i

Sociocultural Aspects of Water Usage

amongst the Chagga People in North -
Eastern Kilimanjaro, and implications for
anti-trachoma programmes



Dissertation submission for the degree of
MSc in Medical Anthropology, Brunel University
Academic Year 2002-3

Robert Th
ii
ACKNOWLEDGEMENTS

This research project has been prepared, conducted and completed with the support of the following
institutions and persons:

LONDON, UK

Brunel University

Dr. Nicholas Argenti (For bibliographic assistance)
Dr. Suzette Heald (For additional fieldwork advice)
Dr. Melissa Parker (For her invaluable feedback and supervision)

The London School of Hygiene and Tropical Medicine

Dr. Hannah Kuper (Project and finance co-ordinator)
Sarah Pollack (For her preparatory work on water usage in Rombo and technical assistance in the
field)
Carolina Abuelo (For her observations and suggestions in the field)
Rosa Arques (For her attentive support in sourcing trachoma-related literature)

Acknowledgement is hereby made to the LSHTM for their generous funding of the research project.

KILIMANJARO REGION, TANZANIA

Kilimanjaro Centre for Community Ophthalmology (KCCO) Tumaini University, Moshi

Dr. Paul Courtright (For providing feedback and support in the field)
Robert Geneau (For his insightful anthropological comments and advice)

Rombo Trachoma Research Project, Mkuu, Rombo District

Patrick Massae (For his patience, ready assistance and technical support in the field)
Mosoi Balthazar (Without his invaluable cultural and linguistic assistance, this work would not have
been possible)

Heartfelt thanks are also extended to the members of Urauri Village who warmly welcomed me into their
community and provided me with the opportunity to observe and participate in their daily lives, whilst
politely humouring my stumbling Swahili and undoubtedly strange Western ways.

Finally, mention must be made of my friends and family, especially my wife, Adel, who quietly endured my
long absence as a consequence of my involvement with this research project.
iii
ABSTRACT


The focus of this ethnographic study is to examine the sociocultural aspects of water usage amongst the
Chagga of North East Kilimanjaro.

Issues concerning water access involving economic and geographical inequalities, as well as threats to water
access in the form of alcoholism will be considered, and various aspects of water usage practice will be
explored in detail. A discussion of these findings will follow with reference to trachoma control programmes
which are being implemented throughout the region.

The study will conclude with some recommendations for said programmes based on the insights gained
from this research.












































iv
LIST OF FIGURES AND TABLES

Figures

Figure 1: Map of Tanzania
Figure 2: A long history of markets on Kilimanjaro
Figure 3: Christianity - a colonial legacy
Figure 4: Kilimanjaro - tradition and modernity
Figure 5: The long, slow march towards development
Figure 6: A Chagga funeral of a twenty-seven year old man
Figure 7: Typical forest zone scene on Kilimanjaro
Figure 8: Districts of Kilimanjaro
Figure 9: Map of Rombo District showing Mkuu and Urauri
Figure 10: Urauri Village
Figure 11: The road less travelled - a typical Urauri village scene
Figure 12: Sorting out coffee beans domestically prior to selling them at the
Village Office
Figure 13: Palisades keep intruders out
Figure 14: House valuables can be locked
Figure 15: Village priest performing mass
Figure 16: Detail from interior of house
Figure 17: Illegal logging on the slopes of Mt. Kilimanjaro
Figure 18: Dry river bed
Figure 19: Mkuyu, a traditional tree
Figure 20: GPS identification of water source and indicative distance to Urauri village
Figure 21 - 23: Various photos of rain harvesting, iron roofing, gutters and storage containers
Figure 24: Traditional rainmaker
Figure 25: Comparison of water usage between piped and un-piped households in Tanzania.
Figure 26: The top end of the village socio-economic scale
Figure 27: The bottom end of the village socio-economic scale
Figure 28: Filling up at a standpipe
Figure 29: Mother and son returning from water collecting in Kenya, exhausted
Figure 30: Village childrens drawings of themselves carrying water
Figure 31: Collecting water by bicycle
Figure 32: Hand-driven cart
Figure 33: Collecting Fodder
Figure 34: Threshing beans
Figure 35: Water collecting
Figure 36: Queues are usually lengthy
Figure 37: Masai warrior
Figure 38: Drinking to celebrate: a village wedding
Figure 39: Drowning sorrow: a Chagga funeral
Figure 40: Males drinking at a bar
Figure 41: Chagga fathers drinking habits
Figure 42: Chagga mothers drinking habits
Figure 43: Gendered drinking not without its risks
Figure 44: The consumption of alcohol can occur at the expense of other activities
Figure 45: Water stored inside the house
Figure 46: Children looking after children
Figure 47: Too large a burden?
Figure 48: Mother chops wood alone
Figure 49: Water: essential not just for human survival
Figure 50: Boy carrying banana leaves for livestock




v
Figure 51: Domestic busa making
Figure 52 53: Preparing and washing plantains for cooking
Figure 54: Chai for two
Figure 55: Percentage of water consumed for cooking and drinking in relation to total water
available in a household with piped/unpiped water
Figure 56 57: Children wearing soiled, ragged clothing
Figure 58: Schematic of Chagga domains of hygiene
Figure 59: Village women wearing clothes for the road
Figure 60 - 61: Public event = Public display of cleanliness
Figure 62: Dirty clothes awaiting their destiny
Figure 63: Washed clothes. Note the multiple washing lines
Figure 64: Washing the patio
Figure 65: Sweeping the compound
Figure 66: Watering plants
Figure 67: In some households not a temporary state
Figure 68: Group face-washing
Figure 69: Mother teaching her children the importance of hygienic practices
Figure 70: Foot scrub
Figure 71: Eldest daughter washes a younger sibling
Figure 72: Impact of water supply on washing and bathing
Figure 73: Jiggas in a hand
Figure 74: Jiggas in a foot
Figure 75: Ringworm
Figure 76: Neglected child plays in the dirt
Figure 77: Flies feeding off the tear duct of a child

Tables

Table 1: Estimates of global morbidity and mortality of water-related diseases
Table 2: Top ranking villages for Trachoma prevalence in 2002
Table 3: Summary of weekly research activities in the field
Table 4: Indicative list of local prices
Table 5: Highest trachoma prevalence in Rombo district
Table 6: Tanzanian population figure and water supply coverage
Table 7: Thirty year comparison water access and usage across range of indices for
households lacking piped water facilities
Table 8: Standing taps in Urauri
Table 9: Allocation schedules for standing taps
Table 10: Chagga definitions of water typologies




6
CONTENTS



Acknowledgments p. i

Abstract p. ii

List of figures and tables p. iii - iv

Contents p. v

Chapter 1 Introduction p. 1

Chapter 2 - Methodology p. 14

Chapter 3 Ethnographic findings p. 34

Chapter 4 Discussion p. 114

Chapter 5 Conclusion and recommendations p. 122

Bibliography

Appendices

Appendix 1: Global Water Issues
Appendix 2: Water Availability In Africa
Appendix 3: WHO Trachoma Grading Card
Appendix 4: ITI Anti-Trachoma Programme Poster
Appendix 5: Information Sheet For Village
Appendix 6: Research Schedules
Appendix 7: Standard Questions for Household Interviews
Appendix 8: Focus Group Discussions Indicative Questions
Appendix 9: Tanzanian Political Organisation and Political Offices
Appendix 10: Political Organisation of Rombo District
Appendix 11: Threats to Mt. Kilimanjaro Forests
Appendix 12: Kiliwater Contract
Appendix 13: Glossary
Appendix 14: Water conflict chronology
7
INTRODUCTION

Global freshwater: a 21
st
century issue

>1% - A limited Supply

Although our planet is mainly composed of water - 1,386,000,000 km
3
- only a small fraction of
that water (2.5%) is freshwater. Unfortunately, in terms of human needs, most of that
freshwater nearly 70% - is inaccessible, locked up in the form of ice so that out of the total
freshwater available globally, only a small fraction, some 0.7825% exists in a form accessible and
usable to humans (see Appendix 1).

Climate change

Whilst the debate still rages as to the legitimacy of global warming as a scientifically proven
phenomenon, many people are noting through personal observation and experience that weather
systems are changing both locally and globally, and that with these changes have come
significant changes to established patterns of precipitation
1
. Thus as the 21
st
century unfolds, we
are likely to see some traditionally wet areas experiencing a decline in rainfall patterns to
become significantly drier, whereas dry areas may experience the opposite, turning into areas
with high precipitation, as the weather becomes more extreme in nature.

Although greater precipitation might appear to be a solution for countries with expanding
populations and insufficient water to meet their various needs, the additional precipitation may
often take the form of sudden massive downpours which may occur in areas lacking the
infrastructure to collect and channel the water to human populations residing far away. This
water would therefore remain unavailable for human usage. Alternatively, high rates of
precipitation may occur within areas of human population, but in amounts simply too large for
the current water infrastructure to absorb, with potentially catastrophic effects. One only has to
think of the devastation caused in 2002 when more than a months worth of rain fell on Prague in
just a few days, overwhelming the grounds capacity to naturally absorb the rainfall which
resulted in local rivers becoming over-burdened, bursting their banks and unleashing terrible
damage to a historical capital.

Environmental degradation and its impact on available freshwater

Some argue that recent climactic changes are part of a large natural cycle, but it is more than
likely that the accelerated human economic activity that has occurred across the planet during
the past 100 years and the attendant environmental destruction this has unleashed is a major
cause for what is being experienced today. Natural forests have been steadily encroached upon
and depleted, without necessarily understanding their importance in larger hydrological cycles or
in stabilizing weather systems. Deforestation also acts to change patterns of precipitation, may
cause soil erosion and has been implicated as a major trigger for the onset of desertification.

Economic development has also meant the destruction and loss of nearly half of all planetary
wetlands which function to filter water and serve as buffers against flooding. Their loss therefore
means that the sea threatens to encroach upon coastal freshwater aquifers, contaminating them
and resulting in the permanent loss of a valuable resource. However, should sea levels continue
to rise according to current projections, then contamination and loss of coastal aquifers may be a
certainty rather than a likelihood.


1
Source: www.pacinst.org/naw.htm
8
Increased human demands

The main engine for economic activity and environmental degradation has been driven by an
unparalleled explosion in the size of the global population over the past one hundred years. But
more mouths to feed translates into more and more increasingly unsustainable demands being
made on a finite resource
2
. For example, while the worlds population expanded three-fold during
the last century, the demand on freshwater increased six-fold.
3
The world population is therefore
not just expanding: it is getting thirstier.

The combination of increasing populations and economic development translates into increased
agricultural and industrial demand for water which constitutes 70% of all human demand for
freshwater
4
. The consequence can be seen in the status of great rivers like the Colorado, the Nile
and the Ganges which, dammed, diverted and over-subscribed, are now slowly dying. For
example, in 1997 the waters of the once mighty Yellow River failed to empty into the sea for a
combined total of seven and a half months (220 days).
5


Once seemingly inexhaustible freshwater sources like the Aral Sea are also showing marked
changes; images from NASA and European Space Agency (ESA) satellite pictures taken in 1985
and 2003 respectively reveal that Aral Sea has shrunk dramatically over this period,
6
with what
were thriving fishing villages located by the edge of lapping waters now kilometers away from
the new shoreline, the result of the Arals waters being increasingly diverted for the purposes of
cotton production.
7


Other important sources of freshwater, aquifers bodies of water-bearing rock are also being
increasingly stripped of their valuable contents. Deep aquifers play an important role in the
hydrological cycle since they act in dry periods to slowly release water into rivers, lakes and
wetlands, but absorbing water to prevent flooding during rainy periods of rainfall.
8


Aquifers are an important source of freshwater, with perhaps a quarter of the world getting its
supplies from these sources, which provide perhaps as much as 10% of global freshwater
consumption annually.
9
But the over-abstraction of ground water is increasingly common which
has resulted in many water tables falling rapidly, in some places dropping a meter a year, as
water is taken at unsustainable rates - up to ten times faster than it can be replenished.
10
The
danger of this for the environment and for freshwater supplies in particular is that if too much
water is abstracted at too fast a rate, this can mean that rivers gradually dry out and become
exhausted.

Inequalities in supply and access

Whilst global demand for freshwater has increased rapidly, supply, access and usage patterns are
unevenly distributed. Urban population centres often take water from rural areas
11
so that big
differences may emerge in access patterns within a country. However, water usage can vary just
as significantly between different countries: for example, the average person in the United States
has been calculated to use between 2-300 liters a day, whilst the average person in Somalia uses

2
Source: http://news.bbc.co.uk/go/pr/fr/ - /2/hi/science/nature/2943946.htm
3
Source: http://watervision.cdinet.com/execsumm.htm
4
Source: http://news.bbc.co.uk/go/pr/fr/ - /2/hi/science/nature/2943946.htm
5
Source: http://www.guardian.co.uk/worldsummit2002/earth/story/0,12342,77661,00.html
6
Source: http://news.bbc.co.uk/2/hi/science/nature/3107915.stm
7
Source: http://news.bbc.co.uk/2/hi/science/nature/3107915.stm
8
Source: http://www.guardian.co.uk/worldsummit2002/earth/story/0,12342,77661,00.html
9
Source: http://watervision.cdinet.com/execsumm.htm
10
Source: http://www.guardian.co.uk/worldsummit2002/earth/story/0,12342,77661,00.html
11
Source: ibid
9
a mere 8.9 liters of water per day less than 5% of an Americans usage.
12
So whilst some parts
of the world possess seemingly abundant supplies of water, and can use them somewhat
profligately, for many this is not the case, with 40% of worlds population
13
living with
inadequate water availability and some 1.2bn people lacking access to clean drinking water.

Different levels of water availability which indicate cause for concern are known as: water
vulnerability, water scarcity and water stress: The shift from possessing sufficient water without
any cause for concern to beginning to experience occasional shortages is known as water
vulnerability, defined as access to 2,500 1,700m
3
(6.85 4.65 m3/day)
-
per person per year;
water stress occurs between 1,700 and 1,000m
3
per year (4.65 2.74 m3/day)

and water
scarcity occurs when water availability falls below 1000 m
3
per year (> 2.74 m3/day)
14
.

A resource worth fighting for

Throughout much of human history, water resources have not only been a prime target in human
conflicts but also an occasional source of conflict in their own right with nation states prepared to
go to war to secure access to water supplies (Appendix 14). For example, Ariel Sharon has stated
that the real date of the Six-Day War was not the 5
th
of June 1967, as is officially claimed, but
some two and a half years earlier when Israel decided to act against the diversion of the Jordan
River.
15
But rather than being a thing of the past, conflict over water is an ever-present danger,
as disagreements mount amongst neighbouring countries over how shared rivers should be used,
the Nile, Euphrates and the Mekong being prime examples of current potential flashpoints.

However, more worryingly, major shifts in recent US military thinking
16
whose stated aim is full-
spectrum dominance
17
of traditional (land, sea and air), as well as new strategic domains (outer
space, cyberspace), would seem to indicate that control of water resources may now be
considered a legitimate, although somewhat unconventional battlespace
18
, one, however, which
other countries may now start to formally include in their national security strategies. Higher
water demand yet decreased availability may mean that in the near future there is a high
probability that blood will flow where water does not.

Creating Solutions

Some analysts believe that water availability is not necessarily the key issue, nor that the world is
running out of water
19
; rather, it is the un-sustainability of current usage patterns, coupled with
uneven water distribution, inadequate delivery infrastructure and poor water management which
results in wide-scale denial of access for many around the world, a major factor in creating and
sustaining poverty as well as compromising the well-being of millions
20
.

Whilst there are calls for the creation of specific international water-conflict resolution
mechanisms, and greater cooperation at wider basin-level over shared international aquifers to
balance the dominance of any single nation state in terms of water access and defuse potential
future conflicts
21
, many believe that the best way to prevent such conflicts from arising as well as

12
Source: www.pacinst.org/water_facts.htm
13
Source: http://society/guardian.co/uk/society/guardian/story/0,7843,916631,00.html
14
UNEP Figures
15
Source: http://news.bbc.co.uk/go/pr/fr/ - /2/hi/middle_east/2949768.stm
16
While the terminology and concepts may appear novel, they were perhaps first conceived and articulated by Sun Tzu in
his work, The Art of War. Of note are the following observations: Chapter 1 Laying Plans Observation # 16: All warfare
is based on deception. Chapter 3, Attack by Stratagem Observation # 2 Break the enemys resistance without fighting.
17
Source: http://russfound.org/Enet/FSD.htm
18
US Department of Defense Term. Source: www.dodccrp.org/ucNature.htm
19
Source: http://www/guardian.co.uk/analysis/story/0,3604,903045,00.html
20
Source: http://www.gci.ch/GreenCrossPrograms/waterres/wwf_03/wfpopening.html
21
Source: ibid
10
ensuring adequate water access and supply for all is through global market-driven water
management.

Arguments for the privatisation of global water assets

This argument is predicated largely on the fact that governments have shown, particularly in
non-industrialised, post-colonial countries, that they have not been able to adequately develop
and maintain national water infrastructures. Part of the reason for this has been inadequate
funding (often because such countries have small GDPs, have huge international debts to service,
and may not have adequate financial control systems in place, allowing the possibility of
embezzlement and corruption to take place). This has meant decaying, inefficient water systems
with poor connectivity or supply for many citizens. Often this inadequate service provision is
tolerated by end-users because of its perceived cost either free or very inexpensive. However,
such low costs obscure the fact that national water systems are often subsidized by governments
which draw the subsidy from the taxes levied on the general population.

Operating, developing and managing a water system costs money and that money has to come
from somewhere, either through indirect taxation or by tariffs. Supporters of privatisation argue
that many governments have reached a point where they are either are no longer able to fiscally
provide subsidies, given the precarious states of their national accounts, or that if they do, they
do so at the expense of investment in the water infrastructure, which will inevitably decline and
cause even more distress for end-users. Either way, government-run water infrastructures
increasingly appear unviable, unable to provide extensive connectivity, adequate service or the
ability to update the water systems for future requirements. The only other option, therefore, is
for privatisation to provide the capital, knowledge and expertise necessary to develop, operate
and maintain large water systems efficiently and effectively, enlarge the number of people
connected to national and local water grid, as well as make large scale infrastructural
investments.

In addition, privatisation can help conserve water resources in two ways: Firstly, decaying
infrastructure can be repaired or replaced which will mean an end to the major source of water
loss within a water system: leaky pipes. Secondly, current household water usage practices and
attitudes often change, becoming less wasteful as people learn to value and be more responsible
for a resource they have to pay for.
22
The key message is that water infrastructures can be run
efficiently and effectively so that long-term water management becomes a reality rather than a
dream. However, this needs to be purchased rather than hoped for, and end-users must accept
that water service provision can not be free.

Already various privatisation efforts have been rolled out often to much local acclaim such as the
Bolivian city of El Alto jointly run by Suez and Aguas de Illimani,
23
which shows that public-
private partnerships can be made to work successfully. Furthermore, new management and
payment structures have meant that millions of poor people across the world now have access to
basic water services, whereas this was perhaps not always the case under previous government-
run water systems.






22
Source: http://news.bbc.co.uk/go/pr/fr/ - /2/hi/science/nature/2943946.htm
23
Source: http://society/guardian.co/uk/society/guardian/story/0,7843,916631,00.html

11
The opposing view

However, these arguments are not universally accepted. Many object on ideological grounds to
the commodification of water as an economic good (Marlow 2002)
24
, a valuable natural resource
that can be sold for profit, arguing that the United Nations recognition of the right to water
(November 2002) constitutes a basic human right, not something that can be simply treated as
an economic good.

Global organisations that ostensibly sound democratic such as the World Water Council (WWC)
do represent a variety of interests, but are dominated by the World Bank - whose policies are
widely considered an embodiment of the Washington Consensus (Huntingdon 1996: 184) - and
transnational water companies, may explain why the WWCs recently published vision for future
water provision and management is that of water as a need that can be best served by the
private sector.
25


Arguments for the privatisation of water that are drawn on economic grounds alone are
considered by privatisation-skeptics to mask a projection of Western capitalist ideology and
power (Ikenberry 2002: 46). At best, comprehensive privatisation could result in a cartelisation of
the worlds water delivery and wastewater systems. At worst, ownership of national water assets
by (Western) private companies could potentially form an integral part of covert efforts to
achieve full-spectrum dominance on a global scale by establishing economic control over key
national infrastructural elements
26
.

There are also fears that privatisation will bring more disadvantages to the people it is meant to
serve than benefits. Firstly, there is the question of whether privatisation benefits its customers:
Multinational companies know that third world governments are cash-strapped and are able to
offer low tenders and extract maximally advantageous contractual conditions. Furthermore,
transnational lending agencies such as the IMF can suggest that markets are deregulated which
may mean that permitting equitable access to water for the poor and improving long-term water
use efficiency and productivity may become a secondary concern for private water companies in
comparison to the primary objective of profitability. In addition, such contracts may allow private
water companies to cherry-pick the most lucrative services to provide whilst either failing on
agreed obligations to provide adequate water provision and connectivity to new users or rolling
out a token service to rural communities which may, like the UKs experience of tele-
communications and postal service privatisation, eventually lead to systematic rationalisation
such as the withdrawal of pay phones and local post offices in unprofitable rural areas in the UK.
Unfortunately, privatisation often brings with it a lack of accountability as well as a lack of
transparency about how contracts are selected and how policies will be developed and
implemented. In short, privatisation is often characterised by the existence of a closed door
culture, where affected parties may be shut out of the decision-making process completely.

Profits generated through privatisation efforts rarely remain in the country they were generated
in, and are often sent back to the home country where the private water company is based, or
placed in a tax haven. However, these profits are sometimes generated through huge rate

24
Commodified water has rapidly acquired such a market value that some commentators refer to it as blue gold

25
Source: http://www/guardian.co.uk/analysis/story/0,3604,903045,00.html

26
The nature of war has changed over the last thirty years from a conception of balanced forces engaging each other in
physical battle with established rules of engagement to a conception where any domain is a legitimate arena to be
contested, be it economic, military or informational. The wars of the future are not necessarily going to be fought on a
distant battlefield; they are being offensively fought right now on a wide variety of fronts in ways we would scarcely
recognise as constituting warfare if understood using traditional conceptions, but which make perfect sense if viewed
through the conceptual framework afforded by the doctrine of Full-Spectrum Dominance.
12
increases. In Ghana, for example, water bills doubled when Ghanaian national water assets were
privatized, hitting the poor hard (Pilger 2002: 121). Often these increases can mean that water
purchased water for survival only may account for a significant proportion of household income
within a marginal household economy. This, as Sandy Cairncross noted in his study in urban
Sudan amongst people living in the squatter camps, may unfortunately mean stark choices have
to be made: enough water to survive or adequate nutritional intake for every household member
(Cairncross 1989: 183). With privatized water representing up to 50% of a households income,
the real price for water was often paid by children with their lives (ibid: 188).

Whilst those in favour of privatisation often cite the positive water privatisation experience of the
Bolivian city of El Altos, not all privatisation experiences match this profile in terms of customer
satisfaction. As part of fiscal controls and demands made by the World Bank in return for
renewed loans facilities
27
, the running of the water infrastructure of Cochabamba, another city in
Bolivia, was taken over by a private international water consortium, San Francisco-based Bechtel
Inc. and London-based International Waters Limited who sought an average 16% annual return
on their investment, immediately hiking up water rates upon commencing the contract (before
making any improvements in water quality or delivery). These increases represented of the
average adult monthly wage
28
and this triggered a general strike in protest, city-wide riots with
50 injured and 1 death
29
, a declaration of a state of emergency and imposition of martial law,
and the hurried rescinding of the contract by the Bolivian government which found itself
overwhelmed by the level of public anger and hostility towards water privatisation.

Whilst critics are not blind to the shortcomings of particular governments and their ability to
deliver adequate water services, they argue that privatisation is not the only way to overcome
these limitations. Subsidies could be phased out replaced with staggered increases which would
give end-users time to adjust to increasing prices. In addition, tariffs could be structured
according to usage: large industrial users, for example, paying a much higher tariff given the
volume of water consumed than small householders. Such measures and differential tariffs could
also help to cross-subsidise the cost of water, would encourage responsibility for the water
consumed and help conserve the resource.

Whilst overcoming various structural problems may be a difficult and arduous task, many,
including leading water analysts, believe that it is a task that must be undertaken no matter
how unrealistic it may initially seem. Whilst opponents of water privatisation may differ on
various issues, the central belief they all share is that the management and delivery of freshwater
should not exclusively be in the hands of private industry. According to Dr. Peter Gleick, President
of the Pacific Institute and author of the annual report, The Worlds Water, this is because,
Water is far too important to human health and the health of our natural world to be placed
entirely in the private sector. Market forces alone cannot fulfill the varied roles that water plays
socially and ecologically; privatisation ends up fragmenting the integrated management of global
freshwater as a resource, obscures larger ecological and long-term concerns, channels water
towards profitable domains only, may impose additional burdens on the poor, thereby possibly
aggravating their marginality. What is needed, it is argued, is for control of water assets and
infrastructure by nation states as well as local community-based and transnational NGOs
30
to help
plan for, articulate and sustainably manage the responsibility to deliver water as a human right.






27
Source: http://www.americas.org/News/Features/200004_Bolivia_Water/Shultz_and_Kruse.htm#Bolivia%20Water%20War%201
28
Src: http://www.miami.com/mld/miamiherald/news/world/americas/4271738.htm/4271738.htm?template=contentModules/printstory.jsp
29
17 year old Victor Hugo Diaz killed by a live bullet
30
The Green Cross website lists various recommendations for NGO involvement: http://www.gci.ch
13


The Future of Freshwater

We have already noted that population increases have resulted in increased demand for water, but given
climate changes, projected population growth figures, inadequate infrastructures, increased consumption -
doubling every 20 years
31
- new sources harder to find and declining availability of accessible freshwater,
the future of freshwater access looks bleak for many.

Whilst greater numbers are being connected to national water infrastructures across the world, this
proportion of connected vs. unconnected is increasing because the rate of population growth outstrips the
rate of connections made.
32
Furthermore, it is calculated that as much as 70% of all accessible freshwater
will be used by 2025
33
, leaving half the worlds population, mainly in Africa, Middle East and South Asia to
face severe water shortages, making freshwater access the most pressing environmental issue of this
century with scores of countries making the transition from water vulnerability to water stressed within
decades (Appendix 1 and 2). No region will be spared from the impact of this crisis, laments Koichoro
Matsuura, Director General of UNESCO.
34
Declining water availability because of climate change or human
usage will result in compromised food production
35
and security, under nutrition and starvation and
widespread misery and ill-health (Rosegint 2002).

The lack of (clean) freshwater already takes a heavy toll on the worlds poor in the form of various water-
related diseases which are caused by either lack of access, contaminated sources or insufficient amounts for
daily requirements. A study conducted by Pruss estimated water, sanitation and hygiene factors accounted
for 5.7% of total disease burden, and 4% of all deaths worldwide (Pruss 2002). The following table presents
a snapshot of the global prevalence of water-related diseases in the early 1990s.

Disease Morbidity (episodes/year or people infected Mortality (deaths/year)
Dengue Fever 1,750,000,000 20,000
Intestinal
Helminthes
1,500,000,000 (people infected) 100,000
Diarrheal Diseases 1,000,000,000 3,300,000
Malaria 400,000,000 1,500,000
Schistosomiasis 200,000,000 (people infected) 200,000
Trachoma 150,000,000 (active cases) ---
Bancroftian
Filariasis
72,800,000 (people infected) ---
Onchocerciasis 17,700,000 (people infected) ---

Table 1: Estimates of global morbidity and mortality of water-related diseases. Source: WHO 1995

Based on these figures, every six seconds someone somewhere dies of a water-related disease, a figure
that is likely to increase as more and more countries experience water stress during the next twenty years.
Whilst arguments rage about the best way to tackle the impending water-crunch, the stark reality is that
worsening inequalities in water access will continue to have a considerable impact on health and sickness.
However, the expenditure of considerable biomedical effort in trying to control and reduce the prevalence of
fatal and non-fatal water-related diseases may help to reduce mortality rates and alleviate, if not prevent,
the chronic suffering and pain experienced by millions as the result of infections and their attendant
sequelae.



