tourists visiting to climb
the mountain, and other businesses catering to tourists. A variety of other income-
generating activities include beer-brewing, butchering, tailoring, marketing agricultural
products, and tree and flower nurseries.

Mt. Kilimanjaro has long been important to Tanzania for its coffee production,
coffee being the largest source of foreign exchange in Tanzania (estimated at 40% in
1979 (O’Kting’ati and Kessy 1991, 71)). Smallholders on Kilimanjaro contributed 30%
of the nation’s mild-arabica coffee in 1985-86 (O'Kting'ati and Kessy 1991, 72). Between
1900 and 1960, much of the land on Kilimanjaro was appropriated by European colonists
for large coffee estates, by missionaries, or by the national government for forest
reserves. The forest reserve above and the coffee estates below restricted the area
available for expansion by smallholders (Gamassa 1991, 7). A similar and perhaps more
severe Situation of land appropriation and squeezing of Tanzanian smallholder farmers
obtained on Mt. Meru near Arusha (Spear 1997, 90). The famed Kilimanjaro Native
Cooperative Union (KNCU), formed in 1932, enabled Chagga farmers to enter the coffee
market against the resistance of colonial settlers (von C lemm 1964, 100).

The Kilimanjaro Forest Reserve gazetted by the British colonial government in
1921 occupies 1078.8 km2 (Lema 1997, 105). The Kilimanjaro National Park was
established in 1973 and constitutes 756 km2 of the combined forest reserve and national

park. The 1900m countour forms the lower boundary of the forest reserve and national

l3

park (Fernandes et al. 1984, 74) and the upper boundary of Lyasongoro. Harvesting of

live wood in the reserve was banned in 1984 due to over-exploitation (Gamassa 1991, 8).

1.3 Chagga Homegarden Agroforestry System

The Chagga homegarden agroforestry system is the predominant agricultural
system in the high-potential zone from mid- to high-elevation on the southern and eastern
slopes of Mt. Kilimanjaro (the “banana belt”). It is an extremely intensive farming system
that has evolved over the last 200 years (O’Kting’ati and Kessy 1991, 76). The study of
the ecology and sociology of homegardens worldwide is still in its early stages. This
system is an exemplary model of a homegarden system.

In general, homegardens are tropical multi-story agroforestry systems, but this
category encompasses a wide range of forms throughout the tropics. The following
description serves as a good working definition:

...a piece of land with definite boundaries, usually near a house, occupying an area

generally between 0.5 and 2.0 ha, species rich, usually woody perennials

dominant, stratified, resembling natural forest structure (Gliessman 1990, 161-
162).

Other characteristics of tropical homegardens include Spatial and temporal complexity,
continuous groundcover and production in some form throughout the year, and a wide
variety of human use functions (F emandes and Nair 1986, 280-281). They tend to arise in
regions of high population density (Fernandes and Nair 1986). Homegardens vary
tremendously, both between and within types. They may be established from bare soil or
by gradual replacement of natural forest. They have been described as requiring low
management intensity (Perera and Rajapakse 1991; Wojtkowski 1993) or intimate
management (Michon et al. 1983; Femandes et al. 1984). There are contradictions in the

literature on homegarden efficiency, similarity of Species composition across geographic

14

regions, Spatial pattern or lack thereof, and whether homegardens are continually
modified or not (Wojtkowski 1993, 215). For example, while Chagga farmers reported
the labor intensivity and strictures of their homegarden system, many homegardens in
southeast Asia reportedly require very low labor inputs relative to production with
flexible calendars (Torquebiau 1992, 197). Both the complexity of homegardens and lack
of appreciation for their productive and ecological potential have led to a lack of study,
both empirical and theoretical.

Femandes et al. (1984) and O’Kting’ati et al. (1984) provide detailed descriptions
of the Chagga homegardens and of the plant species contained in them, respectively. Due
to the great diversity of crop types and plant uses, I will provide only a general
description drawn from my field research and will refer the reader to these sources for
greater detail. The livestock component of the Chagga homegardens includes stall-fed
cattle (traditional, modern, and cross-bred varieties), goats, chickens, ducks, and other
minor livestock. Major crops include coffee, bananas, maize, beans, cocoyams, cassava,
potatoes, sugar cane, squash, and other vegetables and fruits. Fodder grass is an important
component of the homegardens, as will become clear below. Tree-climbing lianas (vines)
producing fruits or vegetables constitute another homegarden component.

Trees and shrubs are critical to the homegarden system. In addition to coffee and
bananas, trees are used for fuelwood, fodder, building material, livestock bedding, fruit,
shade, soil improvement, beekeeping, medicine, beer-brewing ingredients, banana tree
props, and other uses. An inventory of plant species in the homegardens found a total of
100 plant species (53 tree species, 29 food crops. 21 non—woody plants of economic

value, and 8 weed Species) on 30 farms (O’Kting’ati et al. 1984. 177). My inventory of

15

plant species on two of the farms in this study found 37 and 23 tree Species, respectively.
The Chagga homegardens usually have three or four vertical strata, from the herbaceous
layer to intermediate layers of coffee and fruit trees to the upper canopy stratum (Lyimo
et al. 1999, xix).

The composition of the Chagga homegardens varies across the mountain,
depending on altitude, soils, rainfall, and local preferences. In Lyasongoro, the colder
temperatures and less fertile soils result in slower growth. Root crops grow well and
some farmers are attempting to grow temperate fruit trees (such as apples) and flowers
(such as roses). Figures 6 and 7 illustrate the vertical strata and horizontal arrangement of
the homegardens.

With this introduction to the national, regional, and local context of this study in
terms of political-economic, social, and ecological factors, I now turn to the literature that

frames this study of gender, land, and labor relations on Kilimanjaro.

2 Literature Review

Various models have been proposed to explain the relationship between labor and
land, that is, between changes in population density and environmental conditions. Here I
review a select sample of classical and contemporary models that especially influenced
my research approach and analysis. Traditional models have assumed a straightforward
causal relationship between population pressure and environmental degradation.
However, recent literature, especially in the last decade, has begun to complicate this
relationship. While several models counter Thomas Malthus’s assertion that population
growth necessarily results in environmental degradation, fewer consider that populations

and households are not homogeneous in terms of access to land. Fewer still examine

l6

{KW ‘ "A ‘. "'
1. a-‘J’Oversm

 

Figure 6. Chagga homegarden vertical strata.

House/Sheds/
”Kitchen Garden

A
N.
. a

Overstory/COffeel‘ _ 1'
Bananas ,

Maize/Beans -

.'>~

Fencerow Trees

 

Figure 7. Chagga homegarden spatial arrangement.

differential labor responsibilities at the sub-household level, where gender relations can
be seen to interact with land and labor relations, as well as with ecological factors,
producing bi-directional environmental and social dynamics. Assumptions, predictions,
and emphases of these various models are summarized for cross-comparison in Figures 8,
9, and 10. By extending these theories of population and environment relations and
incorporating perspectives on gender, households, and environment, I develop the
conceptual framework for this study.

Before proceeding, attention is required to the question of what is meant by
“scarcity,” particularly of labor, land, and other resources. Often, the term is used as if
scarcities are objective, isolable, and quantifiable facts, mechanically linked to changes in
population Size, much the way environmental degradation has been assumed by some to
be a “natural” result of population growth. Most of the population-environment models
discussed below refute the naturalization of this assumption. However, none make
explicit that scarcity itself is an outcome of multiple social ecological relations
encompassing technology, ideology, and ecology, among other relations. For example,
the literature on the social construction of needs suggests that this dimension of scarcity,
for one, is not universal (e.g. Cheal 1996).

These relations are contingent on particular temporal and spatial scales. Seasonal
scarcity is just one well-known example. Furthermore, scarcity is usually predicated on
the imposition and naturalization of temporal/spatial boundaries. This becomes obvious
when we consider extreme cases, such as the vast quantities of grain left unconsumed in
the US. while farmers on Kilimanjaro struggle to collect enough fodder to feed their

cattle. However, these boundaries are often taken for granted, even at the local level, and

19

Model Population: Labor/Land Land Base
Ratio

 

 

 

 

 

 

 

 

. Increases Increases Degraded
‘ exponentially
~f Increases Increases Maintained
' Increases Increases Maintained
Increases Increases Improved
Increases Increases Degraded
Increases Increases Maintained or
improved, both
on- and off-farm
May increase, but May increase, but Variable with
may be less may be due to context, but may
relevant than power inequitable access occur under
relations to land or rising, declining,
migration or no population
pressure
Decreases Decreases Degraded
Increases, but Both land and Degraded
patterns of labor scarce, may
» migration and decrease

_ inequitable access
to land critical

 

 

 

 

 

Figure 8. Population-environment models in comparison: population, land/labor ratio,
and land base.

20

: a 1.. ”:3 5w “up

.-..n. .. . - . _
.hilrw‘fi .

: thy,-

.. . ( .. ,(m . ._,. . -.
' "fifth—ml. ~'-1 - 'I
I

t.
r.

*r

v V

v"?!
Ir.-
§—:
:17
.7 (

 

 

 

 

 

 

 

 

 

 

' may be insufficient

labor to implement
appropriate
technologies

 

 

 

Static - Declines Declines
Innovations - Intensifies with More likely to
increased decline than
innovation and increase
labor/ land ratio
Highly labor- None Intensifies Marginally
intensive declines
innovations
; (involution)
Introduced and Yes, both Intensifies Increased
indigenous inputs and
_( agricultural and quasi-fixed
' social innovations capital
Marginal None Intensification, Declines
modifications followed by
declining
productivity
-. Innovations Yes, whether Intensifies Maintained
inputs or
quasi-fixed
capital
‘7 Variable, but may Variable with Variable with Variable with
“7;; be innovations in context context context
' technology and
‘ social institutions
Insufficient labor - Declines Declines
_ “ to implement
* . appropriate
technologies
Innovations, but Variable Declines Declines

 

Figure 9. Population-environment models in comparison: agricultural technology, capital
inputs, and production per area and per person.

 

b Defined by Clay et al. (1996, 10) as nonlabor inputs such as fertilizer or organic matter
and “quasi-fixed capital” such as soil conservation structures, which may be purchased or

produced.

21

 

Model

 

 

 

 

Standards of Carrying Key Factors

Living Capacity

Decline until Constant Natural population

misery promotes increase

population

control

Improve Variable Innovation

Gradually decline Variable Involution (agricultural
and social system able to
absorb excess labor)

Improve Variable Imported innovations,

external markets,
institutional context

 

Decline

Variable, within
narrow limits

Lack of capital inputs,
only labor added

 

 

Improve Variable Capital inputs to
maintain soil fertility, not
just labor; income
diversification

Variable Variable Historical and political-

economic context, non-
homogeneous population,
local knowledge and
innovations

 

Decline

No Simple linear
relation between
population and
degradation

Labor for degradation-
ameliorating practices

 

Decline

 

No simple linear
relation between
population and
degradation

 

Migration patterns, off-
farm income, gendered
relations of labor

 

 

Figure 10. Population-environment models in comparison: standards of living, carrying
capacity, and key factors.

22

the Spatial/temporal distribution of scarcity too often goes unquestioned. With this caveat,
I continue to use the terms land and labor scarcity, but with the reminder that in this
study, as well as in the models discussed below, scarcity should not be uncritically
understood as a given, but rather as produced by particular material/ideological relations

contingent on Spatial and temporal scales.

2.1 Population, Environment, Land, and Labor Models

2.1.1 Thomas Malthus: Population Growth and Environmental Degradation

The model of population-environment relations proposed by Thomas Malthus in
1798 (Malthus 1959) continues to be influential today. In Short, the Malthusian model
holds that the pr0pensity for human populations to increase rapidly, at geometric rates,
while the land base sustains only incremental, arithemetical increases in food production
results in land scarcity and resource abuse. Environmental degradation and starvation
ensue until p0pulation size decreases to a sustainable Size (Malthus 1959, 5).

Critics of the Malthusian model (including those cited below) argue that it
exaggerates and over-simplifies the link between population increase and environmental
degradation. Factors such as technology, culture, and mode of production can intervene
to modify, slow, or reverse environmental degradation. Kilimanjaro is a case in point.
Despite high population growth rates over many decades, Chagga farmers not only
survived, but prospered relative to Tanzanian farmers in less-densely settled regions due

to their elaborate farming system and the favorable coffee market.

23

2.1.2 Ester Boserup: Agrarian Change Under Population Pressure

In 1965, Ester Boserup proposed a model of population-environment relations
that Sharply countered the Malthusian model. In Boserup’s model, population increase is
an independent variable that serves to stimulate agricultural innovation in the form of
intensification of the cropping system. As Tiffen et al. wrote, “Boserup differed from
Malthus in seeing the adoption of new technologies as impelled by population growth and
made feasible by additional labour” (Tiffen et al. 1994, 264). Higher population densities
are thus maintained on the same land area via intensification and increased total
production, resulting in lower output per unit labor but in higher standards of living
compared to more sparse populations (Boserup 1993, 11, 36). In this model, rather than
necessarily resulting (in degradation of the land base, population increase leads to
productive permanent agriculture, with a general progression from under-utilization of
land for agriculture to swidden cropping to permanent cropping to intercropping
(Boserup 1993, 15-16).

Boserup’s model is clearly applicable to the case of Kilimanjaro, where land
scarcity is believed to have motivated the development of the intensive homegarden
agroforestry system. However, Boserup treats intercropping as the most intensive form of
agriculture, as the peak of agricultural innovation. This model does not address further
intensification in the form of vertical stratification, year-round production, agroforestry
systems, or homegardens. Hence, my original research question was formulated to ask
how processes of innovation and intensification operate in an already highly complex and

intensified farming system.

24

2.1.3 Clifford Geertz: Agricultural Involution

Like Boserup, Clifford Geertz (1966, 77-78, 80-81) also proposed a model in
which natural population increase results in agricultural intensification as a response to
land scarcity. However, Geertz’s model was based on highly intensive rice paddy
production in an already densely populated region of Java. This led him to emphasize
labor surplus alongside land scarcity, together which stimulated labor-intensive
innovations that could absorb excess labor and marginally increase production to sustain
the growing population, but at an ever-declining standard of living. He called this process
agricultural involution. As in Boserup’s model, degradation of the land base is not an
inevitable result of population increase in Geertz’s model.

Geertz’s model is an important extension of Boserup’s, addressing an instance of
intensification of an already intensive agricultural system, resulting in declining, rather
than rising, standards of living with population increase. However, the marginal
production gains, albeit with diminishing returns, from labor-led intensification that were
possible in the case of rice paddy production in Java may not be applicable to all
agricultural systems. Whether labor-led intensification might be possible in the Chagga
homegardens is an unanswered question, Since labor is not in excess, as will be seen

below.

2.1.4 The Case of Machakos

Tiffen et al. (1994, 13) provide a model that runs directly counter to the
Malthusian suggestion that population increase leads to environmental degradation. In a
longitudinal study of dryland farming in Machakos District, Kenya, the authors found

that population increase at rates as high as 3% per annum (Tiffen et al. 1994, 261)

25

resulted in agricultural intensification. This was due in large part to an increased labor
supply, among other factors such as invention and introduction of new technology and

market factors (Tiffen et al. 1994, 262-265). As the authors write:

It is argued in this study that the growth of population, working conjointly
through an increase in the labour force and the growth of markets, has driven
agricultural intensification on the smallholdings, and that this intensification has
characteristically taken the form of investments in sustainable technologies and
management (Tiffen et al. 1994, 28).

Both agricultural productivity and environmental sustainability improved as a
consequence.

The authors differ from Malthus in their emphasis on the variable sociopolitical
and institutional contexts of population change. For example, they compare the
contemporary context, which is more conducive to self-determination, with that of
colonial times when labor for soil conservation projects was coerced and local resistance
resulted in only partially successful conservation (Tiffen et al. 1994, 178, 256). Further,
they argue that their study refutes the neo-Malthusian notion of a population carrying
capacity.

There has long been a belief that a given agro-ecological zone has a population-
supporting capacity which cannot be exceeded without environmental
degradation... The dangers of exceeding this ‘scientifically’ assessed carrying
capacity are used to justify government intervention to regulate land use. It is
clear that agricultural output per unit of land varies according to agro-ecological
zone if technology, crop mixes and capital investment are held constant.
However, rising population density combined with access to new knowledge
means that these things do not stay constant; rising population density facilitates
access to new markets, new knowledge and new technologies (Tiffen et al. 1994,
14-15).

They also contrast their model with that of Boserup:
[Our study] makes apparent the importance of two factors given little attention by
Boserup: the external market and inflows of new ideas from external sources,

both of which reinforced the impetus which population growth gave to local
technology generation (Tiffen et al. 1994, 267).

26

A parallel can be drawn between the communal soil conservation activities in
Machakos and the elaborate system of traditional irrigation ditches which route water
from the upper Slopes to the middle and lower slopes on Kilimanjaro. These waterways
still exist and continue to be maintained on a group labor basis, but many are now in
disrepair or have dried up (Lyimo et al. 1999, 106). As predicted by this model,

insufficient labor may be one cause of this decline.

2.1.5 Models of Intensification

Building on Boserup’s model of agricultural intensification, Clay et al. (1996,
10), among others, delineate two broad pathways of intensification of production in
response to population increase and land scarcity. These are termed “capital-led
intensification” (nonlabor inputs such as fertilizer or organic matter and “quasi—fixed
capital” such as soil conservation structures—these may be purchased or produced) and
“labor-led intensification” (only labor added to “crop more densely, weed and harvest
more assiduously, etc.”) (Clay et al. 1996, 10, 57).

The authors argue that the labor-led path to intensification is often not
environmentally sustainable, and they particularly single out the East African tropical
highlands with heavy rainfall and steep Slopes (Clay et al. 1996, 11, 58). They claim that
without some capital-led intensification, especially application of fertilizer, intensifying
production will likely lead to degradation of the land base (Clay et al. 1996, 85). The
predictions of this model contrast with the Geertzian model in which labor alone results
in marginal but sustainable intensification. It would apply best where the limiting factor

of production is soil fertility, where productivity levels are already close to maximum

27

sustainable yields, and where the relevant management techniques for increasing
production are capital-led.

However, the definitions of capital-led and labor-led intensification used by Clay
et al. are problematic in that so-called capital-led strategies are defined to include quasi-
fixed capital, such as soil conservation structures, that can be produced using labor.
Hence, the distinction confuses any comparison of strategies requiring cash versus labor
versus both, and confers unwarranted strength to the argument that sustainable agriculture
necessitates cash cropping and chemical inputs. The definitions also ignore the potential
for what might be termed “knowledge-led” strategies, such as the innovative technologies
that were important in the Machakos case and the crop combinations or other system-
specific knowledge that allow vertical stratification and intensification in the traditional

Chagga homegardens.

2.1.6 Political Ecology

Another type of approach to demographic and environmental change has been
labeled political ecology by Piers Blaikie and Harold Brookfield (1987). It “combines the
concerns of ecology and a broadly defined political economy. Together this encompasses
the constantly shifting dialectic between society and land-based resources, and also
within classes and groups within society itself” (Blaikie and Brookfield 1987, 17).
Political ecology is an umbrella term for analytical approaches which share these
concerns; which also include the historical, political, and economic context in analysis;
and which place the land manager at “center stage” of analysis (Blaikie and Brookfield

1987, 239).

28

The implication of this approach for population-environment relations is that
“population” cannot be treated as a single, independent causal variable (Sage 1994, 39)
and populations themselves cannot be treated as homogeneous, Since power relations
among groups are central to social—environmental dynamics. In the political ecology
approach, land scarcity may be the result of unequal access to land or migration patterns,
not only natural population increase. Production problems and environmental degradation
may be due to unequal access to capital and resources, not only land scarcity. Depending
on the local context, land scarcity and labor surplus may stimulate innovations by
farmers, whether by intensifying production or by creating local institutions or social-
political mechanisms to cope with constraints. These may be what I have termed
knowledge-led strategies, or innovations in management techniques within an intimately
understood indigenous knowledge system. For example, Chagga farmers’ knowledge of
crop associations appropriate to light/Shade requirements and to local soils constitutes
knowledge which can lead to intensification of agricultural production. These innovations
may or may not reverse or ameliorate environmental degradation, depending on the

particularities of the local, regional, national, and global political contexts.

2.1.7 Depopulation and Environmental Degradation

With the exception of political ecology, the preceding models of population and
environment relations assume a growing population and a labor surplus. However,
patterns of rural to urban migration and the impact of HIV/AIDS on rural agricultural
populations, particularly in sub-Saharan Africa and parts of Asia, demand that greater
empirical and theoretical attention be paid to the relationship between depopulation and

environmental conditions.

29

Blaikie and Brookfield (1987, 34) argue that environmental degradation can be a
result of declining population, particularly when labor required to maintain conservation
structures in sensitive systems is no longer available. Likewise, a study of rural villages
in Oaxaca, Mexico, found that national structural adjustments led to increasing semi-
proletarianization and urban migration, resulting in depopulation of the region. With
insufficient labor to maintain elaborate terracing systems, depopulation led to a decrease
in per capita food resources (Garcia-Barrios and Garcia-Barrios 1990, 1569; see also
MacDonald et al. 2000). These findings regarding depopulation are relevant to the
current situation on Kilimanjaro where, although population density remains high, labor

is scarce and is a factor contributing to resource degradation and poverty.

2.1.8 Jane Collins: Simultaneous Land and Labor Scarcity

Finally, of all the models of population, environment, land, and labor relations,
only the analysis developed by Jane Collins explicitly addresses the possibility of
simultaneous land and labor scarcity. Collins argues that land scarcity does not
necessarily translate into labor surplus at the household level and may, in fact, lead to
labor scarcity when coinciding with other political economic factors. She explains:

While land scarcity and labor scarcity might appear to be in contradiction, they
actually may occur together. First, land scarcity may arise from processes of land
transfer and encroachment as well as demographic growth. Secondly, whatever
the cause of land scarcity, the downward pressure it exerts on household income
may force productive members into other activities, or to leave rural areas
permanently. There is no guarantee that demographic growth and rural exodus
will balance one another in the end. The latter depends on a complex of factors
including the availability and reliability of alternative work, the skills such work
requires and the possibilities of acquiring them, the recruitment and employment
practices of industries involved, cultural definitions and expectations of work, and
a range of market forces (Collins 1987, 22).

30

Thus, unequal access to land, off-farm sources of income, patterns of rural to urban
migration, and various local and extra-local political economic factors must be taken into
consideration.

Furthermore, in her research in the Peruvian highlands, Collins found that the
convergence of land and labor scarcity frequently led to environmental degradation,
which in turn led to even greater demands on household labor.

The need to participate in these diverse activities is frequently related to a

diminishing land base or to declining terms of trade. This dynamic may lead to

problems of labor scarcity. Labor scarcity may, in turn, lead to patterns of poor
resource management, thus creating a vicious circle of impoverishment. As
natural-resource quality declines, families must struggle even harder to maintain

constant levels of income. In consequence, still more labor may be drawn away
from the farm (Collins 1987, 33).

In contrast to this analysis, the models of Boserup, Geertz, and Tiffen et al., address cases
where land scarcity does not result in environmental degradation because sufficient labor
is available for sustainable intensification of agricultural production. While Collins finds
that population growth may lead to environmental degradation, this is far from a simple
or necessary outcome, but rather is mediated by complex relations of land, labor,
migration, and ecology in interaction with the political economic context—particularly
the declining terms of trade faced by so many small farmers worldwide.

The prediction that agricultural production will intensify given land scarcity is
premised on what Collins calls the myth of labor surplus. She traces this myth to the
period of the 19603 and 19705 when:

Given the high rates of population growth in developing nations, it seemed

inconceivable that complex interrelationships between population growth and

population movement could give rise to regionally and temporally specific
problems of labor scarcity (Collins 1987, 26).

31

What is absent from many traditional population-environment models is an understanding

of the importance of off-farm activities and household reproduction activities, in addition

to agricultural production, when assessing labor availability.
In view of the frequent assumption that labor is the resource most freely available
to small farmers in developing nations, the questions raised in this chapter may
seem to contradict some of our basic intuitions about the nature of rural poverty.
Yet a careful reading of the available literature reveals that the crises of
reproduction increasingly experienced by rural households in many parts of the
world involve a complex relationship between declining terms of trade,
intensification of production, and migration and off-farm labor. As producers
attempt to maintain constant levels of income and to allocate labor between new

and old forms of economic activity, ecological decline is not an uncommon result
(Collins 1987, 20).

This analysis implies that labor-intensive strategies for rural development may be
inappropriate in such contexts, which may be increasingly relevant given declining terms
of trade and increasing diversification of activities for small farmers worldwide.

In addition to patterns of migration and off-farm activities, Collins highlights the
labor required for household reproduction as key for conceptualizing simultaneous land
and labor scarcity (Collins 1987, 24). All three of these processes are gender- and age-
differentiated in ways specific to local or regional contexts and in sub-Saharan Afiica (as
in most of the world (Deere and Ledn de Leal 1981, 339)) women are typically
responsible for household reproductive labor. In the context of a declining land base and
environmental degradation, more labor iS required to meet the same level of subsistence
for household reproduction. These subsistence activities typically depend not only on the
farm plot itself, but also on natural resources available off-farm in common areas, such as
water, fuelwood, fodder, and medicinals. While the gender division of labor in rural
contexts varies, where women are responsible for household reproductive activities these

have tended to be ignored in past economic analyses (Mbilinyi 1997, 326). Traditional

32

 

economic analyses that take household reproduction for granted have perpetuated the
myth of labor surplus and have distorted understanding of population-environment
relations and simultaneous land and labor scarcity.

Of the eight models of population-environment relations above, only the model
described by Collins accounts for the simultaneous land and labor scarcity existing on
Mt. Kilimanjaro today. To the best of Collins’s knowledge and mine, no other literature
explicitly addresses even the possibility of Simultaneous land and labor scarcity. This
study of Kilimanjaro therefore provides an empirical case of this under-examined
dynamic. It also provides a theoretical extension of Collins’s model by focusing more
explicitly on the gender relations of labor at the household level and the implications of
these for Simultaneous land and labor scarcity and for social-environmental interactions.
Collins’s model implies, though does not spell out, the importance of the gender and age
organization of household labor, off-farm labor, and migration. These prove critical for
understanding the Situation on Kilimanjaro. The literature on gender, environment, and
development, which begins with women’s experiences as environmental and household
managers, thus intersects with Collins’ model. I review this literature in the next section
as a basis for understanding the role of gender in population, environment, land, and

labor relations on Kilimanjaro.

2.2 Gender, Agriculture, Environment, and Household

Academic awareness of the importance of gender in theorizing processes in less-
industrialized countries began with Boserup in the field of gender and agriculture,
particularly the gendered division of labor. Boserup’s work began to be expanded in the

19705 by socialist feminist analyses of agriculture. which tended to be highly structural

33

critiques of capitalist agriculture and the exploitation of women’s reproductive and
productive labor (e.g. Deere and Leon de Leal 1981; Beneria and Sen 1981). These
developments coincided with attention to “women in development” more generally and,
in the 19803, were followed by growing awareness of the Significance of gender in
relation to environmental/natural resource issues.

Both early socialist feminist analyses of agriculture and the early gender and
environment literature rested on assumptions of a clear productive/reproductive divide.
Much of the early gender and environment literature contrasted with the socialist feminist
literature in its lack of a structural analysis, material emphasis, or strong empirical basis.
It also tended to portray women as victims of environmental degradation rather than
agents of environmental change and natural resource management in their own right.
However, its emphasis on the reproductive Sphere did eventually bring to light the
gendered character and importance of activities such as fuelwood, water, and fodder
management. Factoring the ecological and labor dynamics of these activities into
Collins’s model of labor and land scarcity forms the core theoretical framework
developed and applied in this study.

Furthermore, the gender-environment field is diverse and more recent scholarship
has made strides in theoretical rigor. Rocheleau er al. identify six major schools of
feminist thought on the environment: ecofeminist, environmentalist, feminist
poststructuralist, feminist environmentalist (after Agarwal 1991, 10-11), socialist
feminist, and feminist political ecological (their own perspective) (Rocheleau et al. 1996,

3-4). Of these schools, this study is most strongly informed by the latter three, which tend

34

to be more applicable to the context of less-industrialized countries and which emphasize
the material basis of the gender-environment relation.

Incorporating gender into development theory dramatically altered analyses of
poverty, land, labor, and productive/reproductive relations. It did so first and foremost by
exploding the myth of the homogeneous household, previously assumed to be the
appropriate unit of social and economic analysis. Gender analysis implies attending to
intra—household relations and sub-household dynamics in addition to household level
processes, which is the approach taken in this study. Speaking of the Tanzanian context,
Koda’S (1995, 149-150) work illustrates that this is requisite not only to understand the
heterogeneity of Tanzanian households, but also the dynamism and flexibility of intra-
household, gender, and age relations.

Given that current changes in household relations are closely linked to the context
of economic reform and social and economic crisis outlined above, a review of the
literature on gender, agriculture, and structural adjustment in Tanzania is called for.
Mbilinyi (1997) and Meena (1991) offer complementary analyses which illustrate how
important a gendered perspective is to understanding household economy.

Meena Shows how each of the conditions imposed by the IMF (devaluation, wage
freezes, end of price controls, increase of producer prices and interest rates, removal of
import controls, and cuts in government expenditures) failed to consider that effects
would not be equitable for all producers, particularly given the gender division of labor
and differential access to credit, land, and other resources (Meena 1991, 170). Rather,

“the IMF package has been totally gender blind; i.e., women seem to be absent from the

35

reform measures” (Meena 1991, 186). The burdens of reform measures have therefore

tended to fall more heavily on women. Currency devaluation is one case in point:
Devaluation has been eroding the real wages of both rural population and urban
population alike. When devaluation erodes the real income of peasants, there is a
tendency for the male population to migrate to urban areas to seek wage
employment. But even if wage employment is available to a Significant proportion
of the population, wages alone do not enable wage earners to subsist. The female
adult population that remains in the rural sector often increases its workload, not

only in order to fend for the family, but also to subsidize the wage earner in the
urban area (Meena 1991 , 170).

Cuts in government expenditures in social sectors also weigh more heavily on women
and children. Meena argues that SAPS exacerbated already existing crises in education
and health (Meena 1991, 177-179). For example, in the health sector:
The IMF recommended that the community Should take a greater role in the
provision of its own health services. This recommendation was made at a time

when the real income of the people had deteriorated and the state had already
invested in infrastructure for health services (Meena 1991, 177).

Meena argues that women’s health intersects with the high demands on women’s labor,
also exacerbated by the economic crisis, and points out the absurdity of reducing
investment in “nonproductive” sectors such as health for the sake of productivity when
productivity in the Tanzanian context (particularly in agriculture) is so dependent on
human labor power (Meena 1991, 179).

Mbilinyi, focusing on gender and agriculture, likewise argues that tendencies to
homogenize households, large- and small-scale agriculture, and gender, ethnic, race, and
nationality differences, result in distorted analyses of structural adjustment policies. She
finds that SAPS “ultimately favour the foreign-owned, large-scale sector, and benefit men
more than women” (Mbilinyi 1997, 317-318). Having documented the increased use of
women’s labor in large-scale commercial agriculture under SAPS, Mbilinyi laments the

’99

“blinkered focus on ‘indigenous’ peasant ‘households by donors, the government, and

36

researchers in Tanzania, a myopia which is again rooted in outdated assumptions of
homogeneous households (Mbilinyi 1997, 332). Hence, the absence of analysis of
dynamic and complex gender relations in the discourse surrounding SAPS actually serves
to exacerbate gender and class inequities in the agricultural sector. Although this research

9”

does, for better or worse, focus on “‘indigenous’ peasant ‘households, the use of a
gender analysis which takes into account the heterogeneity and changing dynamics of

households in the context of economic crisis helps to problematize the assumptions of

development and economic reform discourse.

3 Research Questions, Design, and Methods

3. 1 Research Questions

Initially, my research question was formulated as follows: How do the processes
of innovation and intensification, first predicted by Ester Boserup (1965) and Clifford
Geertz (1966), play out in an already highly complex and intensified homegarden
agroforestry system, in an already extremely densely populated region?

When labor turned out to not be in surplus, and therefore intensification of
production to not be occurring, this question was modified to focus on innovations or
techniques to minimize or halt the decline of production and maximize the efficiency of
labor. The paradox of how labor and land could be Simultaneously scarce became a new
research question, with gender, migration, and the gender division of labor as central
variables. This paradox can only be explained by understanding that population pressure
is mediated by complex and gendered land and labor relations, as well as by external

economic forces which encourage off-farm enterprises and out-migration.

37

3.2 Research Design and Methods

The framework used primarily to inform the choice of methods in this study was
farming systems methodology. A farming systems approach encouraged viewing Chagga
farm households through as wide-angled a lens as possible (e.g. Bawden 1995). Each
component of the household and community economy, culture, and biophysical
environment was considered to be potentially relevant to the others. Studying the farming
system as a whole also more closely approximates the farmers’ own perspective and
highlights systemic issues such as labor or cash flow constraints, household relations, and
cultural practices which might influence the farming system. The multiple types of
agricultural components, agronomic strategies, and household livelihood strategies of
typical Chagga farmers made a systemic approach especially appropriate and necessary
for understanding relationships among complex, inter-dependent factors. Farming
systems methodology also provided a framework for integrating social and ecological
data, by incorporating both “hard systems” and “soft systems”, as well as farmers’
subjective experience (Bawden 1995, 14).

Thus, the Chagga system was studied from the perspective of the farmers
themselves. This approach is compatible with feminist and participatory research
approaches, which advocate beginning with the subjective experience and expressed
priorities of research participants and building from there to an integrated, social
theoretical interpretation. Drawing on feminist approaches to research, Chagga women’s
experiences were valued and explicit attention was paid to household gender relations

and cultural gender dynamics.

38

This methodological framework implied the use of primarily qualitative methods,
in particular repeated, open-ended interviews, walking tours of farms, and participant
observation. Formal interviews were conducted with an opportunistic sample of ten
households in Lyasongoro, comprising twelve individual interviewees: ten women
ranging in age from 18 to about 60 and two men in their 503 or 60s. The sample was
skewed towards female interviewees and female-headed households (whether de jure or
de facto) due to my gender and expectations that I would want to learn about women’s
issues. This over-sampling was an advantage rather than a disadvantage, as constraints
faced by women did become a central concern of the study. Two group interviews were
conducted, an initial one to ascertain the characteristics of the village and invite farmers
to participate in the research and another with a village women’s business enterprise
group. Interviews were conducted and transcribed in Swahili and field notes were taken
in both Swahili and English. Additional ethnographic data were collected through
participant observation and casual conversations. I resided in a village five kilometers
from Lyasongoro and gathered field notes on experiences there and in other neighboring
villages. An ecological inventory was conducted on two of the farms. I lived on
Kilimanjaro intermittently from March through July 1999. Time conducting research in
the field totaled approximately three months.

Due to the complexity of the Chagga homegardens, farm tours and participant
observation proved very valuable for gaining experiential knowledge through daily
encounters with different farms, farmers, locales, and weather. For example, farmers may
not consider specific non-crop plants worth mentioning, but upon encountering them will

elaborate their uses or their interactions with other plants. Farmers may have so many

39

species of trees that they will recall them all only on a farm tour, not in an interview. The
labor-intensiveness or Significance of a particular farm management operation might only
be apparent when encountered in season.

The research was originally conceived as an exploratory study. As such, the scope
of research topics was initially broader than the period of research in the field could
support for in-depth study. The topics included in this paper represent a subset of the
issues addressed in the exploratory phase of the study. The initial broad approach
provided an overview of significant issues and the repeated, in-depth interviews provided
rich, narrative qualitative data. However, the small sample size of both interviewees
(twelve) and villages (one, with some comparison based on secondary literature and
visiting other villages) and the choice of an opportunistic rather than statistical sampling
approach strictly limit the generalizability of this study. The scale at which primary data
were obtained is restricted to the village, household, and individual levels, thus limiting
the extent to which these processes can be linked to zonal, regional, national, and global
processes. I chose the village of Lyasongoro because of its accessibility and because it is
adjacent to the forest reserve. I wished to explore the relationships between the reserve
and the village.