31
Source: http://www.guardian.co.uk/worldsummit2002/earth/story/0,12342,77661,00.html
32
Source: http://news.bbc.co.uk/go/pr/fr/ - /2/hi/science/nature/2943946.stm
33
Source: http://www.guardian.co.uk/business/story/0,3604,778688,00.html
34
Source: http://news.bbc.co.uk/go/pr/fr/ - /2/hi/talking_point/2963656.stm
35
Between 1981 and 1983 food production in the African countries most seriously affected by drought dropped by 15%
(Abel Smith 1986: 202)

14
Trachoma: A major health problem

One of the most widespread and distressing non-fatal water-related diseases is trachoma which
prevalent in 48 countries with 150,000,00 people thought infected. It is found within dry, dusty
places lacking in adequate water supply, and is endemic in Africa, Eastern Mediterranean Region,
Australia and parts of SE Asia (Munoz, B & West, S. 1997). Trachoma is also one of the leading
causes of blindness worldwide, with six million people blinded some 10-15% of recorded cases
of blindness globally, and the leading cause of preventable blindness (Thylefors 1990),

Understanding trachoma as a disease process

Young children (typically between 2 and 5 years of age) constitute the main reservoir of the
active disease and infection which generally diminishes as an active disease with age. Infections
from Chlamydia Trachomatis occur within the ocular region and can repeatedly either occur
asymptomatically or as visibly inflammatory form. The trajectory of trachoma as a disease
process is a slow, chronic one with easily identifiable elements and stages (see Appendix 3).
Active infection takes two forms: mild infection trachomatous infection follicular form (TF) is
identifiable as the presence of five or more follicles with are 0.5mmm each which are found
behind the top eyelid (the upper tarsal conjunctiva). Acute or intense infection is known as
trachomatous infection intense form (TI) which is identifiable as visible inflammatory thickening
of the tarsal conjunctiva. As children mature and become teenagers and young adults, they may
experience re-infection many times which can causes the conjunctiva to become scarred
trachomatous scarring (TS), prolonged scarring may result in eyelids slowly turning inwards
which can cause the eyelashes to rub against the eye surface, a stage known as trachomatous
trichiasis (TT) which often causes painful corneal abrasions. Over a prolonged period such
constant abrasions may result in corneal opacity and blindness.

There are, therefore, four clearly identifiable stages within the trachoma disease process:

Infection
Scarring
Corneal abrasion
Loss of vision

Tackling trachoma

Trachoma is a preventable disease, but the later it is tackled within the disease process, the
harder and less practical it becomes to intervene successfully. Attempting to reclaim vision lost
because of corneal opacity is the most unrealistic juncture for medical intervention; tackling
trichiasis is more realistic since it can be treated through relatively simple surgical techniques.
The optimum juncture, however, to tackle trachoma is in its initial stages, attempting to deny it a
foothold amongst the young. To do that, the mechanisms for infection need to be understood
and appropriate interventions designed so that the possible infection vectors are reduced and
controlled, diminishing the likelihood of the emergence of later, more serious stages of the
disease.
15
Trachoma transmission vectors, risk factors and distribution

Infection spread from infected eyes to non-infected eyes can be transmitted by a variety of
vectors:

Fingers
Dirty clothes
Flies

Risk factors for its prevalence include:

Poor access to water supplies
Lack of facial cleanliness
Lack of household latrine
High fly density within household

Distribution

More often than not found in poor rural communities with sub-optimal standards of
hygiene.

Also found clustered within villages and in households.

Associated with poverty even within a single community, trachoma is more commonly
found in poorer socioeconomic households.

Children particularly vulnerable.

Adult women are three times more likely to suffer from Trichiasis the lashes-in-the-eye
stage than adult men making them more vulnerable - reason being greater and
prolonged intimate contact as primary caretakers (Congdon 1003: 386).

Devising a coherent trachoma prevention and control strategy

The most comprehensive strategy which has emerged from the body of trachoma research so far
conducted is known by its acronym, SAFE. This is a four element programme designed to combat
and control the transmission and development of blinding trachoma, which has been adopted by
the World Health Organisation as an integral part of its Global Elimination of Trachoma by 2020
campaign 2020 (GET 2020) (WHO 1997). SAFE stands for:

S = Surgery for Trichiasis
A = Antibiotics to reduce the prevalence of active disease reduce the
pool/reservoir of active disease
F = Face washing to reduce transmission of trachoma
E = Environmental (control) to reduce transmission of trachoma

***






16
Whilst the evidence base for the success for the use of surgery and of antibiotics is strong, there
are questions as to whether the usage of antibiotics is truly sustainable in controlling trachoma,
principally because of issues of possible resistance emerging and the long-term costs involved.
Furthermore, some researchers point out that the prevalence of trachoma as an active disease
fell in Europe and elsewhere before the development of antibiotics, which indicates that the
control of trachoma could be successfully achieved through other means, principally through
designing intervention strategies based on environmental improvements and behavioural factors
such as hygiene practices (Lynch 1994: 513)

These form the last two components of the SAFE strategy which have been designed to interrupt
trachoma transmission by focusing on face-washing to reduce transmission of trachoma, and
environmental control efforts to reduce transmission of trachoma. However, the evidence base
for these elements is weaker than for the first two elements, and whilst a moderate and positive
association between the lack of a clean face and trachoma prevalence has been demonstrated,
face-washing behaviour has not been shown to diminish the prevalence of active trachoma, and
only minimally diminish the prevalence of severe, active trachoma
36


Bearing this evaluation in mind, the findings of other studies provide grounds for quiet optimism:
satisfactory personal hygiene was reported to have a positive impact by reducing trachoma
prevalence by 48% (Mathur & Ramewshwar 1970), and facial cleanliness was reported by 5
studies to be inversely associated with the prevalence of trachoma in children: (Taylor 1989;
West 1989 ; West 1991; West 1995; West 1996). However, these and other studies have been
criticized for their methodological flaws, making it difficult to assert with confidence that facial
cleanliness programmes act to substantially reduce the prevalence of trachoma in communities
(Bailey 2001: 234).

Despite these reservations, many researchers still believe that water supply and usage patterns
need to be explored in depth to verify the validity of face-washing as an intervention strategy,
pointing out that face-washing behaviour is still widely considered a powerful mechanism to
break the cycle of re-infection and interrupt the transmission of the disease, particularly amongst
children. It is for this reason that, despite a seemingly negative evaluation of prior face-washing
research findings, interest is still high in exploring issues of water access and water usage
patterns and how these relate to trachoma.

The role of Medical Anthropology in exploring issues of water access and
water usage patterns

Methods and insights derived from the discipline of Medical Anthropology constitute a valuable
adjunct to biomedical knowledge and epidemiological understandings since they can help to
provide a detailed understanding of local conceptions and processes which were sometimes
previously overlooked in the design of health-promotion interventions (Marx 1989: 26), but
which, however, were critical to the success of health education programmes which attempted to
introduce new hygiene behaviours and water usage practices within local communities (McCauley
1990: 1233; Sutter 1983: 1813).

Since the barriers that restrict the acceptance of trachoma prevention measures are often
cultural, there is a need for medical anthropologists to conduct qualitative research which
provides insight into the conceptual framework which underpins local conceptions of what is and
isnt important in their lives. For example, S.D. Lane and others conducted research into the

36
Source: Kuper, H. et al. A critical review of the SAFE strategy for the prevention of blinding trachoma, The Lancet, Vol.
3 # 6, June 1, 2003.

17
perception of eye disease and disability within an Egyptian village and found that the villagers
lacked any conception of a) the causative factors of trachoma and b) did not see the child form
as being related in one continuum to the adult form (Lane et al 1993). Here, trachoma
interventions could be focused on education of those factors and attempting to impart the notion
that prevention in childhood can prevent visual impairment or blindness at a later stage.

Another under-explored area in the formulation of trachoma prevention programmes has been
the role of household and gender dynamics. In her study of the Gogo people of Tanzania,
McCauley found that whilst women were cognizant of the importance of face washing and were
responsible for the collection of water, they were not responsible the decision as to how water
was allocated and used. That was a male decision (McCauley 1992). Here, public health
education about face washing is only partially successful unless it realises the importance of
discovering who is responsible for decisions about water usage and targets its message
accordingly.

Anthropological data, therefore, can help biomedical professionals design intervention strategies
which are based on local knowledge and understandings of possible difficulties. It is to this end
that this M.Sc. study has been commissioned
37
by the Department for Infectious and Tropical
Diseases at the London School of Hygiene and Tropical Medicine (LSHTM) which is conducting
evaluative studies
38
on behalf of the International Trachoma Initiative (ITI) who are funding
research and programs into the SAFE strategy for Trachoma control. This study is also being
conducted in association with the Kilimanjaro Centre for Community Ophthalmology (KCCO) and
the University or Tumaini, Moshi, Kilimanjaro.

***

This research builds upon extensive work on trachoma. Early studies focused on charting the
extent of the trachoma problem, identifying various transmission vectors and identifying the most
effective strategies to limit and control the disease. However, whilst these aspects are well
understood by the biomedical community, problems in implementing anti-trachoma strategies
indicate a need for further research on the very communities which suffer the most from the
disease yet which are least enthusiastic in adopting the necessary health measures. Within one
population, that of the Chagga people of the Rombo region in Tanzania who suffer from a
moderate prevalence of trachoma, there exists a reluctance to engage with various aspects of
the SAFE strategy.

With changing and declining rainfall patterns, water - like in many other parts of Africa - has
become a scarce resource within the Rombo district of Kilimanjaro, and its supply increasingly
aleatoric: piped water is limited to certain locations and sporadically for short periods (Grove
1993). When available in this form, demand can be fierce and waiting periods extensive. Aside
from piped sources, water can also be found at various other locations around Mount Kilimanjaro
but locating it is an uncertain business and often involves long walks, sometimes crossing the
border into Kenya.

Ethnographic studies of the Chagga are somewhat thin on the ground: Raums classic Chaga
Childhood, whilst detailed and informative, describes a world of tradition that belongs more in
the past than in the present (Raum 1940). In contrast, the most recent studies focus on
contemporary concerns and issues rather than attempting to be encyclopedic in scope.

37
A total of 2,250 was paid for research costs.

38
The Trachoma Initiative in Monitoring and Evaluation (TIME) is designed to evaluate community perceptions of
trachoma and the SAFE strategy, as well as identifying reasons for non-compliance with the SAFE strategy.
18
M. Howards study chronicles the impact of neo-colonialism and the increasing social stratification
of Chagga society, and its consequences for child malnutrition and mortality (Howard 1997).
Philip Setels work on AIDS traces the complexity of sexual dynamics and beliefs on the mountain
(Setel 1996,1999), and Amy Stambachs work examines issues of gender and education within
Kilimanjaro (Stambach 2000).

However, apart from the classic Drawers of Water and its follow-up study (White 1972;
Thompson 2001) which examined water usage patterns within the East African countries of
Kenya, Tanzania and Uganda, there has been little published which provides quality information
about water access and usage within Kilimanjaro or amongst the Chagga in detail. This study
therefore represents an attempt to gather ethnographic data which will be of use to public health
professionals wishing to understand the milieu into which they seek to introduce the SAFE
strategy as well as researchers interested in water as an object of study, or the Chagga as a
people.

Local beliefs and practices can then be taken into account when implementing the SAFE strategy
within the region, optimising the probability of maximum uptake and long-term participation.
However, it is possible that, as we have already seen that since more and more countries will be
experiencing water scarcity and water stress in the immediate future, that the ethnographic
findings of this ethnographic study inasmuch as they relate to trachoma prevention may
applicable in other contexts elsewhere, and the LSHTM have therefore requested that the
research findings be made more generally available, and that the study be rewritten as an article
and submitted to an appropriate academic journal for publication.

Study Objectives

As an attempt to improve access to water, new water resources are being developed within the
Rombo district which are hoped will be more accessible and dependable, and provide more water
for householders in absolute terms. However, the possible effect this might have on hygiene
behaviour is an unknown. Will it result in increased facewashing, no change or decreased
levels? What implications are there for increased water supply for the reduction of trachoma?

In order to answer these questions, I prepared a wide range of questions to explore prior to
leaving for the field which I hoped would give me the best perspective to understand the various
dynamics involved in accessing and using water amongst the Chagga study population.

Indicative questions:

Who is responsible for collecting water? Are they male, female, young, old, single, married?
Where do they collect water from?
Are there sources where water is available but not used? How often is water collected?
Under what conditions is water collected and how frequently is water collected?
How far do individuals have to travel to acquire water? How much can they carry?
What is the water used for, and in what amounts?
Do any activities receive priority with regards to water? Is it cooking, or drinking, or
irrigation, or brewing, or hygiene? Or perhaps some other activity?
Are decisions about water usage made at a village level, a household level, or an individual
level?
What factors are involved in making decisions about water usage?
What might the Chagga do with more water? Would their priorities change? Would more
water availability encourage more face-washing?
How might extra water impact upon self-hygiene practices?
19
Are health education messages about the importance of water in prevention of trachoma
understood by the Chagga audience?
If understood, how much water is estimated to be necessary for effective face-washing?
Will putting in taps or water wells change water usage behaviour, especially with regards to
personal hygiene and face-washing of children?

It is hoped that such research will constitute a modest contribution to anthropological knowledge
by providing a detailed picture of the various dynamics involved in the collection and allocation of
water as a resource within a target population. In addition, it is hoped that an understanding of
the possible obstacles in the form of sociocultural constraints to increased face-washing
behaviour - even after the introduction of new water sources - will allow for the formulation of
informed strategies that seek to find solutions that honour local sensibilities.

Structure of the study

Organisation

This study is structured in five parts:

Introduction
Methodology
Ethnographic findings
Discussion
Conclusion

The methodology section following this introduction provides an overview of the various issues
involved at various stages this study, from its preparation to the actual fieldwork itself and the
final creation of an ethnographic text. It is then followed by a chapter outlining the ethnographic
findings which is divided into three sections:

1) An overview of the ethnographic setting which situates the people studied within various
geographic spaces on the mountain including the field site, Urauri Village, as well as
exploring water availability on the mountain

2) An exploration of the various issues in water access, focusing on how economic and
geographic factors can result in inequalities in water access.

3) An exploration of various water usage practices, which will illustrate how water is used
across a range of households, and will draw on case studies to exemplify some notable
patterns.

The ethnographic findings will then be reviewed in light of the trachoma control literature to
examine how water usage patterns may possibly be implicated in the epidemiological distribution
of trachoma. Finally, the study will conclude with some recommendations for trachoma control
programmes which are being implemented throughout Rombo District and further afield.







20
METHODOLOGY

This section presents an overview of the formulation and execution of the research project, and
outlines the key issues during each stage of the project: preparation (understanding the context),
engagement (fieldwork) and lastly reflection (analysis and representation).



Preparation

Literature review

Pre-fieldwork consisted mainly of comprehensive and thorough examination of secondary
literature conducted not only at Brunei and LSHTM, but also at the RAI, LSE and SOAS libraries
as well as by drawing upon the Web of Science and Medline databases. This was essentially to
become familiar with the literature not only about Tanzania as a nation, but also the history and
culture of the Chagga people. In addition, a substantial amount of literature concerning trachoma
and trachoma-related issues was consulted in order to ground the study in relation to previous
trachoma research.

Discussions were also held at the London School of Hygiene and Tropical Medicine (LSHTM) with
a staff researcher, Sarah Pollack, who debriefed me on her Rombo District water survey earlier in
the year, which proved highly informative and provided me with a comprehensive overview of
water availability and sourcing strategies within Rombo district.

Language learning

Attempts to learn Swahili through self-instruction (book and CD) were limited in success. Without
a teacher or the benefit of daily exposure, this was only valuable in as much as it sensitized my
ears to the sounds and rhythm of the language, arming me with some basic phrases to facilitate
introductions and navigate simple social situations. Language issues were to be a constant
frustration in my desire to understand the fullness of my informants sometimes lengthy accounts
which were unfortunately often rendered all too concisely by my field assistant.

Development of a protean research question

Water as a research focus was jointly suggested by Kilimanjaro Centre for Community
Ophthalmology (KCCO) and the LSHTM who were interested in exploring the relationship of
water usage and trachoma to discover whether ethnographic research could assist in the control
and elimination of trachoma.

Whilst it is generally accepted that research questions formulated prior to entering the field rarely
remain unchanged by the reality found there, the formulation process did prove invaluable since
it allowed me to anticipate the issues that I would tackle and how best to tackle them, and
leaving me as prepared as one can be for a new continent, country, culture and language.







21
Engagement

Choice of setting

The choice of research setting was determined by both KCCO and LSHTM and only confirmed
after I reached Mkuu, the administrative capital of Rombo district which is located some seventy
kilometers from Moshi, capital of the Kilimanjaro region. Prior to this, I had only been aware that
the site was going to be based somewhere in Rombo District.

Upon my arrival, Patrick Massae, head of the Rombo Trachoma Research Project (RTRP), a
LSHTM funded organisation, based at Haruma Hospital in Mkuu, informed me that the study site
was going to be Urauri village. Not only had this village been identified by Sarah Polack of the
LSHTM as suffering from water scarcity, but the recent 2002 Rombo District Trachoma Survey
had also revealed that Urauri possessed the highest rate of trachoma infection in the district.
These factors meant that the site was considered ideal to explore how issues of water access and
usage were related to the problem of trachoma.

Village Total
Registered
Total
Examined
TF Percentage
Urauri 260 236 79 33.47%
Leto 309 303 89 29.37%
Ngareni 188 179 52 29.05%

Table 2: Top ranking villages for Trachoma prevalence in 2002 (Source: Rombo District
Trachoma Survey)

While it could have been possible to have been based in Mkuu and conduct research (as some
LSHTM students previously choose to do), this would have involved a daily commute of 80
kilometers, some 2 hours on rough, difficult and dangerous roads, and would have diminished my
ability to observe and develop an informed feel for the game obtained by being continuously
present. Despite my keenness to be based in the village, Patrick expressed doubts about whether
it was going to be possible to find adequate accommodation. I told him I didnt care and could
handle rough conditions, if necessary.

Access to the field

However, prior to sourcing accommodation, we had to first obtain formal permission at different
political levels to be able to conduct the study. The main gatekeeper was the Rombo District
Commissioner who not only willingly granted permission, but dismissed any questions about my
security, and said that it would be fine to reside in the village. Following this, we spent the
remainder of the first week visiting lower officials, including the village chairman, to inform them
of our intentions and ask their permission as a matter of courtesy. We were fortunate in that not
only was permission granted freely, but no formal conditions were imposed other than supplying
the District Commissioner with a summary of the finished research, leaving me free to conduct
our research as I saw fit, free of any official control or surveillance.

Accommodation arrangements

A wealthy village businessman was found who was willing to provide accommodate both myself
and my field assistant, Mosoi Balthazar, offering us a shared single room in his house. Located on
the border between Mhaka and Nessae subvillage, it proved a relatively central and comfortable
base from which to conduct research, as well as an opportunity to live in close proximity with a
22
village family, albeit perhaps - in their affluence - not very representative, living in a concrete
house and possessing both water storage facilities and electricity.

Timescale

The fieldwork took place over seven weeks from the 2
nd
of June to the 23
rd
of July 2003. I felt
that this was a sufficiently long period to establish a basic understanding of the villages water
issues, as well as enjoying continuity in the fieldwork and the opportunity to absorb the rhythms
and concerns of village life. A detailed daily breakdown of field activity is included (Appendix 6),
but this provides a summary:

Week # Activity
Week one Approaching the field, negotiating
gatekeepers
Week two Entering the field proper. Interviewing
village political officers
Week three Exploring local experiences through focus
groups
Week four Household observations begin semi
controlled
Week five Observations of trachoma positive
households
Week six Observations of trachoma negative
households
Week seven More trachoma positive household
observation and focus groups on household
management and parenting
Week eight Final observations - trachoma negative
households. Final interviews. Exit the field

Table 3: Summary of weekly research activities in the field

In total, fifteen focus group discussions, several key informant
39
interviews and ten household
observations were completed during this period.

****

The fieldwork was conducted in mid-winter at the start of the dry season. When I arrived the
rainy season had just concluded and the flora in the village was verdant and lush, making it
difficult to believe judging by the vegetation that there could possibly be a water shortage. Most
harvesting had been concluded; agricultural activity in general had begun to tail off, and
household water recycling efforts associated with the dry season only just beginning to emerge.
However, in the forty odd days I stayed in the village it rained just twice, once heavily
throughout the night and once finely for a couple of hours only, providing some indication of just
how little rain was to fall in the subsequent months.






39
Defined as an informant selected on the basis of possessing special expertise or knowledge and understanding of a
topic which allows insight into a specific/various issues under consideration.

23

Fieldwork facilitation: Field Assistant

Part of the KCCOs conditions for making a field assistant available to me was that there would be
some transfer of anthropological skills, which I thought reasonable and readily agreed to. This
meant that I spent a lot of time continually explaining my motives for asking a particular question
or series of questions to my field assistant. Initially, my questions often struck my field assistant
as somewhat pedantic or perhaps bizarre since they concerned the minutiae of Chagga
common-sense practice. However, with time he began to understand my persistence, why I
would seize upon small inconsistencies in accounts, or why I would contrast and compare
different informants accounts.

A large part of my understanding of the field is directly traceable to the long debriefing sessions
that Mosoi and I would have after a long day observing where we would discuss elements of
what we had seen, and try and mutually elucidate what drove a certain behaviour. Sometimes,
like a dog with a bone, we would mull over a troublesome issue, refusing to let go of it until we
had made some breakthrough in understanding, both of us exchanging ideas and opinions,
sometimes heatedly. In the end, having Mosoi with me was a blessing since not only was he
highly articulate in English and very good with making informants feel comfortable, but he was
also able to share the sense of adventure in solving anthropological problems.

I felt very fortunate in having Mosoi assigned to me, since he was intrigued by the work, was
keen to learn from the experience, was very good with reading and handling people, and was
also very easy to work with. Without him, I doubt I would have obtained as much data as I did,
or had so many insights into Chagga culture. And for him, I think it was an eye-opener to see
how Urauri village operated, apparently in many different ways from his own village, and to
witness a wide range of households and behaviours.

Mosoi also spoke Kichagga, the language of the region. This was a bonus since he was able to
switch from Swahili into Kichagga to make people feel more comfortable by conversing or joking
with them. This was also useful when informants had quiet conversations amongst themselves
when sensitive issues arose which were, of course, lost to me with my rudimentary Swahili, but
transparent to Mosoi who would later explain the details of these conversations.

However, having a field assistant who doubles as an interpreter also had its downside: it
prohibited me from engaging directly with the informants and practicing my Swahili. On a few
occasions when he was absent, this forced me to converse directly with my informants, albeit in
a limited fashion, which, I feel, served to strengthen my connection and rapport with them.

Although Mosoi had been a member of the RTRP team, his contract had ended, and his wages
were paid exclusively by me for the duration of our stay in the field. But despite this financial
relationship, I sought to treat him as a colleague, rather than his employer, and this was
achieved by sharing the same living conditions, cooking together and resolving small personal
frictions through mutual discussion.

Field Notes Primary Data Collection

Field notes took various forms, from brief jot notes, expanded, detailed notes, personal journal
entries and weekly field reports to both my Brunel Supervisor, Melissa Parker, and Hannah Kuper
of the LSHTM. Data was collected using a wide variety of tools from the most basic - pen and
paper - to the most sophisticated: GPS technology.

24
The focus of my note-taking in the field was initially very wide, becoming progressively more
refined as themes and categories emerged from subsequent interviews and participant-
observation. However, I strove to capture as much detail as possible, looking to be able to recall
the fullness of the contexts I observed at a later point. In addition to hand-drawn sketches and
maps, I noted the space in which activities occurred, actors, actions, conversations,
circumstances, observations and analytic ideas, my reasoning being the richer the data capture,
the more opportunity to clearly envisage the field once home.

In addition to capturing details from the field on a daily basis, I also noted personal impressions
and possible further avenues for exploration, and used the evenings for reflection on the days
events. Reflection represented something of an iterative, dynamic process (Hammersley & Atkins
1995: 24) as I searched to resolve puzzling issues, or identify gaps and inconsistencies in my
data and lines of possible inquiry that needed further exploration (ibid: 191) all in the attempt to
develop as accurate a picture of village life inasmuch as it related to water. This process
resembled Kolbs experimental theory with its cyclical four stages: systemic information
gathering; reviewing information; reflecting on findings and presentation of results, and problem
identification (Kolb 1984). However, given the physical demands of concentrating on observing a
household from the early hours to dusk, it must be admitted that this wasnt always possible
every day.

Field notes were taken during interview sessions as well as during observational periods. As I
became more acclimatised to the field, the initial wide focus of my note-taking became
increasingly defined as I knew what to look for based on accumulated field experience, reviewing
and updating past notes in the light of new information.

I would also chart analytic ideas which would help me to make sense of what I was observing,
but there were times when I was wary of the categories that I had to understand my experience,
reluctant to impose these on my subjects. Instead, I would feedback my idea to the subjects to
see if it was congruent with their experience and categories, a good example being my
understanding of one household as being negligent but hesitant to label it as such
40
.

THE RTRP office in Mkuu also gave me access to electricity and a laptop which I could use to
prepare standardized forms for semi-structured interviews as well as to update my notes and
store them onto disk for protection. It was on these occasions away from the field that I could
reflect further, remembering additional details, sometimes seeing new patterns or even chancing
upon unexpected insights.

In addition to the traditional tools of pen and paper, I used a personal digital recorder to capture
data and observations on the fly. A small tape recorder was also used to record interviews which
helped me to gauge the competency of my field assistant to accurately interpret and capture the
most important elements of field interviews at the earliest stage of the research, having him
transcribe various interviews and then checking what had been noted down in our original notes
to see if there was any major difference between the two. Fortunately, there was very little, and
this confirmed my confidence in his ability to distil the essence and important points of any
interview.

The RTRP also provided use of a Canon camcorder which possessed a digital capture mode,
allowing me to take digital photographs of the scenes we were observing. Although of basic
quality, these could further refined using Photoshop

software to enhance detail, and serve as a
useful reminders of field conditions as well as serving as a visual counterpoint to textual narrative

40
Always careful not use leading questions in any way.
25
and analysis. A small analogue camera was also used as a technological back up, and subsequent
negatives were scanned and imported digitally for incorporation into the research.

Finally, mention must be made of the Trimble GPS machine provided by the Rombo Trachoma
Research Project which served as an invaluable technical resource which allowed us to obtain
detailed maps of both the location of the water sources in neighbouring Lucia forest as well the
village with its boundaries and main internal roads.

Field methods

Semi-structured interviews

My entry into the research process proper began with holding a series of daily semi-
structured interviews
41
during my first two weeks in the field. The purpose of these
interviews was to develop an overview of the history and issues of most concern to the
village, especially those concerning water.

I began by selecting key informants such as the Village Chairman (VC), Water Committee
Chairman (WCC), Village Health Officer (VHO), and worked my way down the political
hierarchy interviewing the three subvillage leaders, and then five balozi leaders from
each of the subvillages who were selected by the subvillage leaders according to criteria
which I thought would most provide a representative sample
42
, each subvillage leader
left to select suitable informants as they saw fit.

Whilst I began with some key ideas that I wanted to cover
43
, discussions would often
emerge about other related issues that would be explored in turn. In addition, these
would often generate debates within the groups where informants would disagree with
each other, a process I encouraged whilst underlining that all viewpoints had equal
validity. This interview process was invaluable since I was able to obtain detailed and
extensive in-depth information about the village and its difficulties with sourcing
sufficient water, as well as opening up various possible lines of future questioning.

After interviewing the balozi leaders from each subvillage, I thought it would be a good
idea to explore the same terrain I had covered in my first week but with representative
villagers. I then turned to the balozi leaders to ask their assistance to source five male
and five female informants from the households they represented using the same criteria
given to the subvillage leaders.

Using question schedules drawn up from information gleaned from the initial set of
interviews, I was then able to ask targeted questions covering a wide range of topics:
landholdings, wealth indicators, crops grown, livestock held, sources of income, detailed
questions about water usage, hygiene practices, knowledge of trachoma prevention and
membership of any support organisation, religious or otherwise. These provided me with
ample data which I could use to discern patterns, and evaluate the data obtained from
household observation.


41
A process characterised by the informal but systematic asking of questions.

42
I.e. an interview group should include at least one of each kind of person: highly educated; little education, highly
prosperous; not prosperous, possessing a large family; possessing a small family.

43
Using a mixture of open and semi-closed questions, but always careful not introduce any leading questions.
26
In addition to interviewing adults, I was keen to get a non-adult perspective of the
impact of water scarcity within the village, and also arranged a series of interviews with
local primary school children, which did provide me with additional informational that was
valuable in that it contradicted some adult accounts.

Although the bulk of interviewing was done in the first two weeks, I drew on this
technique at various points of the research, interviewing such key informants as a
rainmaker, traditional healers, Kiliwater representatives and officials from anti-trachoma
organizations.

Semi-structured interviews were extremely useful since they not only allowed us to
rapidly gain in-depth local knowledge about water related issues, but findings were also
easy to document. Their only limitation was the large amount of time needed to cover
the topics in sufficient depth, which may have proved tiring to some informants.

Systemic walkabout

This is a term derived from Hygiene Evaluation Practices (HEP) which essentially means
the identification of water sources within a fieldsite. Whilst I did not visit every standing
pipe within the village, I did visit several, often during times of usage which allowed me
to gain a first-hand understanding of the conditions under which they were used, as well
as developing a feel for the village.

Household observation

In addition to interviewing, I drew upon the technique of participant-observation which
allowed me to not only be able to generate thick description and emic perspectives, but
also provided me with another reference point to compare data obtained from interviews.