All interviews were conducted in Swahili without an interpreter. Swahili is not the
first language of either the respondents nor myself, but proved adequate for discussion of
most issues related to the agricultural system, household livelihood strategies, and social
and ecological change. Most formal interviews were tape-recorded or else notes and
quotations were recorded by hand. These data were transcribed to electronic text files for

coding and analysis. The major data analysis tool was Atlas/ti® textual analysis software,

40

which was used both during and after field research to organize and code field notes and
transcribed interviews. This software is designed to support inductive, iterative research
in the tradition of grounded theory (Corbin and Strauss 1990). Microsoft Access database
software was used to record mostly numeric household demographic and farming data.
The approach taken to interpretation of the data from this study prioritizes
farmers’ own perceptions, accounts, and narratives regarding their world and their lives
and experiences. I take these accounts as valuable and meaningful regardless of whether
there can be or have been verified as “factual.” Therefore, throughout the study I will
commonly use phrases such as “reportedly” or “the farmers reported” to indicate that the
analysis is based on the statements of farmers themselves, supplemented by field
observations and secondary data. Quotations used within this document are translated to
English based on meaningful, rather than literal, translation. However, the literal
transcription in Swahili is provided in the endnotes. Interviewer’s dialogue is provided in
brackets and notes are provided in parentheses. Because Swahili has a gender-neutral
pronoun, this is variably translated as feminine, masculine, or neutral, depending on the

CODICXI.

4 Results and Analysis

4.1 Overview

The social and environmental changes seen on Mt. Kilimanjaro in the last twenty
years are exceedingly complex. They are influenced not only by a large number of factors
at the local level, but also by the regional, national, and international political-economic

context. Some important developments at these wider scales, such as declining terms of

41

trade for farmers and economic reforms, were addressed in the Introduction (Section 1.1).
These larger trends should be borne in mind even while I turn here to one set of critical,
inter-related trends particular to Kilimanjaro.

The dynamics of and evidence for these trends will be discussed in detail below,
but first I will briefly articulate the overall argument. First, the average Size of kihamba
plots on the mountain is declining. This is due primarily to land tenure patterns (that is,
the culturally normative practice of bequeathing a portion of kihamba land to each son) in
conjunction with population increase. A simple model of population-environment
relations might predict that land scarcity would lead to labor-led agricultural
intensification in order to increase productivity on smaller plots of land. However, in
addition to small plots, economic forces are driving local Chaggas, especially youth and
men, out of farming and into other types of enterprises and/or off the mountain to seek
livelihoods elsewhere. As foremost among these forces, I would highlight declining terms
of trade for farmers and increasing household needs for cash, both of which have been
linked to changes brought about by structural adjustment programs (Mbilinyi 1995;
Meena 1991). Due to these dual processes of out-migration and diversification of
household strategies to include off-farm enterprises, household agricultural labor has
become scarce. Therefore, both land and labor are limiting factors and labor cannot be
expected to easily compensate for land scarcity through intensification of production
(Figure 11).

These critical trends concern not only land in crop production, but also the on-
and off-farm natural resources necessary for household reproduction (Figure 12). Fodder

for stall-fed cattle is particularly critical in this regard, due to the importance of cattle

42

 

Land

 

 

 

 

 

 

 

 

 

 

 

 

 

Scarcity
Gender
Population Environmental
Pressure Degradation
Gender
V
Labor
Scarcity

 

 

 

 

External Economic Forces

 

 

 

Figure 11. Model of Simultaneous land and labor scarcity.

43

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Non-Agricultural Income Production Reproduction
Land F uelwood
Gender
Labor Fodder

 

 

 

 

Declining Terms of Trade

 

 

 

Declining Land Base

 

 

Figure 12. Model of factors affecting simultaneous land and labor scarcity.

44

 

manure in maintaining soil fertility and crop production. In the past, fodder grass was
readily available in the interstitial spaces within or between farms, but farmers report that
it is increasingly difficult to obtain. Scarce land as well as inadequate labor for traveling
increasing distances both limit the availability of fodder and hence the number of cattle
which can be maintained per household. The quantity of livestock reportedly declined by
50% between the 19505 and 19805 due to lack of fodder and increasing labor demands
(O’Kting’ati 1985, iii). Other evidence of the declining availability of fodder is that in the
19605, the sale of fodder was virtually unknown, but now people travel up to 10km to
obtain it (O’Kting’ati and Kessy 1991, 77). The declining quantity of livestock kept has
various consequences for household sustenance, but particularly relevant to the trends in
land and labor is that the cattle manure available for fertilizer is inadequate to maintain
soil fertility at past levels, much less to intensify production on smaller plots of land.

This simplified outline of the major inter-dependent social and environmental
trends on Kilimanjaro illustrates how social changes at multiple scales have contributed
to on- and off-farm resource degradation, which in turn has put further constraints on the
ability of households to sustain themselves. The scope, quantity, and rates of these
changes on Kilimanjaro are important questions that have been and continue to be
examined by various researchers (e.g. Lyimo et al. 1999). The purpose of the present
qualitative analysis is to elaborate these processes in greater detail as they relate to each
other and as they are experienced by the farmers themselves. While clarity of exposition
requires that different processes be addressed in turn, the highly relational character of

these processes makes this exposition challenging. It will become evident that the way

45

farmers themselves talk about their lives indicates a high awareness of the complex and

tightly bound relationships among the processes that affect them.

4.2 Land Scarcity

I begin this analysis with the critical trend of land scarcity, already well-
recognized and established in the literature on Kilimanjaro (e.g. Femandes et al. 1984;
Gamassa 1991; Lema 1997; Lyimo et al. 1999; O’Kting’ati and Kessy 1991). For most of
the farmers interviewed, small plot Size and unavailability of local land was indeed
identified as a primary constraint on household sustenance. However, rather than
accepting the assumption that natural population increase directly causes land scarcity, it
is necessary to understand the land tenure system, land inheritance, variability of
landholding size, and land fragmentation in the local cultural context as critical mediators

affecting land scarcity.

4.2.1 Land Tenure, Transfer, and Fragmentation

The Chagga have a unique cultural tradition associated with land tenure and
inheritance. This tradition is embodied in the term kihamba which, though often used in
the scientific literature to refer to the Chagga homegardens themselves, more accurately
refers to the class of mountain land which is considered ancestral land. When asked
whether kihamba land is owned by individuals, clans, or the government, most farmers
responded that it is the property of clans or, in certain senses, individuals, despite the
legal status of all land in Tanzania as property of the government (Shivji 1998, 81).

Vihamba land is the property of the clan, it’s not permitted to sell. Each person

has an inheritance, his child has an inheritance, and so on. It’s not the property of
the government.l

46

 

Indeed, the Chagga have succeeded historically in maintaining clan control over occupied
land on the mountain, even while villagization programs were carried out in other parts of
Tanzania by the early socialist government (Shivji 1998, 13(fn)). This resistance to
villagization means that the traditional dispersed pattern of homesteads surrounded by
agroforestry plots has been maintained to this day.

Clan control of land also results in highly secure land tenure, encouraging long-
term investment, without, however, the uncontrolled right of individuals to transact land.
The right to transact land is often part of economic definitions of security of land tenure
(e.g. Clay et al. 1996, xii, 25), but is not requisite for secure tenure, as the case of
Chaggaland illustrates. Title deeds are usually unnecessary, Since the borders of property,
marked by a particular tree (sale—Draecaena afromontana). Sale of kihamba land was
reported to be rare in Lyasongoro. It usually occurs between clan members, and only with
the sanction of the clan for some Special circumstance. When I asked why Chagga who
have jobs elsewhere do not sell their land, one farmer responded with extreme surprise,
“To sell your farm?! Ha! Where will you live?”2 When I said that these people already

live elsewhere, She said, “It’s very hard for a Chagga to sell his farm.”3

She pointed out
that even if he does not live on the farm now, he may move there in retirement, his son or
grandson will want to farm it, and he will want to be buried there. If he were to sell, it
would only be to a relative.

Strong cultural and Spiritual meaning is attached to the particular ancestral lands
of each Chagga. This is evidenced by massive annual pilgrimages (at Christrnastime, and

smaller pilgrimages at Easter) to the mountain by Chagga who live away, and by the

strong desire of the Chagga to be buried on the land where their ancestors are buried.

47

Also, critical for understanding the process of land fragmentation, farmers reported a
strong cultural value attached to each son inheriting a piece of land from his father, even
if the small size of the plot forces him to make his living elsewhere. When asked whether
one of her sons would need to obtain another farm or find other work, given the small
size of her farm, a farmer replied, “Even if he will work and get another farm, he must
get his share.”4 Similarly, another farmer elaborated:
Even if it’s a small farm, each one gets his space to build a house. He gets at least
a place to put a garden. If it’s a small farm, indeed, this is how it’s arranged. If
he’s able, if he has his livelihood like this one, he can buy a farm someplace else.
But here at home he gets his space to build his house. Indeed, this is how it’s done
in our parts. If there’s a person with a small farm and each one is allotted his
small place, he puts his house there and his little garden. Then if he goes where

there’s work and he has cash, he buys a farm some other place, but here at home
he has a place to hold on to.5

As one villager succinctly put it, “We don’t have farms here, we just have little plots to
live on.”6

Farmers claimed that in most cases this cultural value attached to land inheritance
overrides the desire to maintain consolidated plots large enough to viably support a
family, though fathers may decide to allocate larger plots to some sons and smaller plots
to others. According to tradition, the larger plots are given to the first- and last-bom sons,
the last-born being given the responsibility to care for his parents and to inherit their
portion of the ancestral land at the time his father chooses. One respondent argued that in
the next generation, children will be forced to leave for lack of land, despite the
reluctance of Chagga to leave Kilimanjaro. I remarked that otherwise it might become
like a city on the mountain. He replied, “It already is! If you cut down all the trees you

would see that it already is a city. The houses are just hiding in the trees.”

48

As Collins (1987, 22) points out, land scarcity may result from processes other
than population density. It may, for example, be a consequence of inequity in landholding
size and corresponding social inequity which looks a large number of smallholders out of
access to lands controlled by a small number of large estates. This is the type of social
dynamic brought to the fore by a regional political ecology approach to population-
environment relations. However, according to the respondents, this dynamic of inequity
in landholding Size has little role to play in land scarcity, at least in the vicinity of
Lyasongoro, because a) individual holdings larger than five acres are few to none, with
most estimated at two acres or smaller, and b) wealth is no longer strongly associated
with landholding Size, but rather with success in off-farm enterprises or employment.
Nevertheless, it must be borne in mind that large holdings such as coffee plantations
restricted expansion of farmers on Kilimanjaro in earlier periods (Gamassa 1991, 7).
Within the narrow range of five acres or less, inequities in landholding Size have resulted
from initial differences in land allocation when Lyasongoro was first settled
approximately a century ago (farmers report that those ancestors with more livestock to
exchange with clan leaders were given larger plots) and from the differential inheritance
among sons described above. It was reported that farmers in Lyasongoro with excess land
(larger plots claimed more recently at the highest elevations) may allow relatives or
neighbors to farm it without compensation, other than the labor invested in improvements
such as terraces.

Due to the gender-differentiated processes of out-migration and off-farm
employment alluded to above and described more fully below, access to land for women

is increasingly problematic. In Chagga society, which is strongly patriarchal, women are

49

not expected to inherit land, but rather to have only usufruct rights to land via their
fathers, their fathers-in-law, their husbands, and their sons, in succession during the
course of their lives. Unfortunately, in this traditional scheme unmarried women, women
bearing children out of wedlock, women who do not have male children, women who are
widowed at a young age, divorced and abandoned women, and abused women who wish
to leave their husbands do not necessarily have access to land.

Girls are not given vihamba, only boys. When you get married your husband is

given a kihamba by his father. If he doesn’t have much he’ll give according to

what he’s able. If there are many married sons, you must divide the land so that
each one gets a little bit. If there are few children, you get land accordingly.7

While these problems are not unique to Kilimanjaro, there does seem to be greater
resistance to women gaining ownership of vihamba land than to women holding title to
newly settled land. Even in the latter case, women face barriers and inequities. A widow
recalled that at the time of her husband's death, he had recently purchased land in West
Kilimanjaro, but because the title was not to be found she lost all rights to that land.
Migration patterns and changes in modern society have probably increased the
proportions of unmarried and divorced women. Perhaps for this reason and perhaps due
to increased sensitivity to issues of gender equity, several interviewees agreed that
daughters Should be able to inherit land for their own security, but doubted that this
would occur with much increased frequency due to the strength of the cultural tradition.
Only fathers with very large farms were thought likely to bequeath kihamba land to a
daughter. One elderly farmer related that her sister had been given shamba land (plots in
the lowlands) by her father as a form of security, because he recognized that her husband
was abusive. This farmer herself was widowed at a young age. Because She had already

borne male children, she was allowed to remain on her husband’s land but received no

50

assistance from her in-laws. Therefore, her father gave her a gift of cattle to provide some
additional security. When asked if her daughters would receive an inheritance of kihamba
land, she replied:
No. No. They won’t get a portion. Females definitely don’t get anything. There
are some who get a portion, like in Himo, in the lowlands, like my birth father
gave to my Sister. And me, my father gave me a cow. . .. Now other fathers know,
they don’t look at whether this one is male or this one is female. It depends on

that father. Other fathers give out portions even for those who are female, they get
their portion, their inheritance.8

The contradictions contained in this answer are indicative of the social changes and
tensions surrounding the issue of gender and land. Many farmers are willing to
acknowledge that women need more equitable access to land, but few are willing to alter
tradition to give land to their daughters.

To summarize, I would argue that natural population increase on Kilimanjaro is
not the singular cause of land fragmentation and land scarcity, as implied by traditional
population models. Hypothetically, in a different cultural context population increase
could be off-set by outmigration to urban or to less densely-populated agricultural areas
in Tanzania. However, the cultural preference among the Chagga for each son to inherit a
plot of land on which to build a home, even if not to live and farm, results in reduced plot
sizes regardless of outmigration. Hence, the local cultural context and land tenure system
are critical mediators influencing the relationship between population growth and the
creation of land scarcity. Since access to land is locally structured by gender, marital
status, age, and other related circumstances, these factors also mediate land scarcity.
Furthermore, outmigration is also an age- and gender-differentiated process which, as
will be detailed below, helps to explain how both land and labor can be simultaneously

scarce.

51

4.2.2 Land Scarcity as a Constraint

With the specific cultural context of land scarcity on Kilimanjaro in mind, the
question of the relative significance of land scarcity as a constraint to household
sustenance needs to be asked. In particular, land scarcity as a factor influencing
agricultural management, particularly homegarden stratification, diversification, and
spatial arrangement, needs to be explored. Respondents consistently pointed to land
scarcity and small plot size as a major constraint, requiring trade-offs between desirable
management practices. For example:

...if there is a large area of trees, s/he will remove trees in order to get fodder

grass for cows. This too s/he reduces in order to make space for farming. Because

the number of children is increasing. These farms you found in the old days, our
ancestors had big farms. Now that you have many children, indeed, it’s

increasingly divided up. In the end what’s left over iS small, each person just gets
his small plot. It’s like a building plot.9

In addition to trade-offs between land uses, farmers attributed declining soil fertility to
land scarcity. As a farmer explained, even though the soil on Kilimanjaro is carefully
tended, it is tired because it gets used every year, year after year. In the past, there was
more land with space available for grazing, farming, fallowing, etc.

As expected based on the ecology of homegardens, farmers positively associated
land scarcity with the extreme diversity, mixing, stratification, and spatial and temporal
heterogeneity of the Chagga homegardens. Perhaps more surprising is that farmers
invariably identified land scarcity as the primary or only reason for mixed cropping
(kilimo mseto) in the homegardens, claiming that there is no other reason or benefit for
mixing crops and that crops always produce better when grown apart. For example, one

farmer said that the reason for mixed farming is that there is not enough space and that

52

farmers would get better yields if they planted each crop apart, if there were only enough
space to do 50.

One benefit of homegardens frequently cited in the literature is the efficient
capture of light by multiple strata despite restricted space. When asked whether farmers
take light utilization into account in managing their homegardens, respondents replied
that farmers know inadequate light can be a problem. However, they attributed
inadequate light to the mixed cropping system, rather than viewing the mixed cropping
system as a means to efficiently utilize light. The following dialogue provides an example
of this perspective.

[But do you see any benefit to using light well by mixing everything, in order to

use all the light?] Really, if you mix everything, nothing will get enough light. Eh.

Because when you want coffee to reach here, and bananas are up above, now the

coffee doesn’t get enough light. Now, if we had big plots, we would put coffee

here, bananas over there. Now we have small plots. Mm. [Okay Is there any
benefit to having shade for any crops? Or no?] Eh. There are areas like there in
the field, you have to plant, for example, beginning there by the boundary, you
plant one tree there for purposes of shade. Eh. There’s a period of sun here that

really burns, January and February. Now that sun is really bad. It scorches the
earth. And then that soil has no fertility. Eh. It dries up very quickly.10

Thus, for the most part, inter-mixing of crops seemed to be perceived as a
necessary evil rather than as an efficient system with ecological benefits. Many farmers
said the mixed farming system is used simply because it was established by their
ancestors. After hearing this explanation, I began to ask farmers why, in that case, did
their ancestors, the inventors of the mixed homegarden system, initially mix crops if they
had a good deal more land. This question received various contradictory responses.
Several farmers replied that their ancestors had started mixed farming simply because
they didn’t know any better. Another mentioned that while the elders who developed the

agricultural system had been knowledgeable, nowadays farmers just place crops here and

53

there wherever there’s space, mixing crops because they have no space to do otherwise.

Other replies included:

I wouldn’t know, truly. We found mixed farming on the day we were born.ll

Because in the past they had very large farms. They planted, they put trees for
borders, and so on. Now these days plots have become very small here. They
divided plots for the children to sleep there.12

You can separate each thing, but land is scarce. We have allocated that your space
is small, so you have to mix everything here. And you apply manure. But if you
apply manure, things go just fine. Eh. ...Indeed, the ancestors had big farms, but
they didn’t know the meaning of agriculture. [The meaning of agriculture?] Eh.
They had not yet recognized the meaning of agriculture. [Really?] If they had
recognized that, agriculture where you put maize here, you return there, you put
this, you return, you put beans here, if they had planted this way. But they didn’t
know its meaning. Eh. Nowadays people have planned with intelligence. They
know things. And they’ve studied and so on. Eh. They have already studied and
they know that if you mix it ruins things. It’s ruined, some things choke out
others. Mm. ‘3

None of the farmers interviewed (an admittedly small sample) seemed to be aware
of the many reported mechanisms by which intercropping may improve productivity per
unit land area and general ecological sustainability. These mechanisms may include
complementary use of soil resources, provision of beneficial partial Shade to coffee,
efficient light capture, layered distribution of root systems (Fernandes and Nair 1986,
112), conservation of soil and moisture, efficient cycling of nutrients, and plant diversity
(Torquebiau 1992, 192-9). The Chagga homegardens are often cited in the literature as a
particularly fine example of a sustainable traditional agroforestry system, at least under
past conditions (e.g. Femandes et a1. 1994). Yet today Chagga farmers seem to view
intercropping as an inevitable, rather than ingenious, method for dealing with land
constraints and they do not seem to have a clear understanding of why their ancestors

may have developed the method.

54

It is difficult to ascertain where the ideological preference for monocropping in
the village comes from. Certainly monocropping is strongly identified with “modern”
agriculture, suggesting that national-level discourse on agriculture has perhaps neglected
the potential of intercropping systems, even while many Tanzanian agronomic
researchers study and promote agroforestry and intercropping. During the colonial period,
Chagga farmers were pressured to grow coffee only in monocrop, and therefore a
monocropping preference might have derived from that period. However, on the whole
Chagga farmers successfully resisted this pressure, continuing to provide beneficial
partial shade to coffee by mixing with bananas and other trees (O’Kting’ati 1985, 36).

This raises a number of interesting questions. Why, then, did respondents not
acknowledge any productivity or ecological benefits to intercropping? Are these
ecological mechanisms not highly relevant for productivity per unit land area in this
case? This explanation is unlikely. O’Kting’ati (1985, 114) reports that the Chagga
homegardens have higher land expectation values than monocultures, even when
production from minor crops is ignored. Were the early Chagga farmers aware of these
benefits and did they intentionally mix crops despite having relatively larger
landholdings? Have Chagga farmers gradually lost local knowledge of these benefits due
to farm management practices being made tacit or rote? Or has this knowledge
disappeared through collision with external knowledge systems that continue to promote
monocropping as “modern” agriculture? Unfortunately, definitive answers to these
questions are beyond the scope of this study, but the assumptions of one farmer about the

presumed inferiority of mixed cropping are apparent in the following discussion:

55

(Mixed agriculture) is traditional agriculture. (laughs) Because this spatially
segregated kind is expert agriculture. Or no? [I don’t know, it depends, I think]
To separate crops isn’t expert agriculture?”

However, the tone of her query to me suggests that while experts might view mixed
cropping as inferior, she herself was uncertain.

It is also important to note that while farmers seemed to view inter-mixed
cropping as a necessary evil, they placed a very high value on diversified agriculture, that
is, growing a wide variety of crops on their plots.

These coffee and banana trees together, we were, the farm came with these coffee
and banana trees. Therefore we can’t remove them. Because land is scarce.
Indeed, this is why people mix. But if we had enough land to be able to put
banana trees over this whole farm, and another crop here like coffee, it would be
better. Eh. Because the coffee would produce a lot. And a lot of bananas would
bear, if we weren’t mixing them. But now we’re obliged to mix. You mix this,
you bring that, you put this, you put that, in order to get something to eat. Without
mixing, you don’t get food.1

In sum, the farmers viewed land scarcity as a critical constraint on the farming
system, forcing them to make trade-offs among necessary land uses. They viewed
intercropping as a tradition put in place (literally, given the location of tree plantings) by
their ancestors and simultaneously as a necessary evil forced on them due to land
scarcity, but resulting in lower productivity. They did not recognize any benefits to
intercropping in terms of productivity per unit area, beneficial crop interactions, or other

ecological benefits, but they did value the diversity of foods and other products produced

in the homegardens.

4.3 Labor Scarcity

AS I discussed in the introduction to Section 2, the terminology of “scarcity” is
problematic as it has typically been used in development discourse. The previous section

demonstrated how, although farmers’ plots are small and space is indeed cited as a

56

constraint, this condition is not an objective, given fact, but rather the product of
historical and material processes, including cultural factors. This section relates the
extreme distress expressed by Chagga farmers, especially women, in the face of
insufficient time and help in terms of labor to meet the productive and reproductive needs
of the household. However, this conceptualization of labor scarcity, like land scarcity,
Should not be taken as an isolable and inevitable fact, Since labor scarcity is conditioned
by a great many contextual factors. These factors range from structural trends, such as
declining terms of trade, to cultural factors influencing gendered divisions of labor and
“needs” definition to socioeconomic changes driving choices about out-migration and
education. Thus, when farmers speak of insufficient labor, this must be recognized as a
Simultaneously real and socially-constructed outcome of historical and material processes

influencing individuals’ decision-making processes.

4.3.1 Challenging the Divides

Before turning to the question of how labor can be scarce in a context of land
scarcity, I would like to first address some problematic theoretical issues. Chagga
households’ livelihood strategies are extraordinarily complex and the trend appears to be
towards further diversification of activities, each requiring household labor. In this
complex and dynamic context, the traditional division between productive and
reproductive activities and between cash crops and subsistence crops becomes very
ambiguous and, 1 would argue, difficult to justify. Likewise, as discussed in the
introduction, the necessity of diversification of income sources also has led to the
increasing importance of non-farm income and ambiguity in the divide between

agrarian/non-agrarian households. In the past. feminists and other theorists strived to

57

distinguish even “subsistence” agricultural production from domestic work and
reproduction (Deere and Leon de Leal 1981, 360). Today, Mbilinyi argues for a gender
analysis that links production and reproduction and “rejects the false dichotomy.” She
points out, “The national economy depends on the unpaid work of women (and to a lesser
extent, men) to produce food for household subsistence, to provide fuel and water, to
maintain the household and family on a daily basis and thereby reproduce the labour
force” (Mbilinyi 1997, 322).

This has important theoretical implications because much economic analysis of
rural and agricultural sectors has been predicated on these divisions. Commonly, the
exclusion of reproductive activities (often primarily the work of women) from these
analyses has resulted in skewed interpretations. In fact, this exclusion is one of the
reasons suggested by Collins (1987, 24) for the assumption of excess labor in rural areas,
which in turn leads to the failure to recognize the phenomenon of Simultaneous land and
labor scarcity. Excluding reproductive activities from analysis also excludes much of the
environmental dynamics of agricultural livelihoods, since natural resources gathered and
used for purposes of household reproduction are considered irrelevant to production. The
analysis here does not rely on a strict division between productive and reproductive
activities, nor between cash crops and subsistence or food crops, in part because of the
theoretical implications and in part because such a division cannot be sustained,
particularly from the perspective of farmers themselves. Women farmers say they value
food cr0ps for household subsistence, but also value coffee for needs such as school and
medical expenses, and for buffering seasonal fluctuations—in other words, also for

household subsistence.

58

Bananas provide an example of the ambiguity of these divides. One farmer
explained that in the past bananas were once so plentiful that there was no local market
for them. Then there came to be a market for bananas, on and off the mountain, and they
became both a cash and food crop. Now that productivity has declined and is especially
low at high altitudes, they have become only a food crop for many families. Bananas
remain a profitable cash crop for farmers who grow them in excess of their needs,
especially in the productive middle zone, but in contrast some farmers in Lyasongoro
have to purchase them in local markets to meet their own household needs. Research
indicates that bananas are gradually being replaced by maize and rice as a staple food
(Lyimo et al. 1999, xix). In the case of one family, which grew very few bananas because
they lacked the labor for planting new trees, cassava was grown as a substitute staple.

Similarly, leaf and vegetable crops, which are often considered subsistence crops,
are now grown both for consumption and profitable sale while maize, commonly a cash
crop, might be sufficient only for household consumption. Only coffee is grown solely
for sale and never for household consumption, but even this income is often required to
meet household maintenance expenses, such as food purchases, school fees, or health
care, as opposed to luxury items or investment in farming or business enterprises. Many
farmers complain of the declining productivity and profitability of coffee, but one woman
farmer, in clarifying the reasons why she will nevertheless not reduce the number of
coffee trees, explained the value of coffee in subsistence and cultural terms:

Coffee is a commercial crop, it helps with the children’s education. You can’t

remove it completely. Even if you get only a little, it helps, to get money. These

bananas are just for food. But (coffee) is for commerce. If people get two buckets,
like four kilos, four kilos you get money that helps the children to go to primary

school. [Even to reduce it a little bit isn’t good?] Coffee? (laughs) The men love
coffee. If you even break a branch! (laughs)l6

59

The ambiguity of the productive/reproductive divide is also evident in the case of
fuelwood, which is grown on—farm for household cooking but is also gathered in the

forest, split, hauled, and sold for cash income, primarily by young women.

4.3.2 Non-Agricultural Income

In the context of small plot sizes per household, declining productivity, and poor
terms of trade for coffee and other crops, Chagga farmers are increasingly incorporating
non-agricultural income into their livelihood strategies. AS one farmer put it:

Agriculture doesn’t pay in Chaggaland, not to depend on. This is because plots
are small.17

Truly, here in Chaggaland people really use their intelligence. You have to be
clever to find a way to live.1

This trend of farm households engaging in a range of non-agricultural activities is not
unique to Kilimanjaro. Rather, it is arguably the trend throughout sub-Saharan Africa,
though probably under-reported due to entrenched assumptions about sectoral
deve10pment (see Section 1.1 and Bryceson 1997). Non-agricultural income may take the
form of various business enterprises or employment engaged in by household members
living on the mountain or by household members out-migrating to work elsewhere but
maintaining ties with home. Illustrations of the first case will be elaborated here and the
latter case will be addressed in the subsequent section.

Farmers clearly articulated the impossibility of relying on only one type of
income activity, agricultural or non-agricultural. For example:

You can’t say that you’ll just farm, you’ll fail to have cash to buy supplies and so

on. You can’t say you’ll just have a job, you’ll fail to have produce for the

children, or little things for eating. Therefore, both of these are equally
important.‘9

60

When asked how a woman farmer saw the importance of the farm compared to her
husband’s work in the city in terms of obtaining household necessities, she replied:
Truly, in our region, a person with a small area has trouble. Because the value of
Tanzanian currency is low. Now if you won’t use these kinds of things
(commodities) for your home, you will definitely have trouble, you will live only
with difficulty. If you depend only on that cash, cash with which you buy
vegetables, you buy sugar, you buy meat, etc., it’s not enough. Now you see the

reason why we farm a bit; you get potatoes, you get squash, you get a little of
everything here on the farm, bananas, coffee. 0

Thus, despite the trend towards non-agricultural income, agricultural production
remains central for household sustenance. Several farmers who had attempted to start
their own shops reported that this was not reliable as a primary source of income.

You find that as soon as you’ve opened a business, immediately you are defeated,
you’ve quit it.2|

The store isn’t enough to meet our needs and to tend to the family like this
farming. It’s just something to help out, to get a little extra cash.22

By now a theme of the need for cash on Kilimanjaro should be apparent. Over the
last century, roughly, the region has undergone a transition to a cash-dominated
economy, yet one where agricultural production for household consumption remains
important. This transition has had far-ranging consequences for social and ecological
relations on the mountain, including changing modes of social support and subsistence,
and dependence on cash crops or non-agricultural income. With the advent of economic
reform and “cost-Sharing”, as discussed in Section 1.1, households need cash for school
fees and medical expenses that were formerly provided by the state. While a cash
economy does, in theory, allow households increased flexibility to purchase goods or to
buffer seasonal shortages via savings, reports indicate that increasing needs for cash have

weighed heavily on the poorest segments of the population, who frequently lack access to

61

cash income sources, capital, or loans. The Kilimanjaro Regional Nutrition Programme
reported:
Though Kilimanjaro is not a deprived region as far as social and economic
infrastructure is concerned as compared to other regions, there are some
extremely socially and economically poor households in the region. Such

households have low access to resources, especially land and all other basic
human needs such as health, housing, and food (KRNP 1990, iv).

Differential access to cash income in an increasingly cash-dependent economy, probably
more so than landholding size, helps to explain the high levels of social inequity on
Kilimanjaro today, a relatively wealthy region, though other factors such as cultural
norms also contribute. “Quite often material progress in ‘best’ villages is not matched
with social progress” (KRNP 1990, Preface).

Commoditization is an example of one outcome of the growing dependence on
cash. Salati, or watercress, is a very nutritious wild vegetable that was traditionally a
common resource available in streams and irrigation ditches. Today it is sold in the
market as a commodity. In Lyasongoro, an individual dammed off a section of a stream
with a bank of stones in order to lay private claim to the salati grown there (Figure 13).
Henceforth, this source of nutrition will only be available to others via the market. AS
commodities replace traditional means of meeting household needs, certain forms of
expertise and knowledge are reportedly being lost, leading to further dependence on cash.

Now you come to find people are going with the times. Traditional houses and

grass aren’t to be found, they use tin sheets. What causes it is this, there are many

uses of cash. In the past, many needs were met by craft expertise. Expert
knowledge that didn’t require any cash.23

Knowledge of local medicinal plants and their uses is another example of this

increasingly rare traditional expertise.

62

 

Figure 13. Privatization and commoditization of salati (watercress) on common property.

63

Cash should allow households some room to negotiate seasonal demands. On
Kilimanjaro, women have traditionally been the market sellers. They continue to use the
markets to balance cash and food demands and buffer seasonal fluctuations. However, as
will be elaborated below, farmers report that they are not able to save enough cash or
food stores to see them through the season of scarcity. The regional economy also
depends heavily on cash income from tourism, which unfortunately is also highly
seasonal in character. To make matters worse, the low season for tourism coincides with
the agricultural off-season. A farmer with male relatives who worked occasionally in the
tourist industry described the role of tourism in the village economy as follows.

A lot of the wealthy people’s houses are for things to do with the mountain, tours,

hotels... Yes, it’s a big support. If it weren’t for the climbing season, it would be

very difficult... [Does this cash get spread around?] Yes, it circulates around in
the village... Climbing businesses truly help a lot. If you yourself aren’t able,
your husband goes, or your child, or another person, but that person when S/he
gets money, you have your business, s/he comes with it, you get a customer.

Therefore, this cash circulates... But money available in large quantities here on
the mountain, there isn’t any, in these parts there isn’t any. And farming doesn’t

pay.24

Thus, some of the cash income from tourism reaches villagers directly or indirectly, but
others reported that much of the cash leaves the mountain, especially when spent on

luxury items.

4.3.3 Migration, Work, and Residency Patterns

Non-agricultural household enterprises—which are increasingly necessary in the
context of the cash economy, declining agricultural productivity, and land scarcity—are
drawing household labor away from agricultural activities. At the same time, some
household members are removed from agricultural labor entirely due to migration to

nearby urban areas (predominantly Himo, Moshi. and Arusha) or to other regions.

64

 

 

Because of the historical influence of missionaries and settlers on Kilimanjaro, the
Chagga had a reputation for being highly educated and migrating to urban centers to
work in government and business, long before land scarcity became critical. The Chagga
diaspora is fascinating in itself, but it is beyond the scope of this work to detail these
migration patterns. Here I will focus on the implications of out-migration for household
strategies, labor patterns, and land scarcity on the mountain.

Given the severity of the land shortage on the mountain, it is at first glance
surprising that more Chagga do not simply migrate to other regions where agricultural
land is abundant. Farmers identified several reasons for a general resistance to out-
migration, foremost among them a cultural attachment to living on the mountain.

On Kilimanjaro, land is scarce. You have already been told that for the Chagga

it’s very hard to move to many other areas. They don’t like to at all. Except that

the state of life these days makes people learn they must move to other parts,
because you find the family has insufficient space for a person, to help his family.

Indeed, so many these days are raising themselves up this way. But you will find

that he won’t let go of the section that he was given by his parent. He can’t let go

of the place, he can’t leave it empty like desert. You will definitely find that he
tends it. If not himself, he’ll hire even a goat. . .. But for land to be empty without
anyone, it’s not easy, it’s not easy.25
As this explanation suggests, even absentee owners will usually arrange for some
cultivation, or at least grazing, of their land. Often the arrangement will be with
neighboring family members or with a hired laborer to maintain a homegarden or, as in
one case I observed, simply to clear the land for seasonal maize planting which requires
less day-to-day maintenance than a homegarden.

Class and opportunity structure probably play the most important role in

determining who will out-migrate, despite the cultural attachment to the mountain.

Wealthier, educated Chagga will migrate to where job and business opportunities are and

poor, landless farmers will also migrate to where land is available. Age is another

65

important factor, as younger people perceive farming as a less desirable occupation,
given the shrinking land base and declining terms of trade.

Children often don’t want to busy themselves with farming matters...they are
bound to go to the city.26

Truly it’s tiring, life here in the countryside is hard. Because you’re just chasing
after money. If you don’t, you use up all your strengths. So now we’ve started to
compare rural life and city life. City life is better I think, because I’ve already
stayed in the city. And it’s not just me who’s stayed in the city, even others like it,
but when it has to be the farm, 1 stay on the farm.27

Many of the older children, both male and female, of the farm families were in vocational
training in the cities.

Statistics on the precise gender, age, class, and education make-up of the part of
the population out-migrating are unavailable. Of the farm families sampled, 8 women and
8 men of the younger adult generation (about 36% of the total number of adults) plus one
man of the middle generation lived primarily off the mountain during the period of
research. No one from the eldest adult generation lived off-mountain. Farmers reported
that wives often accompany husbands who out-migrate. These families where all adults
are off-mountain would not appear in my sample of residents. However, wives do not
generally out-migrate without their husbands, supporting my impression that out-
migrants are disproportionately male and of younger generations. Other reports also
confirm this impression that mostly married women are left behind to work in agriculture
(KRNP 1990, 66). Migration and residency are further complicated by every variety of
migration pattern—seasonal, weekend, holiday, for temporary employment, for
education, for retirement, residing with relatives, etc.