In total we observed ten different households, a tiny sample
44
and hardly ideal but one
which I strove to make both as representative and as extreme as possible; representative
in the sense that we chose some households to function as baseline households, neither
too rich or too poor, neither too large nor too small; extreme in the sense that we then
sought households that were at the top end of the bell curve in terms of trachoma
prevalence as well as households that were free of trachoma in order to discover
whether there were any water usage practices that would explain these differences
between these households.

Household observations were invaluable since they allowed me to gain first hand
experience of what happened at a domestic level. My initial strategy was to spend long
periods of time within a household from early morning to early evening, observing for
two or three days. This allowed me to build up a picture of a typical range of activities
and behaviours as well as habituate the subjects to our presence, in the hope that this
would help to diminish the observer effect
45
. Surprisingly, this was effective with the
interest and attention paid us by informants on the first day soon giving way to the
practical demands of running a household.


44
The short duration of my stay within the field meant that I had to implement a purposeful (not random) sampling
strategy so that I had the widest and most representative range of opinions and viewpoints possible, which would ensure
that the research was as accurate and fair given the limited timeframe and would meet the studys objectives.

45
Observer effect the effect of being watched, behaviour that emerges through being watched.

27
The initial observations also allowed us to compare subsequent households, discovering
similarities as well as differences in practices, and allowing us to evaluate different
accounts more accurately. We also spent time interviewing householders using the
question schedule developed for informants in the second week of interviews which
provided further details to contextualise our household observations, as well as a means
to achieve consistency in our data collection, allowing for cross-evaluation of data
obtained from multiple sources.

In addition, the more time we spent in the households the more we were able to develop
something of a checklist, of things to look for in a household: where was the water
stored, what container was used to collect water, where was the latrine, did the
household possess a clothes line? This allowed us to conduct semi-structured
observations within the context of participant-observation, allowing us to identify
household water usage practices efficiently and quickly. Without sounding too rash, I feel
that after spending two weeks observing various households, we had developed a good
sense of what to look for, and decided that spending long periods of time in any
particular household would not necessarily lead to greater or better quality of data. We
therefore progressively reduced the amount of time spent in individual households with
no appreciable decline in the quality of data obtained
46
.

Focus Groups

Information derived from household observations would then be checked by holding a
focus group whose purpose would be the exploration of a theme, or various themes so
that I could ascertain the validity of my ideas, resolve contradictory information, allowing
me to refine my working model further. Examples of focus groups held would be:
discussions with village officials (VC, WCC & VHO) of village history and issues, views on
village political leadership with a variety of representative informants and household
management, childcare, gender and hygiene issues with groups of female informants.

Once the ice was broken and rapport established, these were often lively and informative
sessions, which generated large amounts of further information, deepening our
knowledge of local issues considerably.

Overall, focus groups were extremely useful for gauging the range of opinions and beliefs
in the topic of enquiry; for exploring emergent issues during the execution of the study
and for interpreting data derived by other methods in the final stages of the study, all
helping to illuminate various beliefs and practices as well as providing insight into the
various constraints physical, social, cultural and economic which influenced the
management of water.

Data Triangulation

Using multiple research techniques allows a problem domain to be investigated from a
variety of angles, generating different kinds of data. Collectively, multiple research
techniques act to provide insight into four different levels of data usually not available
through one method alone that is:




46
This process of progressive focus is defined as possessing a characteristic funnel structure (Hammersley & Atkins
1995: 206)

28
What people say they believe, think or do;
What they actually do;
What they really believe or think;
The context where all these occur
47
.

Whilst using a multiplicity of techniques generates a level of comprehensiveness, multiple
techniques can in themselves be problematic since it is rarely the case that informants
accounts match each other exactly, it then being difficult to evaluate which account is
the most accurate. However, discrepancies within the data can be used positively, with
informants being allowed to comment on other informants perceptions (anonymously, of
course)
48
. Discrepancies between data obtained can also be valuable since they can point
to new areas which merit further exploration, which may represent significant advances
in the development of a hypothesis or understanding of local issues.

This process has been called triangulation, a method widely used within qualitative
research, which acts as a form of data quality control, ensuring that the study design and
execution are competent by allowing for findings and working hypotheses to be validated
or contested by informants, revealing which inferences are valid and helping to guard
against flawed analysis.

Triangulation can be applied not only to informants claims, but also to the data collection
techniques themselves which can be compared for consistencies and discrepancies
amongst the various concepts and issues explored. This process can, in itself, be a
source of additional information as new information can be generated through this
sorting out and evaluating, and surprising insights obtained.

***
Peer and supervisory feedback

I was fortunate whilst in the field to have access to various people who were professionally
involved in either ophthalmology or water research and able to act as sounding boards or provide
suggestions as to possible areas for investigation. Both Carolina Abuelo and Sarah Polack were
conducting research in Rombo on behalf of the LSHTM, and I was able to draw on Dr. Paul
Courtright of the KCCO in Moshi for ideas.

Whilst contact was continually maintained with UK based supervisors during the field research, I
was also fortunate to have Dr. Hannah Kuper of the LSHTM visit the field, and be able to share
ideas in real time with her, which proved most useful.

***

Access within the village

In addition to being given political clearance at district level, within the village I experienced
nothing but a positive relationship with village officials who supported the research fully by
helping to organise various meetings and focus groups that were to be held in the village office
meeting room, as well as assisting with sourcing a wide range of informants.




47
This model is sourced from Helman 2000: 265

48
This is called respondent validation (Hammersley & Atkins 1995: 230)
29
We were free to conduct our research wherever we wanted, but it soon became apparent during
the course of the fieldwork that we would not be able to walk around the village casually
dropping in on villagers, visiting them in their homes. Even if we had the name of an individual
household and knew approximately where they lived, it was often very difficult to find out exactly
where their homestead was located by asking people we encountered on the road. They were
loath to give exact directions, often feigning ignorance or uncertainty.

Household observations, therefore, were of necessity mainly arranged through the subvillage
leaders who would often transmit our desired intentions to the parties in question and organize
an appropriate day for us to visit.

But even if we were already conducting a household observation, it was very difficult to visit
neighbouring households in the immediate area to ask impromptu questions. Mosoi, my field
assistant, was loath to approach an unknown household without prior communication, making
identifying and accessing possible informants a much more difficult process. On the one occasion
we did try an impromptu interview with a neighbouring household when our original subject was
found to be absent, we were met with suspicion and reticence on the part of the householder
when we both appeared and asked if we could ask some questions. This served to reinforce my
impression that the Chagga household was something of a protected space, which could not be
casually approached. This therefore necessitated the constant assistance of subvillage leaders
who acted as intermediaries, facilitating micro-access to these guarded spaces. However, this
also meant that we were frequently unable to develop any relationship with the households prior
to the day we were scheduled to observe.

Working through intermediaries to organize our research instead of freely being able to do
ourselves felt restrictive, cumbersome and distancing, and occasionally this indirect
communication system would break down, with households failing to be informed by balozi
leaders that we were scheduled to observe them on a particular day, which put both the
householders and ourselves in a difficult and uncomfortable position when we arrived ready to
find them anything but.

Furthermore, we had cause to question how we were being represented by the subvillage leaders
and balozi leaders to their constituents. Even though Mosoi would introduce me as a daktari -
doctor - since this was more comprehensible to the local people than an anthropologist, more
often than not we would have villagers ask us when we would find water, or when we would be
supplying the village with water, indicating that they thought we were water engineers of some
sort rather than humble anthropologists. I found this puzzling since the village was no stranger to
trachoma related projects, Rombo Trachoma Research Project (RTRP), International Trachoma
Initiative (ITI) and Christoffel Blinden Mission (CBM) having visited various times during the past
couple of years.

Whilst this was frustrating since I wanted to be open and honest as to why I was there, and for
informants to understand my purpose in being there, I realized that their misconception actually
was hardly unusual (Hammersley & Atkins 1995: 81), and actually revealed a great deal about
the flow of official information within the village, which could be best described as haphazard.
Whilst it was widely known that there was a mazungo a white man in the village, this had not
been formally explained or discussed at any village meetings so that the average villager was
free to project whatever hearsay or wishful thinking upon myself and my field assistant when
they encountered us in the flesh. This served as an indication that even when information is
officially promulgated, it is not a certainty that it will reach its target audience either accurately or
promptly.


30

Informants

Those who were invited to take part in a focus group were given 500 Tanzanian shillings which
was a practice begun by the RTRP over the past couple of years. This was in recognition of the
fact that they had to walk to the village office from their homes (often a considerable distance
away) and that they would not be economically productive during this period of interview.
However, those who participated in household observation were not remunerated in any way.

It was my working practice to be as open and informative about our presence and research
intentions to everyone we met in the field, and to answer and address whatever questions or
concerns they might have had.

Choosing the households

The first two households were chosen from the initial set of focus group discussions (week 3).
The focus of our initial observation was to ease our way into the process of participant
observation, and also establish some kind of baseline for our subsequent observations. Both
households observed were prosperous and economically active and being present allowed for
insight into the rhythm and pattern of daily life in Urauri village.

Shortly after the initial observations, I examined the 2002 Rombo District Trachoma Survey files
which gave a breakdown of the survey for Urauri. Based on a sample size of 10 Balozis, the
results of the survey indicated that there was a predominance of trachoma in Nessae and Mhaka.
Balozi leaders and individual households were identified with the identities of individuals surveyed
and their trachoma status. The discovery of this survey constituted a valuable resource inasmuch
as it allowed us to work more systematically and in a more focused man by examining the
records and identifying specific households which had been exposed to trachoma education. It
was then an easy task to separate trachoma positive and negative households and solicit the
relevant subvillage leaders assistance in communicating with the relevant households seeking
permission to observe.

We focused more on the subvillage of Nessae since the incidence of trachoma infection was
significantly higher than in the other subvillages. However, we also spent time in Mhaka in order
to be able to identify any similarities or differences between the two subvillages. Unfortunately,
constraints of time meant that we were unable to spend time in Reha subvillage household
observation, a limitation which could possibly be rectified by a further study.

Temporal framework

When conducting the household observations we would arrive at the house concerned in the
early hours of the morning and spend the whole day, observing the ebb and flow of daily activity.
We paid particular attention to early morning, midday and late afternoon activities which were
usually characterized as periods of high water usage. However, one unfortunate limitation was
that as dusk fell between 6:30 and 7, lack of electricity precluded further detailed observation of
household activities. A further issue was one of personal security. Households were often located
far from our host household, and returning often involved negotiating isolated dirt tracks and
paths: an unattractive prospect given the darkness and my field assistants unwillingness to be
expose us to potential dangers. At this point it must be said Urauris location on the border with
Kenya means that there have been violent and armed incursions before and that despite the
District Commissioners assurances about our safety, there was nothing to guarantee our security
either from Kenyans or opportunistic local street boys looking for an easy target under the
protection of darkness. Regrettably, therefore, we were not able to stay beyond sunset to gain a
31
complete empirical understanding of household water usage. However, we were able to question
householders about these missing hours, and often physical evidence would remain of any
nocturnal water usage (damp ground etc.,).

Impression management

I was constantly alert to impression management, and even managed to circumvent it during
participant-observation by visiting households we had already visited unannounced to confirm my
suspicions as to whether impression management had occurred first time round, or whether what
we had been presented with was in fact the true picture. More often than not, my initial
impressions of over-management i.e. the presentation of an inaccurate idealised picture of a
well-managed household were confirmed upon finding the household in a chaotic state upon
our visit. Other times, however, our initial perceptions that a household was well organised and
managed were confirmed upon our subsequent visits. Most of the time, though, impression
management was easy to detect, especially when a family would act as if face-washing was a
normal, everyday activity, and the children (watoto) would betray this as a lie by bawling at the
top of the lungs when their mother would attempt to clean their face hardly the daily habit
claimed.

In a sense I was able to hide behind my field assistant, Mosoi, in my efforts to become invisible.
I began to rely on him as an intermediary not only linguistically, but also culturally and socially to
engage with informants and ask questions on my behalf, something he was happy to do since he
was very much a people person, able to make people quickly feel at ease as well as keep them
engaged in conversation for hours. This meant that he was able to involve informants in
extensive conversations while I faded into the background, free to take notes and observe daily
life unobtrusively.

The evolution of the research question

As I spent more time in the field, the process of data collection began to turn the initial set of
general foreshadowed questions into more focused lines of enquiry. However, whilst often
allowing me to glimpse a wider perspective, the data would sometimes contradict or even
overturn previous data, so that rather like opening a Russian doll, something new to be
discovered would be revealed at each turn. As Malinowski pointed out, it soon became a question
of moulding theories according to facts, and of seeing facts in their bearing upon theory
(Malinowski 1922: 8-9).

At times I felt like a detective, having to reconcile conflicting accounts, trying to evaluate what
was true from the mass of information. At others I would challenge front stage management
through persistent indirect questioning until chinks would appear in the wall of presentation and
stories would shift subtly, almost without informants being aware of what was happening. One of
the initial difficulties of this process was that it was very hard to know what was relevant since
everything in the field is potentially relevant. I therefore had to rely on a combination of intuition
and serendipity in the data collection process to reveal the most plausible avenues for
exploration, an activity described by Dollard as informing myself incidentally of many apparently
remote matters (quotation in Hammersley & Atkins 1995: 30).

As it turned out, I often had to be patient and rely on my memory of what had earlier been said
or claimed by one informant in order to fully realize the importance of what a different informant
might say later. Sometimes information was presented in a non-linear way or in a highly
fragmented way by different informants. This made the evolution of a single line of enquiry
difficult, forcing me to tolerate uncertainty and a certain degree of anthropological freefall in
order to allow the field itself to reveal its patterns and secrets. As such, the research question
32
was continually subject to much reformulation so that it might be fair to say that there was one
major theme, one destination: Water Access and Usage, but the journey involved many different
paths. But all in all this helped identify various issues with increasing clarity and precision.

Viewing the breakdown of my research activities (Appendix 6) may provide the impression that I
developed a highly structured prearranged research schedule prior to the field. The truth is that
appointments with informants were scheduled one week ahead which provided enough structure
for the research process whilst allowing complete flexibility to respond to new information as it
emerged and organize additional interviews to explore emerging themes.

Issues of reflexivity

All anthropologists have to contend with the fact that they are not objective tools of research,
and that their subjectivity does exert a considerable influence on how research is formulated,
conducted and written. Personal characteristics are perhaps the most important factors in
determining what domains an anthropologist may have access to, how informants will view and
respond to them and how data will be obtained, interpreted and presented. Within the context of
this research I found the issues of gender, race and religion key elements in my dealings with
informants.

Gender

Chagga society is highly differentiated according to gender. As Hammersley notes: the
world of women may be somewhat difficult to access (Hammersley & Atkins 1995: 93),
and I did find that at public events such as weddings, village council meetings or even
funerals I was steered away from the womens section and placed firmly within the
mens section. However, we were able to establish rapport with the women in the
households we observed by engaging with them on a wide range of domestic topics, a
subject which they are often expert on.

Even my gender was a subject of much discussion and mirth amongst the villagers who
were unsure how to classify me. With my long(ish) hair and lack of stubble (Asian genes)
was I really a man? How could I be with hair the length of a womans
49
. Whilst calling my
masculinity into question was not particularly amusing to me at the time, I appreciate
how this ambiguity
50
may have helped me transcend rigid gender boundaries, if only
slightly, and make village women somewhat more receptive to me, or at least less
guarded.

Race

Living in cosmopolitan Britain, my racial background is rarely an issue. My looks and
linguistic capability have allowed me to blend somewhat into the background in quite a
few places whilst travelling around the world. In Kilimanjaro, this wasnt an option, my
multicultural origins reduced to a single classification. Here, for all intents and purposes I
was a mazungo a white man - and for perhaps the first time in my life I truly knew
what it meant to stick out, to be singled out as different.


49
Chagga men invariably sport very short hair.

50
This ambiguity may have intensified when word spread around the village that I cooked, an activity firmly associated
with a womans domain.

33
I also had the dubious honour of being the first mazungo to have lived full-time in the
village. To be sure, there have been European visitors before, Catholic priests from
Germany and Norway, but they were always just that, transitory. However, I became a
common sight in the village, tramping up and down the lanes with my field assistant,
probably a peculiar sight with my baseball cap, daypack and obligatory bottle of mineral
water, and no doubt fulfilling their stereotypes of how a mazungo should be like.

School children would stare, transfixed; young girls would giggle; sometimes very young
children, surprised at my appearance would scream and run away in terror. And it was
then that I discovered that the word mazungo occupies a similar role in Kiswahili as the
bogeyman does in English (Raum 1940: 56)
51
. On Kilimanjaro, however, the word
mazungo is used liberally by parents to scare young children into compliance. Dont be
bad, or the mazungo will come and take you away. Particularly difficult children might be
brought under parental control with tales of white people eating children (ibid: 299),
which might perhaps explain the stricken expression on some of the young faces I
encountered.

But being white meant that I was always visible, always prominent. This had its
frustrations, especially when greetings were always prefaced with white man, which
began to sound progressively more annoying as the weeks rolled by. Whilst I tolerated
this outside in the field, in my hosts home I objected to being greeted as the white man
by my host family or their staff since this to my ears had all the equivalent finesse of
addressing an African-American as nigger, and hardly a term of address I wanted to
hear after-hours. However, once I explained to the household members the reason for
my discomfort with such usage, it soon stopped, and I was addressed by my first name
instead.

In retrospect it might have been somewhat nave, but whilst in the field I was keen to try
and diminish as much as possible public perceptions of my mazungo-ness, to seek
(somewhat optimistically) to be accepted for individual rather than stereotypical qualities.
However, as I deepened my relationship with the field, I realized that this desire to be
seen under a particular gaze reflected my personal concerns, and that in fact potentially
valuable ethnographic data could be obtained from observing how villagers came to
negotiate a relationship with me under their terms, not mine.

Religion

In the UK, religious issues barely register on my personal radar. This is perhaps because
this we are predominantly a secular society, and religion is largely seen as a private
matter. In the village, however, the Catholic church is an overwhelming presence, and I
found myself being very surprised by my reactions to it. For me, the influence it wielded
and the absolute faith villagers placed in the church seemed somewhat misguided; the
church to me resembling more a vehicle of ignorance, the antithesis of progress. For
example, the churchs stance on family planning raised (and still does) extreme ire in me
since I was seeing first-hand on a daily basis within the village the consequence of a lack
of family planning: parents that could not cope economically or emotionally with the
parental demands placed by having so many children, the children suffering by lacking
clean clothes, sufficient food - their bellies swollen from either malnutrition or worms
and clearly receiving inadequate attention or stimulation.


51
A word derived from the Bugis people, a seafaring people whose reign of terror attacking spice ships in the sea lanes of
Eastern Indonesia secured their eternal fame, finding its place in Western folklore and countless bedtime moments.
34
My anger intensified when I heard stories of intentional disinformation from various
sources; of how the church informed their parishioners that condoms had holes in them
and that they were bad for ones health. In the context of Africa where AIDS is the most
significant problem facing the continent this was nothing short of socially irresponsible
and at the very least morally questionable, I felt. However, I never betrayed my true
feelings whilst in the field, always being outwardly respectful towards all forms of
religious practice, shows of devotion and members of the church.

Whilst the faith expressed by the villagers seemed genuine, I felt it was misplaced and at
risk of being abused, which made me even less well disposed to the Churchs presence.
At times I wondered what lay behind my antipathy, and then one day I made the
connection: my family comes from a Spanish Republican family, my grandfather
executed for organising a plot against general Franco in 1947. I had grown up with
stories of abuse and hypocrisy by Catholic priests: little wonder that I was so poorly
disposed towards its presence in the field. However, once I acknowledged the origins of
my negativity towards the church, it became easier to understand where this antipathy
came from, and to be able to separate my personal feelings and my role as a
professional anthropological investigator. In the end I made my peace. Despite my
personal feelings, I could recognise that the Church played an important function/role in
the community which, regardless of my personal feelings or misgivings, was my duty as
an anthropologist to understand. I thus began to see the Church in a more objective
way, trying to recognise its positive function within the community - even attending mass
on a couple of occasions the first time in thirty years.

Participant or Observer?

My position as researcher was that of an adopted position of comparative detachment, so that
one could say that my position as a professional stranger was more observer-as-participant than
participant-as-observer. I wasnt interested in participating as a villager: my status or role was
that of researcher and this certainly precluded me from participating in the normal activities such
as fetching water, or cutting fodder or cooking, which were considered largely womens work.
Furthermore, with such a short research time frame I needed to understand what was going on
and I needed to be highly focused, gathering data and testing analytical ideas formulated to
understand accumulated data, rather than working to build long-term relationships. This meant
that in addition to the normal adjustment issues involved in negotiating a new environment in its
myriad aspects, I also felt under considerable pressure to extract cogent information from the
field, to elucidate from the mass of village activity around me the basic social facts and rules
which underpinned daily life. This, understandably, elicited no small degree of anxiety as I
wondered continually whether I could do it or whether the two month time frame would be
sufficient. However, the positive side to this is that it probably drove me to be vigilant in
exercising due diligence, and expend extra effort in my quest for understanding.

Ethical considerations

Ethical clearance for commencing the project was obtained in London prior to leaving for the
field, whilst clearance for conducting the research project in Kilimanjaro was obtained by KCCO in
Moshi. Giving due consideration to possible issues prior to departure proved useful, but like all
real world situations, the field proved to contain a few unexpected obstacles. Here, then, are
some of the major issues I thought important to consider and the stance I took in relation to
them, as well as the key ethical dilemma I encountered whilst conducting fieldwork.



35
Informed consent

In every encounter with a potential informant, I endeavoured to explain the purpose of
the research as fully as possible: why we were requesting their assistance, as well as
clarifying that an informants contribution was entirely voluntary and that they could
withdraw from participating at any time. This was done with the assistance of an
introductory leaflet I had written and had translated into Swahili and printed out before
entering the field (see Appendix 5). If a person was illiterate or had difficulty reading, our
research purpose would be verbally explained. However, it must be noted that no written
consent form was sought as a permanent record of informed consent since I felt that this
would have been excessive to the immediate purposes and scope of this project; verbal
consent was deemed sufficient instead. Permission was also sought for the taking of
photographs, and heeded if denied.

Lastly, informants were also encouraged to participate in the ownership of the research
process by being invited to feel free to ask any question of us during or after the
interview/observation process, should they wished to do so.

Overtness

Whilst our research was conducted overtly, there were limits to just how overt we could
be.

I found that I was encouraged by my field assistant to NOT expand or extemporise on
various aspects of our research purpose. This was because such explanations would not
be seen as an elaboration of the complexity of the project, but rather as inconsistencies
in accounts, in other words lies, and would encourage a perception amongst our
prospective informants that we were up to no good. Therefore, despite my willingness to
be forthcoming and explicit about the various issues involved, field conditions meant that
I had to present a simplified and general explanation of what we were doing and
interested in, even though this did not necessarily reflect the full picture of our interests.

I experienced the truth of Mosois warnings when once I added more detail about our
research focus to one householder who immediately challenged it, since they had heard
a different story from a neighbouring household we had observed. So the danger here
lay in being too forthcoming, expecting individuals to see the nuances of research
interests as just that rather than discrepancies or lies. It is because of this that some of
my explanations were not as overt as I would have wished them to be, the restriction
stemming from the villagers perceptions and pre-existing notions of what I was there for
(Hammersley & Atkins 1995: 265).

Maintenance of privacy and confidentiality

Whilst my field data readily identifies individual informants by name, in this dissertation I
have instead decided to use pseudonyms which draw on typical Chagga names to ensure
informants anonymity and protect their privacy, as per my original agreement in the
field.

Furthermore, at times I was privy to sensitive information communicated to me in
confidence. If clearly identifiable, such information could potentially cause unwanted
problems for various individuals in the community, and as such will be withheld or
transmuted into a generalised statement to protect both the source and the persons
concerned.
36

Exploitation

There is no denying that I stand to benefit academically from my research on this Urauri
community, if only to pass the M.Sc. Indeed, a few villagers were keen to point that out,
and inquired as to how the village would benefit from my conducting research there.
Whilst I was always clear that I couldnt bring any specific, immediate tangible benefits
to the villagers, I did state that I wanted there to be some reciprocity, that I wanted the
research to aid the village, if only to identify problems and posit solutions which would
have a beneficial impact on the health of the community, both in terms of trachoma
reduction and more generally. As such, it is my hope that my findings will be of use in
understanding the issues and complexities involved in water usage in water scarce
communities as well as potentially helping to refine future anti-trachoma programmes
which will benefit not just the Urauri community, but others further afield.

One possible way this could happen is to draw upon this research to write a paper for
publication in an appropriate journal so that the wider academic community can have
access to this information, and that the LSHTM can implement its findings into its long-
term evaluation of the SAFE trachoma strategy.

Ethical challenges

The main ethical challenge I faced in the field was when I came across a young man of
nineteen being abused one evening in the courtyard of the very house I was staying in.

Whilst some ethnography textbooks take an absolute stance in cases of abuse, arguing
that human rights take precedence over notions of objectivity and research, and
advocating intervention in such cases, I agree with Hammersley and Atkinsons position
that ethical dilemmas need to be recognised as being situated, that is: what might be
appropriate in one context might be inappropriate in another.

To give an example, if I saw someone being unfairly physically attacked, it is quite likely
that I would step in and intervene. And indeed in the past I have, intervening to stop a
man beating a woman in London. However, sometimes what seems to be the right
course may not be appreciated as such by those involved. In the case of the woman
being beaten by the man, rather than thank me for stopping him from beating her
further, she told me to Fuck off and mind my own business. If, within the context of a
culture I am familiar with, I fail to get it right, what guarantees are there that in an alien
context I would be able to accurately read what was happening, and intervene
appropriately, or even not put myself in danger?

As the beatings progressed, I realised that my presence as an outsider or even as a
mazungo was not acting as any kind of restraining influence. Furthermore, when it
became apparent that the men abusing the boy were getting ready to take their anger
further and perhaps kill the boy, I was paralysed by my position. What was I to do?
Should I step in and stop eight angry men, one armed with a handgun, from inflicting
whatever they had in mind? But I hesitated: I didnt understand the context. What had
the boy done? Did he merit this? Who was I to intervene where it wasnt any of my
business? Was this a case of me imposing my morality on others? Fresh in my mind was
a particularly gruesome story of a burglar being hacked to death in a neighbouring
village the week before, yet those who related the story to me didnt seem shocked by
the story, or regard this as murder rather as anything other than swift and appropriate
justice. So how could I intervene and say I knew better? The truth is I didnt.
37
Furthermore, much as I would have liked to have intervened, I was transfixed, more in
shock than anything. As it happens, the boy was taken eventually to the police, but it
had unsettled me, more for the fact that it did bring up issues of what was the best
course of action ethically, a decision that has to be negotiated on a case by case basis, I
feel, rather than by appealing to any absolute guidelines such as the AAAs code of ethics
(Hammersley & Atkins 1995: 277).



Reflection

This final section seeks to outlining some of the key processes involved in creating an
understandable model of complex social experiences, as well as the representational choices
made during the writing process so that the reader can judge the merits or otherwise of the
approaches taken within this ethnography.

The process of analysis

It would be a misconception to think that analysis is a process which occurs exclusively away
from the field and only in the study or library. On the contrary, attempting to understand the
data obtained in situ as it is collected is an intrinsic part of fieldwork, and I constantly found
myself refining my ideas, generating fresh working hypotheses and discarding older ones as they
were invalidated by new findings whilst in the field, creating dynamic models of what was
experienced and observed, updated by new events, accounts, perceptions and insights, which
would then have an effect on what information was collected and regarded as relevant or
potentially useful.

However, away from the field the kind and perhaps quality of analysis that can be performed is
very different; largely, I think, because once the data collection is finished, not only does the
assembled corpus of data allow one to gain a very different perspective or purchase
(Hammersley & Atkins 1995: 218) on processes, events, relationships and structures that were
observed in the field, but also being away from the field often means there is time and space for
the process of focused reflection which is rarely available when one is in the midst of the
ethnographic action.

Some ethnographers are keen to fit their data into a single theoretical model, be it a functionalist
reading, or symbolic interactionist. My position is that I think that one should attempt to honour
the phenomenon studied by using the data obtained to suggest or generate a theoretical model
which best explains the patterns and inconsistencies found therein, an approach has been called
grounded theorizing (Glaser and Strauss 1967). In other words, the model should fit the data
rather than the data fitting the model.

Whilst I have been much impressed by the insights afforded by Critical Medical Anthropology
(CMA) (Baer & Singer 1982; Singer 1986a, 1989, 1995) and its usage of political economic
models to explain differentials in health experiences, I am equally aware that it is just that: a
model which seeks to explain phenomena but which, like all models - as Korzybinski noted -
possesses certain limitations in its ability to explain and account for all the data obtained in the
field since the map is not the territory (Korzybinski 1933). Whilst I freely acknowledge that
much of my initial analysis was inspired by CMA, ultimately I consider myself a servant to the
data, free to draw upon a multiplicity of theoretical models which could provide the best insight
and analysis (Hammersley & Atkins: 214), as well as drawing upon the data itself to generate a
model that explains village processes and relationships largely on their own terms.