Complex rural-urban migration patterns are being reported worldwide as part of

the diversified strategies of rural families in the context of declining terms of trade for

66

small farmers. Remittances from the urban employed to their wives or parents in rural
areas have been crucial in some cases for rural families’ survival. From this small study,
however, the importance of remittances from urban areas is not clear. Women in the
study with husbands working off the mountain did usually receive cash support.
However, parents reported that they did not receive significant support from their
children residing in towns and that, in fact, some children were being supported during
their education by cash earned and food grown on the mountain. This implies a strategy
of sending young people to urban centers in order to foster their independence from the
family and their freedom from not very promising futures in agriculture, rather than with
the expectation of Significant future remittances. The following discussion explains this
dynamic:

[And your children who stay in Arusha?] They just depend on us, they haven’t
earned much cash yet. If you get a little, like 2000 (TShillings), you have to send
it to them, so on and so forth. I see s/he has trouble, s/he endures it. (laughs) S/he
endures it. Especially the girl, I tell her, she just started (vocational school) last
year, March. She hasn’t been there very long. So, she has to pay the costs of the
vocational lessons, soap for washing, oil for cooking and lamp oil, that is, she’s
bound to make demands on us here at home. [Do you think there are more
necessities (living) in the city than in the village?] There are more. Especially the
food there. And here we get greens (mchicha) from the garden. They have to buy
it. There in the city it’s for sale, maize we can grow, we use produce from the
cows, there they buy, everything just for food. [How do you view life in the city,
or how do they see it?] Very difficult, truly difficult. But you teach them how to
be able to endure until they can succeed. [Do you think they’ll return again to the
village?] I think they’ll make this decision themselves. If s/he succeeds in their
work, if s/he comes to practice it here in the village, it’s s/he herself/himself who
arranges it. If s/he desires to work there (in the city), it’s s/he herself/himself who
will plan it. We don’t make their plans.28

Another farmer, who had herself lived in the city in the past, described a more equitable
relationship between household members living in the city and the Village.
But also, city life without the farm wouldn’t be possible for people... Because

people in the country get cash from those in the city. And those in the city get
things from those in the country. These things help out.29

67

It is important to note that the families interviewed in this study were confined to
ordinary farmers, though in general Chagga farmers are considered wealthy by Tanzanian
standards. Most of the younger generation in Lyasongoro have cement-block homes
rather than wattle-and-daub construction or traditional Chagga round houses. However,
dispersed around the mountain are the occasional large, modern homes built by Chagga
successful in business, government, or academia, both in Tanzania and abroad. With the
exception of those who live on the mountain and invest in tourism, creating employment
opportunities, these very wealthy Chagga reportedly have little impact on the rural
economy.

[You said that cash doesn't circulate here. How do you view the wealthy people

who live outside, but have a house here? Do you think they help with

development on the mountain, or...] Wealthy people who live outside, likewise,
due to understanding, people don’t come to understand from a distance. They just
put up their house, but to depend on him/her to make things productive here is

difficult, because S/he already knows that income here is very hard (to get). It’s
better to go to the city where you can seek and find income.30

[What do you think of the wealthy people from Kilimanjaro who live outside, do
you think they help here in the village or they don’t help?] Truly, not all of them.
Maybe you find your relative or your brother, S/he can help you. But others less
close it’s not very easy. They help their own kind with whom they have a
relationship.’ ‘

AIDS is another factor related to migration and social change which has impacted
Chagga households both in rural and urban areas. In at least two, and perhaps three, of
the 10 households sampled, children orphaned because of AIDS were dependents of their
extended family. The land inheritance of the deceased may be available for the use of the
extended family, slightly alleviating land scarcity (though this does not apply to the case
of orphans living with their maternal grandparents). However, this probably does not

compensate for the proportional loss of productive, adult family members and the

68

increased proportion of child dependents. Regarding AIDS, as one farmer put it, “They
are being cut down like grass; they are laying down like grass.”32

Understanding these migration and work patterns, in combination with the
traditional system of land inheritance in Chaggaland, helps in large part to explain the
phenomenon of simultaneous land and labor scarcity. AS would be expected in a context
of declining plot sizes, declining productivity, and declining terms of trade, many Chagga
are out-migrating. This does not automatically translate to more space for those
remaining, however, because cultural norms require that even those sons out-migrating
receive a plot of land, even if it’s only big enough for a house and garden. Thus, some
agricultural land is lost to house-building and other areas may go uncultivated if no one
from the family resides there. If someone does reside there, it is likely to be women,
children, and older generations, who must replace the agricultural labor provided in past
generations by resident men in the household. Given smaller plot sizes, it might be
assumed that overall labor requirements would be lower than in the past and that,
therefore, despite out-migration, labor should be in surplus. This assumption is predicated
on a narrow view of mainly “productive” household activities, ignoring the labor,
typically women’s labor, required to reproduce the household. This reproduction in turn
depends in part on natural resources available off the farm and held in common, such as
fuelwood and fodder. Land scarcity and land degradation make these resources less
available and severely impact household labor requirements. The next section will detail

how households negotiate competing demands for their labor, and how gender and other

social relations impact this process.

69

4.3.4 Division of Labor and Gender Relations

With increased out-migration of men, a basic typology of three household types
must be considered. First, there are male-headed households with the husband present
and active in day-to—day farm and household management. Second, there are households
where the males are absent but have the final say in certain decisions. These might be
considered de facto female-headed households. Finally, there are de jure female-headed
households. The proportions of the last two categories seem to be increasing with
increased out-migration of men.

Historically, the general gender division of labor in Chagga culture was that
women were responsible for household reproduction and homegarden agriculture and
men and boys were responsible for livestock. When coffee was introduced, men became
more involved in management of and labor in the homegardens, particularly the
profitable components. Today, however, most of the men in the families interviewed
pursue other occupations or have out-migrated. Some elderly men and capital-poor men
are full-time farmers, but their labor tasks are somewhat differentiated from those of
women in the household. Husbands, sons, and schoolchildren may help with agricultural
tasks occasionally or seasonally. However, in Lyasongoro women perform the majority
of farm labor and farm management tasks. Today there is also some increased flexibility
in the traditional gender division of labor. For example, some women are learning to
prune coffee trees and to work as mountain climbing porters, and some men are willing to
cut fodder grass and clean cow stalls.

Women are primarily responsible for family subsistence and increasingly for

“productive” activities as well, whether agricultural or non-farm enterprises. They are

70

therefore the first to feel labor, cash, and transport constraints as well as grass, fuelwood,
and food Shortages. Although here I focus on labor constraints as expressed by women
farmers, it should be evident from the following quotations that labor shortages lead to
and link systemically with other problems and constraints. For example, although soil
fertility and production can be improved by application of cow manure, families often
have insufficient labor to collect fodder to maintain more stall-fed cattle.

[You don’t want more than one cow?] I’m not able to because it’s a lot of work.

[Work to get fodder grass, or everything?] Now. If you have two or three cows,

they can prevent you from doing the farm work, like cultivating, and when it

comes time to harvest coffee you won’t have the time. If I had family enough to
do the work, I would add two or three cows, we would help each other out. Yes.33

Note that this farmer must consider seasonal labor constraints (planting season and coffee
harvest season) in deciding how many cattle to keep. In this family, half of the children
were attending trade school off the mountain, the others lived at home but attended
school locally, and the husband worked frequently as a porter on the mountain. Thus, the
wife bore primary responsibility for household and farm work, in addition to providing
full-time care for her ill, elderly mother-in-law.

In theory, labor constraints could be ameliorated by hiring laborers, but for many
cash-poor families this is not feasible.

Now we have no cash to hire laborers to trap those fuko (type of rodent pest)

down below. If you plant banana trees and one is eaten, it falls, and if you

plant. . .it goes on this way. So, there’s no assistance with labor to help here on the
farm.

Note that while best management practices for controlling fuko were known, the labor to
implement them was insufficient. Furthermore, those women farmers attempting to raise

cash through commercial enterprises, such as small stores, reported difficulty balancing

71

labor requirements on the farm and at the store. One woman determined that the time

invested in the store had not proved worthwhile.
But the way work has started to squeeze now and then, and much of the time I
don’t have the opportunity to watch the store and get such a small amount of
income. Now here we’re going into May, and continuing into June, indeed, it’s
the time for cultivating. When you’ve finished hoeing, you are planting, when
you’ve finished planting, you’re weeding, even twice. Therefore, the time I have
to look after the store is scarce, it’s the work for food crops so you don’t lose

them. That is, the store isn’t enough to meet our needs and to support the family
like this farming. It’s just something to help out, to add cash.35

Therefore, she decided to withdraw her involvement from the store, a group enterprise.

In the traditional gender division of labor, men perform the annual pruning of
coffee branches (critical for maintenance and productivity). The technique requires some
special training, perhaps three months at the local trade school. In households without
resident men who can perform this task, women farmers may hire male relatives or
neighbors to assist them. Although farmers invariably reported that it is not wise to skip
the pruning one year if labor is unavailable, because it will hurt production in future
years, one household headed by a widow reported that late in the season they still did not
have sufficient cash to hire a neighbor to prune. She said that she would like to obtain
training in coffee branch pruning herself, but had not had the opportunity.

Implicit in the reports of many women farmers is that labor is even more of a
constraint than land, sharply countering the assumptions of traditional population-
environment models. One family, with a relatively large farm in the higher elevation area
of the village where land was more recently allocated, reported insufficient labor to fully
utilize the land.

...there’s no one to help us, if the children are doing their own work. For us,

we’re defeated. [So, you have land, but no labor.] Land is there, eh, there isn’t
enough labor.3‘6

72

However, this family occasionally farmed a plot in the lowlands, requiring only seasonal
labor (rather than year-round maintenance) which a teenage daughter provided, although
she normally attended trade school in a neighboring village. One farmer’s husband
earned sufficient cash income from his off-farm employment to purchase land in the
lowlands, but the wife reported that she would not have enough time to manage it (even
though she already employed a male laborer for assistance with her homegarden plot).

I don’t have another farm plot. If I needed it I would get it, but I would get very
tired. I can’t do it.37

However, the priority of land Shortage versus labor shortage varied with circumstances
from household to household and only a few farmers identified one over the other as a
primary constraint. This key point indicates the simultaneity of land and labor scarcity at

both the regional scale and the household scale.

4.3.5 Social Support and Stratification

Gender, age, and labor relations are closely linked to issues of wealth
stratification, marginality, and social support in the village. Understanding the continuing
significance of the homegardens in this community requires addressing these issues, since
marginalized households appear to be more dependent on the diversified homegarden
than wealthier households. Although the economic importance of homegardens on the
mountain is decreasing for some farmers (relative to maize-bean farming at lower
elevations), mixed farming on the kihamba plot is still the main source of income and
subsistence for others, especially those without business enterprises, lowland plots, or
easy access to capital.

To understand the range of options farmers have for coping with livelihood

constraints, I asked about the extent to which individuals could call on others for social

73

support as various forms of material aid. In Section 4.4.3 (Migration, Work and
Residency Patterns), I discussed the generally limited extent of social support provided
by community members living off the mountain. Here I discuss the range of social
support within communities and households on the mountain.

Most farmers reported that each adult individual has his or her own cash account,
unless a married couple shares one account. Married couples of each generation typically
have their own plot of land.

[Do your in—laws farm together with you, or each one has their plot?] Each person
has just their small plot, garden-Size. It’s like a garden, it’s just a small plot. Eh.
Each person has their little plot.38

[And do you all Share one fund? If one person needs cash or aid, do you help each
other out or does each person have a separate fund?] Each person has their fund.
Eh. Each person has their fund. There’s no mutual aid. [Therefore, even if your
(adult) male children help with the farm, it’s just to help out, they won’t get any
profits from the plot? It’s yours?] No. It’s each person. Like the way these girls
help, they just help out. [But this son has already married, so his wife has her
portion of land?] He’s already married. She has his portion, and his cows. She
does her work by herself. [It was like this even in the old days?] Even in the old
days. Eh. Even in the old days.”

When women are widowed or divorced, they may face extreme isolation if neither
their husband’s nor their natal families are willing to provide support. The widow who
already had male children was allowed to reside on her husband’s land, but received no
other help from her in-laws.

When my husband died, I was busy with the children all by myselfl Without help

from in-laws, without the aid of inheritance, without any help at all, it was that I

had moved here because my husband died. I took care of myself by myself, there

wasn’t help, if you request aid they don’t give it.40
This level of individualism, even between generations of one family living on

neighboring plots, is quite distinctive to Chagga culture in contrast to other Tanzanian

ethnic groups where pooling resources is more common.

74

Substantive social support between more distant relatives or neighbors is likewise
reportedly rare, especially between members of different class strata. Most of the
households in the sample were relatively poor. One farmer described her relations with
wealthier people as follows:

You will find wealthy people will look after their fellow wealthy people. But a

wealthy person won’t meet a person of a low class. If he knows that a wealthy

person will help a poor person, many times even if he gives to you it’s just a
little.“

Hence the forms of social support typical in the village do not appear to serve as
redistribution mechanisms, since the main forms are either mutual assistance between
people of a Similar class or else very limited assistance from wealthier to poorer
households. The same farmer also reported that requesting aid from wealthier people
could invite derision42 and shame; the wealthier donor may never let the poorer
supplicant forget the favor.

These findings concur with a study of class relations on Mt. Kilimanjaro, in which
Chagga cultural ideals of wealth and entrepreneurship, in combination with the influence
of introduced ideologies of development and modernism and the erosion of social safety
nets and kinship obligations, were found to negatively impact childhood malnutrition
(Howard and Millard 1997, 11-12). These findings point to the contradiction between the
international reputation of the Chagga for wealth and education and the reality of a sector
of poor, marginalized farmers, linked to the pressures of a tightening cash economy and
limited distribution of resources from wealthy to poor.

At the same time, however, some forms of mutual aid between members of the
same social class reportedly make a significant difference in household livelihood

strategies. Examples 1 observed or that were related to me included giving seeds or

75

cuttings, sharing extra fodder grass, trading advice, cooperating in business enterprises,
assisting with house building, managing lowland plots, gift giving at times of celebrations
or funerals, and grandchildren providing labor and care for grandparents. Also, as
reported above, some landowners with excess land may make it available to land-poor
relatives. The relative importance of these forms of social support for household
livelihoods is difficult to measure and warrants further study, but farmers appeared to

value such exchanges highly.

4.3.6 Seasonality

Seasonal variability in the economy and the farming system on Kilimanjaro is a
source of hardship, especially for the poor. It is necessary to understand this dynamic
since one benefit of the mixed homegarden system, with its variety of seasonal and year-
round crops, is that it helps to buffer this seasonality.

”43 occurs, all farmers agreed that the

When asked when the “season of trouble
worst season runs from the beginning of the long rainy season (monsoon rains—masika)
until July when the rains end and the tourist season resumes. During this time, farm
production and income are low, savings from the coffee harvest have usually already
been spent, and the tourist season is low. This is also the maize-bean planting season in
the upland zone, when women have the most heavy work to do, and it is the cold season
when more food is required to fuel this work, especially at high altitudes. The confluence
of these factors results in overall food scarcity, cash scarcity, and economic constriction
on the mountain.

It is interesting that the tourist (mountain-climbing) season affects villagers to

such an extent, since a large proportion of the income from tourism probably does not

76

remain on the mountain——for example, park fees and payments to outside tour
companies. Some Lyasongoro men and men from other villages on the mountain earn
income as guides and porters, but village leaders and women complained that much of
this income is spent in the cities and on luxury items, including alcohol and cigarettes. I
believe that if a greater proportion of tourist money were captured by or redirected to
village women responsible for household subsistence, it could make a large impact on the
household cash supply and village development.

When asked how households cope during this season, one farmer replied that you
just have to get by, Skipping meals if necessary. 1974, 1983, and 1997-1998 (El Nifio) are
remembered as very bad years when people went hungry. Hence, in addition to
seasonality, weather and pest variability from year to year also have serious
consequences for household survival. The Kilimanjaro Regional Nutrition Programme

6

reports that the region maintains food storage ‘godowns,” but that storage at the

household level is lacking due to pest problems. Forms of household storage include
drums, silos, sacks, cribs, and attic spaces (KRNP 1990, 21-22).

Farmers did not consider reserving cash savings or selling off livestock to be
realistic or reliable solutions to the cash and food constraints of the low season.

Here there isn’t anything that we can sell in order to save up that cash. Or there’s
nobody to buy their necessities from us. We don’t get anything, even the crops
which we want to harvest here on the farm, they don’t last for very long. If you
harvest them here on the farm, immediately they’re just finished. Perhaps I don’t
know if it will reach July, if they (crops) reach then they’re already finished.
Because the farm closes up (dries up) a lot. And the maize doesn’t produce. It
doesn’4t4 produce the same in our parts (our elevation). Until August, it doesn’t do
much.

The coffee money has been taken by CBD. Now you can say you’ll save some of
that cash from coffee if you get it, if there’s no CBD, then you have it. For us, if
the cash isn’t available because of CBD, or if you have your cow, you sell it. If
you don’t have a cow, you don’t have coffee, indeed, you just have trouble.45

77

Furthermore, farmers said that even if they had livestock to sell during the low season,
they would be forced to sell at low prices, since the cash constriction would translate to a
severe scarcity of buyers on the mountain. Most farmers were adamant that they could
not hope to save enough cash to see them through the difficult season, even in years of
favorable weather. However, they agreed on the importance of having a savings account,
if possible, especially for medical emergencies. Interestingly, a sermon at the village
church counseled members to keep a savings account46 and to make a plan for managing
the household economy, particularly given the declining state of the national economy
and the inability of the government to provide support.

While cash and food are scarce during the long rains, several farmers noted that
the total number of cattle that can be maintained is limited by grass availability in the dry
season as well as by labor available in that season to search farther for grass, if necessary.
The number of cattle kept, in turn, limits the quantity of manure available for soil fertility
maintenance and thus overall crop productivity. Many farmers dry maize stalks that can
be stored for up to two years to use when fodder is scarce or when labor for cutting
fodder grass is unavailable.

Grains, legumes, and bananas can also be dried and stored for food, but supplies
for canning or otherwise preserving food are not readily available. Thus farmers are often
forced to sell produce at low prices when it comes into season at once (especially
tomatoes) and to buy produce out of season that they are accustomed to use in cooking
(such as various vegetables). Cassava tubers (mhogo) can be left unharvested for over a
year, providing another form of storage and buffering of seasonal scarcity. Bananas

usually produce year-round, with a low season lasting only one-quarter of the year during

78

which some households buy bananas in the market. However, banana production varies
annually if the weather is severely wet or dry (bananas produce well in wet weather and
poorly in dry weather, while maize, coffee, and bean crops may suffer in very wet years).

Farmers reported that kisori (Panama disease) increases in hot, dry years.

4.4 Homegardens, Environmental Dynamics, and Natural Resources
4.4.1 Homegardens

The analysis above outlines the importance for Chagga farmers of complex and
changing land, labor, and gender relations. In this section, a focus on the homegarden
system itself provides an illustration of how these social dynamics play out in systemic
relation with each other. In addition, analyzing the homegarden system makes it clear that
these social dynamics cannot be isolated from environmental dynamics, Since each
impacts the other. In the current context of change on Kilimanjaro, social changes have
contributed to environmental decline, which in turn puts further pressure on the
socioeconomic system.

A major benefit of the homegarden agroforestry system, described in Section 1.3
above and in the literature, is that mixing various crops and trees enables vertical
stratification which makes efficient use of scarce resources, particularly light and space.
As explained in Section 4.2.2, farmers cite land scarcity as the major motivation for
mixed cropping in the homegarden, and also express a preference for a diversified
production system to meet household needs. However, the small size of plots today has
pressured farmers to seek other sources of food production and cash income.

Well me, the way I see it, because this mixed agriculture truly has a lot of benefits

in it. There are many benefits in it, but if you have a big area, mm hm....It’s not
enough to meet the needs of the family. It’s not enough.4

79

While acknowledging the benefits and necessity of mixed cropping, many farmers
compared it negatively to other forms of agriculture considered more modern, such as
monocropping or less intensive intercropping. One farmer said, “Mixed farming is only
with difficulty, hassles.”48

Chagga farmers’ perceptions of this discourse of “expert” versus “traditional”
agriculture (also discussed in Section 4.2.2 above), and development discourse generally,
impacts their attitudes about homegarden agriculture and their livelihood strategy
choices. When asked about the invention of the homegarden system by elders in the past,
a farmer replied:

Eh, they were using mixed agriculture. But it was for household uses, not for

development. Not development. It was for the sake of household requirements.

But they were clever because the youth we have now, you find that development
began in the past. They were clever.49

In this way, the Chagga try to balance a respect for their ancestors and for cultural
traditions with the ideologies of modernity and development. which sometimes conflict
with these traditions.

We are going with the times. We can’t go back again. You can’t tell someone to
go backwards. Indeed, even life these days.50

The form of mixed cropping that makes homegardens a distinctive agroforestry
system is the intimate mixing of crops and trees of various heights, forming vertical strata
and complex horizontal arrangements. To explore farmers’ views on and understanding
of the ecology of this form of mixing, I asked them about their practices of planting trees
in the homegarden and management of shade. Most farmers reported that scarcity of
space and competition with crops, including banana and coffee tree crops, constrained

tree planting in their homegardens.

80

Farmers’ understanding of the dynamics of light competition and their active
management of shade is evident in the following explanation.

But we also care about that distance (between trees) because of when it squeezes
together banana trees and coffee a lot. Likewise, you come to hoe for crops, you
see that other plants like maize and beans aren’t coming. Eh, because of that
shade, if it closes up. Or a period like this one right now, the time of preparations
for planting maize...the places that have a lot of banana trees they are preparing to
reduce them.5 ‘

In response to a question about whether it is possible to incorporate more multipurpose
trees in the homegarden by pruning them to reduce shade, a farmer replied:

If you increase the number of trees on the farm, you don’t get coffee, the banana
trees get spindly, even if there are very many roots underground, it’s not good for
farming then, to plant maize, cassava, even beans. Because if you plant a lot of
trees... [you won’t get a lot of produce]. You know our agriculture is the mixed
kind. [YeS.] Therefore, if there gets to be too much Shade, the beans don’t come,
even cassava, bananas will get spindly, even the coffee.52

Another farmer emphasized individual preferences in making shade management
decisions and tree-planting decisions. When asked if a farmer needs to know how far
apart to plant coffee and banana trees, she responded:

Now this just depends on your space, if you want them far apart, if you want them
close together. One person you find doesn’t like to crowd those trees so that grass
doesn’t become scarce. When you have them close, then the shade has spread
over even the grass. Therefore, the grass below doesn’t grow. Eh. Places with
Shade. A person plants far apart in order to have space if grass is there. Or s/he
puts in other trees, or of another kind. So that depends on the needs of the person.
If it’s an area in the valley that isn’t appropriate for grass, or a person has a big
farm—you see you find they’ve already planted lots of cypress trees (Cinara
cupresii)! Another person you find plants far apart and mixes with grevillea
(Grevillea robusta). If it’s a section that isn’t located close to the plot. Because
those cedars you can’t put in places with bananas or coffee, you don’t plant
amongsfit3 the cedars. They must be on the Side, like a place that’s extremely
rolling:

In contrast, other farmers emphasized the inheritance of tree planting patterns

from previous generations.

81

And us, for example those of us who were children in those days, children of
people of the old days, we haven’t planted banana trees or coffee to tell you the
truth. We found it this way. Our work is just to weed. We found it. Say there is
one that dries up, a banana tree, you take it out and you return one to that very
place. But we’ve only put in banana trees.54

Several respondents said that the general trend has been for farmers to reduce the number
of trees in their homegardens in comparison to previous generations, to make Space for
other uses.
Now you find if he has planted many trees, his father has given him a portion with
many trees, and now those trees are for his service. He wants to build a cow shed,
or a kitchen, or a house for sleeping—to build with bricks, even if it’s bricks,

now, there are sections of timber. So. If it’s not to build, you’ll find he’s reduced
for reasons of shade, he’s reduced in order to plant fodder grass, or to farm.55

However, the participants in my study, who may have self-selected for interest in
environmental concerns, all expressed willingness to plant at least a small number of new
tree seedlings. Even within the sample, interest varied widely, however, with some
farmers planting or tending a larger number and variety of trees. I asked one farmer why
she has so many different species of trees while other farmers may grow only a few
species. She said it depends on “each person’s diligence and drive to plant trees.”56 The
number of trees and number of species did not appear consistently correlated (in this
small sample) with household characteristics such as farm Size or wealth.

As explained in the overview above, soil fertility was cited as a severe constraint
on productivity in the homegardens, in addition to constraints of Space, light, pests, and
disease. Most households enumerated steep declines in crop productivity over the years,
mostly attributed to soil fertility decline. As one farmer put it, the soil “has already been
cultivated a lot. These days it doesn’t have the strength it had in the past.”57 Another

remarked that because the soil is tired, if you don’t wait a month or so to properly cure

82

green manure before application, you won’t get a good harvest, whereas in the lowlands
you can plant the day after you turn the soil.
Without being prompted, farmers were able to give quantities of production in the
past compared to the present, using local units such as gunia (90 kg sack) and debe (18
litre tin). For example, regarding coffee, one farmer obtains only one gum'a today, but
remembers harvesting 3O gunia from her grandfather’s plot in the same village in the
past. Another reported She had harvested only about 4-5 kg of coffee per year in the last
three years (due to cold weather which exacerbated CBD), but in the past, before CBD
and before dividing the kihamba between herself and her son, she obtained five gunia. A
study in another village in the district reports that farmers are producing 50kg/acre of
coffee compared to 250-750 kg/acre obtained in the 19705 (Lyimo et a1. 1999, 6). One
farmer reported a reduction in maize production from six gunia to one-and-a-half or two
gunia and a reduction in beans from two to one gunia, while another reported a reduction
in maize from three to two gunia and in beans from two to one debe. Several farmers
pointed out that in the past even farmers at high altitudes were able to grow bananas in
excess of their needs, enough to sell. A farmer’s husband reported that when his elderly
mother used to go to market, she needed five helpers to carry her sugarcane to sell,
whereas now the farm does not produce excess sugarcane, reportedly due to declining
soil fertility.
Significantly, soil erosion has been a relatively less widespread problem in the
village, despite steep slopes. The homegardens serve as excellent controls on erosion
because they provide good ground cover, especially mulch and close grasses and crops

(Anderson 1982, 116). Also, contour terraces are built on Slopes (many of the rock

83

terraces were built by previous generations). In contrast, some lowland areas with less
steep slopes, but with open, monocropped fields, are more susceptible to severe erosion
(Anderson 1982, 33), now being mitigated by the promotion of grassy contour bunds.
There are reports of extensive soil erosion along waterways and steep upland areas in
other parts of the mountain, however (O’Kting’ati and Kessy 1991, 80; Anderson 1982,
33).

As alluded to above, soil fertility has been maintained traditionally by application
of cow manure from stall-fed cattle. However, land and labor scarcity both limit the
availability of fodder grass, which limits the number of cattle that can be kept. Thus, soil
fertility decline can be attributed, at least in part, to the systemic effects of the
simultaneous shortage of land and labor. The following scenario is typical of how these
multiple constraints play out. One farmer with an especially small farm usually collects
fodder grass from the farm plot itself, but when it is scarce, she has to collect it from the
forest reserve or even purchase it. She tries to buy it from fi'iends or relatives who might
give her a good price. She would like another cow, but the farm does not have enough
grass. Yet the one cow she has does not produce enough manure to maintain soil fertility
and production on the plot. Very few village farmers can afford to buy commercial
fertilizer.

The critical role of cow manure, but also labor, appears in the rich description
offered by a farmer of her brother-in-law’s strategy for maintaining soil fertility and
productivity.

Yes, you can increase fertility through management practices. [By keeping cows

or. . .?] Keeping cows and even on the plot, by farming the plot and raising cattle

that produce a lot of manure. You spread it, if there’s something that you are
planting, you plant it with a lot of manure. Here’s the fertility, and there’s a lot of

84

manure. You know here at my in-laws there are banana trees. Because they have a
lot of cows, he has like four cows. Four. Now, those banana trees are big ones. So
he has a lot of manure, and his farm isn’t big. It happens that it’s not big in size.
Here he plants bananas and it’s he who stands with the farm, to work by himself.
The manure is plentiful, it’s Spread everyplace, and he has farm work to do every
hour. To expand, every minute, and every minute he is busy. If it’s branches
(coffee) he does the work himself, every minute he knows to keep himself busy.
Many plants are going in, so, with the bananas, you find it’s with manure,
manure, manure. So, even in the lower altitudes, when you’re renting, you do it
this way. They have enough cows for manure.58

She does not explain how her brother-in-law is able to obtain fodder for four cows if his
farm is not big, but since these farmers reside in the upland part of the village where plots
tend to be relatively larger, the farm may not be exceedingly small or it may be easier to
obtain fodder grass from the neighboring forest reserve. In addition to the key role of
manure, an interesting aspect of this description is that it is very reminiscent of
involution, the development of extremely intensive labor inputs to offset productivity
decline, as described by Geertz (Section 2.1.3).

The limited scope of this study resulted in a focus primarily on the homegarden
component of the farming system, but it is important to acknowledge the role of farming
in lowland mashamba (fields) in systemic interaction with upland homegardens. Another
study lamented the neglect of this interaction in past research (Lyimo et al. 1999, xx).
With respect to soil fertility, other authors have raised concerns over the practice of
transporting crop residues from the lowlands to supplement organic manure in the
highlands, resulting in degradation of soil in the lowland fields (e.g. Cook and Grut 1989,
24). However, of the ten households in this study’s sample, at least four owned no
lowland fields at all and ecological interactions between lowland and upland fields
appeared minimal, perhaps because of the high altitude of Lyasongoro and constraints of

transportation. Economically, seasonal maize-bean cultivation in the lowlands appears to

85

be increasingly important in livelihood strategies of those households who can afford the
labor and cash inputs (seed, fertilizer, and pesticide as necessary).

It is also arguable that soil fertility decline is not solely an outcome of current
land and labor scarcity, but may be a trend in the homegardens over the very long term,
exacerbated by recent developments. While it is not possible to definitively assess the
sustainability of the Chagga homegardens as they were cultivated long ago, a nutrient
cycle analysis of a contemporary homegarden would help to assess the Significance of
some identifiable changes in the system. For example, while farmers emphasize scarcity
of fodder and manure, other changes in the nutrient cycle include the increased export of
bananas from the mountain and the fact that human waste is now disposed of in latrines,

whereas in the past it was recycled in the homegardens.

4.4.2 Forestry

Analyzing the social and environmental dynamics of the homegardens reveals that
these changes impact forest resources both on and off farms. In particular, declining
homegarden Size appears to lead to three major trends in forest resources:

1) increased pressure on resources in the Kilimanjaro National Park and Forest

Reserve;

2) decreasing tree density at the landscape scale; and

3) changing tree density, vertical distribution, horizontal distribution, and species

distribution at the homegarden scale.

First, as the size of homegardens declines, households have less space overall for
trees and often food or cash crops take priority over fuelwood, fodder, or timber trees.

O’Kting’ati (1985, 38) estimated that farmers obtained only about one-third of fuelwood

86

needs on-farm even prior to 1985. Land pressure has also reportedly reduced the area of
common lands and interstitial spaces where cattle could be grazed or firelwood or fodder
could be collected in the past. While some households facing scarcity turn to purchasing
fodder or fuel, others in Lyasongoro travel into the neighboring park and forest reserve,
where collection of fodder and dead wood for fuel is allowed. Several farmers, including
two who are members of the village forestry committee (Kamati ya Misitu), reported
observing degradation of forest resources in recent decades (e.g. smaller and sparser
trees, rarer species disappearing, and less wildlife). This degradation may be a result of
illegal harvesting of live wood.

With increasing constraints on cash income, some villagers in Lyasongoro have
turned to splitting dead wood in the park to haul and sell by the headload for small
amounts of extra cash. In the past, young men tended to practice this form of income
generation, but today it is primarily young women, even girls of primary school age.

A second trend resulting from the declining size of homegarden plots is that
overall tree density (excluding coffee and banana trees) at the landscape scale (below the
forest reserve) is reportedly declining. This is due to the pressure to harvest available
trees for fuelwood or timber in conjunction with an apparent decline in tree planting.
Concern over this trend has led to both regional and village-level tree-planting
campaigns. The village forestry committee was initiated in recent years with government
funding. Its members occasionally patrol the forest to discourage illegal cutting of live
wood (as do park employees) and to encourage tree planting in the village. The farmers’

awareness of landscape scale impacts of deforestation is evident in the following

statement.

87

If we ruin the forest, indeed, we ruin ourselves. Eh, and water is increasingly
drying up because of the degradation of the forest.59

The village also prohibits felling of trees near riverbanks, even though the trees are
considered privately owned.

A third set of trends occurs at the homegarden scale. In addition to an average
decline in tree denstiy, as homegarden size declines, tree planting is increasingly removed
to the farm margins, resulting in a less vertically-stratified homegarden system. Also, the
age and species structure of trees in the homegardens is probably changing. Farmers
noted that rarer species they remembered from their younger years were no longer to be
found. As pressure for fuelwood and timber increases, fewer trees reach maturity and

fewer are replaced as they are harvested or senesce (Figure 14).

5 Summary and Conclusions

5. 1 Theoretical Implications

The case of Mt. Kilimanjaro extends theories of population, environment, land,
and labor relations. It illustrates how, in the current context, even while land is scarce,
labor too can be in shortage. The case of Mt. Kilimanjaro extends theories of population,
environment, land, and labor relations. It illustrates how, in the current context, even
while land is scarce, labor too can be in shortage. This is not to say that some households
are short on labor while other households in the region are short on land, but rather that
most households simultaneously experience both constraints. This indicates simultaneous
land and labor scarcity at both the household and regional levels. The context I refer to
includes economic changes brought on by economic reforms, increasing needs for cash,

declining terms of trade for farmers, pressures toward diversification of income and non-

88

 

Figure 14. Senescence of a mature tree.

89

farm activities, out-migration, a shrinking land base, declining soil fertility, and
degradation of natural resources, both on- and off-farm. The gendered division of labor,
the age- and gender—differentiated character of out-migration, and the inter-dependence of
productive and reproductive activities are central to these processes, which result in a
continual squeeze on farmers’ labor and capital resources.

While the case of Kilimanjaro and the Chagga homegardens is unique in terms of
particular cultural, historical, and ecological circumstances, these same processes are
occurring in other rural locations. Theories of land and labor relations in rural economies
cannot continue to neglect the trend towards de-agrarianization and diversification of
income, even while agriculture remains of critical necessity for rural households’
sustenance. Nor can they neglect the centrality of gender relations for understanding
differentiation within rural households, and labor constraints in particular. These findings
are no surprise to Chagga farmers. While they may be significant theoretically, they are
most interesting in that they reveal the biased assumptions, particularly gender-blindness
and reductionism, that prior scholarship brought to analyses of population, environment,

land, and labor relations.

5.2 Recommendations

The farmers in this study were, for the most part, aware of “best management
practices,” but most of these are labor-intensive and many are capital-intensive. In terms
of practical implications, this research suggests the need for appropriate technologies
which are labor-saving and do not require high capital investment in order to improve
productivity on small plots. Perhaps even more critical than new technologies would be

changes in local institutions that would help to alleviate the strain on individual

90

households, such as access to credit, pooling labor or funds, or redirecting some of the
income from tourism to village women for use for development projects (e.g. child care
nurseries, flour mills, food processing and storing facilities, fuel-efficient stoves, or
income-generating activities). Given the size of the tourist trade on Kilimanjaro and the
extreme need for cash income, even the diversion of a small proportion of tourist income
to women’s efforts at village development could make a significant impact on household
livelihoods. A number of organizations already promote tree nurseries on Kilimanjaro
and the government has initiated a nationwide tree-planting campaign, with one million
trees expected to be planted in Moshi District (Lyimo et al. 1999, 6).
This study also suggests a number of questions for future research. For example:

0 How is tree density, spatial distribution, and species distribution in the
homegarden belt changing over time?

o How do changes in the homegardens affect the forest reserve?

a What are the precise ecological, economic, and social linkages between
farming in the uplands and in the lowlands. within and between
households?

0 How have recent social and cultural changes impacted management of the
homegardens? (e.g. changing gender relations, out—migration of youth,
AIDS)

0 How do local-, regional-, and national-level policies, discourse, and
ideology affect Chagga farmers, economically and socially? (e.g.
modernism, development, structural adjustment, monocropping,

privatization, cost-Sharing)

91

o How has the unique Chagga diaspora influenced processes on
Kilimanjaro?

Some of these suggestions for future research concur with a recent report by the Selian

Agricultural Research Institute, which also suggests many additional questions Specific to

crop, livestock, and tree research needs on Kilimanjaro (Lyimo et al. 1999, 104-108).

5.3 Conclusion

The story I have related here is that of simultaneous land and labor scarcity on Mt.
Kilimanjaro in a social and economic context of increasing constraints. There is a larger
story, however, to which I have alluded above, but which I choose to revisit here in hopes
of leaving its lasting impression on the reader. This is the story of the serious and critical
impact, on real people in particular locales, of the economic and agricultural development
policies generated at national and international levels. In assessing the impact of
structural adjustment policies on agricultural livelihoods in Tanzania, Forster and
Maghimbi reach a sobering conclusion:

There is little optimism for the future. The overall message seems to be that there

were many mistakes in the past but that the new system has done nothing to

improve the situation. There is much recognition of the ineffectiveness of top-
down measures and bureaucratic excesses, and of the need to listen to the
peasants: but this is to be seen in the overall context of entrenched elite privilege
which also enables the process of law to be used for personal advantage or even to
be circumvented. Unregulated capitalism has not made the lot of the peasant any
easier, and has deprived many of their means of livelihood. There clearly is a need

to improve peasant agriculture, and culture involves some constraints as well as
sometimes being responsive to innovation (Forster and Maghimbi 1999, xxiii).