38
In common with a lot of ethnographers, I returned home burdened with interview notes,
informant accounts and detailed observations, to find that my conceptual cupboard was
somewhat bare, and that organising my findings coherently was going to be a necessary but
challenging step in the process of understanding the data and creating an ethnography proper.
This involved not only reading and re-reading the data to discern patterns and inconsistencies: to
think with the data (ibid: 210), but also revisiting the literature to review previous ethnographic
findings in the light of my recent experiences. These processes helped me to begin to
differentiate the data obtained into categories that would help to explain the phenomena
observed and data amassed, bringing order to chaos, form to the amorphous.

The process of creating categories abstracting key features from the body of data which are
representative of central features of the phenomena being studied - is an intrinsic part of model
building and ethnographic analysis. Often this is a dynamic and organic process, concepts
emerging, being refined and perhaps discarded when a more suitable or more valid category
develops based on additional understanding. Even the interplay between categories can help to
identify new patterns, necessitating the creation of further categories to explain new findings or
articulate new insights, bringing greater coherency to the overall analysis.

The ultimate aim of the formulation of categories, then, is that they should help to explain the
data obtained, to be able to provide additional insight into what was observed, so that as
understanding deepens, the conceptual framework becomes sturdier, more powerful and more
accurate in its claim to be able to represent the social reality in this case - of a small rural
village located in the shadow of Mount Kilimanjaro.

Representing the other

As anthropologists have become more reflexive and open to the most valuable insights afforded
by postmodernism concerning issues of representation and the nature of the text, the notion of
returning from the field to simply write up ones material has become somewhat outmoded,
perhaps even nave. Early and mid-twentieth century anthropological writings were written in a
style which acted to perpetuate the illusion that the translation of field data to text was a matter
of mechanical conversion alone (Hammersley & Atkins 1995: 245), a process of taking various
raw ingredients collected from the field and transforming them into a palatable dish
52
.
So complete was this illusion and entrenched in the anthropologists unexamined worldview that
little consideration was given to the fact that such writings constituted a specific style or that
they themselves were a kind of situated cultural product with an implicit ideology. Embedded
within such ethnographic writing, defined by Van Maanen as realist, is the idea that the text is
an accurate rendering of the social world studied, i.e. isomorphic with the reality described (Van
Maanen 1988). Also implicit within such a style is a larger argument about knowledge in general
that there exists a tangible objective reality which can be observed and accurately represented.
Furthermore, choosing realism as a writing style encourages self-occlusion, thereby creating an
impression of scientific authority; one advantage of this approach being that it allows the writer
to make assertions about the culture observed, which can encourage a narrowing of possible
interpretations a reader can make of a text.

The unhappy truth, however, is that realist writing is, in one sense, a lie; realist ethnographies -
like all ethnographies - are representations, constructions, fictions (albeit often convincing ones),
drawing on well established writing conventions and narrative traditions for their authoritative
gaze and claims to be able to accurately render the real world.


52
With apologies to Levi-Strauss

39
Far from being a 1:1 representation of complex, multidimensional, dynamic social realities, the
writing of any ethnography of necessity involves making a choice as to what is relevant and
irrelevant; including here, deleting there, to finally present a highly focused account or model
which is inherently reductionistic.

One way to exemplify some of the more complex and ambivalent realities that do exist in
fieldwork is to draw on an impressionist style (ibid 1988), where narrative techniques drawn from
descriptive and fictional writing are employed to evoke a sense of place as well as render key
events more tangible to the reader. Such a style can be valuable in that they can help to shake
up the single, dominant voice (Hammersley & Atkins 1995: 253) found in realist texts by
providing the reader with some insight into the multi-vocality of the setting, as well as making
events in the field more understandable by revealing their more enigmatic attributes, something
that a realist style may sometimes overlook.

During the writing of the ethnography I decided to combine elements of realist and impressionist
styles since I felt that these approaches were not necessarily antithetical or antagonistic to each
other; rather, they could emphasize different aspects of the field experience. I was also aware
that I was only in the field for a very short time, and that my understanding of it did increase and
deepen the longer I remained, but that it was still in a state of flux and the disparity between
what I would have liked to have understood and what I had actually grasped by the time of my
departure all too large. Thus my apprehension at assuming a mantle of anthropological authority
which I was dubious I deserved to wear, to claim a position of omniscience and the authority to
speak unequivocally of and for the people in question (ibid: 253). Furthermore, I was aware of a
responsibility to portray as accurately and fairly as possible the lives of my informants, believing
that non-Western informants, normally denied the opportunity to voice their concerns and
experiences within Western discourses, should at the very least be rendered as fully as possible
by those who claim to study and represent the intimate details of their lives. In this I agree with
Richardsons observation that:

If we wish to use our privileges and skills to empower the people we study, then we should
value the narrative. (Richardson 1990: 133-4)

Whilst one could emulate Pat Caplans approach in African Lives: African Voices to allow
informants to speak largely for themselves (Caplan 1997), I chose instead to moderate the
potential abstract harshness of a realist style by interweaving it with impressionistic splashes,
informant quotes and vignettes within the discursive space (Hammersley & Atkins 1995: 241).

Whilst possibly laying myself open to charges of hubris and self-importance, I believe that with
the passing of time, this study may even become something of a historical document, a snapshot
of a people living under difficult conditions at the turn of the 21
st
century. It is my sincere hope
that I have done them justice.


******
40
1 ETHNOGRAPHIC FINDINGS

The general setting

Kilimanjaro
53
is home to the Chagga people who occupy all the slopes of the mountain range,
except the northern
54
. When asked about their origins, they will often reply that they have always
been there, but this is not the case; linguistic evidence connects Kichagga, the language spoken
in various forms around Kilimanjaro with older Bantu forms (Nurse 1979) and it has been much
speculated that the Chagga were part of the second major Bantu migration which occurred in the
fifteenth century of people seeking to escape drought, disease, the threat of violence and
searching for lands that offered safety and a sustainable life (Setel 1999: 44). In Kilimanjaro they
found a place which not only provided sanctuary from the numerous threats of the plains, but
also offered abundant water and fertile land; food for families and fodder for livestock.



Fig. 1: Map of Tanzania (Kilimanjaro Region in red)

53
Kilimanjaro is one of twenty-one regions in Tanzania comprising: Arusha, Dar es Salaam, Dodomo, Iringa,
Kagera, Kigoma, Kilimanjaro, Lindi, Mara, Mbeya, Morogoro, Mtwara, Mwanza, Pwani, Rukwa, Ruvuma, Shinayanga,
Singida, Tabora, Tanga & Zanzibar (including Pemba & Mafia islands)

54
The geographical area occupied by the Chagga on Kilimanjaro is sometimes referred to within the anthropological
literature as Chaggaland.

41

The presence of linguistic islands within the Kilimanjaro region, namely the Nilotic forms spoken
by the Masai would indicate that migration has been recent, occurring from a variety of directions
(Stahl 1965: 37). As migrants settled in North Eastern Tanzania and into South Eastern Kenya,
further linguistic variation occurred which encouraged the formulation of a variety of ethnic
identities, (even if they were historically, culturally and linguistically related), a process
encouraged by the heterogeneity of the local topography: each ridge, each spur, each valley
giving rise to a sense of identity that differed from its neighbours. Inevitably, though, such
geographical differences lead to political divisions emerging which were locused around individual clans
(ukoo) led by hereditary chiefs (mangi).

Rebmann, the first European to visit the region in 1848, noted the dynamic and open quality of
Chagga society (Stahl 1965: 35), with a flourishing trade of food and commodities amongst
various groups on the mountain and from further away.



Fig. 2: A long history of markets on Kilimanjaro

In addition to economic openness, the presence of a hundred or so chiefdoms on the mountain
at the beginning of the nineteenth century also meant that the mangi were also eager to
capitalize on the arrival of new information, technology or people if it meant the extension or
consolidation of their political power (ibid: 39): superior spears, for example, could prove decisive
in local conflicts whilst the possession of guns would dissuade the Swahili from attempting
slaving raids (ibid: 40).

Whilst the European presence brought key institutions to Mt. Kilimanjaro that transformed
Chagga life: cash economy, connection to a world market, Christianity and formal education
(Howard 1997: 48), economically, the colonial presence may not have been as positive for
Kilimanjaro as for other parts of East Africa since while the German imperial project considered
its territories as colonies, and made investments accordingly, following the end of World War One
the British administered the German territory as a protectorate, unlike Kenya, which made a
significant difference in the level of investment, and arguably subsequent development.
55



55
For further discussion of the impact of European imperialism in Africa in general as well as in East Africa in particular,
see (Packenham 1991; Rodney 1981)
42


Fig. 3: Christianity - a colonial legacy

Some scholars have argued that the pre-colonial economic system based on subsistence farming
and kinship-based social relations and obligations which engendered mutual aid (Masao 1974: 1)
prevented extreme poverty, with chiefs expected to protect those most needy from experiencing
hardship (Howard 1997: 46), but the advent of colonialism and a shift to a new way of life based
on a cash economy (ibid: 7) led to a worsening of the situation of the poor, greater
socioeconomic differentiation (ibid: 51), labour surpluses and economic exploitation on
consolidated coffee estates (ibid: 206) as well as a decline in the power, relevance and protection
offered by the chiefdoms which numbered just fifteen by the beginning of independence in 1961
(Stahl 1965: 35).



Fig. 4: Kilimanjaro - tradition and modernity

Modernisation through changes in agricultural practice, education, even adopting new religious
affiliation were enthusiastically embraced by the Chagga which meant that by the time of
independence in 1961 although they were politically disenfranchised, the Chagga were
economically empowered and had the benefit of being one of the most educated ethnic groups in
Tanzania. Nyerere had witnessed the strained legacy of British colonial administrations and the
resentments and violence that unfolded from perceived colonial ethnic favouritism and swore to
avoid the same bloody fate befalling his fledgling nation through inter-ethnic rivalry breeding
political violence and instability. For this reason, he enforced a policy of positive discrimination
against the Chagga so that they wouldnt have access to political power or be able to build any
43
power base within the new Tanzania
56
. The consequence of this policy was not only a notable
decline in the use of Kichagga amongst the Chagga in favour of Swahili
57

58
, but a lack of
investment and economic development of the region for many years.

Across Tanzania, traditional political structures were modernized; the mangis replaced by newer
political offices (Stahl 1965: 38). At village level a village chairman was elected by the villagers
and was responsible for managing a village which was divided up administratively even further
into subvillages. Village households (nyumba) were organized in cells of ten called a balozi and
were represented by a balozi leader, chosen from amongst the householders themselves.

Inspired by Soviet and Maoist models of collectivization, Nyerere also instigated a policy of village
co-operatives - ujama. This policy represented a major upheaval for many in Tanzania because of
the relocation and consolidation of lands, but on Kilimanjaro nearly all the land in the kihamba
belt was already divided among contiguous farms, allowing no room for new villages
59
, thus
sparing the region from the economic damage and upheaval often caused by the policy.



Fig. 5: The long, slow march towards development

While Kilimanjaro has long been romanticised by Europeans, seen as the promised land by some
early writers (New 1873: 392) or as the ultimate must-do for many contemporary backpackers -
a beacon for the intrepid who seek to prove their individuality and mettle by reaching the peak,
the mountain range is seen in a very different way by those who live within its shadow. Some
Chagga derive their income from the influx of tourists seeking to challenge the elements and
climb Africas tallest mountain, or who wish to experience the splendour of the Serengeti. But
most of its inhabitants have a more ambivalent relationship with the mountain. At times, it is the
very source of life - its waters, fertile earth and moderate climate allowing for the cultivation of a
wide range of crops and keeping of livestock; at others, the mountain is less bountiful, increasing
population and land pressure, climatic changes and decreased rainfall, and the every-present
threat of market fluctuations to cash crops can mean privation, hardship, disease and sudden
death.


56
This has since changed with the advent of multi-party democracy and the successor to Nyereres old political party
(TANU), the CCM being dominated by the Chagga.
57
Kichagga may be in the process of becoming an endangered language. Most young people prefer to communicate
amongst themselves in Kiswahili, and seem to be fast losing facility and familiarity with the richness of Kichagga.
58
Whilst interviews were conducted in English and interpreted by my field assistant using a mixture of Kiswahili and
Kichagga, only Kiswahili references are cited within this study to ensure that the ethnographic information is not
linguistically constrained by the terms of one particular locale, and applicable to other sites throughout the region.
59
In the case of Kilimanjaro, it was more a matter of delineating existing settlements rather than uprooting dispersed
homesteads and creating new nucleated settlements (Grove 1993: 438)
44


Fig. 6: A Chagga funeral of a twenty-seven year old man

Kilimanjaro represents, therefore, something of an enigma. Whilst it possesses the highest per
capita income, it also has the second highest rate of child malnutrition (Howard 1997: 16); some
of the richest Tanzanians are Chaggas - as are some of the poorest. It boasts the highest level of
secondary education achievement as well as some of the highest rates of functional illiteracy.



Water supply on Kilimanjaro

Sources of water on the mountain

The inhabitants of Kilimanjaro are entirely dependent on water which is found on the mountain
and derived mainly from precipitation. In terms of available water sources, the mountain can be
divided into four zones:

The summit
The arid zone and alpine meadows
The forest zone
Cultivated lower slopes

The summit

The ice coverage and glaciers on Kilimanjaro which are found at above 5,000m have
been shrinking for the past seventy-seven years (Hastenrath, Greischar 1997), but the
combination of high altitude and lower atmospheric pressure means that most of the
disappearing snow and ice is lost through a high daylight evaporation rate, rather than
as melt-water which makes its way to progressively lower levels of the mountain
(Ramsey 1964: 92).

The arid zone and alpine meadows

This zone situated at between 3,000m and 5,000m seems to spring from a bed of clouds
clustered at the 3,000m mark and, except for a few isolated areas, is generally
characterized by its lack of covering vegetation which reveals a somewhat austere
rockface. Those few clouds which ascend higher than 3,000m have usually deposited
their water content at lower altitudes, and what little remains and falls within this zone is
also subject to high rates of daylight evaporation. This means that this zone contributes
little to the available water on the mountain.
45

The forest zone

A 1964 survey identified 96% of all streams as originating either within the forest zone or
just below it (ibid: 93). Often Kilimanjaro is a brilliant sight in the early morning, but as
temperatures and the humidity rises, clouds can be seen to form and clump together,
becoming denser and denser as the morning progresses until the mountain becomes an
indistinct haze, even completely obscured. This cloud cover between 1,500m to 3,000m
means high precipitation levels.



Fig. 7: Typical forest zone scene on Kilimanjaro

The humidity and high rainfall means that vegetation is abundant, if not well-nigh
impenetrable without a machete. Extensive tree cover and decaying detritus absorb
much of the rain, allowing it to work its way slowly into the subsoil, and ever downwards
through the pull of gravity where it often emerges as a spring source lower down the
mountain.
60


Cultivated lower slopes

The cultivated slopes which are found below 1,500m do not represent a major source of
water on the mountain, receiving a fraction of the precipitation found at higher levels
(which if not pro-actively harvested or used to naturally irrigate thirsty crops, can be lost
for further domestic or productive use); rather, most of the water used on the inhabited
slopes of Kilimanjaro is water that has wound its way through the subsoil from the forest
zone above. Such water was the source of much of the early prosperity of the region,
and was channeled by the intricate and ingenious furrow system (mfongo), a network of
irrigation channels developed by Chagga before the advent of the European presence
which would carry water over tremendous distances, in some cases up to fifteen
kilometers over the most difficult and demanding terrain (Ramsey 1964: 94).







60
Illegal logging can remove vital tree cover which can result in reduced precipitation levels as well as soil erosion and
less water finding its way into extant streams, available for human use.

46
Traditional water management on Kilimanjaro

Traditionally, Kilimanjaro has been well served through its system of irrigation furrows
which diverted water from diverse sources around the mountain, from intakes cut high
up on a mountain stream, often within the forest zone, or taken from main rivers (Masao
1974: 1,3). Channeled water could be transported large distances, managing not only to
supply numerous land holdings with sufficient water for irrigation and the growing of
various crops, but also satisfying the water demands of various communities around the
slopes. (Anonymous 1973: 178)

Prior to independence the furrows were rigorously managed along lineage lines, and
furrow rights held sacred and widely respected
61
- with violence a distinct possibility if
they werent (Masao 1974: 5), which ensured an equitable distribution of water - even
throughout the difficult dry season (Pike 1964: 95), as well as adequate food supply the
whole year round (Masao 1974: 7). But the success of the furrows depended largely on a
co-operative effort: no single man could operate them alone (ibid: 8), and worked well
under clan management.

However, by the mid 1960s population pressure had begun to exert considerable
pressure on the furrows wafongo - so that a farmer might have access to furrow water
for irrigate his holdings only once a week (Pike 1964: 96). In addition, as furrows were
increasingly drawn upon for a multiplicity of purposes, wastage became a considerable
issue, occurring through leakage and evaporation which with increasing division and
subdivision of irrigation channels, which lead to a decrease in the amount of flow within
the wafongo in general which, with greater abstraction at every level, meant less and
less water available for those living lower and further down the distribution chain (Pike
1964: 95).

The arrival of public standing pipes in the 1970s and later of privatization in the mid-90s
has resulted in the wide-scale reduction in the traditional customs which previously
managed and protected the furrow system by providing the infrastructure for the
maintenance necessary for their upkeep. Extensive neglect of the traditional furrow
system has consequently meant that the number of functional furrows has fallen to 50%
of previous levels (Grove 1993: 434) as village councils decided that their upkeep was
economically unviable (Ibid: 435).

In addition, the very sources that feed the furrows have been under attack over the past
thirty years: land shortages (discussed below) have led to extensive deforestation which,
when combined with elevated rates of land tilling, acts to reduce a rivers catchment and
inevitably the amount of water flowing into a furrow, diminishing their usefulness and
making their upkeep increasingly uneconomic (ibid:8).

Furrows are still used on Kilimanjaro, predominantly on the southern slopes, but to a
much lesser extent than previously
62
, mainly helping to supply the water needs of
livestock, and acting as a general safety net in case of interruption to piped water supply
caused by damage to or theft of the pipe infrastructure (Grove 1993: 440). Within the
context of the eastern slopes, however, and Rombo in particular, the focus of this study,

61
The arrival of Christianity has meant an erosion of traditional beliefs of the sacredness of springs and streams and a
decline in the protection thereof at both a local and governmental level, resulting in abuse and neglect of the furrow
system (Grove 1993: 434)
62
Their continued existence may owe a lot to the interplay of formal and informal rules of usage which allow for
flexibility and sustainability (Gillingham 1999).
47
they were not observed to play any role in providing water for domestic or agricultural
needs.

Contemporary water management

The independence of Tanzania from colonial administration lead to an extensive drive for
modernization at a variety of levels, including water supply and management. A
programme of developing community standpipes began in 1966 (White and Bradley
1972: 81), which in Kilimanjaro often drew on the empirical topographical knowledge of
the Chagga whose furrow system was able to use both terrain and gravity for optimum
water delivery, with modern water pipes often laid side by side with extant furrows.

By 1971 the Nyerere government began to promulgate a policy to reduce water
collection efforts by promising access to safe water within 400m
63
by 1991 (Therkildsen
1988: 27) However, sometimes this top-down policy was poorly received principally
because of a lack of consultation by water policy makers and the ultimate end-users
which resulted in a mismatch in expectancies, usage patterns and levels of satisfaction
64

Consequently, many of the standing pipes were sabotaged either through theft or illicit
irrigation, or were allowed to fall into long-term disrepair. The early improvements in the
water delivery infrastructure often floundered because of inadequate funds, poor
management and various levels of corruption (Grove 1993: 447) which meant that
Tanzanias experience of government-run water services has been similar to other
countries across East Africa who have also seen a decline in the reliability and quality of
service of piped water sources over the past three decades (Thompson 2001: 53).
Furthermore, some analysts have voiced skepticism that the problem can be solved by
the intervention of outside donor agencies and governments contributing sizeable
monetary, material and human resources since this do not add any extra sustainability to
the infrastructure (Therkildsen 1988: 5).

The Kilimanjaro piped water infrastructure was privatized at the beginning of the 1990s
with a new company, Kiliwater Ltd, responsible for water delivery and supply. Although
technical assistance and start-up capital was provided by a German company, Kiliwater
was marketed ingeniously as a community-owned resource, and stakeholders (i.e.
villagers) were required to make a one-off investment of 1000 Tsh
65
.

The slow decay of piped water infrastructure and the decline of the traditional furrow
system has meant that those who have borne the brunt of worsening water supplies
have been the end-users average villagers - who have had endured reduced and
irregular supply, stoppages, corruption and competition between villages for access to
piped supply. Kiliwater claim that they are striving to improve the situation, making
material infrastructural investments
66
, pleading climactic and terrain limitations
(inadequacy of gravity-reliant mechanisms) as well as discounting claims that its agents
are responsible for corruption that such actions are conducted illicitly by water
engineers from other districts/companies who seek to capitalize on their technical

63
Maximum walking distance between a water post and a water users home.

64
Top-down and bottom-up approaches to water delivery and management are called turn-key and own-key approaches
within the literature. (Therkildsen 1988: 27)

65
However, interviews with informants reveal little sense of community ownership, more a sense of us and them in the
shape of a distant organization.

66
This claim was contested by an independent water consultant I interviewed in Mkuu, Rombo who asserted that
Kiliwater lacked the working capital for coherent infrastructural improvements.

48
knowledge and expertise to make money. Whether their claims of promised
improvements will be realized is yet to be seen, but in the meantime those who struggle
with gaining sufficient water on a daily basis are those with the least political and
economic leverage, but who pay the highest price in terms of development and health.



The People - considering Chagga identity and culture: a caveat

Whilst people living in the Kilimanjaro region readily identify themselves as being Chagga, it does
not necessarily follow that there is a single, coherent Chagga culture which is understood and
articulated in a homogenous way by those who claim to be of the Chagga people. This, however,
is not the impression given in the ethnographic literature where anthropologists write about the
Chagga as if they were one people united by single, universal culture. When I discussed the
findings of previous ethnographic writings and observations about the Chagga with Mosoi, my
field assistant and others I was informed that they a) didnt understand the linguistic references I
brought up, and that b) they also contested the representation of certain practices as being
familiar to them.

This information encouraged me to recognise that the notion of the Chagga people is one which
is of recent historical vintage, and that prior to the coming of the Europeans, it is doubtful
whether there was widespread recognition of the Chagga as being one people even amongst
themselves. It is quite possible that the inhabitants of Kilimanjaro saw themselves as principally
belonging to a clan/chiefdom rather than as being first and foremost Chagga
67
, a term that may
have been used by outsiders to refer to the collections of chiefdoms rather than by the Chagga
themselves (Kimambo 1996: 75), whose political awareness was bounded by geography, clan
membership and dialect. Many tribes on Kilimanjaro were politically and militarily antagonistic to
each other, speaking differing and not necessarily comprehensible variants of Kichagga.

Given these historical and environmental limitations, therefore, it is not unsurprising that a
multiplicity of Chagga identities co-exist today, sourced from the locale from which they spring,
or that a coherent, unified culture (which if the Germans had not arrived might well have
emerged through an organic process of political unification and consolidation) does not exist.

On historico-political grounds, therefore, previous ethnographic assertions of a homogenous
Chagga are somewhat questionable which disallowed for the formulation of any single political
entity capable of enforcing a standardized vision of what Chagga culture should entail. Add to
this the multiplicity of dialects and the difficulty of terrain, one comes to the conclusion that any
ethnographic discussion of Chagga culture is almost meaningless unless it is made clear exactly
where the evidence is sourced from. Thus I am hesitant to make any statement about Chagga
culture in general, and restrict myself to discussing my observations amongst the Chagga living in
a specific part (Urauri Village) in a specific ward (Najara Ward) in a specific district of Kilimanjaro
region (Rombo). It may be that there are similarities that are shared throughout Chaggaland, but
without further comparative ethnographic fieldwork, this would be hard to establish. I therefore
adopt a conservative position and limit my discussions to what I observed and personally
experienced as occurring in a specific village, rather than generalising for an entire people who
occupy what is, after all, a vast mountainous terrain. For this reason, I consciously limit my
observations as being pertinent to a people occupying a limited and specific geographical area:
that of North-East Kilimanjaro.



67
In Kiswahili the singular form is mchagga; the plural - wachagga
49
Rombo District

Leaving Moshi, Kilimanjaros commercial and tourist capital, one heads northwards to soon
abandon the comfort of the sealed asphalt roads which are replaced by uneven ochre roads that
wind their way from ridge to ridge, from valley to valley. The passing traffic throws up huge thick
clouds of red dust blanketing the lush roadside vegetation, choking the locals as they walk slowly
by the side of the road, mindful of the enormous sacks of maize or buckets of water impeccably
balanced on their heads. We pass village after village, the continual jolting and jarring of the
Land Rover shaking any lingering romantic conceptions of overland travel out of this
anthropologist. Its hard to distinguish the villages at first, but slowly the same motifs appear,
women walking in groups together on some industrious journey; men sitting together in small
groups on wooden chairs; school children shouting and waving as we roar past, a small roadside
duka (shop) selling essentials often accompanied by a lone wooden stall with a handful of
tomatoes or onions for sale - sometimes even avocadoes or bananas; the skin of recently
slaughtered goats drying in the sun.

This is Rombo District, one of the six which comprise the Kilimanjaro Region
68
, which occupies
the Eastern slopes of Kilimanjaro bordering Kenya. Named after perhaps the greatest chief
Kilimanjaro has ever known, Orombo, who at the beginning of the nineteenth century organized
and armed his age-group, leading them into multiple campaigns of conquest so that at the height
of his reign he was master of half of Kilimanjaro (Stahl 1964: 40), the various lands east of the
Kishingonyi hills, subdued and unified by his rule coming to be known by his name.



Fig. 8: Districts of Kilimanjaro

Unfortunately, his empire, like all those built on strong-man politics, quickly disintegrated into
antagonistic factions, and Rombo sank into ignomious decline, eclipsed by other chiefdoms who
were quick to seize the advantages brought by contact with new technologies, showing no

68
The six districts of Kilimanjaro Region are: Hai, Moshi Rural, Moshi Urban, Rombo, Mwanga and Same
50
hesitation in procuring slaves from their weaker neighbour (ibid: 44). Indeed, towards the end of
the nineteenth century Mangi Marealle of Marangu, a master diplomat and strategist, was able to
manipulate his relationship with German officials to further his political aims and eliminate
military rivals. The Germans brought the fractious chiefdoms of Rombo under single
administrative control, yet were content to relinquish day-to-day control to Marealle, who had
won their complete trust.
69
However, Marealles agenda for Rombo was revenge - revenge for the
humiliation wrought upon the people of Marangu by Rombos namesake a century earlier. Rombo
thus became a site of plunder with livestock transferred en masse to Marangu, together with
men, women and children who were often sold as slaves. Those left behind were forced to work
the land or within the households of their new masters.

The effect of these actions was to send Rombo into deeper decline, with those held as slaves in
Marangu finally being freed only with the arrival of the British in the 1920s (ibid: 45). However,
the long-term damage to Rombos development was extensive: politically it was left with a legacy
of suspicion and continual in-fighting amongst local chiefs which inhibited co-operation and
mutual aid; economically, it took some thirty years for the district to begin to recover from the
deep wounds inflicted by Marealle (ibid: 40). Even to this day, though, it could be argued that
Rombo has never fully recovered, its stunted development evident in its poor infrastructure,
widespread poverty and omnipresent threats to person and property
70
.

**

After an hours journeying from Moshi and some 70 kilometers later one reaches Mkuu, the
capital of Rombo district, and location of the Rombo Trachoma Research Project (RTRP), an
organization created in 2000 by the London School of Hygiene and Tropical Medicine (LSHTM).
Mkuu will serve for the duration of my fieldwork as something of an occasional base camp - a
place where I can connect electronically with the various persons interested in this research as
well as friends and family, pick up some provisions as well as collect my thoughts about what the
current weeks insights and observations.















69
These usually involved the betrayal of other rival chiefs through the invention of spurious plots to overthrow the
German presence (Stahl 1964: 44)

70
Car journeys within the region of Rombo are generally planned to conclude before sunset (5 6pm). Travelling after
this time may result in meeting a roadblock and the liberation of the vehicle at gunpoint. Burglaries are also not
uncommon, and can mean bloodshed and death, if interrupted. Inept burglars who are caught may be dealt summary
justice by incensed villagers. One case I heard of in the field was of a burglar who was discovered ransacking a house
who fled the scene. He was pursued by a group of local men, and climbed up a tree to escape. Refusing entreaties to
come down, the tree was chopped down and he fell out of the tree onto the ground. He was then set upon, his Achilles
heels sliced through to prevent his escape, and brutally interrogated as to the identities of his assailants. This obtained,
the villagers then procured a long spear and stuck it in his throat, slaughtering the perpetrator as they would a cow.
51
The fieldsite: Urauri Village

Urauri lies some 45 kilometers due North of Mkuu, a distance which involves negotiating steep
winding roads and frequent reckless traffic.



Fig. 9: Map of Rombo District showing Mkuu and Urauri

Initially, I found accepting Urauri as a village rather difficult since it exploded my preconceptions
about what a village should be like, preconceptions based largely on my previous experiences
within rural England and Spain that is to say, villages that were both compact in size and tightly
woven in terms of interaction amongst its members. To give some indication, Urauri is
approximately eight kilometers wide and three deep and, owing to a slightly irregular shape,
possibly totals an area of about twenty to twenty-three kilometers squared. It is then further
divided administratively into 3 subvillages Kitongoji - whose boundaries are demarcated by
various roads within the village. From left to right respectively the subvillages are: Reha
(pronounced ray-ha), Mhaka (pronounced ma-ha-ka) and Nessae (pronounced nesigh).



Fig. 10: Urauri Village (External and internal boundaries defined by GPS measurements)

52
Even my field assistant, Mosoi, was moved to comment on Urauris exceptional size since any one
of its subvillages was as large as his own village, which was also situated in Rombo. Even using
the main lanes - in effect uneven, boulder-strewn, rutted dirt tracks to travel anywhere within
the village by foot took a considerable amount of time. Travelling within Urauri was therefore
never an easy process - even with the existence of dozens of quiet backwater lanes that
crisscrossed the village, the distance between two points unfortunately usually remaining the
same.



Fig. 11: The road less travelled - a typical Urauri village scene

In addition, the thousand or so households within the village
71
seemed to be organized around
no other principle other than contiguity with each other, each household a collection of family
dwellings surrounded by some arable land (shamba), which subtly segued into a neighbouring
shamba. Fifty to sixty years ago Urauri was a small hamlet consisting of perhaps 200 households,
which essentially were under constant threat from the cattle-raiding Masai until the building of a
permanent police station in neighbouring Tarakea in 1971. But increased land pressure across
the region led to settlers from various parts and various lineages settling within Urauri, following
the classic pattern outlined by Norton of demographic fission.
72
In a sense then the impression
that Urauri is a collection of households thrown together happenstance is perhaps not inaccurate
given the organic process of development it has experienced, which has left it without a clear
geographic center other than the two central landmarks of the wooden building that is the village
office and the local stone church.

My observations lead me to believe that this was a village which lacked a sense of integration as
a community, and that people were more closely bound to their lineage and the subvillage in
which they lived rather than to what was in reality an administrative abstraction. Certainly, I
witnessed several incidents where the local subvillage community responded, rather than the
entire village occasions such as funerals. Even when village meetings were begun, they would
be characterized by call and response type patterns initiated by the Village Chairman,
encouraging villagers present to identify themselves according to their subvillage. Commerce too
seemed to occur in a decentralized, local fashion with various makeshift roadside stalls stocked
with perhaps some beans, candy, toothpaste and other small items; small bars with benches

71
Urauris population numbers approximately 6,250

72
- This is defined as a segment of a community breaking off when the surrounding lands could no longer sustain it any
more. Norton further notes that the needs of farmers and cattlemen who are geographically bounded by having to live
within walking distance of their garden and pastures combined with the nature of land tenure in much of East Africa has
generally acted to encourage recent settlers to live in either small villages or in isolated dwellings on individual farms
while simultaneously discouraging the emergence of towns (Norton 1976: xvii)
53
outside serving freshly brewed pombe from the early morning to late at night were equally
ubiquitous, seemingly on every corner.

Whilst households in Urauri often depend on what they grow on their shamba for subsistence,
shambas are used to produce an income, with crops being grown and livestock raised for sale in
the local small markets. Typically grown are: avocadoes; bananas; beans; cassava; coffee;
maize; millet; lemons; mangoes; oranges; peas; potatoes; sorghum; sunflowers (as edible seeds
and as a source of cooking oil); sweet potatoes; wood from trees. The typical range of livestock
kept include: cows; ducks; goats; hens; pigs; sheep and turkeys. The bulk of most peoples
income in the village is mainly derived from the sale of such crops and livestock, a valuable
commodity sold for breeding or human consumption. However, despite being rather labour-
intensive to cultivate and process, the most profitable crop grown on Kilimanjaro is still coffee,
which is extensively grown throughout the region. After stripping the coffee beans of their
protective husk, the beans are left to dry in the sun and then sorted using a large rectangular
sieve. On Wednesdays, it is a common sight to see men cycling through the village with burlap
sacks filled with coffee beans on their way to the village office which acts as a branch of a district
coffee growing co-operative where the beans are weighed and cash payment made.



Fig. 12: Sorting out coffee beans domestically prior to selling them at the Village Office

In addition to selling produce, adult household members supplement their income through
offering their services as agricultural labour, working in other peoples shambas, or as casual
labour or building projects. There are also individuals who run small businesses in the village,
working as a barber or tailor, a butcher, shop vendor, healer or brewer and vendor of pombe, the
local brew.

Average monthly earnings are estimated at between 15,000 Tanzanian Shillings, (1000 Tanzanian
shillings = US$1) equivalent to US$15 a month, or 180,000 Tanzanian Shillings (US$180
annually. We can get an idea of purchase power compared to earning power by examining some
typical local prices for crops, livestock and imported manufactured goods.

Natural
Produce
Price
Tanzanian Shillings
Eggs 50 per egg
Maize 150 per kilo
Beans 400 per kilo
Coffee 450 per kilo
Hen 3,000 each
Goat 15 50,000 each
Pig 50 70,000 each
Cow 70 150,000 each

54
Manufactured
Goods

Radio 20,000 60,000
Bicycle 60,000 80,000
Motorbike 1,000,000
Land
per hectare 1,500,000

Table 4: Indicative list of local prices

***

Despite the natural beauty and majestic presence of Kilimanjaro which overshadows all human
activities within the village, life for many is undeniably a precarious affair and it is a common
sight to see telephone wires hanging from a pole, useless, their inner copper wires long since
stripped and sold for a handful of shillings, perhaps to feed a desperate family or for personal
economic advancement, or to walk into any number of households and see a collection of small
graves, those of children cut down by any number of diseases, or to hear of yet another funeral
six (an average of one a week) during my stay alone. Electricity is found only in the wealthiest
permanent homes (which can be counted on one hand) and even these are subject to frequent
black-outs, neither are there any telephones to be found, fixed or otherwise. A few individuals
possess ageing motorbikes; Chinese-built bicycles more evident, but the majority of villagers
transverse the village by foot. Most major thoroughfares, therefore, are somewhat busy with
mainly women visible who walk with heavy burdens upon their head
73
. In the morning they can
typically be seen sporting a delicately balanced hoe or empty buckets, later they return either
with implausibly large sheaves of fodder, dozens of planks of wood, filled water buckets or even
30-40 kilo bags of maize all remaining perfectly balanced no matter how imperfect the road
underfoot. Children too can be seen, either walking to school or running some household errand.
Male villagers are mostly seen on bicycles which they sometimes use for domestic chores such as
gathering fodder, firewood, water or food supplies. But the overwhelming impression is one of
incessant female activity driving the village.

But it would be misleading to paint a picture of a village entirely isolated from the larger world.
Lack of telecommunications does not preclude access to battery-run radios which are used to
tune in to mostly Kenyan stations to listen to news and discussions about global affairs. Bars too
are a popular place to meet, the sonic backdrop being a mixture of US Hip Hop, and its
Tanzanian equivalent, Bongo Flavour, with the occasional Zairean Highlife thrown in for good
measure. Western iconography can be found in the most unusual settings: at one wedding I
attended the walls of the makeshift pavilion where the bride, groom, respective families and
honored guests would sit were adorned with various Western posters including Sylvester Stallone
as Rambo and - somewhat surreally - a poster of an Iron Maiden album cover. Western branded
goods are also present within the village: Coca-Cola, Fanta, Head and Shoulders, Dove Soap
which are often wrapped in Arabic newspaper after purchase - another sign of how deeply
enmeshed even a seemingly remote and rural setting is within todays global markets, both
materially and culturally.

Urauris location by the Kenyan border means that villagers can freely enter Kenya, and many do:
some to till shamba land they have rented from Kenyans; others to fetch water (see below), but
equally this means that Kenyans are free to do the same, and it is not uncommon to see Masai
warriors, morani, walking in small groups with a few donkeys within the village en route to
Tarakea market to do business, or Masai women who have come to trade some produce or visit a

73
In one study Chagga women spent 20 hours a week, on average, carrying grass bundles weighing as much as 120
pounds each from the lowlands (Howard 1997: 148).

55
traditional Chagga healer
74
. However, a porous uncontrolled border also means that the very
problems with violence and gun proliferation that afflict Kenya are also beginning to be felt in
Tanzania, Urauri being no exception. Thus by day the village is a friendly, warm place difficult to
traverse without being hailed and greeted a dozen times over; at night, however, Urauri becomes
a different place, with little human traffic except for those returning home from roadside bars
and householders shutting their gates, hiding within their wooden palisades or locked doors from
the hearts of darkness that emerge with nightfall.

Sometimes the threat is external, coming from neighbouring villages or from across the border.
Armed strangers have been known to suddenly descend in the still of the night, sweeping a
household clean of all its material goods, down to forks and spoons, threatening violence if
opposed. Faced with such threats, it was unsurprising to learn that certain households carry
firearms to protect their property and family, since the police in Tarakea were too far away to
respond to any situation should it occur.

However, sometimes dangers come from within the village itself. Street-boys - vijana wa sururaji
may sometimes use the cover of darkness to attack drunken passers-by, beating them and
relieving them of any money carried; sometimes the street-boys look for easy money by burgling
a household, causing havoc, and in some cases coming away with nothing more than beans.
Jobless yet work-shy and needing money to buy alcohol and bangi marijuana some turn to
opportunistic theft, stealing whatever is not locked down and guarded.
75


Apart from armed robbery and associated violence, the biggest fear in the village is theft of
livestock, which represents a considerable investment of time, energy, food and water resources
as well as capital. It would be fair to say that household wealth within the village can be
measured by the amount of livestock kept, which explains why, even though they represent a
health risk and make for a difficult unhygienic environment, goats and cows are kept within the
same household as the family. Sometimes a family might employ a high school student to act as
night watchman, if the livestock are kept separate from family sleeping quarters. Loss of livestock
is thus a source of much apprehension amongst villagers since their loss can represent a major
economic blow, and can be guaranteed to provoke fury amongst those affected.
76


But the threat to livestock does not lie with opportunistic stealing alone; witchcraft against
livestock (and people) is also much feared, and apparently much practiced too. Often the
motivation may be jealousy, jealousy of a neighbour who is perceived to be prospering, to be
progressing materially. Village informants relate that witchcraft very often takes the (classic East
African) form of a naked man banging his buttocks against the household door three times and
then running away if the door is opened. Even if the householders want to give chase, they find
they are immobilized and cant. Often after a household has been hexed, animals may sicken for
no apparent physical cause, and may even die. Consequently, many households call on the
services of traditional healers to counter such hexing. This involves invoking the power of the

74
The Masai women who attend Chagga traditional healers often come for infertility treatment, and may even stay for a
few days within the healers compound.

75
One incident shocked the immediate community I stayed in occurred when a hoe that was used to dig a grave was
stolen from the graveside the day of funeral.

76
I witnessed just how much fury when a young man was discovered to have opportunistically stolen and sold a piglet
that had escaped from a sty. Severely beaten and abused, his armed interrogators attempted to stuff him bodily into a
burlap bag (which they would have carried to the roadside and shot bullets into). Instead they threw a rope over a
wooden beam and strung him up by his hands. He was finally released when the master of the house a cultivated, well
groomed, educated, rich local businessman then proceeding to head butt, punch and kick him for having dared to steal
from him. They left the young man trussed up for the night on a cold concrete floor and then took him to Tarakea police
station the following morning to be charged.

56
household ancestors to protect the house and its inhabitants from future attack. It is an
expensive service (30,000 Tsh)
77

78
, but apparently is effective and protects both livestock and
family members, especially children who are less able to fend off the effects of hexing and may
inexplicably sicken and die, if traditional medicine is not brought in quickly.

Jealousy is such a powerful force that even distinguished figures may not be immune from its
effects. One subvillage leader, an articulate and wealthy man, recounted to me how he was
threatened many years ago in no uncertain terms by his neighbours with death should he
attempt to build a permanent house on his land. Apparently, he thought the threat was real
enough to heed their warnings and continued to live in his wooden house, sinking his wealth
instead into educating his sons and acquiring livestock. Fear and loathing thus constitute
powerful forces on the mountain, and acting to create a climate of wariness and suspicion.
Households are often protected by a palisade which acts to keep unwanted visitors out - even if
invited inside a compound, all interactions with strangers are conducted outside the dwellings, it
being considered bad form to enter a Chagga house. Furthermore, livestock are kept out of sight,
hidden in stalls inside a separate building or the main house itself so that visitors are unable to
estimate the extent of a familys wealth, and valuable items such as stores of grain locked up.



Fig. 13: Palisades keep intruders out Fig. 14: House valuables can be locked away

In addition to fearing theft and the effects of jealousy, villagers in Urauri are also very wary of
outsiders, especially government officials, and are often very unwilling to divulge information,
even when pressed.
79
This affected me directly since even though word quickly spread as to the
presence of a mazungo (white man) in the village, very few knew why I was there and while
outwardly friendly were often reluctant to give directions to anyones household
80

81
, constantly

77
Anti-witchcraft rituals involve the slaughter of a goat and a mixture of prayers to both ancestors and Christian powers
such as Jesus. Should a witch approach the house with malicious intent, the protective hex will act to shroud the house in
darkness and cause temporary amnesia. Witchcraft in Kilimanjaro has a subtext of a fight against good and evil, the
witchcraft being motivated by human jealousy, bad spirits and the devil.

78
1,000 Tsh = approx US$1

79
For example, when the RTRP Land Rover came looking for myself and my field assistant one day in a remote area of
the village, the villagers immediately scattered since they thought the car might contain government officials who were
going to make them sell their land for the weather station based in the village, an indication of the high level of distrust of
the government and its agents.

80
Traditionally, within Chagga culture, it was deemed very unlucky to give out the details of someone who was wanted
and who might later be killed which would bring bad luck to the person who had done this. This would explain why no
one wants to be held responsible for having leaked personal information. Furthermore, a general fear of theft of food
stocks or livestock and a very real fear of being a target of witchcraft (Raum 1940: 110) would go a long way to
explaining why there is a tradition of secrecy and suspicion and of misleading outsiders (ibid: 193 4). Our requests
whilst searching for individuals were thus invariably met with a whos asking? response, and we would have to
57
checking my story or explanations as to why I was there with other people to see whether I was
lying about my purpose there. The result of this was to make ad-hoc interviewing very difficult
we couldnt casually drop in on anyone for interview purposes everything had to be organized
well in advance (even though this meant relying on the village administration in the form of
balozi leaders to communicate on our behalf, which often result in misunderstandings) and I
also had to be very careful to maintain consistency in my explanations of purpose).

Despite the threat of theft and the fear of actions motivated by jealousy, when questioned as to
what the greatest problem was, the villagers consistently stated water represented the biggest
problem facing the village. For the majority, sourcing enough water on a daily basis was a
constant worry and absorbed a significant amount of time, energy and financial resources,
particularly in the dry season when the rain they had previously harvested had been depleted,
standing pipe sources reduced to an occasional trickle, and household water use severely
restricted. The problem of water was seen to be a major obstacle for economic development,
and was also widely recognised as responsible for a wide range of health problems. In addition to
diarrhoea and various skin diseases, many villagers also suffer from eye diseases. Indeed,
according a recent 2002 Rombo-wide trachoma prevalence survey conducted by the RTRP team,
Urauri ranks as the village possessing the highest rate of prevalence for both TF and TI forms of
the disease (See Appendix 3).

Village TF % TI %
Urauri 33.47 7.62
Leto 29.37 3.30
Ngareni 29.05 2.56

Table 5: Highest trachoma prevalence in Rombo district

Source: Rombo District Survey 2002

This level of prevalence means that examining issues of water access and water use within the
village are of paramount importance.


****************************************





laboriously explain why we were looking for such and such an individual. This surprised me since I thought that my
presence as a foreigner would be public knowledge, with people already informed as to why I was there. That this was
not the case may serve to highlight the problem of diffusing information throughout the village.

81
It is also possible that there are illegal activities occurring within the village that people are aware of, and hesitant to
reveal to outsiders. This might explain why contact with officialdom was shunned, and why people were often somewhat
less than open.

58
2 WATER ACCESS

This section will provide an overview to understand heterogeneity within the village with regards
to difference in access to water. Within the Kilimanjaro context, access to water is largely
determined by: the size and growth rate of the population; how much water is available and how
much demand is made on water sources; what kind of infrastructure exists for its supply and
deliver; how ecological and structural limitations are responded to across a variety of levels;
differential access generated by geography; socioeconomic status and family structure, and lastly
household
82
strategies for sourcing water.

I begin my examination of the issue of water access with the question of population growth and
its impact on water resources.

Increasing population: increasing demand for water

Figures supplied by a joint WHO/UNICEF report on water supply within Africa indicates that the
population of Tanzania expanded rapidly during the 1990s, increasing by nearly a million year-on-
year.

Year Total Pop. Urban Pop. Rural Pop. % of urban
water
supply
coverage
% of rural
water
supply
coverage
1990 25,470,000 5,298,000 20,172,000 80 47
2000 33,517,000 11,021,000 22,456,000 80 47

Table 6: Tanzanian population figure and water supply coverage (Source: WHO and UNICEF 2000)

Whilst the majority of the increases occurred in urban centers, more than doubling in size over
the period, the rural population also increased by 2 million. However, what these figures dont
indicate is that the increase in urban population may not entirely stem from additional urban
births, but migration from rural areas to cities in search of employment and a better life. This
may act to obscure an even higher rate of population increase in the rural areas. However, lack
of capital, political transition from a one party state to multi-party democracy and an extensive
water privatization programme has meant that both urban and rural water supply coverage have
remained at 1990 levels despite additional demand from an enlarged user base. This has only
served to intensify demand for the water that can be accessed, and to make water an even more
valuable resource than previously.

Chagga population growth

The last decade has also seen increases in the Chagga population of Kilimanjaro region.
However, this growth needs to be understood within the context of a population that has been
continuously expanding since the 1920s with the economic and material benefits accrued by the
cultivation of coffee as a cash crop after its introduction by Sir Charles Dundas at Kilema Mission
at the beginning of the 20
th
century (Brewen 1964: 115), and a fall in both adult and neonatal
mortality (Koponen 1996: 22).


82
The basic unit of analysis in this study is the household, which is defined here as the bounded geographical space in
which people live and sleep where duties and resources are shared on a routine basis, serving as a focus for both family
and economic life (Anonymous 1973: 178)
59
An increasing population meant that demand for land similarly increased. Just how much can be
seen by examining changing practices in kihamba allocation: that is the Chagga practice of
allocation of land to the landless. In 1925, the average size of a newly granted plot was some
three acres; twenty years later in 1945 this had fallen to one hundred square meters. However,
despite reduced land availability, the Chagga population showed no sign of slowing its growth
curve. On the contrary, the next twelve years from 1945 to 1957 witnessed an increase of nearly
50% (Von Clemm 1962: 32), resulting in population density rising from 600 people per square
mile to 950 people per square mile during this period.

However, in addition to improved material conditions which encouraged the initial increase in the
Chagga population, other factors continue to play a role in its continued rise up to the present
day:

Preference for large families

The first is a stated traditional preference amongst the Chagga for large families.
Certainly, thirty years ago a study indicated that large families were the norm with an
average number of 7.9 births per woman who had completed childrearing (Egeno and
Henin 1973) It is thought that having two children is a sign of deficiency, that something
is not quite right with the parents; that it is more acceptable to have three or even four
children. It is the African way, announced one informant who then told a story of how
one family had only had two children, out of choice, and that when these children had
suddenly died, the parents found themselves in the unenviable position of being unable
to conceive further children, and were thus condemned to an old age of penury, stripped
of the support a larger family would have supplied. To make sure I understood, the
informant added, Children are our wealth, our insurance.

This story can be contrasted by examining another informants account which
emphasizes the social dynamics involved in having a large family. Traditionally, the first
son is said to represent the continuation of the male lineage; the first daughter the
continuation of the mothers. Care of the first and fourth child is given to the fathers
parents, whilst care of the second and third child given to the mothers parents, with the
children being named after the grandparents as reaffirmation of this relationship. The
first four children were not considered to belong to the parents, but to the grandparents.
From the birth of the fifth child onwards are the parents considered to own any
subsequent offspring
83
whilst the grandparents involving themselves in the socialization
of all the grandchildren (Marealle 1961: 56; Setel 1999:35). Whilst this practice seems to
have largely fallen into disuse, this account helps to explain part of the contemporary
expectancy and acceptance of producing a large family, even though this may put the
family itself under considerable economic strain and pressure. However, even this
limitation can be turned on its head: many returning migrant workers, often seek to
prove that they are bwana mkubwa a big man (Howard 1997:68) by the production of
many children (Setel 1999:76), proof of their new wealth and their own sexual prowess
measured in terms of offspring (Howard 1997: 122).








83
See Raum 1940 for further discussion: 156-158

60
The Role of the Church

This traditional expectancy is reinforced by a very strong grassroots adherence to
Catholicism, which is the dominant Christian sect within the region, with up to 95% of
the population following Catholic teachings at village level with mass held once a day
Monday to Saturday, thrice on Sundays.
84




Fig. 15: Village priest performing mass Fig. 16: Detail from interior of house

The introduction of Christianity within the region may have helped to reduce death rates
through the introduction of information about hygiene and nutrition, and may have also
been a factor in stimulating a rapid rise in Chagga birth rates. (Koponen 1996: 35) as
well as seeking to promote the acceptability of reduced birth intervals (Moore 1986: 110,
122) which resulted in Chagga women giving birth every year (Koponen 1996: 34).

Currently, despite the everyday difficulties involved in providing for a large family,
support for large families and ideological opposition to family planning is extensively
promulgated from the pulpit, and the villagers are loath to incur the ire of the local priest
(which carries with it the power to cause death through cursing) by being seen to limit
their fertility in any way
85
.

Decline of traditional birth spacing

An additional factor which may be responsible for a rapid increase in the expansion of
the Chagga population has been the abandonment of traditional birth spacing practices
under pressure from the Catholic church which sought to forbid such customs (Howard
1997: 66), which not only acted as a form of family planning, but also spread the
economic impact of a rapidly increasing family longitudinally, keeping families in
manageable proportions (Setel 1999: 45), as well as helping to maintain population
growth at sustainable levels (Koponen 1996: 33).

84
Most daily activities are usually conducted with some religious reference or appeal: meals are begun with grace; a car
journey with a prayer for safety. At school, children are frequently reminded by teachers to thank Jesus for the smallest
things - asante Yesu); adults comment on gross misfortune as mungo ni kubwa translation: God is mighty/ all-
powerful/works in mysterious ways.

85
Fertility is often contested within households. Whilst women bear the brunt of the burden of child-rearing, they are
sometimes not free to control their own fertility even when they wish to do so (Howard 1997: 121), having to run the
double gauntlet of both husband and church.

61

Traditionally, mothers were prohibited from conceiving until three years after birth or
until the child was able to walk unaided (Howard 1997: 57). Failure to do so was
regarded hostilely since it not only implied a lack of restraint on the part of the parents,
but a fetus was thought to be able to invoke misfortune on a child still breastfeeding
(ibid: 125). Kwashiorkor - kuvimba (meaning swelling) in children is thus seen as a sign
of punishment for a familys non-adherence to birthing norms (ibid: 106).

These practices involved the maintenance of two separate houses within one compound,
the husband eating and sleeping with his father in one house and the wife doing the
same with her mother-in-law in another in a house constructed after the birth of first
child (ibid: 118), with family meals prepared and jointly consumed outside. This
arrangement precluded sexual activity before the child was weaned and the
transgression of these mores was thought to raise the ire of the ancestors
86
. With the
arrival of Christian missionaries, however, concerted efforts were made to eradicate local
superstitious practices which were unpalatable to European sensibilities and religious
beliefs (Koponen 1996: 36)
87


This process continued under the modernizing influence of the German and English
colonial authorities (Setel 1999: 192), but the increasing difficulty of maintaining two
separate households in a cash economy meant that, in conjunction with pressure from
the church, such traditions which encouraged birth spacing have effectively disappeared,
and it is now the norm for husbands and wives to co-habit. However, the absence of any
cultural impediment to a natural reproductive cycle, and increased opportunity for sexual
congress, means that women can fall pregnant more frequently than before, and the
maturational rate of subsequent generations is squeezed into a population bulge which
then goes on to produce its own offspring with similar birth-spacing.



Land pressure, inheritance and migration

We have alluded above to how an increasing population meant that there was less unused land
available for distribution. However, the combination of a rapidly expanding population and
reduced land availability - out of a total of 550 square miles (Maro 1988: 273-282; Setel 1999:65)
- coupled with a traditional system of patrilineal inheritance resulted in the fragmentation of
family land holdings into ever-smaller plots (ibid: 12), which were insufficient to adequately
sustain the nutritional and economic needs of a single family (Arens 1979: 53). The land
increasingly worked ever harder and allowed less time to lie fallow and recover (Spear 1996:
221), resulting in ever-smaller crop yields (Howard 1997: 58).


86
The use of the word ancestor in English is misleading when applied to the Chagga context. Rather than implying a strict
genealogical lineage, the Chagga refer to warumu as departed members of the family who are still intimately connected
to the family (Marealle 1961: 58). They are, as it were, the living-dead who maintain an interest, are able to influence
family affairs, and occupy a functional role as intermediaries between the living and the traditional Chagga conception of
the godhead, Ruwa. (Lema 1999: 45-46). According to the Chagga, mankind is thought to have its origins in Ruwa who is
thought to live in the skies in Nginenyi (Marealle 1961: 57)

87
The influence of Christianity can be measured by the decline in the practice of male circumcision amongst the Chagga
(according to various informants), which traditionally acted to define the transition from boyhood to manhood. A study
conducted amongst Urban women in Kilimanjaro revealed that FGC is still relatively common, but less prevalent amongst
young (> 25 years old) educated Christian women who were 50% less likely to have been circumcised than women over
34 years old, indicating significant changes in attitudes amongst women towards the practice (Msuya 2002: 159-165).

62
Such factors explain why a montagnard people with a preference for the well-watered fertile high
slopes and dependable food supply (Kimambo 1996: 72) should have begun to move to the lower
slopes (shamba)
88
and plains nyika at the end of the 19
th
Century (Setel 1999: 46), areas
traditionally shunned and considered unsuitable for settlement because of infrequent rainfall,
poorer soil - suitable only for drought-resistant crops, bananas and grasses (Howard 1997: 43) -
and at constant risk from attack by Masai cattle raiders (Kimambo 1996: 76), and whose
perceived occupants, watu wa porini backward people of the plains acted as the traditional
object of Chagga humour. However, the need for more lebensraum was intensified by the
pressure for land caused by the coffee-rush in the early 20th century (Moore 1986: 111) which
saw Europeans occupying as much as 30% of the most fertile land in the Kihamba belt by 1913,
and a struggle between Europeans and Chaggas for the right to produce coffee on the mountain
(Howard 1997: 51).

Ecological pressure on natural resources

Whilst migration may ease pressure on the land spatially, it does little to attenuate the
demand for other natural resources, such as firewood and water which rises in line with
an expanding population. Despite Kilimanjaros montane and submontane rainforests
being extremely important water catchments for the immediate population, and as such
designated protected reserves (Bjorndale 1992), they are vulnerable to extensive,
environmentally destructive activities and are the prime target for resource stripping such
as illegal logging (see Appendix 11) This has caused considerable concern at
governmental level because it is unknown the extent to which how this impacts soil
erosion, microclimatic change and precipitation patterns.



Fig. 17: Illegal logging on the slopes of Mt. Kilimanjaro

In addition, movement to lower areas was accompanied by the wholesale cutting down
of trees for the creation of shambas, heightening environmental degradation.
Furthermore, the growth in the Chagga population and has meant that increasing
numbers of people have become dependent on open water sources which can then
become unable to sustain water demand and, in conjunction with climatic change, can
rapidly deteriorate and become exhausted (Thompson 2001: 50).





88
The word shamba has undergone a transformation in its original conception as a term in binary opposition to kihamba,
land located on the middle high slopes of Kilimanjaro, as land located on the lower slopes. Its colloquial usage indicates
that it is used more in the sense of a plot of land attached to the household and usually located in the immediate vicinity.