The lot of farmers on Kilimanjaro certainly does not seem to be easier as a result of
structural reforms, which have failed to take into account the realities of those,

particularly women, coping with shortages of land, labor, and cash. Forster and

92

Maghimbi’s assessment is fair in that it takes into account the complexities of
interactions between local cultural factors and “top-down” policies.

15 more research into these complexities the solution to the problems of rural
people in Tanzania? While some research may be part of the solution, it must be
acknowledged and seen clearly that development research itself is embedded in the same
material and discursive relations that shape development policies. Throughout the course
of this research, I felt acutely the contradictions of my presence in Tanzania as a white,
female, graduate student researcher from the US, one of the wealthiest and most
powerful countries in the world. I found my role to be incompatible with the needs I was
hearing expressed by Tanzanians for self-determination, autonomy, and fair access to
resources. As an elderly doctor, who had long experience with foreign student researchers
on Kiliamanjaro, put it, “Let’s be honest. These students come, they make their
recommendations, and they leave, but the recommendations never come with any
resources for implementation attached, so the same recommendations continue to be
made.”

I close with this moment of self-reflexivity because I ask the reader to likewise
question your own relationship to these processes, even to the villagers of Lyasongoro,
Kilimanjaro, Tanzania. Mbilinyi reminds us:

The relations of production, consumption and distribution within which rural

women work are therefore not limited or confined to the peasant household. They

include TNCS, the World Bank, other donors, experts, consumers, taxpayers,
politicians and others located in both the North and in Tanzania. Residents in the

North are part of the politics of peasant agriculture in Africa. The coffee and tea

they drink, the clothes they wear, the taxes they pay and the policies their

governments promote implicate them in the deepening exploitation and
oppression of Afiican workers and growers (Mbilinyi 1997, 340).

93

REFERENCES

Agarwal, Bina. 1991. “Engendering the environment debate: lessons from the Indian
subcontinent.” CASID (Center for the Advanced Study of International
Development) Distinguished Speaker Series. Michigan State University, East
Lansing.

Anderson, GD. 1982. “A survey of the soils and land use potential of the southern and
eastern slopes of Kilimanjaro, Tanzania.” Dar es Salaam: Institute of Resource
Assessment and Clark University.

Bagachwa, Mboya S.D., Fanuel C. Shechambo, Hussein Sosovele, Kassim A. Kulindwa,
and Alexis A. Naho, with Elizabeth Cromwell. 1995. Structural adjustment and
sustainable development in Tanzania. Dar es Salaam: Dar es Salaam University
Press.

Bawden, Richard. 1995. “On the systems dimension in FSR.” Journal of Farming
Systems Research-Extension 5: l -1 8.

Beneria, Lourdes, and Gita Sen. 1981. “Accumulation, reproduction and women's role in
economic development: Boserup revisited.” Signs 7:279-298.

Blaikie, Piers, and Harold Brookfield. 1987. Land degradation and society. London and
New York: Methuen.

Blaikie, Piers M. 1994. Political ecology in the 1990s: an evolving view of nature and
society. East Lansing: Center for Advanced Study of International Development,
Michigan State University.

Boserup, Ester. 1993. The conditions of agricultural growth: the economics of agrarian
change under population pressure. London: Earthscan Pubs. Ltd.

Bryceson, Deborah Fahy. 1997. “De-agrarianisation in sub-Saharan Africa:
acknowledging the inevitable.” Pp. 3-20 in Farewell to farms: de-agrarianisation
and employment in Afi-ica, edited by Deborah Fahy Bryceson and Vali Jamal.
Aldershot, UK: Ashgate Publishing Ltd.

Bryceson, Deborah Fahy, and Vali Jamal (Eds). 1997. Farewell to farms: de-
agrarianisation and employment in Africa. Aldershot, UK: Ashgate Publishing
Ltd.

Cheal, David J. 1996. New poverty: families in postmodern society. Westport:
Greenwood Press.

Clay, Daniel C., Thomas Reardon, and Asif Shaikh. 1996. Population, environment, and
development in Afi'ica: dynamic linkages and their implications for filture
research and development programming. East Lansing: Institute of International
Agriculture, Michigan State University.

94

Collins, Jane L. 1987. “Labor scarcity and ecological change.” Pp. 19-37 in Lands at risk
in the Third World: local-level perspectives, edited by Peter D. Little and Michael
M. Horowitz, with A. Endre Nyerges. Boulder and London: Westview Press.

Cook, Cynthia C., and Mikael Grut. 1989. “Agroforestry in sub-Saharan Africa.”
Washington, DC: World Bank.

Corbin, Juliet, and Anselm Strauss. 1990. “Grounded theory research: procedures, canons
and evaluative criteria.” Qualitative Sociology 13(1):3-21.

Creighton, Colin, and CK. Omari (Eds). 1995. Gender, family and household in
Tanzania. Aldershot, UK: Avebury Publishers Ltd.

Deere, Carmen Diana, and Magdalena Leon de Leal. 1981. “Peasant production,
proletarianization, and the sexual division of labor in the Andes.” Journal of
Woman in Culture and Society 7(2):338-360.

Femandes, E. C. M., O. Kting' ati A, J. Maghembe, and J. A. Maghembe. 1984. “The
Chagga homegardens: a multistoried agroforestry cropping system on Mt.
Kilimanjaro (Northern Tanzania).” A groforestry Systems 2(2):73-86.

Femandes, E. C. M., and P.K.R. Nair. 1986. “An evaluation of the structure and function
of tropical homegardens.” Agricultural Systems 21 :279-310.

Forster, Peter G., and Sam Maghimbi. 1999. “Introduction.” Pp. xvii-xxiv in Agrarian
economy, state and society in contemporary Tanzania, edited by Peter G. Forster
and Sam Maghimbi. Aldershot, U.K.: Ashgate.

Gamassa, D.M. 1991. “Chapter 1: historical change in human population on Mount
Kilimanjaro and its implications.” Pp. 1-8 in The conservation of Mount
Kilimanjaro, edited by William D. Newmark. Gland, Switzerland and Cambridge,
UK: IUCN—The World Conservation Union.

Garcia-Barrios, Rail, and Luis Garcia-Barrios. 1990. “Environmental and technological
degradation in peasant agriculture: a consequence of development in Mexico.”
World Development 18:1 569-85.

Howard, Mary, and Ann V. Millard. 1997. Hunger and shame: poverty and child
malnutrition on Mount Kilimanjaro. New York: Routledge.

Kilimanjaro Regional Nutrition Programme (KRNP). 1990. “Report on situation analysis
of food and nutrition problems facing women and children.” Moshi, Kilimanjaro:
Kilimanjaro Regional Development, Director's Office, Tanzania Food and
Nutrition Centre (TFNC).

Koda, B0. 1995. “The economic organization of the household in contemporary

Tanzania.” Pp. 139—155 in Gender, family and household in Tanzania, edited by
Colin Creighton and CK. Omari. Aldershot, UK: Avebury Publishers Ltd.

95

Lema, Anderson J. 1997. “Land degradation on the southern slopes of Mount
Kilimanjaro and the prospects for sustainable development.” Pp. 101-118 in
Gender and agrarian change in Tanzania with a Kenyan case study, edited by
Suleiman Ngware, Rie Ordgaard, Rose Shayo, and Fiona Wilson. Dar es Salaam:
DUP Ltd.

Lyimo, S.D., C.J. Lyamchai, M.Z. Owenya, M.N. Kingamkono, F.S. Ngulu, H. Mansoor,
and S. Bitende (Eds). 1999. A participatory rural appraisal (PRA) of the Chagga
homegardens in Moshi District, Kilimanjaro Region, Tanzania. Arusha, Tanzania:
Selian Agricultural Research Institute (SARI).

MacDonald, D., J.R. Crabtree, and G. Wiesinger. 2000. “Agricultural abandonment in
mountain areas of Europe: environmental consequences and policy response.”
Journal of Environmental Management 59(1):47-69.

Malthus, Thomas Robert. 1959. Population: the first essay. Ann Arbor: University of
Michigan Press.

Mbilinyi, Marjorie. 1997. “Beyond oppression and crisis: a gendered analysis of agrarian
structure and change.” Pp. 317-364 in Engendering African social sciences, edited
by Ayesha M. Imam, Amina Mama, and Fatou Sow. Dakar, Senegal:
CODESRIA.

Meena, Ruth. 1991. “The impact of structural adjustment programs on rural women in
Tanzania.” Pp. 169-189 in Structural adjustment and African women farmers,
edited by Christina H. Gladwin. Gainesville, FL: University of Florida Press.

Michon, G., J. Bompard, P. Hecketsweiler, and C. Ducatillion. 1983. “Tropical forest
architectural analysis as applied to agroforests in the humid tropics: the example
of traditional village-agroforests in West Java.” Agroforestry Systems 1:117-129.

Newmark, William D. (Ed.). 1991. The conservation of Mount Kilimanjaro. Gland,
Switzerland and Cambridge, UK: IUCN—The World Conservation Union.

Ngware, Suleiman, Rie Ordgaard, Rose Shayo, and Fiona Wilson (Eds). 1997. Gender
and agrarian change in Tanzania with a Kenyan case study. Dar es Salaam: DUP
Ltd.

O'Kting' ati, A, J. A. Maghembe, E. C. M. Femandes, and G. H. Weaver. 1984. “Plant
species in the Kilimanjaro agroforestry system.” Agroforestry Systems 2(3):]77-
186.

O'Kting' ati, A, and H. O. Mongi. 1986. “Agroforestry and the small farmer: a case study
of Kilema and Kirua Vunjo in Kilimanjaro.” International Tree Crops Journal
3(3):257-265.

96

O'Kting'ati, Aku. 1985. “An analysis of the economics of agroforestry in Kilimanjaro.”
Doctoral thesis in the Faculty of Forestry. Morogoro, Tanzania: Sokoine
University of Agriculture.

O'Kting'ati, A., and JR Kessy. 1991. “Chapter 8: the farming systems on Mount
Kilimanjaro.” Pp. 71-80 in The conservation of Mount Kilimanjaro, edited by
William D. Newmark. Gland, Switzerland and Cambridge, UK: IUCN—The
World Conservation Union.

Perera, Ajith H., and R.M.N. Rajapakse. 1991. “A baseline study of Kandyan Forest
Gardens of Sri Lanka: structure, composition and utilization.” Forest Ecology and
Management 45:269-280.

Redclift, Michael, and Colin Sage (Eds). 1994. Strategies for sustainable development:
local agendas for the southern hemisphere: John Wiley and Sons, Ltd.

Rocheleau, Dianne, Barbara Thomas-Slayter, and Esther Wangari (Eds). 1996. Feminist
political ecology: global issues and local experiences. New York: Routledge.

Rocheleau, Dianne, Barbara Thomas-Slayter, and Esther Wangari. 1996. “Gender and
environment: a feminist political ecology perspective.” Pp. 3-23 in Feminist
political ecology: global issues and local experiences, edited by Dianne
Rocheleau, Barbara Thomas-Slayter, and Esther Wangari. New York: Routledge.

Sage, Colin. 1994. “Population, consumption and sustainable development.” Pp. 35-59 in
Strategies for sustainable development: local agendas for the southern
hemisphere, edited by Michael Redclift and Colin Sage: John Wiley and Sons,
Ltd.

Sankhayan, P. L. 1995. “Effect of structural adjustment policies on resource management
at peasant household level in Tanzania.” Pp. 13-30 in Seminar on structural
adjustment policies and the management of soil and forest resources in Tanzania,
2 3 August 1994, Sokoine University of Agriculture, Morogoro, Tanzania.
Norwegian Journal of Agricultural Sciences. Supplement No. 21.

Shivji, Issa G. 1998. Not yet democracy: reforming land tenure in Tanzania. Dar es
Salaam: IIED/HAKIARDHI/Faculty of Law, University of Dar es Salaam.

Spear, Thomas T. 1997. Mountain farmers: moral economies of land and agricultural
development in Arusha and Meru. Berkeley: University of California Press.

Temu, A. A., and P. S. Mamiro. 1995. “Structural adjustment programme: preliminary
situation and outlook of coffee-based small scale farmers in Kilimanjaro.”
Agricultural Economics Analysis and Rural Development 5(1):51-59.

Tiffen, Mary, Michael Mortimore, and Francis Gichuki. 1994. More people, less erosion:
environmental recovery in Kenya. Chichester, UK: John Wiley and Sons.

97

Torquebiau, Emmanuel. 1992. “Are tropical agroforestry home gardens sustainable?”
Agriculture, Ecosystems and Environment 41 :189-207.

von Clemm, Michael. 1964. “Agricultural productivity and sentiment on Kilimanjaro.”
Economic Botany 18(2):99-121.

Wojtkowski, RA. 1993. “Toward an understanding of tropical home gardens.”
A groforestry Systems 24:215-222.

World Bank. 1999. Tanzania: social sector review. Washington, DC: The International
Bank for Reconstruction and Development/The World Bank.

98

END NOTES

 

I Vihamba ni mali ya ukoo, hairuhusiwi kuuza. Kila mtu anarithi, mtoto wake
anarithi, na kadhalika. Serikali, siyo ya serikali.

2 Kuuza shamba lako?! Ha! Utaishi wapi?

3 Ni ngumu sana kwa Mchagga kuuza shamba lake.

4 [...pengine, mtoto wa kiume mmoja, itambidi kuenda kutafuta shamba lingine,
kutafuta kazi nyingine .7] Hata kama atafanya kazi na kupata shamba lingine, lazima
apate sehemu.

5 Hata kama ni shamba dogo, kila mmoja anapata mahali pa kutengeneza
nyumba. Anapata tu mahali pa kuweka bustani. Kama ni shamba dogo, ndiyo ni hivyo
wanapanga. Kama ana uwezo, ana kibarua chake kama huyu, anaweza kununua shamba
lake pahali pengine. Lakini pale nyumbani anapata mahali pa kutengeneza nyumba yake.
Ndiyo hivyo huko kwetu wanafanya. Kuwa na mtu ana shamba dogo na kila mmoja
anagawiwa mahali pake, ndogo tu, aweke nyumba yake na kibustani choke. Halafu
akienda kwenye kibarua akiwa na hela, anunua shamba mahali pengine, lakini pale
nyumbani ana pahali pa kushikia.

6 Hatuna mashamba hapa, tuna kiwanja kuishi tu.

7 Watoto wa kike hawapewi Vihamba. Ni watoto wa kiume. Ndiyo unapoolewa
mume wako anapewa kihamba na baba yake. Unakuta kama hana atakupa kulingana na
jinsi anacho. Kama watoto wanaoa ni watoto wengi, lazima ugawe ili kila mmoja apate
kipate kidogo. Kama watoto ni wachache, unapata sawa sawa ulivyokuwa.

8 Hapana. Hapana. Hawatapata sehemu. W a kike hawapati kabisa. Kwa kweli
kuna wengine wanapata sehemu, kama, kama ni Himo na shamba lako liko porini, ni
baba yangu mzazi alimpa dada yangu. Na mimi, baba, alinipa ng'ombe.... Sasa baba
wengine wanajua, hawaangalii huyu ni wa kiume au huyu ni wa kike. Inatokana na yule
baba. Baba wengine wanatoa yaani sehemu hata kwa wale wa kike wanapata sehemu
yao, urithi wao.

9 Halafu unakuta wengine kama kuna eneo kubwa la miti, anatoa miti ilikuwa kwa
ajili ya kupata majani ya ng'ombe. Haya anapunguza kwa ajili ya nafasi ya kulima pia.
Kwa sababu watoto wanaongezeka. Mashamba haya ulikuta zamani, mababu zetu
walikuwa na mashamba makubwa. Sasa unapokuwa watoto wengi ndiyo inazidi
kugawanyisha. Mwisho inayobakia ni kidogo, kila mtu anapata sehemu kidogo tu. Kama
vile ni kiwanja.

'0 [Lakini unaona faida yo yote kuhusu mwanga, kutumia vizuri mwanga, kwa
ajili ya kuchanganya kila kitu, ili uweze kutumia mwanga yote?] Kwa kweli,
ukichanganya kila kitu, havipati mwanga wa kutosha. Eh. Kwa sababu unapotaka
kahawa imefikia hapa, na migomba iko kule juu, sasa kahawa haipati mwanga wa
kutosha. Sasa tungekuwa na mashamba makubwa, tungeweka kahawa hapa, migomba
hapo. Sasa mashamba ni madogo. Mm. [ Okay. Kuna faida yo yote kupata kivuli kwa
mazao yo yote? Au hamna?] Eh. Kuna sehemu kama huko shambani, lazima unaotesha
kwa mfano kuanzia pale mpaka hapa, unaotesha mti mmoja hapa kwa ajili ya kivuli. Eh.
Kuna kipindi jua huko kwetu linawaka sana, Januari. na mwezi wa pili. Sasa jua lile ni

99

 

libaya kwa kweli. Linaunguza ardhi. Na ardhi ile haina rutuba. Eh. Inakauka haraka
sana.

” Sitajua kwa kweli. Eh. Hata tulipozaliwa sisi tulikuta mseto siku ile.

12 Kwa sababu ya, waliotesha kwa sababu walikuwa na, zamani walikuwa na
mashamba makubwa sana. Waliotesha, wakaweka mpaka, na hivi. Sasa wakati hivi hapa
tumekuwa vidogo. Eh. Wakuja kukata watoto wakalala huko.

'3 Unaweza kutenga kila kitu, lakini ardhi ni ndogo. Tumetoa sehemu yako ni
kidogo, lazima uchanganye kila kitu hapa. Na uweke mbolea. Lakini ukiweka mbolea,
vinaendelea tu vizuri. Eh. ...Ndiyo, walikuwa (wazee wa zamani) na mashamba
makubwa, lakini walikuwa hawajui maana ya kilimo. [Maana ya kilimo?] Eh. Walikuwa
hawajatambua maana ya kilimo. [Kweli?] Wangekuwa wametambua ya kwamba, kilimo
ukiweka mahindi hapa, urudi hapo. uweke hiki, urudi uweke maharagwe hapa,
wangeotesha hivyo. Lakini sasa walikuwa hawajui maana yake. Eh. Sasa hivi watu
wamepanga kwa akili. Wanajua nini. Na kisomo na nini. Eh. Wameshasoma wakajua
uchanganya inaharibu. Inaharibika, vingine vinanyonya vingine. Mm.