63
Increasing expectancies

One final point that should be borne in mind about the Chagga people is that their
narratives about themselves often include the word civilized, which is used to contrast
their previous state prior to the colonial encounter. In essence, this word seems to be
used by the Chagga as a synonym for modern, a notion which includes intellectual
attributes such as literacy, access to formal education (elimu), scientific thinking as well
as personal attributes such as being appropriately clothed and being clean (Howard
1997: 77). However, embedded within the Chagga use of the word, civilized is the
notion of a people transformed by their encounter with Europeans, and a people whose
expectancies are now informed and fueled by contact with an increasingly globalized
world. However, as part of a national discourse concerning progress, maendeleo (Howard
1997: 7, 77) which is predicated on a cash economy, the articulated desire to be seen as
civilised necessarily entails imposing an increased burden on their environment in order
to sustain those desires and heightened expectancies
89

90
. This is reflected by the
increase in average daily per capita use which has nearly doubled in thirty years at Mkuu
from 7.8l to 14.2l. (see table 7 below)



Table 7: Thirty year comparison water access and usage across range of indices for households lacking
piped water facilities:(Source: Thompson 2001: 110)



Tanzanias population has increased rapidly within the past decade which has resulted in
increased demand for water, particularly within rural areas, which has not been sufficiently met
in terms of additional water supply coverage. Part of the recent national expansion includes
changes in Chagga demographics. However, these changes can be situated as part of a larger
historical demographic process which has been driven by a variety of sociocultural factors.
Growth in Chagga population has resulted in extreme land pressure, forced migration and an
increased demand for natural resources within the Kilimanjaro region, particularly water.




89
Not all progress and development is viewed positively. The emergence of AIDS within Kilimanjaro region in 1984 was
viewed principally as a disease of development and the disease served to highlight various debates and contested issues
about social change. (Setel 1996: 1169 1178).

90
Post-Socialist liberalisation has resulted in new forms and objects of competition within Kilimanjaro, with the emergence
of new political parties within a multi-party framework, power struggles within the Lutheran church, even within clans, all
of whom are in dialogue with a historical past but at the same time forced to respond to pressing contemporary demands
and major structural changes (Moore, S. 1996: 587 606)

Tanzania-wide
1970
Tanzania-wide
2000
Mkuu, Rombo
District 1970
Mkuu, Rombo
District 2000
Exposure to
education (years)
5.8 8.4 6.6 6.7
Average per capita
use l/day
14.3 23.5 7.8 14.2
Average time per trip
(minutes)
13.9 34.4 28.6 31.2
Average nos of trips
3.1 3.8 3.0 3.9
Reported diarrhoea
incidence during
previous week
13.9 22.2 12.5 18.2
64
Global climate change and water availability

At the same time that the Chagga population has experienced its most rapid growth (from the
mid-twentieth century onwards), the weather systems that deliver the water necessary for life
and continued agricultural activities on the mountain have been changing.

Several elderly informants recount how they had known for years that this was happening, and
that it was the mountain itself that showed this to be true. Many recalled a Kilimanjaro of their
youth whose snow peak resembled a large white cap, but since 1961 this had begun to recede,
leaving behind exposed rock, the first clue to the changes to come
91
.

Shortly afterwards, they began to notice other changes, some subtle at first; a winter which
didnt force them to wrap up so tightly as usual; a rainy season that didnt last as long as the
last, and then changes which have come with greater intensity and frequency the unusual
gradually became the norm: unbearable summers that stretch for months; enjoyably mild
winters; rainy seasons that not only bring less rain, but last only two months instead of the usual
six.

And then there were the maverick years years such as 1984 when the rain never came and the
crops failed or when sometimes too much came, destroying crops before they could be harvested
(Grove 1993: 441; Howard 1997: 100). These, they now know, are the el nio years, and are
spoken about in subdued tones
92
. Whatever the cause, rainfall patterns have become both
unpredictable (Howard 1997: 42; Kabanda 1999: 231 -241) with declining precipitation levels
increasing the burden on mountain water sources already over-stretched and sometimes
completely exhausted by an ever-growing population.



Fig. 18: Dry river bed


91
Using maps dating from 1912 and more recent satellite imagery (1989), the total ice cover on the summit of
Kilimanjaro (the Kibo cone) has been found to have contracted from 12,058 x 10(3) m(2) to 3,305 x 10(3) m(2) over a
period of 77 years, with considerable disintegration of numerous smaller ice fragments of previously large ice bodies.
Given this rate of shrinkage, it is highly probable that the ice cover will have disappeared from the summit in anything
from between ten to fifteen years. (Hastenrath, Greischar 1997; Irion 2001)

92
Anomalies in both the short and long rain seasons have been correlated to climactic anomalies elsewhere on the planet,
indicating that the East African rainy seasons are enmeshed in highly complex and sensitive global climactic relationships
(Clark 2003: 548 555; Kabanda 1999: 231-241; Zorita 2002)




65
Local influences on climate

Agricultural practice and local climate change

In addition to large-scale climate changes, local agricultural practices have also had an
impact on local precipitation patterns. Thirty years ago villagers across the region often
had big traditional trees such as the Mfuruwanji, Mfumu, Mtembwe and Mkuyu (see Fig.
20 below) growing amongst the crops in their shambas. However, following
independence, agricultural officers instructed the villagers to cut down and clear the
trees. The thinking was that these big trees were stopping the cash crops from receiving
vital sunlight, and that if the trees were removed the crops would grow both faster and
stronger. However, what the agricultural officers didnt realize at the time is that these
trees performed a vital ecological function, acting to absorb and retain rainfall locally.
Furthermore, when the trees were cut down it was observed that less rain seemed to
fall, and the land seemed also less able to retain what little rain fell, becoming noticeably
drier. Lastly, whilst it was true that the trees did act to block sunlight and perhaps
constrain the growth of crops, with their disappearance there was nothing left to protect
the exposed crops in the dry season from the twin ravages of drier soil and the
unmitigated power of the scorching sun.



Fig. 19: Mkuyu, a traditional tree

Kilimanjaro: its micro-climates and precipitation

Unsurprisingly, as the continents highest free-standing mountain, Kilimanjaro has a very
wide base, and its slopes are characterised by hundreds of ridges and valleys, each of
which varies from its neighbour, not only in elevation but also climate and precipitation
levels (Kimambo 1996: 73). For example, during my stay in Rombo I observed that it was
perfectly possible for it to be overcast for several days in a row in Mkuu, the district
capital, with daily sudden snap showers, whilst over the same period in Urauri village 30
kilometers away the skies were reported to be clear and the ground ever-thirsty.

So rainfall was seen to vary extensively from area to area around the slopes of
Kilimanjaro, with the southern slopes experiencing the most rainfall of all (Howard 1997:
56). To illustrate, it rained many, many times in my first week in Mkuu but only once
during the seven I was in Urauri. This clearly indicates that geographical location can be
of great importance in determining the amount of rain that falls because this not only
66
affects how it replenishes and feeds the main water sources that serve a locality, but also
limits how much rain can be harvested, no matter the level of preparedness.

Micro-micro climates, rainfall and marginality

In addition to variation in precipitation between different parts of the mountain which
has a bearing on water availability for villages who receive less rainfall, mention must be
made of subtle variations within a specific locale which can be significant for those
affected. For example, within Urauri it was common knowledge that Nessae subvillage
was subject to significantly less rainfall than other parts of the village, a fact deduced by
long-term observation by the villagers. Less rainfall at a subvillage level placed the
people of Nessae in an even more disadvantageous position than their neighbours in
Mhaka and Reha subvillage, and acted to underscore the precariousness of survival in an
already marginal situation. Thus it was common knowledge that Nessae was less
developed than the other parts of the village both economically and materially simply
because making it on that particular land was especially difficult; many had tried and
given up, each surrender and abandonment accentuating the differences between a
subvillage that was advancing, no matter how slowly, and one that seen as going in the
opposite direction.



We have dealt with the impact that an expanding population has had in terms of water demand,
and also looked at how climate changes at various levels can impact precipitation levels which
not only allow for agricultural productivity but also sustain the water sources necessary for piped
water. I now turn my attention to the role the existing infrastructure for water supply, delivery
and coverage plays in determining water access, and the various ways one village has responded
to these structural limitations.



Infrastructural limitations

The problems that Urauri faces in terms of water scarcity lie largely out of the control of the
villagers themselves, and stem from the infrastructural limitations that have meant a decline in
investment and connectivity for rural people across the region. Not possessing any wells or
boreholes within the village itself means that, apart from rainfall harvesting, Urauri is wholly
dependant on external sources for its water.

Whilst the political stability engendered by Nyereres one party system did much to reduce the
prospect of internal conflict and upheaval following independence, on some levels there was a
lack of inertia to implement the necessary steps to ensure that future needs would be adequately
addressed. The legacy of this lack of vision at the level of state planning means that Urauri is not
alone in facing the daily struggle for water.

***

The roots of the current problems lie in the past. Fifty years ago Uraruri was originally a very
small cluster of households, a frontier post at the edge of the Chagga world; more a hamlet than
a village, Water was obtained from the well at the neighbouring village, Msangae, which had
been built by the British colonial authorities. However, as Urauri attracted more migrants who
settled and began farming, the Msangae villagers began to resent Urauri villagers from using
their water and blocked access to the well.
67


Fig. 20: GPS identification of water source situated in Lucia Forest and indicative distance to Urauri village

The first administration, perhaps busy just contending with the task of learning how to run a
newly independent country, did nothing to heed the demands of the villagers for alternate water
sources, and they had to draw their water from the Kishalu area instead of from Msangae.

With the accession of the second administration, the minister for Rombo District, Leonsi Ngalai,
ordered the construction of pipes in 1966 so that water would be brought to the village with the
intention that every villager should fetch water from a public tap. Water was mainly sourced from
the Kamdavi stream in Lucia Forest (See Fig. 20 above), and at the time, the water was
supposed to be for Urauris use only, but soon water began to be diverted by competing villages
before it reached Urauri, who would tap into water pipes to clandestinely irrigate their own
shambas, or draw water off for their own various needs.

Whilst the government was ostensibly responsible for the upkeep of the water delivery pipes that
delivered water and would repair damage when reported, they seem to have been somewhat
laxer about the standing pipes which were allowed to gradually fall into disrepair and disuse so
that out of the thirteen original pipes only four are in current working order (see table 8 below).

In 1987, the problem of illegal diversion of water for irrigation purposes grew so large that the
Village Committee banned its practice outright, but still had to contend with the effects of its
practice elsewhere.

In 1992, responsibility for water delivery services was taken over by Kiliwater, and water was
declared by the Rombo District Commissioner to be a commodity, something which the end-user
had to pay for at a flat rate of 10 TSh per 20l or TSh a liter, which would remain constant,
even in the dry season.




68


Subvillage Name of standing
pipe
Operational Source of tap
Reha Mawene No Msangae
Reha Moruru Yes Msangae
Mhaka Maweni Yes Msangae
Mhaka Tobiasi Marki Yes Msangae
Mhaka Coleman Stephans No Msangae
Mhaka Gabriel No Msangae
Mhaka Reha Primary No Msangae
Nessae Mwingene Yes Moshi
Nessae Ladislaus Salema No Moshi
Nessae Mkindi No Moshi
Nessae Aloyce No Moshi
Nessae Epifani Mtasania No Moshi
Nessae Selestini Mtauli No Moshi

Table 8: Standing taps in Urauri

Kiliwater attempted to improve the water situation by focusing on repairs and improving flow,
something which was of no concern to previous government representatives. While initial water
flow was poor, it briefly improved as Kiliwater installed new practices and updated part of the
infrastructure. But soon people in other villages were diverting water again and the water
pressure fell once again and flow slowly diminished, resulting in slower, larger queues. In 1994
the village formed a Water Committee to represent and lobby for its interests, its main benefit
being that it acted as an official body able to engage Kiliwaters attention. Unfortunately, no
amount of lobbying was able to halt declining water supplies and flow so that Kiliwater began to
implement an allocation schedule in 1996 to ensure that all villages would be served equally,
which while fair only served to exacerbate the pressure to source adequate supplies of water.

Urauris problems with accessing sufficient water can be seen to stem from historical causes: a
legacy of poor planning, lack of infrastructural maintenance, poor management and insufficient
investment.

Issues in supply

Seasonal fluctuation

Under normal conditions, the sources (Kamdavi) from which the village piped water flows
are sufficient to provide some semblance of supply, albeit restricted. However, during the
dry season these sources suffer from being heavily taxed because of a lack of
replenishment during the preceding rainy season and evaporation. Furthermore, Urauri is
in the unfortunate position of being last in the local water distribution network so that
before it even reaches the village taps the available water has already been sourced
many times over so that what should ideally be a gush of water becomes a frustrating
dribble that can take hours to fill just one plastic container.

Demand vs. availability

There are four working taps within a village of approximately one thousand households.
Under ideal conditions, these taps are meant to operate twice a week between the hours
of 9 am and 2 pm, a total of five hours per tap or 40 tap hours per week. If every
69
household was to use the taps, each tap would be responsible for serving 250
households, or 50 households per tap hour. This translates into 1.25 minutes per
household - insufficient time under the best conditions; pointless during the dry season
when tap hours contract to perhaps two per day (10 12 am) and flow slows to a trickle,
or stops entirely.

Subvillage Name of Standing
pipe
Allocation day Times available
Reha Moruru Tuesday only 9 am - 11 am
Mhaka Maweni Tuesday and Saturday 9 am - 2 pm
Mhaka Tobiasi Marki Tuesday and Saturday 9 am - 2 pm
Nessae Mwingene Friday only 9 am - 2 pm

Table 9: Allocation schedules for standing pipe

The system of allocation days as it currently operates fails to adequately serve the basic
needs of the community, its limited flow acting as a structural obstacle to water access
for the majority of households, forcing them to formulate alternative water sourcing
strategies.

Private connections

Kiliwater has established contracts with individuals rich enough to pay for private
household supply. However, these private supplies inevitably mean that community
water is diverted before it reaches the users its meant to be supplying. Furthermore,
there is a great deal of suspicion expressed in the village about the real nature of
Kiliwaters contractual agreements because these same individuals with private
connections are often also private water vendors, and the accusation commonly voiced is
that these individuals are paying kickbacks to Kiliwater engineers to turn the supply on
during the night so that they are able to have water to sell the following morning.

But even these households are not assured of constant supply, often flow can occur for
anything between 1 and 5 hours a day (Thompson 2001: 54).



Competition for water with neighbouring villages

We have seen that Urauri has experienced competition for water in the past, but water is still a
resource that is much competed for, and the village often suffers as a consequence.

Urauris standing pipes are not directly connected to the water source in Lucia Forest. Urauris
water has to travel through other villages first which not only means that these other villages
have considerably better water pressure, being closer to the water source and first in line for
access, but that this translates into the advantage of faster filling times and the ability to source
all the water they need most of the time, whilst Urauris villagers suffer as their supply dwindles
to a trickle, taking large amounts of time to fill a container, sometimes hours sometimes not at
all, if pressure or supplies at source are inadequate.

In addition to the woes caused by being last in a distribution chain, sometimes Urauris access to
these pipes is shut off by individuals in neighbouring villages shut off. This happens for one of
two reasons: either there is an element of jealousy, of resentment stemming from the perception
that Urauri is somehow taking away from the water available in the neighbouring village, or
70
water entrepreneurs in neighbouring villages are trying to control supply and demand a and force
Urauri villagers to come to them to buy privately supplied water. This is unfortunately a regular
occurrence, especially for Nessae Subvillage whose access to standing pipe water is via a pipe in
Kahe, which is regularly closed off early on Friday mornings
93
. This means that Nessaes
subvillage leader has to go to neigbouring Kahe that day to physically turn the water tap back on,
to allow the water to flow, or rather trickle once more.



Responses to water scarcity

Water scarcity affects not just individuals, but an entire community. It is instructive to
differentiate some of the responses it generates to see not only how the village responds on a
formal bureaucratic level, but also informally at different levels within various networks and levels
of organization. Thus we see that there are multiple responses, not just at the official basis but
through various forms of social association locused around a co-operative ethos, religious
affiliation or kinship ties, and then at a lower level there are responses to the crisis at a
household and individual level.

Village level response

There is little evidence of organized effort to provide any major new improvement by any means
other than waiting for government action and supervised from the outside (White 1972: 244).

The village leadership has taken some small steps such as trying to raise collective awareness of
allocation days so that people know when they are at least entitled to water. They have also
tried to engage the villagers in becoming pro-active in immediately reporting damage (natural or
otherwise) to the water pipes, or indications of leaky pipes.

Whilst at one level rhetoric of community ownership by Kiliwater was meant to convey a post-
socialist impression that a privatized asset somehow belonged to the people, some members of
the community took that rather too literally as they plundered bits of pipe, meters and taps as
they saw fit, thinking only of their own self interest and economic progression regardless of the
hardship and disruption this would cause the wider community as water supplies were effectively
cut off until the missing parts could be replaced. Unfortunately, this is an ongoing problem
tackled mainly through village meetings where balozis are instructed to inform the householders
they represent of the importance of not stealing, a message reinforced from the pulpit every so
often.

However, the fact is that Urauri lacks a strong political leadership, able to pro-actively address
problems before they need urgent redress, and unable, according to many informants at a village
meeting I attended, to even translate their promises into concrete results. The problem of water
scarcity is not a new one; however, it wasnt until 1994 that a water committee was formed to
tackle the problem. This delay undoubtedly had repercussions for the village since in the interim
other villages had successfully brought their own needs before the attention to Kiliwater, and the
specific problems that Urauri faced were not addressed.






93
This began in January 2003 and is still an ongoing problem.
71
The post of Water (Committee) Chairman is a voluntary one, therefore there is no economic
incentive for the person holding this post to expend the same effort as someone in a paid
position. After all, every hour he/she spends fighting for the villages cause is one hour less spent
on his/her own shamba, advancing his/her economic prospects. Therefore, it is understandable if
this individual hasnt invested as much effort into addressing the villages problem as could have
been spent.

Efforts have been made sporadically by the village leadership, but the problem is largely seen as
one that is not controllable by them, and they have settled into something of a wait-and-see
attitude, waiting for some concrete action from Kiliwater that will solve the problem in one
stroke, or money from a donating agency - wafadhili - to provide the funds and means to
resolve this situation somehow.

Non-official responses from within the community own key endeavours

Water is essential not only for sustaining individual life, but also for permitting economic activity
and development. Given the failings of the local political infrastructure, it is not surprising that a
variety of non-official responses which attempt to pro-actively respond and find solutions to the
problem of water scarcity have emerged within the village. Here we examine how groups of
households have joined together to form different associations to tackle the problem, pooling
financial resources. We then explore how individual households respond in different ways to the
lack of an adequate water delivery infrastructure.

Kiaranos

A Kiarano is an organization composed of between 30 and 40 households which acts as a
cross between a micro-finance co-operative and a social support network, assisting in
times of hardship assistance with school fees, medical or funeral costs etc. Membership
is open to married couples only and membership fees are paid monthly which are
collected into a giant fund. The exact numbers are not known, but there are estimated to
be anywhere between 150-200 Kiaranos in Urauri village, with sometimes a household
belonging to more than one Kiarano. Many Kiaranos are planning to tackle the issue of
water scarcity by levying 10,000 Tsh from each of its members, which will be held in
trust. This will allow the construction of two lesser quality cement storage tanks (kisima)
or one high quality tank per year at a cost of 4000,000 Tsh (US$400). The name of the
recipients for the cement storage tanks are drawn at random in order to prevent
corruption. One drawback to this system is the timescale involved for every member to
benefit from the scheme. Another factor to consider is that the Kiaranos tend to be
organized according to economic status so that rich households join together with other
affluent households and can afford to raise more capital for the kitty thereby perhaps
responding more quickly to issues of water scarcity, and providing its members with
quicker solutions than more Kiaranos whose membership is less well-off.

Religious Prayer Groups

First founded in 1982 within Urauri, Catholic religious prayer groups, while principally
focused on the dissemination of religious teachings and principles, have also begun to
similarly respond to the problem of water scarcity by trying to raise funds from
participants. These tend to be smaller in size than Kiaranos, and membership is not
predicated on socioeconomic status or the ability to pay membership fees. However,
therein lies their weakness since sometimes participants can refuse to contribute
financially, yet unlike the Kiaranos will not suffer subsequent exclusion, which means that
72
raising funds and implementing development plans can be something of a difficult
process.

Clans

In addition to Kiaranos and religious groups, households are often members of a clan.
These meet weekly and are characterized by extensive group drinking, segregated by
gender. Membership of a clan involves the responsibility to host a meeting and provide
sufficient alcohol for the members at least once a year
94
.

Clans have dealt with the issue of water scarcity by focusing on the issue of rain
harvesting. However, instead of aiming to construct cement storage tanks, the clans
typically have begun at a more small-scale manageable level by requiring their clan
members to contribute one corrugated iron sheet each 5000 TSh (US$5) so that one clan
household in rotation can have its house equipped properly for rain harvesting. However,
clan meetings can devolve into male-orientated drinking sessions, and female clan
members have often set up networks within the clans with the intention of raising
rotating finance amongst themselves for the purchase of 120l plastic containers to store
water within the household.

Womens groups

Within the village there are a few groups of women who have formed self-help
groups/co-operatives. These are a recent development within the village, and have not
stabilized into being a permanent feature of the social landscape. Plans are afoot,
though, to begin saving schemes to fund the purchase iron sheets and plastic containers
for rain harvesting, with perhaps an eye to one day building a cement storage tank. It
remains to be seen whether these groups flourish, and whether they can help ease the
difficulties caused by a water stress. One commentator has observed that such village-
based women's groups can serve as vehicles to enhance social status and economic
position within the framework of Chagga discourse concerning maendeleo.
Mercer further argues that participation in such groups, which can be over-represented
by wealthier female villagers, can act to exclude those who could most benefit from
access to micro-finance: poorer women. The danger of groups which are used to
articulate the Chagga politics of development may be that existing inequalities become
legitimized and further entrenched (Mercer 2002).

Household Responses

Rain harvesting - maji ya mbua

Rain harvesting is extensively used throughout the village as a supplementary source of
water, and works by using corrugated iron sheets lain perpendicularly across the entirety
of a buildings thatched roof. Rain runs down into iron gutters suspended at the end of
the sheets to collect and channel captured water into 60 120l iron or plastic barrels or
cement urns.

Rain harvesting is widely considered to be an effective source of collecting water, one
iron sheet possessing the capacity to harvest at least 40l of water during a shower of
20mm. But despite recognition within the village of its effectiveness as a method, there
was a great deal of variation in rain harvesting facilities amongst the households
observed, with not every household necessarily practicing rain harvesting.

94
Clan members usually bring fold-away wooden seats and calabashes bearing their name to such meetings beforehand.
73

One reason for this variation is economic: covering a roof with corrugated iron can be an
expensive business. At 5,000 a sheet and between 8 10 sheets needed to cover an
average roof, this translates into an outlay of between 40,000 50,000 TSh -
approximately 3 months earnings. However, iron sheets alone are insufficient; gutters as
well as a suitably sized container to store harvested water need to be purchased,
representing additional expenditures.



Fig. 21 Fig. 22 Fig. 23

Various photos of rain harvesting, iron roofing, gutters and storage containers

Furthermore, given the increasing shortage of the rainy season and decreasing
precipitation levels, some informants have been noted to speculate whether this was
rainwater harvesting represented a worthwhile investment; (Drangert 1993: 153). Such
economic obstacles and doubts about its strategic utility has meant that households
rarely invest in making every roof rain harvest ready, which means that only a fraction of
what could be potentially harvested is captured.

Rainwater is principally collected during the rainy season, and not used until the arrival of
the dry season. This last-resort strategy can help to provide an additional source of water
during times of water stress, but two observations must be made about this practice.
The first is that it was rare to see a rainwater container covered. Normally, they were
exposed. Not only does this mean that collected rain water can evaporate more easily,
resulting in a substantial loss of water collected, but also there is a substantial risk of
contamination either by dirt particles, bacteria or mosquitoes
95
, which means that the
water can then pose a significant danger to the health and well-being of the household.

Recycling

Recycling is another widespread household response to water scarcity in the village, but
will be dealt with more fully in Section 3 below which includes it within a general analysis
of water usage patterns.




95
One example we witnessed of mosquito contamination was at Nessae Primary School where the water collection
cement storage tanks were exposed and filled with thousands of mosquito larvae, which represent a significant vector for
malaria transmission in the area.

74
Economic prioritization

One variable in household responses to water scarcity is at an economic level or rather
the extent to which households vary in not only how much money they will allocate to
the purchase of water from public or private sources or to funding rain harvesting and
storage solutions. Most households are said to prioritise their finances for the purchase of
water and will even sell various crops or valuable livestock in times of need in order to
possess sufficient water. However, not all households manage their finances in as equally
a pragmatic way, sometimes mis-spending or allocating resources, experiencing uneven
cashflow, perhaps sometimes not having enough money for the purchase of either food
or water for the household. In these cases sometimes private arrangements can be made
with private water vendors for water-credit: that is, borrow water now, pay later,
though it may be expected that there this particular facility is provided at a premium.

Household finances are traditionally managed by the male head of the household (mzee)
within Chagga culture; women are rarely entrusted with making choices about household
money. Women are supposed to not enquire about their spouses income (Howard 1997:
147), and may threaten their authority by discovering how much they earned (ibid 149).
This puts the female head of the family (mama) at a disadvantageous position if the
mzee is somewhat profligate or imprudent financially
96
.

Seeking alternative sources of water

In addition to home-based solutions, households can also look to external sources other
than standing pipes for their water needs.

Water sourcing strategies include:

Going to neighbouring villages
Going to Kenya
Using private water vendors
Begging for water

The first two choices will be dealt with more fully in the section on water sourcing
strategies below, but here we can explore why private water vendors are drawn upon so
extensively.

Private vendors

Private water vendors which are composed of house-based and peripatetic vendors are
extensively used by villagers when sourcing water, despite the cost being substantially
higher than that charged by Kiliwater. During the rainy season water is sold for between
5 and 10 Tsh for 20l; during the dry season for between 15 - 66 TSh for 20l, a clear case
of profiteering. Despite this elevated cost, they remain a popular option and an important
source of water, particularly since standing pipes dispense water only twice a week and
households need water all-week round.
97
According to informants, they can represent a
convenient way to save both time, energy and, in the long term, money (in the form of
otherwise lost economic capacity) in using their services rather than fruitlessly searching

96
Having said that, it must be recognised that each household must come to its own understandings about finance
control regardless of public norms, and it may be that in private it may be the mama who more often than not controls
the purse strings rather than the man.
97
Some villagers, especially those with large numbers of livestock may have water delivered by car/truck on a bimonthly
basis since their daily water needs (140/day) exceed the delivery capacity of any bicycle-based water vending service.
75
or queuing for water. Furthermore, using their services reduces exposure to danger,
particularly if going to Kenya for water sourcing is an option.

Peripatetic vendors, in the form of young men with bicycles laden with large 60l plastic
containers who travel within the village supplying both regular and opportunistic
customers, are certainly more convenient to use than having to physically travel to the
house of a private water vendor, and return burdened with 20 or more liters. However,
peripatetic vendors have a limited capacity and can only serve a small customer base.
Furthermore, ability to supply is widely seen as generally less reliable than house-based
vendors who, by dint of good water management and rain-harvesting (rather than
having a permanent private water connection) are able to provide a regular supply of
water to a considerable customer base.

Begging/borrowing and water altruism

The flip side of profiteering from water scarcity can also be found in the various accounts
I heard of needy villagers appealing to their neighbours for occasional assistance, and
being responded to altruistically. Sometimes this appeal would be in the form of begging
a simple appeal for water generosity by either children looking for water on behalf of
their school or wajame the elderly, unable to source water for themselves, who may
petition for drinking water, or those with scant material resources from those better
provisioned, with no return for such action.

On other occasions appeals for assistance can be more temporary a household acting
to borrow water rather being given it outright, particularly if a household has had no
luck sourcing water on a particular day and needs to borrow a an amount sufficient for
cooking purposes only, forgoing washing and personal hygiene needs. Such strategies
and small scale interchanges seem to be relatively common, and such strategies of
reciprocity may constitute an intelligent long-term safety net for those who participate
since no household is ever immune from the threat of water scarcity and so lending of
surplus can also work in a households favour over time.



Appealing to the supernatural: Rainmaking

Though not openly spoken about, it is a fact that Chagga children are commonly given two
names upon birth: one a public, Christian name; the other a traditional Chagga name which is
only known to those close to the family, and which often serves to strengthen the relationship
between the ancestors and the living family (Raum 1940: 297)
98
. This practice exemplifies a
pragmatic ambivalence of a people who are devout Christians in public, but whose private beliefs
and practice arouses fierce disapproval by the Catholic Church who warn that those who indulge
in pagan practices consort with the devil (Howard 1997: 155).




98
A Chagga firstborn couldnt be named without the offering of a sacrifice - Ndasoe (Marealle 1961: 59). First sons were
named after the paternal grandfather; first daughters after the paternal grandmother, and the second born of both sexes
after maternal grandparents. (Howard 1997: 127).

76
Hence whilst traditional Chagga religious practices are still widely followed
99
, they are often
surreptitious, clandestine affairs. For example, my host family slaughtered a goat mbuzi - as
part of a traditional ceremony, but failed to inform either myself or my field assistant that they
had done so, even though we were present within the household compound when it happened.

Such ceremonies - tambiko - often involving animal sacrifices occur annually both at a household
and clan level and function to feed, remember, honour and appease the ancestors (mizimwi) who
have the power to intervene in human affairs (Howard 1997: 111). Indeed, they can be called
upon by a traditional healer to help combat suspected cases of witchcraft or even resolve a
mysterious chronic illness. Thus at the height of the dry season (usually sometime between
September and November) when the issue of water scarcity is one that is being confronted by
the majority of villagers, it is hardly surprising that the ancestors should be called upon as a last
resort when all other human agency and efforts have been exhausted. In this case a ritual
specialist, the rainmaker, under normal circumstances an otherwise humble elderly farmer, is
called upon to perform the rainmaking ceremonies which involve the wearing of sacred goat
skins, the intoning of special magical chants and the pouring of some alcohol (pombe) into the
ground by a specially designated tree, with the remainder consumed by those in attendance.
100




Fig. 24: Traditional rainmaker (with grandson)

Should the ancestors (who are thought to live below ground) accept and enjoy the offering, this
will be demonstrated by small insects congregating in the area where the pombe was poured. A
few days later a brief but loud whooshing sound is said to originate from the ground after which
the drought is broken with the arrival of rain. Sometimes, however, the rituals are performed, the
pombe poured, the ancestors sign their acceptance of the offering, yet no rain comes. In rare
cases this is understood as a sign that the elder has somehow been abused, and that no rain will
come until the abuser is identified and offers a goat (mbuzi) for slaughter and apologises to the
elder who then pronounces that the rain will surely come.



99
Entry of a dead parents spirit into the ancestral realm can occur by ritual sacrifice conducted by a son. Such rituals
allow the departed to locate their kin amongst the dead rather than being lost (Marealle 1961: 58) and also serve to
emphasize the Chagga importance placed on (first) male progeny (Howard 1999: 140), and perhaps serving to
functionally explain the widespread practice of such ceremonies despite contemporary discourses emphasizing
development, civilization or progress, maendeleo.

100
A variation of this procedure was related to me by another informant who stated that instead of pombe being poured
onto the ground, a female goat would be strangled, the blood drained and drunk. The eyes of the goat would be left on
the ground by a specially marked tree, and if they had disappeared by morning this would then be taken as a sign of
acceptance by the ancestors who are able to eat and drink human offerings (also noted by Marealle 1961: 58 - 59)
77
While there is a history of infrastructural failures which directly impact the village as a whole, the
village has responded in a variety of ways and at a number of different levels, ranging from
passivity on the part of the village executive to pro-activity at grassroots level, both in terms of
informal organizations as well as household responses.

But again, differences emerge in a households ability to respond which are predicated on its
socioeconomic status: those households which have more disposable income - i.e. either
wealthier or better organised financially - are able to not only join more effective Kiaranos (as
judged by speed of response and results), but are more able to maximize household finances for
the provision of rain harvesting. Within the village, therefore, major differences exist as to how
water is accessed.

What determines these differences is the subject of the next section.

*****

Inequalities in water access

In addition to infrastructural limitations and the varied responses to both institutional failings and
water scarcity, mention must be made of differences amongst village households in the degree of
water access which are predicated on three different factors: geographical location, socio-
economic status and deviance from the average household profile.

Geographic differentials to access

Water availability from public taps varies from subvillage to subvillage. In Reha, for example,
they have been effectively cut off from receiving water supply for the six months preceding the
water study. This is because of the theft of a water meter which meant that no further supply of
water was allowed by Kiliwater until the meter was replaced by the person responsible for the
tap. However, villagers in Reha are able to access a private water vendor within the subvillage
and others in neighbouring villages. Nessae, on the other hand not only has sporadic flow
because of water competition from other villages, but also lacks any nearby private water
vendors, and the alternative, Kenya, is not only fraught with dangers (see below) but is also
physically very demanding (a 3 kilometer uphill return trek laden with whatever water has been
drawn).

As already mentioned, very small variations in a micro-climate can result in uneven precipitation
patterns locally, which can have a substantial impact on a households economic prospects. For
example, since the total banning of using water piped to the village for irrigation purposes in
1987 by the village committee, households have been entirely dependent on rainfall for irrigation
of their shambas. A householder located in Nessae may find that their shamba is significantly
drier and less productive than an equivalent household in neighbouring Mhaka, which can
translate into reduced household income. Furthermore, lower precipitation levels can also mean a
decreased rain harvest, which can detract from the total amount of water available to a
household.

Secondly, a households geographical position within the village itself also merits consideration. A
household located close to a working standing pipe may be able to not only arrive and draw
water earlier than more distant households, thus saving both time and energy, but may be in a
more advantageous position because they may be able to draw water more frequently, and thus
bring more water into the household than a household located much further away from the pipe.
In addition, geographical proximity may also act to leverage a householders time which can be
then spent in economically productive activities rather than in dead time sourcing water.
78

A minor point, but one which merits mention is the observation that geographical location within
the village and distance from the main byways and principal flows of human traffic, may not only
mean increased distance from available water sources, but also reduced access to markets,
centers of social networking, and the passing water vendors who occasionally ply their trade
along the main village lanes. Therefore, one must also consider a households geographical
position and degree of connectivity within a local transport network in evaluating questions of
water access.

Economic differentials

The socioeconomic status of a household is one of the key factors which helps us understand
differences in access to water between households within the village. At one of end of the
spectrum, the very richest houses those of wealthy men wasumba are able to afford a piped
supply, and enjoy almost constant availability of water. Such households are characterized by
informants as possessing:

Large land holdings
Large numbers of different livestock
Quality housing, if possible permanent cement house
An abundance of food and enough money to sustain family needs
High quality, expensive clothes

Furthermore, they might have a house girl or house boy (or both) working in the household,
electricity, motorized transportation etc.,. Such wealth not only allows for the installation of a
piped supply of water with nearly three times increased usage of water compared to a household
lacking such a connection (Fig. 25 below), but also the construction of extensive water rain
harvesting capacity, the building of cement storage tanks and possession of multiple 120l plastic
containers. Access to a piped supply also means that water can be stored and held back in
reserve, and drawn upon during the dry season when water pressure and flow is low, and rain
harvesting impractical, thus making water available within the house when there is precious little
elsewhere, and representing a distinct advantage over households who are unable to afford such
storage capacity.

Unpiped
households
Piped
households
S1
22.4
68.3
0
20
40
60
80
litres per
capita per
day
Household usage of water


Fig. 25: Comparison of water usage between piped and un-piped households in Tanzania.
Source: Thompson 2001: 80
79

Should piped water supply be unavailable, then affluent houses can either choose to use the
transportation available to them, often a motorbike (piki piki), and use this to not only go further
to source water, but also carry more than those without such transportation. Alternatively, they
could arrange for private delivery of water either by car or through a regular arrangement with
peripatetic water vendors. Despite the higher cost, such arrangements represent proportionally
less of a drain within an affluent household than for a less well-off household, and can be better
absorbed as an outgoing expense. However, the ability to have water delivered straight to the
house circumvents the need to spend time and energy looking for and carrying water, and
leverages a households ability to maximize its economic productivity.

Lastly, it must not be forgotten that wealthy households benefit from associating with other
wealthy households and are able to raise more funds within a Kiarano (see above discussion) and
minimize the effect of water shortage during the dry season.



Fig. 26: The top end of the village socioeconomic scale

In contrast, the poor maskini - those with meager material resources - are highly
disadvantaged in terms of water access and are identified as such by informants as having:

Small or no land holdings
Few or no livestock
Poor housing
Not always sufficient food available
Not enough money to sustain family needs
Poor quality clothes often in need of repair

Such households (a substantial number within the village) are placed in a disadvantageous
position with regards to water access since they are economically limited by what they can
produce and sell, may be unable to afford sufficient rain harvesting and storage capacity, may
have to wait longer periods on average for assistance from any Kiaranos they belong to, and
are forced to constantly look for water, which can detract from time spent being economically
productive elsewhere. Such households may also not be able to afford a bicycle (60,000 80,000
Tanzanian TSh (US$60 80) nearly half a years earnings for some), which can restrict the amount of
water available to a household. Lastly, even if a household possesses a bicycle, purchasing water
from a private vendor constitutes a greater proportion of a households income and may be a
factor in limiting how much is bought.
80

At the lowest end of the economic scale are those who live in abject poverty, referred to
derogatively as hohehahe. Poverty is typically characterized as:

No source of income
No shelter (are forced to begging neighbours for shelter)
Little or no clothes (reduced to begging)
No food (reduced to begging)



Fig. 27: The bottom end of the village socioeconomic scale

Such individuals are least able to access water, and are largely dependent on the goodwill of
others, which places them in a similar category as other vulnerable groups (below) and most
likely to experience long-term inequality in water access.

Discussions with various informants and focus groups about why people were poor in the village
revealed some interesting perceptions. Whilst some thought it was a case of individual
responsibility a question of hard work and not being lazy; others indicated other factors as
being responsible for locking people into a cycle of material scarcity and negative economic
development:

Lack of education
Small or no land holdings
Dry land, little access to water
Ill health and sickness
Alcoholism

A lack of education will be discussed in more detail in the social costs of water sourcing section
below but suffice it to say that a lack of education because of a lack of water access can impact
on future opportunities and economic advancement, and thereby act as a constraint to water
access in the future. Clearly, small land holdings act as a constraint to income, exacerbated by an
inability to properly irrigate the land and poor access to water generally. Inadequate access to
water can impinge on domestic hygiene practices and can lead to a higher possibility of ill health
and sickness, which can act as further drains on the household economy, as can alcoholism
(discussed below).

81
One final observation about socioeconomic differentiation: it is possible that the exposition of
these categories can lead to charges that this is too reductionistic a reading, and that would be
partially true. Apart from the very top and bottom socioeconomic ends of this auto-defined scale
which are clearly defined, most households find themselves in an intermediate category, not
necessarily possessing all the attributes of a poor household, yet neither wealthy nor destitute
perhaps a little bit more than poor yet somewhat less than rich. For such households, whilst their
raised socioeconomic status and extra income may provide some assistance to dissipate the
effects of water scarcity, and may even allow some additional access to water supplies, many of
the negative factors experienced by a poor household in relationship to water access are also
applicable to these intermediate or fuzzy households.

Perhaps we should not be surprised by these findings: they were reported in the classic 1972
area study on water usage, Drawers of Water, and its follow-up study 30 years later which both
examined issues of water access within East Africa and came to the same conclusion: that there
was a positive correlation between the wealth of a household and the degree of access to and
availability of water, shown by greater connectivity to the water distribution networks, greater
capture and storage capacity for rainwater harvesting, and superior modes of transportation, and
that households without access to piped water supplies were functioning with quantities of water
(19.7 liters
3
day) which were only marginally higher than those usually used as indicative of basic
human requirements (Thompson 2001: 89).

Vulnerable households

Water access can be particularly problematic for certain households within the village which are
not representative of the average household profile: that is a relatively young (mid twenties -
mid fifty) couple with children who, despite possible geographical and economic limitations, are
physically able to source water. Typically, such households may be composed of an elderly -
wajame - couple who, if they are not connected to a piped supply, are physically limited in their
ability to source water both in terms of travelling distance and physical strength to carry heavy
loads. Often the elderly are found sharing a household compound of their offspring, but there are
households which do not, and as such struggle in their daily attempts to source water, and are
often reduced to begging from neighbours for their basic water needs.

The next vulnerable social group are widows - wazee - and (functionally single) mothers who
have to cope with the demands of running a household without the assistance of a male partner
who, perhaps because of business interests or extra-village employment may be absent (Setel
1999:73), and as such are often forced to cope with a reduced availability of water, increased
demands and no leveraging of time that an extra pair of hands would allow. For a more detailed
discussion of the impact a functional male has on a households access to water, see the
discussion on alcoholism below.
-------

We have explored the main factors that contribute to differences amongst various village
households in the amount of water they can access. We now continue our analysis and explore
the strategies deployed within the village in sourcing and bringing water into the household; how
this effects how much water is brought into the household; the hidden costs and dangers
involved in collecting water, and how problematic drinking can threaten water availability within a
household.
------



82
Water sourcing and water access

For the majority of householders in Urauri sourcing water is a daily challenge. In this section I
will examine the various issues involved in collecting water and the strategies deployed.



Fig. 28: Filling up at a standpipe

Who collects water in the household?

Adults

Although there is, unlike neighbouring ethnic groups like the Kikuyu (White 1972: 62), no
strong cultural prejudice against men sourcing water, the collection of water is widely
described as womans work by various informants. As such, the responsibility falls on
the mother of the house to ensure that there is sufficient water within the household
(Thompson 2001: 59), and indeed it is a common sight within the village to see women
balancing their precious cargo on their head as they return from water collecting.

However, if a household has any means of transportation such as a wheelbarrow or,
more commonly, a bicycle, then very often this will be used by the father. Sometimes the
father takes over sole responsibility for collecting water, freeing a mother from the
onerous burden of having to physically collect and transport water over long distances,
and leveraging her time so that she can be otherwise economically productive. However,
sometimes a mother may still physically collect water since, even though this is a drain
on personal resources, it translates into more water available for domestic and
productive use.



Fig. 29: Mother and son returning from water collecting in Kenya, exhausted.
83

Children

In Drawers of Water 2 it was observed that children have increasingly become involved
in the drawing of water over the past thirty years (Thompson 2001: 59), a phenomena
found equally within Urauri village.

Though only able to physically able to carry small amounts, usually in the form of small
jerricans, children are commonly viewed as being able to make a small but useful
contribution to household water collection efforts. Sometimes children go together with
their mothers, often standing guard over a couple of large containers that a mother may
have left in a queue, filling one up first and leaving the child to fill the second which she
will return for. Sometimes small children go in groups, helping each other carry heavy
pails. This is more commonly seen when a school may often sends its pupils out to look
for water because there it has no water to prepare lunch with or clean classrooms with.

As children grow they may be given the task of using a wheelbarrow or bicycle to fetch
water, a task mostly given to boys, but occasionally to girls. So children have become a
valuable resource in bringing water into the household, and the more they are involved
in the process, the more that can be brought into the house. However, there are costs
involved in the collection of water, and children are as equally, if not more, affected by
these costs than adults. These costs will be examined in detail below.


Fig. 30: Village childrens drawings of themselves carrying water

When

Water collecting is usually an activity performed in the morning, sometimes as early as five. Very
long queues
101
can rapidly form at standing pipes on allocation days so those who get there
earliest can save time which would otherwise be spent queuing, and can therefore have get back
to the household in time to still be economically productive.

Most informants interviewed gave a start time of between six and eight in the morning, with
typically three to four hours (including travelling, sourcing water, queuing and collecting sufficient

101
Queues are generally orderly affairs, and are overseen by a money collector who helps keep the line in check and who
is also responsible for the good working order of the standing pipe (see Appendix 12). Individuals are allowed to fill one
60l container per turn, but are free to begin queuing again if they want more water. Some villagers may also arrive early,
leave their containers in line with a child and return a few hours when they calculate their turn has finally arrive, thereby
freeing up a certain amount of time to carry out other activities.
84
water). Early morning starts can be advantageous if travelling to Kenya since this reduces the
probability of conflict with local Kenyans over water (see below for further details). Children who
collect water often do so before school, often completing water collection after school has
commenced, sometimes as late as midday, thereby missing the bulk of a days lessons.

Those with bicycles can also visit private water vendors or other alternative sources easily
throughout the day since journey times become manageable, even if the queues remain the
same. Some people use the afternoon as a time to source water, perhaps to avoid clashing with
local Kenyans at midday if they choose to go to Kenya to collect water, but the likelihood of
sourcing sufficient water from standing pipes at this time is slim, given that the pressure flow
may have dropped to a trickle or have disappeared entirely, making using a standpipe a highly
frustrating experience. Equally, by this time private water vendors may have sold out or
exhausted their supply of harvested rainwater, though somehow miraculously this always
manages to replenish itself the following day.

The most important point, as informants were keen to make known, is how time-consuming
water collecting is as a process. In the worst cases, such as travelling to Kenya could take up a
significant part of the working day, involving anything between six and nine hours roundtrip.
Needless to say, water collection is in the majority of cases a once a day occurrence, though
those with bicycles may sometimes be able to visit a private water vendor more than once.

The transportation of water

Most water is collected by foot, an arduous process which involves traversing uneven ground
with heavy loads for long period over great distances. The drawback to collecting water by foot is
the time it takes, the amount that can be carried (usually 20l for adults, anything between 4 and
10l for children) and the physical expenditure invoked (see hidden costs below)

As mentioned before, some households possess various means of transport: some possess a
hand-driven cart which can be useful in transporting large amounts of water, but is limited in that
it is primarily akin to collecting water by foot in its speed, and the fact that the cart has to be
steered along difficult road conditions.

Using a bicycle is equally effective in bringing greater amounts of water into a household, and
can represent a great saving or leverage for time resources. Additionally, possessing a bicycle
can mean that water sourcing activities can cover a wider geographic area than by foot or cart,
and that multiple journeys are a distinct possibility, with children also able to use bicycles for
water sourcing. The only drawback to using a bike is the very bone-shatteringness of the
experience as it clatters and jars along unyielding roads, as well as the difficulty involved in
continually navigating steep gradients laden with a heavy payload of water.

85


Fig. 31: Collecting water by bicycle Fig. 32: Hand-driven cart

The most advantageous, if perhaps one of the most inaccessible to the majority of villagers, is a
motorbike, piki piki which travel greater distances, further, faster and perhaps able to transport
greater amounts of water than other modes. The disadvantage of using this form of transport is
that the roads in and around Urauri can be very tough on motorbikes (some riders were very
hesitant, loath even, to go on certain roads for our GPS purposes and had to be cajoled into
doing so). The cost of maintenance and repair to damage caused by poor road conditions as well
as the costs involved in running the motorbike are factors that need to therefore be factored into
the end cost of water collected for a true reckoning.

Factors in sourcing water

Distance

In searching for water, one of the key criteria considered is distance to source. How far is
the water source from the household? However, a source that is located relatively close
by may also be very popular and, unless one can get there early enough, may prove to
be over-subscribed and supplies exhausted. So distance has to be calculated against a
sources perceived level of popularity. A source may also be close by but the gradients
involved may cause individuals to reconsider whether they want to use it since it may
limit how much water can be physically accessed (White 1972: 135).

Quantity

A source may be easy to access, but may be undesirable as an option because of the
quantity of water available. With standing pipes providing small amounts of water, a
householder may choose to avoid using them and go to where they know there will be
larger amounts available, even though this may mean going further or paying more.

Quality

One of the major drawbacks about using Kenyan water sources is that the water is
considered to have a high mineral content, smells (of copper), producing a heavy and
uncomfortable feeling in the stomach when drunk. However, this did not seem to be a
major consideration for water sources elsewhere, and neither did it necessarily inhibit
villagers from going to Kenya to source water. Unlike in other parts of Africa where locals
86
have to contend with brackish or contaminated water, water quality seemed to be a non-
issue: it was generally accepted that locally sourced water was not only potable but was
definable as maji salama, clean water devoid of impurities (see table 10 below for
classifications)

Safety

This is a major factor when visiting Kenya and is discussed below in detail.

Water Sourcing Strategies

Whilst some people may choose to use the standing pipes in the village on allocation days, other
people circumvent them entirely, reasoning that it is better to either search for water in
neighbouring villages such as Kahe, Kibaoni, Msangae, Nayeme, or Tarakea on their allocation
days where the likelihood of finding and being able to collect sufficient water is higher. Such a
strategy involves visiting each village in turn until water is found, which may mean queuing for
long periods only to find that water has run out, and having to start all over again. As such,
water tracing in neighbouring villages cannot be considered a highly effective water sourcing
strategy.

Some people dispense with water tracing strategies (going from source to source seeking water)
entirely and go straight to private water vendors either in Urauri (some in Mhaka and Reha) or
further afield in neighbouring villages where the likelihood of being able to source sufficient water
is very high. However, the problem with the most reliable private vendors is that their reputation
as a good source means that they can be over-subscribed and queue lengths can equal those at
busy standing pipes. Occasionally, private water vendors sell out of water, leaving many people
disappointed and frustrated, especially if they have travelled far or waited long in line.

It is not surprising, then, that many villagers decide to simply head off to Kenya where the water
sources are perennial and guaranteed, making the journey worthwhile despite the distance and
possible risk to personal safety. Furthermore, those people who choose to go to Kenya may
choose to use Mnuyura
102
which is an hour further away than Ndarara
103
and Lasti
104
but less
busy so that despite the additional hour travelling time, villagers may actually save time overall in
not having to queue so long.

In the dry season, water availability drops throughout the region, and water tracing in
neighbouring villages largely becomes an exercise in futility. Water sourcing then becomes a
choice between private water vendors or going to Kenya.

However, having said all the above, it must be said that geographical location within the village
does play a significant role in determining what water sourcing strategy is followed. For those
living in Reha, for example, it is much more practical for them to go to Msangae to trace water or
purchase from a private water vendor than it is for them to journey to Kenya which is a
considerable distance from the subvillage. On the other hand, for those living in Nessae, it would
be absurd for them to go to Msangae when they know they can probably find water in Kahe
which borders Nessae or make the trek to Kenya where they definitely know they will find water.





102
A stream.
103
A deep borehole made by the Kikuyu who charge for water drawn.
104
A stream
87
Hidden costs of water collecting

Drawers of Water and the follow up study both observed that collecting water as a regular
activity often involved hidden costs which could mean financial losses in the present, or at a later
stage, as well as less tangible effects such as diminished well-being (White 1972: 17; Thompson
2001: 67)

Economic advancement

Time and energy given to sourcing water has a tangible effect on the economic development
of a household, and the larger community as well.

This is because time spent journeying, searching, queuing and finally collecting water means
that this time is not available for being economically productive and that there is less
opportunity for income generation. When the time, energy and lost potential income involved
in collecting water on a daily basis are factored to give a true price for the water collected
resource-poor households are found to pay a much higher price for their water than
wealthier households
105
.

Furthermore, wealthier households with superior water access capability, greater access to
water storage capacity and more funds to purchase significant amounts of water are at a
significant advantage in comparison to households which lack these resources. By having a
lower proportion of their resources (i.e. time, energy, money) tied up in meeting basic
survival requirements, the remainder of a wealthier households resources are thus
significantly leveraged, leaving disadvantaged households who are unable to marshall their
time and energy resources in this way or use their financial resources to achieve similar
economies of scale further disadvantaged in economic developmental terms, as wealthier
households steam ahead whilst poorer households flounder in their wake.

Health costs

The roads in Urauri are unsealed, uneven dirt tracks which are physically demanding to
negotiate, with constant ups and downs and places where one might stumble. Add a 20l
bucket which needs to be perfectly balanced on ones head and a roundtrip journey of at
least two hours, and we can begin to see that daily repetition of this physical strain on
the body might easily explain why women who mainly shoulder the burden of water-
collecting complain of often getting headaches and pains in their chest, neck, back and
waist, and of feeling very fatigued. Children who collect water are less used and skilled to
balancing and traveling at the same time and carrying heavy water loads can inflict the
same kind of physical stresses, perhaps worse given their physical immaturity and the
impact on their developmental processes.
106

107

108


105
This difference may be even greater when considering the difference paid for water by those connected to the water
grid and those who remain unconnected. According to Michael Klein, Vice President of Private Sector Development, World
Bank, a lack of connection may translate into having to pay ten times or more for the same amount of water than those
who possess a connection. Source: http://news.bbc.co.uk/go/pr/fr/ - /2/hi/in_depth/2957398.stm

106
Children reported that, in addition to the physical strains of collecting water, their hands ache from holding the
container atop of their head for long periods. Sometimes they would tumble on the uneven dirt path, and drop their
container which could be distressing since not only had they spent a lot of time and energy collecting and carrying the
water, but they now had nothing to show for it.

107
The physical energy spent in collecting water can be expressed in terms of energy calorific expenditure (White 1972:
94)

88

Social Costs

There are two main social costs involved in collecting water: diminished child care and
compromised educational access and achievement.

o Child Care

Mothers who leave the household to collect water may have no other choice but to
leave the youngest children alone for long periods of time. This means that they may
be exposed to a wide range of pathogens and hidden household dangers (wood
splinters, exposed sharp and rusted nails etc.,) and potential accidents that a mother
who was present might be able to identify and prevent. At the very least, lack of
parental supervision poses a significant health risk. Equally too, it can also can that
young children spent long periods of time grossly under-stimulated which can impact
on their cognitive development.

o Education

As mentioned above, children often collect water in the mornings, often frequently
missing a significant part of the school day. In addition, each year grade is asked to
fetch water once a week in order for the cooks to be able to provide school dinner,
but if they cant find water at home, or cant borrow from neighbours, the children
often dont return to the school for fear of being punished, further missing out on
class time, and valuable learning time. If they attend class after collecting water the
children are sometimes so tired that they cant listen and pay attention properly, and
may even fall asleep in class on occasion.

Being disadvantaged in terms of water access can mean that children are enlisted
into a households daily efforts to secure enough water. Whilst there is nothing
unusual within the village in children participating in household activities (see figs. 33
34) such as bearing message, gathering firewood, sweeping a compound, herding
animals (Howard 1997: 142), water collecting can mean missed educational
opportunities and lower academic performances which can translate into
compromised future opportunities and impaired chances for economic advancement
(Howard 1997: 59). These potential disadvantages are not necessarily faced by
households whose children are spared the burden of having to regularly collect
water, and are consequently able to enjoy the time and energy to attend all the
classes and focus fully on their studies.

Education is seen by the villagers as a key for advancement since it gives access to
knowledge, information and the capacity to think and plan ahead. A household which
has no choice but to send its children out to collect water is a household which must,
of necessity, factor the potential loss of education into the true cost of sourcing
water.


108
It is not unusual to see numerous examples of stunted children within the village. I met several boys who claimed they
were 15 or 16 years old whereas I thought they had the physical attributes of a Western 10 year old boy (Howard 1997:
10). Malnutrition is a serious problem for the Chagga on Kilimanjaro (ibid), and much-needed calories consumed in
collecting water may not be available for developmental purposes.
89


Fig. 33: Collecting Fodder Fig. 34: Threshing beans Fig. 35: Water collecting



Fig. 36: Queues are usually lengthy

Dangers of sourcing water

Neighbouring villages

Extortion

Sometimes those with a private supply of water in neighbouring villages use their power
as leverage to force needy villagers (usually female) to work for them in their personal
shamba. Even though money is offered for the water, such vendors see economic
advantage in having a captive labour force able to work in return for water. This is an
additional unwanted cost since in addition to the time and energy expended in walking,
collecting and transporting water, they have also expended additional personal resources
in labouring for someone else. This means that they lose available time to work their own
shambas, and may have decreased energy to do so. This means that individuals
advantaged because of their access to water are able to freely exploit those without, and
this has a direct economic impact on those exploited, compromising their ability to be
contribute to the household economy, compounding any economic difficulties they may
have.
90
The road to Kenya

Wildlife

Urauris proximity to the Kenyan plains is not without its own dangers. It is not unheard
of water collectors to encounter wild animals as they travel, especially early in the
morning or late in the afternoon. Whilst buffalo (nyati) are rarely dangerous on their
own, a herd of wild elephants (tembo), however, have been known to panic and
stampede if they come across humans without warning or are disturbed suddenly, and
this might result in unintended injury. However, one animal that water collectors are
keen not to run into under any circumstances is the lion (simba) which has been
observed searching for water either singly or as a pride during the dry season, and have
been known to have been sighted within Urauri village itself.

Competition for water

Going into Kenyan territory sometimes leads to direct conflict with Masai herders over
water access. Since it is located within Kenya, the Masai view the water as theirs and the
Urauri villagers as unwelcome intruders. Thus at the Ndarara stream they will order them
to leave immediately once they arrive with their cattle. Sometimes, however, the villagers
have not completed filling their containers and have continued to draw water, and have
been attacked. In one incident, a villager ignored the Masais instructions to leave and
was attacked from behind and hit on the side of the head with a Masai club (Rungu)
which perforated his ear drum, resulting in the loss of hearing in one ear.



Fig. 37: Masai warrior (Moran) -
Not always the friendliest welcome for water seekers

One response the villagers have devised is to try and avoid being present when the
Masai typically water their cattle at midday. This means arriving very early in the
morning, or in the late afternoon.