M Ni kilimo cha kienyeji. (laughs) Kwa sababu hiyo kutenganisha ni kitaalum. Au
siyo? [Sijui, inategemea, nadhani.] Kutenganisha si kitaalum?

'5 Hiyo kahawa na migomba pamoja, sisi tulikuwa, lile shamba lilikuwa na ile
kahawa na migomba. Kwa hiyo hatuwezi kutoa. Kwa sababu ardhi ni ndogo. Ndiyo
maana wanachanganya. Lakini tungekuwa na ardhi kubwa ya kuweza kufaa hili shamba
lote la kuweka migomba, na zao lingine hapa kama kuweka kahawa, ingekuwa vizuri
zaidi. Eh. Maana kahawa ikatoka sana. Na ndizi zitatoka sana, bila ya kuchanganya.
Lakini sasa inabidi tuchanganye. Unachanganya hiki, unaleta hiki, unaweka hiki,
unaweka hiki, ili tupate kitu cha kula. Kinyume ya kuchanganya, hupati chakula.

'6 Hiyo (kahawa) ni mazao ya biashara, inasaidia watoto kusoma. Eh. Huwezi
(kutoa) kabisa. Hata ukipata kidogo inasaidia, kupata pesa, hiyo migomba ni kwa
chakula tu. Eh. Lakini, ni kwa biashara...Eh. Watu wakipata sadolini mbili, kama kilo
nne, kilo nne unapata pesa inasaidia kwa watoto kuenda shule ya msingi. Eh. [Hata
kupunguza kidogo, haifai?] Kahawa? flaughs) Wababa wanaipenda kahawa. Hata
ukivunja tawi! (laughs)

'7 Mambo ya kilimo hayalipi kwa sehemu ya Wachagga, kutegemea hapa kwetu.
Haihitaj i kutokana maeneo ni madogo.

'8 Kwa kweli hapa kwetu Wachagga, watu wanatumia akili sana. lnabidi utumie
akili ili uweze kuishi. :

‘9 Huwezi kusema kwamba ulime tu ukose hela za kununua manufaa na nini.
Huwezi kusema ufanye kazi tu ukose mazao ya watoto, au vitu vidogo vidogo vya kula.
Eh. Kwa hiyo vyote viko sawa.

20 Kwa kweli, huko kwetu, hata mtu mwenye eneo lidogo, anapata shida. Kwa
sababu thamani ya hela ya Tanzania ni ndogo. Sasa kama hutatumia vitu kama hivi kwa
ajili ya nyumbani kwako, lazima utapata shida, utaishi kwa shida. Ukitegemea tu ile hela,
hela ya ununue mboga, ununue sukari, ununue nyama, yaani kila kitu, haitoshi. Sasa
unaona maana ya tunalimalima, unapata viazi, unapata maboga, kila kitu unapatapata
huko shambani, ndizi, kahawa.

2 I Unakuta mara moja umefitngua biashara na mara umeshindwa, umeiacha.

100

 

22 Yaani duka haliwezi kutosheleza kwa kutunzafamilia kama hiyo kulima, kilimo.
Hiki ni kitu cha msaada tu, kuongeza hela.

23 Lakini sasa unakuja kwa watu wanaenda na wakati. Nyumba za kuzeeka tena
na majani hakuna, wanatumia bati. Ambalo ni hilo linaleta, kuweka matumizi ya pesa ni
mengi. Zamani walikuwa matumia mengi ni utaalum. Ujuzi wa kitaalum, ambao hakuna
matumizi ya pesa.

24 Nyumba [ya matajiri] nyingi ni kwa ajili ya mambo ya mlimani. Ni mambo ya
tours, mambo ya hoteli.... Ndiyo, ni mtegemeo mkubwa. Lakini kinyume na kuwa na huo
msimu wa mlima, ingekuwa shida sana... [Hela hii inatawanyika?] Eh, ndiyo unakuta
Kama wewe huwezi, akija mume wako anaenda, au mtoto, au mtu mwingine, lakini yule
mtu anapopata pesa, una biashara chako, anakuja naye, anapata mteja. Kwa hiyo hela
hii inazunguka.... Lakini pesa inayopatikana kwa wingi zaidi hapa mlimani, hakuna, kwa
sehemu hizi hakuna. Mambo ya kilimo hayalipi.

25 Eh! Kilimanjaro, ardhi ni ya shida. Eh. Umeshawahi kuambiwa kwamba
Wachagga ni wagumu sana kuhama na kuhamia maeneo mengi. Hawapendi sana.
Isipokuwa sasa hivi inatokea ya hali ya maisha inafanya watu wajifimze kuhama na
kuhamia sehemu nyingine kulingana na familia ya mtu unakuta mahali ambapo
haimtoshelezi, kusaidia familia yake. Basi wengi sasa hivi ndiyo wanainuka. Lakini
utakuta sehemu aliyopewa na mzazi wake haachii, hawezi kuwa apaachia, hapaachii
tupu wa jangwa. Lazima ukute anapotunza. Kama siyo yeye lakini anamkodi hata
mbuzi.... Lakini kwa ardhi ambayo iko tupu haina mtu, si rahisi, si rahisi.

26 Na watoto mara nyingi hawapendi kujishughulisha na mambo ya kilimo...
lazima wataenda mjini.

27 Kwa kweli inachosha maisha ya hapo [shambani] magumu. Yaani unaandama
na pesa tu. Kinyume na hapo unatumia nguvu zako. Kwa hiyo sasa tumeanza
kulinganisha maisha ya shamba na mjini, ya mjini ni nzuri zaidi naona, kwa ajili mimi
nimeshakaa mjini. Na siyo mimi ambayo nimekaa mjini tu, hata wengine wanapenda,
lakini inapobidi shamba, . . . nakuwa shamba.

28 [Na hawa watoto wanaokaa Arusha?] Wanatutegemea tu, bado hawajapata
hela za kuuza nyingi. Lazima ukipata kidogo kama elfu mbili unawatumia unawapelekea,
hivyo hivyo. Naona anajua shida, anaivumilia. (laughs) Anaivumilia. Hasa yule wa kike
namwambia, ameanza tu mwaka jana, mwezi wa tatu. Hajamaliza siku nyingi pale. Kwa
hiyo anatakiwa malipo ya kulipa elimu ya mafundisho, ya sabuni ya kufixa, ya mafuta ya
kupakia na ya taa, yaani ni lazima atakuwa anadai hapa nyumbani. [Unadhani mahitaji
mchicha kwenye bustani. Wao wanainunua tu. Pale mjini ni kununua, mahindi tunaweza
kulima, tunatumia kwa ng'ombe, kule wananunua, kila kitu chakula tu. Eh. [Unaonaje
maisha ya mji, au wanaonaje maisha?] Magumu sana, magumu kwa kweli. Lakini
unamfundisha jinsi ya kuweza kuvumilia mpaka wanaweza kufanikiwa. [Unadhani

kule kule, ni yeye mwenyewe atapanga. Sisi hatumpangii.

101

 

29 Lakini pamoja na hayo maisha ya mjini bila shamba watu hawawezi

kuwepo...Kwa ajili wale wa shamba wanapata hela kwa ajili ya wale wa mjini. Na wa
mjini wanapata vitu kwa ajili ya wale wa shamba. Hivi vitu vinasaidia.

30 [Ulisema kwamba hela haizungukia hapa. Unaonaje matajiri ambao wanakaa
nje, lakini wana nyumba hapa? Unadhani wanasaidia maendeleo ya mlimani, au...]
Matajiri wanaokaa nje hata hivyo kutokana na ufahamu, watu wamekuwa na
hawafahamu wa mbali. Wanaweka tu nyumba, lakini kutegemea kwamba atazalishia
huko ni ngumu, kutokana na ameshajua upato kwa huko ni vigumu sana. Afadhali kule
mjini unaweza kutafuta na kupata upato.

3 ' [Unaonaje matajiri wa Kilimanjaro lakini wanakaa nje, unaonaje wanasaidia
yako, anaweza kukusaidia. Lakini wengine wa pembeni si rahisi sana. Wanasaidia
wenzake ambao wana uhusiano nao.

32 Wanafi/ekwa kama majani, wanalala kama majani.

33 [Hutaki zaidi ya ng'ombe mmoja?] Nimeshindwa kwa sababu ni kazi nyingi.
[Kazi ya kupata majani au kila kitu?] Sasa. Ukiwa na ng'ombe wawili watatu, wanaweza
kukuzuia kuzifanya kazi za shambani, kama kulima, na wakati wa kuchuma kahawa
hutapata nafasi. Kama ningekuwa na familia ya kutosha kufanya kazi, ningeongeza
ng'ombe wawili au watatu, tukiwa tunasaidiana. Ndiyo.

34 Sasa hamna hela za kuweka vibarua vya kukondoa zile fitko huko chini.
Ukiotesha migomba na unaliwa, unaanguka, ukiotesha, ndiyo hivyo. Basi ikawa, hamna
msaada wa kivibarua ya kusaidia huko shambani.

35 Lakini jinsi kazi zinaanza zimesonga mara kwa mara, na sana sina nafasi ya
kukaa dukani na kuwa na mapato ndogo sana. Eh. Sasa huko tunaenda kwa mwezi huo
wa tano, kuendelea wa sita ndiyo ni wakati wa kulima. Umeshalima unaotesha,
umeshaotesha unapalilia, hata mara mbili. Kwa hiyo nafasi ya kuangalia dukani nakuwa
na ni kidogo, ni shughuli ya mazao yasipotee. Yaani duka haliwezi kutosheleza kwa
kutunza amilia kama hiyo kulima, kilimo. Hiki ni kitu cha msaada tu, kuongeza hela.

6 ...hakuna msaada wa kutusaidia, kama watoto wanafanya kazi ya vibarua
vyao. Kwa sisi, tunashindwa. [Kwa hiyo una ardhi, lakini hamna vibarua] Ardhi ipo, eh,
hamna vibarua wa kutosha.

37 Sina shamba lingine. Ningehitaji nitapata, lakini ningechoka sana. Siwezi.

38 [Na shemeji wengine, wanalima pamoja, au kila mmoja ana shamba lake? ]
Kila mtu ana shamba chake kidogo tu, kibustani. Kama bustani, ni shamba kidogo tu. Eh.
Kila mtu ana kishamba chao.

39 [Na nyinyi nyote mnashirikiana na mfuko mmoja? Kama mtu mmoja anahitaji
hela au msaada, mnasaidiana au kilo mtu ana mfuko?] Kila mtu ana mfuko wake. Eh.
Kila mtu ana mfuko wake. Hakuna cha kusaidiana. [Kwa hiyo, hata kama watoto wako
Wa kiume wakisaidia na shamba, ni kama kukusaidia tu, hawatapata faida ya kihamba?
Ni wewe?] Hapana. Ni yeye. Kama hawa wa kike wanavyosaidia, wanasaidia mingi tu.
[Lakini hapa mtoto huyu ameshaoa, kwa hiyo mke wake ana sehemu yake?] Ameshaoa.
Ana sehemu yake, na ng'ombe wake anao. A nafanya kazi yake peke yake. [lmekuwa hivyo
hata zamani?] Hata zamani. Eh. Hata zamani.

102

 

4O Marehemu alipofariki, nilikuwa ninashughulikana no watoto peke yangu!, bila
yo msaada wo shemeji, bila ya msaada wa urithi, bila msaada wo wote, ikawa nilikuwa
nimehamia hapa kwa sababu mume alifariki. Mimi nilijishughulisha mwenyewe, hamna
msaada, ukiomba msaada hawakupi.

4' Utakuta matajiri atafuatilia sana matajiri wenzake. Lakini tajiri hamkuti mtu
wa hali ya chini, akijua kwamba tajiri atamsaidia maskini, mara nyingi hata akikupa ni
kidogo.

42 Watasimanga.

43 “msimu wa shida”

44 Kama hapa kwetu hamna kitu tunochowezo kuuza ili tubaki no zile hela. Au
kutununuzia matumizi, hamna. Eh, hatupati kabisa. hata mazao tunayotoko kuyavuna
huko shambani, hayafikii hata muda mrefu, kama ukiyavuna huko shambani, sasa hivi
yanaisha tu. Labdo sijui kama itofikia mwezi wa saba, yokifikia hapa yameisha. Kwa
sababu shamba ni la kufungafunga sana. No mahindi hayatoki. Hayatoki sawa sawa
huko kwetu. Mpaka mwezi wa none, hayashughulikii.

45 Hela ya kahawa imeshikwa no CBD. Sosa ungesema utabaki no hela ile ya
kahawa kama ungepata, kama si CBD, ndiyo ninayo. Sisi, kama hela haipatikani kwa
CBD, ou kama una ng'ombe wako ukauza. Kama huna ng'ombe, huno kahawa, ndiyo,
shida tu.

4" akiba

47 Basi, mimi ninavyoona kwa sababu hicho kilimo cha mseto kwa kweli kuna
faida nyingi ndani yoke. Kuna faida nyingi ndani yoke, lakini, kama utakuwo no sehemu
kubwa, mm hmm.... Haitoshelezi kwa matumizi yofamilio. Hoitoshelezi.

48 Cha mseto ni kwa shida tu, tatotizo.

49 Eh, walikuwa wanatumia kilimo cha mseto. Lakini ilikuwo kwa matumizi ya
nyumbani, siyo maendeleo. Siyo maendeleo. Ilikuwa kwa ajili yo matumizi ya nyumbani.
Lakini wangekuwa no akili ambao vijona wonao sasa hivi, unakuta maendeleo yameanza
zamani. Waliokuwa no akili.

50 Tunaenda kwa wakati. Hatuwezi tena kurudi nyuma. Huwezi kumwombia
kurudi nyuma. Ndiyo hata maisha ya siku hizi.

51 Lakini pio tunajali umbali ule kutokana no, inoposongamano sana migomba no
kahawa, vile vile ukija kulima unaona mengine kama mahindi na maharagwe hayatoki.
Eh, kwa ajili ya kivuli ile, ikifunga. Au kipindi kama hiki cha soso hivi wakati wo
moondalizi wa kuotesha mahindi...sehemu ambayo ino migomba mingi wonoandao
kupunguza.

52 Ukiongezo kiasi cha miti shambani, kahawa hupati, migomba imekuwa mirefu,
hata kama mizizi ya huko chini ni mingi sana, houfoi kulima sasa, kuotesha mahindi,
yaani mahombo yole, hata maharagwe. Kwa sababu ukiotesha miti mingi [hutapata
mazao mengine]. Unojuo kilimo chetu ni cha mchonganyiko. [Eh, ndiyo] Kwa hiyo,
kivuli kikuwo kingi hoitokei maharagwe, hata mahombo, migomba itakuwo mirefu, hata
kahawa.

53 Hiyo sasa ni nafasi yako, kama unomtaka mbalimbali, kama unatoka
karibukaribu. Mwingine unakuta hapendi kusukumanisha ile miti kwa ajili majani
yosikosekani. Unapo karibu ndiyo teno hata majani kivuli kinokuwa kimetonda. Kwa hiyo

103

 

majani ya chini hayokuwapo. Eh. Kulikuwo no kivuli...Mwingine anaotesho mbalimbali
ili ino nafasi kama majani yawepo. Au anaweka tena miti mingine, au ya aino nyingine.
Kwa hiyo pale ni kwa matumizi yo mtu. Kama ni sehemu za bonde, ombozo hazifoi
majani, au mtu anoyo shamba kubwa, unaona unakuta wameshaotesho miti ya mitorako
mingi! Mwingine unakuta anaotesho kwa mbalimbali onochangonya no meresi. Kama ni
sehemu ambayo haiko karibu no shamba. Kwa sababu ile mitorako huwezi kuweka
sehemu zo ndizi ou kahawa, huoteshi no mitarokwe. Lazima ni kando, siyo kama ni
mabondemabondemobonde.

54 No sisi, kwa mfano sisi waliokuwo watoto sao hizi, watoto wo watu wa zamani,
hotujootesha migomba au kahawa kwa kweli. Eh. Tulikuto. Kazi yetu ni kupalizi tu. Eh.
Tulikuta, sema kuna nyingine inakouko, migomba, unatoa kwenye kina ile unorudishia
pale huko huko. Lakini migomba tu tumeiweka. Eh.

55 Unakuta sasa kama ameotesha miti mingi, baba yoke omempo sehemu ya miti
mingi, na yeye soso ile miti ni kwa ajili yo hudumo yoke. Anataka kutengeneza bonda la
ng'ombe, ou jiko, ou nyumba yo kulola, kujenga motofoli, hata kama ni motofoli sasa
kuna sehemu ya miti. Haya. Kama siyo kujenga utakuta amepunguzo kwa ajili yo kile
kivuli omepunguo ili ooteshe majani ya ng'ombe, ou kwa kulima.

56juhudi zake zo kuotesha miti.

57 Imesholimwalimwo sana. Siku hizi hoina nguvu tena kama zamani.

58 Ndiyo, kwa namna yo utunzaji unaweza kuongeza rutuba. [Kwa kutunza
ng'ombe ou...?] Kutunzo ng'ombe na hata lile shamba, lile shamba kwa kulima no
kutunza ng'ombe ombozo zinatoa somadi nyingi, unotandiko, kama kuna kitu ombocho
unaotesha, unaotesha no mboleo nyingi, hapa rutuba, no mboleo inokuwo nyingi, unajuo
hapa kwa shemeji kuna migomba. Kwa sababu wana ng'ombe mengi, ona ng'ombe kama
wanne. Wanne. Sosa ikawa, no ile migomba ni ile mikubwo. Basi okawo no mboleo
nyingi, holafu shamba lake siyo kubwa. Ikawo siyo yo saizi. Hapa anaotesho migomba no
onakuwo ni yeye anasimama no shamba, kushughulisha peke yoke. [nakuwa mboleo ni
nyingi, inatondikwo kilo mahali, no yeye kilo sao onakuwo no kazi ya kulima. Yo
kupanuo, kilo dokika, no kilo dokika anashughulikia, kama ni motowi anashughulikio
mwenyewe, kilo dokika onoshughulikio anajua, mimeo inaingio nyingi, basi kwenye
migomba, unakuta ni kwa mboleo, mbolea, mboleo. Basi, hota huko chini, unapokodikwa,
unakuwo sasa hivi, ndiyo hivyo. Wono ng'ombe zo kutosheleza kwa somadi.

59 T ukihoribu msitu ndiyo tunajihoribia mwenyewe. Eh, no maji yonozidi kukauka
kwa ajili yo msitu unapunguo.

104

llljljljjljljjj‘jilljjjll