91
Extortion and profiteering

Previously, water was free to Kenyans who crossed over into Tanzania, but it is now
metered, priced and controlled by Kiliwater; in turn, Kenyans sometimes make
Tanzanians pay for the water they take from Kenyan sources. Whilst it is prohibited to
draw water directly from Ndarara stream, there is an unsupervised tap point that anyone
is free to use, and is used in preference to paying for water which is officially charged for
at the Lasti source. However, occasionally Kenyans force villagers from Urauri to pay for
the water they collect, and threaten violence if they dont.

Robbery and assault

Villagers often go to Kenya to buy small items for resale back in Urauri. However, even if
traveling in a group they may be ambushed by Kenyans waiting in the maize field lining
the route, and threatened and attacked by as many as five or six assailants armed with
knives and sticks. The group will disperse and attempt to flee, but some are sometimes
unlucky and are caught and set upon, and later found by villagers who return when the
coast is clear, lying injured and robbed of their goods, money and any mode of transport
they were using usually a bicycle. In one case, an individual ended up with stab
wounds to the head and had to be rushed by motorcycle to Haruma hospital an hour
away for emergency treatment.

Threat of rape

Since 2001 up to the present time there were a spate of rapes near Ndarara. The culprit,
a giant Masai by the name of Ngilisho would catch and terrorise Urauri women. He was
finally captured by a combined mob of Chagga and Masai concerned about this rogue
element, and taken to the Kenyan police. However, his example has since been copied
by other Kenyans, notably groups of Kikuyu street boys (vijana wa sururaji or vijana wa
mitarni ) who follow women into the bush as they return to the village and wait until the
women reach an isolated spot, order them to lie down and then gang rape them. Rape is
seen by the Chagga as a source of great shame, and a woman may go to great lengths
to hide the fact that she has been raped from their husband, fearful that she will be
blamed and that her husband will leave her. However, sometimes rape results in
pregnancy and the resulting offspring is known locally as a mtoto wa maji or water child.

Abduction

In addition to the fear of rape, there is considerable anxiety amongst female water
collectors of being abducted by the Masai who have been known to occasionally abduct
women from Urauri to serve as unwilling first or second wives. According to Masai
informants, these women are given a traditional Masai amnesiastic herbal concoction
which causes them to forget their previous lives.

Responses to the dangers of water sourcing in Kenya

Clearly, crossing the border involves various risks to personal security regardless of whether one
is male of female, and this has resulted in a strategy of meeting at various times during the
morning at pre-arranged spots and waiting until sufficient numbers (five or six) have gathered to
reduce the risk of making an unaccompanied sortie into Kenya.



92
Threats to water access: the spectre of alcoholism

Before concluding this overview of the multiple factors that are implicated in determining access
to water, mention must be made of the role that alcoholism within a community can play in
possibly limiting access to water at a household level.

Urauri village possesses both an ambivalent (O Connor 1975; Howard 1997:153) drinking culture
where the consumption of alcohol acts as a focus of communality
109
(Heald 1989: 7), and is a
principal feature of most social occasions, be they post-baptism celebrations, weddings or
funerals.



Fig. 38: Drinking to celebrate: a village wedding Fig. 39: Drowning sorrow: a Chagga funeral

In addition, regularly scheduled social gatherings such as Kiarano or clan meetings are organized
with drinking as the principal activity, and participants can consume large amounts during the
event, returning home inebriated.

Whilst drinking is segregated along gender lines at such events, with men and women occupying
specific, separate seating areas, women can often drink as much as their male counterparts, and
with equal gusto. Outside these arenas, however, male drinking can become somewhat
problematic. It is not unusual to see groups (vikao) of men sitting at one of the many small bars
that dot the village landscape as early as 8 in the morning, a plastic cup or calabash filled with
pombe, mbege or busa in their grasp
110

111
. Whilst these shops or bars can serve a positive
function (Hendry 1994: 184) as bastions of customary and community life (Howard 1997: 228),
allowing contact with the outside world in the form of a radio, and opportunities to share
information as well as mutual support in difficult times (Howard 1997: 163; Peace 1992: 174),
they can also encourage more negative attributes: lethargy and inactivity. Sometimes upon my
returning home at lunchtime, the same group would be observed to be still sitting there, in an

109
Alcohol can serve as a symbolic instrument connecting the world of the living with the world of the ancestors whose
reported enjoyment of alcohol is acknowledged in Chagga rituals which offer alcohol to them by pouring it into the
ground.

110
There are two kinds of pombe: one made from bananas and millet pombe ya ndizi; one made from maize pombe
ya mahindi. Busa is made from fried flour, water and filtered millet. Gongo is an illegal distilled spirit made from molasses
from sugar cane.

111
The consumption of particular kinds of alcohol can serve to differentiate social status within Chaggaland: bottled beer
is regarded as a rich mans drink; pombe is for the average Joe and gongo for the poverty stricken.
93
even more inebriated state, clearly having drunk throughout the morning. And occasionally male
householders could be seen surreptitiously drinking from small flasks containing stronger stuff
such as gongo a potent alcohol distilled from molasses.



Fig. 40: Males drinking at a bar

Such behaviour is considered problematic by those within the village, who underscored exactly
how problematic by their initial reluctance to talk candidly about its very existence and its effects
within the community. However, persistence with informants and the experience of various
household observations allowed me to unpack some of the unease and ambivalence (Willis 2000:
9) felt about alcohol consumption, which on the one hand is widely condoned and practiced,
even by the priests, yet develop an understanding of how the community viewed this behaviour
as problematic drinking, and how this could and did effect the functioning of a household,
especially with regards to water-sourcing.

Lower wages (Howard 1997: 165) and decreased opportunities for day labouring (kibarua) are
correlated positively with increased problematic drinking amongst men (ibid: 228). In some
cases, it can be a way of coping with the anxiety of economic insecurity and potential economic
failure (ibid: 162/164). Recourse to alcohol acts thus as a salve for such woes: After a hard
days work in the field pombe makes the pain go away (ibid: 158). Sometimes, however,
money that is reserved for important domestic purposes, such as bride-wealth payments, can be
diverted and spent instead on alcohol.

A husband that begins to drink early in the day is a husband that is an economic drain. Not only
are they not available to work the shamba, they are also not bringing in cash to the household
economy through their labours elsewhere. Furthermore, a pattern of consistent and daily drinking
can soon amount to a large outflow of funds from a household. If a household is already in a
marginal economic position and the husband is not actively contributing to the family economy,
then such behaviour can mean that a mother who has to cope with the running of the household
but who does not have control of the finances (this is predominantly a male domain) may have to
make sacrifices within the household budget. Sometimes it might mean fasting for a day because
there is no money for food (Howard 1997: 159), and sometimes it might mean that there is no
money available for buying water the next day.
94
Chagga fathers' drinking habits
38
16
34
62
28
22
0
10
20
30
40
50
60
70
Neighbouring
father
Father of
malnourished
child
%
Excessive
drinking
Moderate drinking
Drinks a little or
none


Fig. 41: Chagga fathers drinking habits. Source: Swantz et al. 1975

A 1975 study of the relationship of drinking habits and malnutrition in Moshi indicated that there
were differentials in the parental profile of malnourished children. Fathers who drank excessively
were more likely to have malnourished offspring than those who didnt drink as much, and those
whose children were not nutritionally deficit were twice as likely to be moderate drinkers (Swantz
et al. 1975) who were more than twice as likely to have fathers who drank excessively, than
children. This suggested that alcohol consumption played a significant role in adequate nutritional
provision of children.

Furthermore, it is important to realize that an economically inactive and absent husband is
unavailable to assist with water collection, and this absence represents a loss of access to water,
not experienced by households where drinking occurs within socially acceptable parameters, and
where both husband and wife work together synergistically in organizing and running various
aspects of a household, including the sharing of water sourcing responsibilities. We have seen
above that if a household possesses a bicycle and a husband uses it in collecting water, this can
act to triple the amount of water normally available to a family in one journey.

However, this picture may seem to suggest that men are not the only gender at risk of alcohol
abuse; the multiple demands placed on Chagga women can also drive some to drink excessively
(Howard 1997:154).

Chagga mothers' drinking habits
14
2
36
12
62
74
0
20
40
60
80
Neighbouring
mother
Mother of
malnourished
child
%
Excessive
drinking
Moderate drinking
Drinks a little or
none

Fig. 42: Chagga mothers drinking habits. Source: Swantz et al. 1975
95

Again, whilst these figures indicate that women may drink less alcohol on the whole compared to
men, the same basic pattern emerges: malnourished children are more likely to have a mother
who drinks excessively than not, and moderate drinking is found to occur considerably more in
households without cases of malnutrition than in households with. Increased consumption on the
part of Chagga mothers can prove a significant factor in the nutritional status of a child;
uncontrolled consumption of alcohol by both parents represents a major risk to dependents well-
being through the limiting of nutritional and water resources.



Fig. 43: Gendered drinking not without its risks

While it may seem unusual to connect problematic drinking or alcoholic (mlevi) behaviour with
decreased access to water, it is a phenomenon which has hitherto been overlooked within the
literature, and a factor which I feel needs to be considered in understanding how sustained
problematic drinking can impact not only a familys finances and socioeconomic progression, but
also result in a significant reduction in the amount of water that could potentially have been
made available for both economic advancement and the well-being and health of the family.



Fig. 44: The consumption of alcohol can occur at the expense of other activities

It is hard to provide an accurate estimate of how extensive this problem is within the community.
Some informants, especially those who represented the village in an official capacity, were quick
to a) deny there was a problem and b) once it was clear that I wasnt willing to disbelieve the
daily evidence before my eyes sought to minimize it by estimating it to effect only a tiny
number of households. However, extensive polling of various informants within different
96
subvillages, asking them to estimate how many households within a balozi context (a unit of ten
households) suffered from problematic drinking, consistently produced estimates of between 3
and 4 households, a figure which not only tallied with my own personal estimates based on close
observation of bar patronage but which, if extrapolated to the whole village, represents a major
social problem
112
, and a significant potential threat to household water availability.



I have explored a wide range of issues concerning water access which included: population
growth, land pressure, climate change, local climatic influences, infrastructural limitations to
adequate water provision, various issues in water supply, competition amongst villages for water,
varied responses within the village to water scarcity, various inequalities in water access, issues,
strategies, dangers and hidden costs of water sourcing, and finally discussed how excessive
drinking can function to compromise water access. I now examine issues and patterns of water
usage within Chagga households.




*****


112
It must also not be thought that alcoholism affects only those at the lower end of the socioeconomic spectrum. It was
not uncommon to find the parish doctor at his local bar at mid-day, completely inebriated, or to hear of the difficulties the
village administration had in getting balozi leaders to discharge their basic responsibilities such as turning up at village
meetings since they were often drunk. However, they may have been taking their cue from the Village Chairman himself
who was widely regarded within the village as a drunkard. My experience of him was of a man showed few leadership
qualities, could be called a powerhouse of indecision and whose clothes, tatty and witness to many weeks heavy
drinking, hardly inspired further confidence.
97
3 WATER USAGE ISSUES AND THEMES

This section examines what happens to water once it has been accessed and brought into a
household, and attempts to illustrate different usage patterns through case studies of such
patterns in households observed.

Factors determining how much water is used by a household

There are two main factors which determine the amount of water used, and how it is used. The
first is socioeconomic status, which as we have already seen, plays an important role in
determining access to water; the second factor is the size of a household, which can be
measured both in terms of number of family members and the extent of livestock holdings

Economic status and water usage

Both the classic study, Drawers of Water (White 1972) and its follow up study thirty years later
(Thompson 2001) made the observation that there was a positive correlation between wealth
and water usage. The wealthier a household, not only did it consume more water in absolute
terms but generally it was able to use water in more ways that is, such households did not
have the same economic constraints limiting how much water could be brought into the
household, and as such were able to meet both economic and domestic requirements without
any restriction.

In contrast, the poorer a household, the poorer its access to water: no piped connection, inability
to harvest and store sufficient rain, restriction on ability to source and carry water all these
factors translate into less water being present within the household, with greater pressure to
pare water requirements, maximize the effective usage of whatever was available either through
frugality, prioritization or consistent recycling strategies.

This is not to say that in Urauri Village wealthy households were never affected by water scarcity.
Even my host household, arguably the richest in the village, found itself short of water towards
the end of my stay and this, coupled with the onset of the dry season and insufficient water
pressure, meant that we ran out of sufficient water for personal hygiene purposes for several
days until fresh supplies could be sourced and delivered.

However, it is fair to say that poorer households, compared to their more affluent counterparts
which may experience occasional interruptions and inconveniences in their water supply, are
generally much more disadvantaged by the amount of water available to them, which in turn can
act to shape and constrain their daily usage of water in ways not generally experienced by
wealthier households

Household size

In the initial Drawers of Water survey, household size was the main factor in determining per
capita use, with a negative correlation found with increasing numbers of children in the
household, and total numbers of household members. A repeat of this survey thirty years found
that household consumption was determined predominantly by household wealth, with cost
factors a secondary factor (Thompson 2001: 3). In fact, the more people a household possesses
the greater the economies of scale, which can decrease the total of per capita usage.

The size of a household can increase how much water a family needs to access and use; the
larger the family, the more food needs to be cooked, more demands may be made for drinking
water, more clothes to wash, more utensils that need washing, more bodies that need to be kept
98
clean. Similarly, the greater numbers of livestock kept, the greater the demand to keep them
watered and healthy.

Two households with equal socioeconomic parity with differing levels of family size and livestock
holdings will require different levels of water for the household. However, this observation needs
to be tempered with the understanding that water demand within a family cannot simply be
straight-lined; not everyone in the family requires or uses as much water, and animals differ in
their water requirements: a grown cow, for example, will require less water than a piglet.

However, it is true that the consequence of larger household can translate into more demand for
water. Whether or not these demands can be successful met is another matter, though; an issue
which can be determined by a households economic status, not just whether it can access
sufficient water, but crucially whether it can afford to pay for it.

For example, in times of acute water scarcity, a household with a marginal economy may find
that the water demands of their livestock holdings exceed their capacity to meet those demands,
either logistically or financially. In some cases it may even be un-economic to continue providing
water for some livestock, which are either sold off or who may sicken and die a case of
selective animal neglect.
113


So size can be a factor but it is not an absolute. It places greater demand on a household, but
depends on whether a family can meet that demand. In the case of wealthier households, they
have the financial resources which allow them to meet the needs of the many, for poorer
households it may be a struggle to meet the needs of the few.

Patterns of household usage

Having examined the general indicators of what determines how much water a household
may use, I now turn to examining specific patterns of household water usage and will look at
who decides how it is used, priorities of water usage, variation of usage within a family and how
water usage can vary during the day. This will be followed by a detailed examination of the
specific uses of water within a village household.

Who decides household water usage?

While the male household head is seen to be person who controls the finances in the house, the
women in the village are the ones who maintain control over the water. This is because not only
are they often the principal water sourcer, but also once it is within the household, it falls within
their provenance because they are the ones who conduct most of the household activities which
involve water. This includes cooking, washing and washing the children. Mens access to water
generally involves giving livestock drinking water and cleaning themselves. So how water is used
is a matter that is controlled by the female head of the house.

However, this control is not necessarily strictly enforced: children are free to access it for drinking
purposes, or if instructed to fetch water for household chores such as washing food or utensils.
Only in one case did I hear of water kept under lock and key (in order to avoid it being
needlessly wasted).





113
After Scheper-Hughes concept of selective neglect from her1999 ethnography, Death without Weeping .
99
Storage




Fig. 45: Water stored inside the house

According to the Drawers of Water 2 Survey, the declining levels of water provision across East
Africa have resulted in changing behaviours with regard to water storage. Whereas at the end of
the 1960s only 3% of households stored water, thirty years later this behaviour was observed in
90% of all households questioned, including wealthy households with piped facilities (perhaps
because of unpredictable fluctuations in supply (Thompson 2001: 82)

Within Urauri village storage vessels can be found in different forms from outside rainwater
containers to large 60l plastic containers positioned within the household, generally near the
kitchen area, often with some form of scoop nearby for convenient transfer be it a small jug, a
pan or mini bucket. However, these are rarely protected from contamination or spoilage by any
lid, and the possibility of increased risk of water-related insect borne diseases (Thompson
2001:ibid p 82). During Focus Group Discussions I learnt that villagers had received education
about adding chemicals to stored water to prevent contamination, but not once did I observe this
practice.

The prioritization of water

Water that is brought into a household is generally prioritized according to functionality. Of
immediate concern are survival issues such as cooking and drinking. Water is then prioritized for
economic production such as rearing livestock or alcohol production, this is often followed by
clothes washing and utensils cleaning with hygiene often low down in a households priorities.
Decisions are made about water usage by mothers, who in their water usage are the greatest
users of water within a household, with men, children and infants following.

Gender roles, work and responsibilities

Chagga society is highly differentiated according to gender roles. It is a society where
ideologically men occupy a structurally superior position to women (Moore 1996:121), a position
which is defined in opposition by the inferior, weaker position occupied by women (Howard 1997:
156).

Gender differentials are learnt at an early age. Whilst children do similar work around the house,
around the age of 5-6 boys begin learning typical Chagga male activities such as clearing ground,
building a house, slaughtering skills from their fathers whilst girls are taught how to become
women through close adherence to their mothers (Anonymous 1973: 178). Furthermore,
100
differential gender roles are constantly encouraged through daily social interactions: boys shaking
hands of strangers when meeting for the first time, girls doing the same but adding a small
curtsey and lift of the left foot; or girls are taught that men take precedence before women, so
that water is poured for men to wash their hands before eating, with women following; or men
are served the choicest morsels before women and children
114
.

Gender divisions are clearly seen on a daily basis not just interactionally, but in terms of the work
carried out within the village by men and women. With men often displaced economically,
unable to find work and no longer the automatic breadwinners, women often assume the burden
of economic productivity, working extremely hard for long durations. Whilst women are
responsible for a lot of domestic activity within a household - fetching water, cultivating crops,
maintaining a home (Thompson 2001), they also play a central role in staple food production -
often bearing sole responsibility for cultivating food crops for the family (Howard 1997: 147),
animal husbandry, the collection of fodder for livestock and coffee farming
115
. Sometimes the
demands involved in being a woman are intensified by a husbands absence either because of
negative factors like excessive drinking which renders them economically redundant, or positive
factors such as migration in search of employment.

Demands of child care

However, multiple responsibilities can often mean that child care can become somewhat
compromised (Howard 1997:13), especially with fathers traditionally remote and uninvolved in
the process of raising children (Howard 1997:136). Whereas previously mothers spent a great
deal of time with their children as part of the child-spacing practice, by the 1970s this was a
declining practice, with mothers leaving home to work in the shamba or sell produce in the
market (Howard 1997: 132). In addition, there was less child care available from kin or from local
families: Women can no longer expect help with a new baby or a child care. Other women have
too many responsibilities to their own families (ibid: 132). Consequently, mothers often have no
choice but to leave children alone at home for long periods while they engage in economic
activities. This was a pattern I saw repeated in several households, even in households where the
grandparents were supposedly meant to be looking after the children, but had to leave for a
social function a funeral - and left the children alone from early afternoon till after dusk.
Such children would spend hours by themselves, unattended and at risk from infection from the
numerous flies crowding their face or the immediate environment. Several mothers revealed that
they didnt like to do leave their children alone, but often had no choice in the matter. They
revealed they were often distressed by what they found when they got home.

We go far to the shamba and while we are gone the child defecates. So we even find the baby
touching and eating the faeces
116


Some mothers attempt to enlist younger children in the care of infants. Sometimes this is a
natural process, often occurring whilst the mother is still within the household, and naturally
continuing whilst they are away from the home. But often once a mother has departed, children

114
Whilst children are encouraged, wherever possible, to attend and benefit from primary schooling, and are treated
equally in this respect, pragmatic considerations about what a girl can achieve, as opposed to what a boy can achieve
and financially bring to a household through receiving further education, may mean that girls from poor households are
discouraged from pursuing secondary education, which reinforces cultural notions of gender inferiority, and denies the
opportunity to break out of cycles of economic hardship found in Kilimanjaros rural areas (Vavrus 2002), further locking
them into culturally defined ideas about gender.

115
Coffee farming is widespread throughout Kilimanjaro, estimated to be practiced by up to 90% of households (Grove
1993: 432). It represents a major demand for water for both growing and processing.

116 From a FGD about diarrhea causation facilitated by LSHTM - M.Sc. student, Carolina Abuelo.
101
charged with looking after their siblings may neglect to feed them properly or follow hygiene
procedures, leaving the child unattended stting in its dirty, sodden clothes, or left to explore a
dirty environment.

Increasing reliance on small children for child care and decreasing availability of mothers add
new uncertainties to child care.



Fig. 46: Children looking after children Fig. 47: Too large a burden?

There was a wide response to childcare observed within the village, with some households highly
dedicated, and some less able to meet family obligations.



Fig. 48: Mother chops wood alone







102
Purposes of water usage

Water brought into a household is used for two purposes: economic production and domestic
water consumption:

Patterns of economic production

Livestock

Livestock are generally kept within the household compound and, as we have seen are kept in
very close proximity to the familys living quarters because of fears of theft and witchcraft. Fed
on cut forage and crop residue (Anonymous 1973: 178), livestock represent a significant drain on
water resources. Not only do they need to be given vital drinking water, but they also need to be
washed occasionally if they are to be kept free of skin diseases. Failure to supply livestock with
enough fluid during the dry season can result in livestock falling ill or dying. Since livestock are a
form of wealth and they are prioritized even in times of water scarcity
117
. However, one measure
that can be taken is to supply livestock with banana leaves. These possess a high water content
and are a suitable substitute for drinking water when there is insufficient supply.



Fig. 49: Water: essential not just for human survival Fig. 51: Boy carrying banana leaves for livestock

Beer making

In addition to the raising of livestock, the brewing of alcohol, be it pombe or busa . is another
activity which is carried out within the household domain. It is an activity not widely seen since
not only is it water-intensive, it also demands a concentrated effort to organize and supply to
local bars. In my period in the village I only encountered one household that regularly brewed
local alcohol for commercial purposes; most other households that make pombe do so in order to
host a social event such as a clan meeting, or an upcoming wedding. Indeed, the subject of local
beer-making was commented on in a village meeting that was convened towards the end of my
residence in the village:

How is it, said one wag that we complain to foreigners of not having enough water in the
village when we obviously have enough to produce and consume so much alcohol?


117
In addition to selling cows for breeding and consumption purposes, cows can be used for dairy products
which can provide a return of around 20% (Mdoe 1997: 75 - 87)

103


Fig. 51: Domestic busa making

Within the household, beer-making is an activity carried out predominantly by women
118
, who not
only take charge of sourcing the water, adding the millet and bananas and straining the
fermented liquids, but are also responsible for its delivery to the purchasing bar (see Fig. 60
below).

Although hard figures were not available, judging from the containers used for its brewing, it is
possible that the average production batch of 80l (four 20l containers) involves the usage of
between 100 120l. This would represent a sizeable water asset for an average village
household, and an additional logistical demand on a household which lacked the ability to source
much water on a daily basis.

However, given the size of the local market and frequency of consumption, household alcohol
production represents a stable income stream which, although margins can be small (Willis 2000:
257) help a household advance economically over the longer term, despite its short-term impact
on water availability for other purposes within the household. Having said that, it would seem
that it may be that a high degree of domestic organization and financial control is needed to
balance the demands of domestic water needs with its usage for economic production, which
may explain why the only observed incidence of alcohol production took place where household
tasks were distributed efficiently amongst family members who acted as one team, and where
household income was prioritized for economic advancement. Within the household, water usage
was maximized through extensive recycling efforts so that brewing could take place, without
detracting from other domestic uses of water. Without the constellation of these various factors:
effective organization, financial and logistical capacity to source water, and successful re-use of
available water, it is doubtful whether a household could successfully carry out brewing on a
regular basis. However, there are some situations like a single mother with children is forced to

118
Chagga women have the right to own property and amass savings independently of their husbands, and can use their
trading skills in local markets selling or bartering produce to generate profits which can be used to purchase food to
supplement harvests or pay for school expenses, or even invest in the purchase of livestock and the development of the
household economy. But in light of generally difficult earning conditions for Chagga men, such entrepreneurial activity
may not be seen entirely positively, and may seem threatening to notions of masculinity predicated in part on economic
performance which may produce conflict and tension within households where females are economically active and
successful. (Howard 1997: 149 151).

104
look at any available means of making money, including making and selling pombe (Howard
1997: 23).

Irrigation

Households variously reported that they are exclusively dependant for the fertilization and growth
of their crops on the rain that falls during the course of the year, and that there is not much they
can do once the dry season begins in terms of additional irrigation.

Traditionally, harvesting began before the full onset of the dry season and so lack of rainfall was
not normally a problem then. However, with its progressively earlier arrival, extended duration
and increased temperatures, the dry season represents a considerable threat to the productivity
of the shambas, and may result in reduced harvests, drier land and more difficult cultivation
conditions.

Whilst they would like to be able to be pro-active, given the difficulties sourcing water in the first
place, and the amounts of water available within the household domain - even with extensive
and efficient re-cycling of water - the villagers feel that there is little they can do. Usage of water
by households for the purposes of crop irrigation within the shamba was not an activity observed
by this ethnographer, although there were apocryphal stories circulated of illegal diversion of
water for irrigation purposes from Kiliwater pipelines, though these were attributed to individuals
in other villages, and an activity prohibited as early as 1987 by the Urauri Village Committee.

Patterns of domestic water consumption

Cooking

According to the Drawers of Water 2, cooking and drinking constitute one of the main daily uses
of water, with a combined total of around 4l per day per person (Thompson 2001: 27). Household
observation confirmed that cooking represent a substantial usage of water within the household,
if utensils washing was also factored.

Village households enjoy three meals a day. Breakfast usually consists of uji, a thin porridge
made of maize and water, and normally consumed by everyone in the household. Lunch and
dinner are not necessarily distinguishable by the meals served since ugali, a maize staple, can be
served at either time as can plantain-based meals (figs. 52-53 below), boiled maize or fried
cassava. These are then complemented more frequently by beans than meat - which is relatively
expensive - but which require more water to cook.


105


Figs. 52 53: Preparing and washing plantains for cooking

The responsibility for cooking is shared by the female members of the family, with daughters
following direction of the mother. Other than the preparation of ugali, a staple dish, cooking can
be a time-consuming process, often taking 1 - 2 hours, involving preparation, washing and actual
cooking (over an open hearth).

Cooking was cited as the number one priority within the household, and even in the most water-
poor households, I did not see any evidence or hear informants talk of any curtailing of cooking
or attendant water usage. Instead, it was seen as something of a fixed overhead around
which everything else had to be negotiated. Whilst high, the water demands made by regular
cooking could be somewhat attenuated by water recycling, examples of which are listed below in
the discussion on water recycling below.

Drinking

Urauri villagers are fortunate that, although water sourcing is a daily challenge, the water
collected is of potable quality. However, daily water requirements rarely draw upon stored water
directly. Adults drink relatively little water, but instead do consume uji (discussed above) and
Chai - a form of spicy tea which is the preferred morning beverage.



Fig. 54: Chai for two
106

Whilst children freely access the water stored within a household to quench their thirst, adults
rarely do so, and were often observed to spend the entire morning without drinking any water,
and perhaps only one glass at lunchtime. Indeed, my field assistant would sometimes go the
whole day without drinking any water, a feat which astonished me no end, given the warm
weather conditions, and my un-hardy dependence on bottled water.

The use of water for drinking purposes forms a relatively small part of a households daily water
usage patterns (Drangert 1993: 107), and it may be that adult water requirements calculated at
being between 1.8 3l per day (White 1972: 8) are initially met by their consumption of tea and
porridge in the morning, and then by the common custom of relaxing at a local bar with a pitcher
or two of pombe at midday, a time when work in the shamba is considered to have been
concluded, and then later in the afternoon when groups of individuals, both males and females
gather again to socialize and drink. Thus it may be that the locally brewed alcohol which is
characterized by its thin, watery quality serves as a major source of water intake by adults who
prefer to obey their pombe thirst rather than quench it with water.

The consumption of water for cooking and drinking purposes varied little across the various
households, a phenomenon described as non-discretionary and averaged at around 4l per capita
daily (Thompson 2001: 27), regardless of household status (Thompson 2001: 90). However, whilst
demand for water for these purposes remains stable, and increased household size not effecting
demand levels because of the negative correlation between household size and consumption
levels due to economies of scale, the impact that such a fixed overhead has on a household can
vary largely.

The relative proportional burden by such a fixed overhead imposed on a household with piped
access is less than that imposed on a unpiped household. Fig. 56 below demonstrates that such
water usage results in less available water for an unpiped household which when, combined with
findings that indicate that such households have less water available for hygiene purposes not
only in absolute terms (quantity) but in relative terms (proportionally)

Percentage of water consumed for
cooking and drinking in relation to
total water available in a household
with piped water
93%
7%
Percentage of water consumed for
cooking and drinking in relation to
total water available in a household
with unpiped water
80%
20%


Fig. 55: Percentage of water consumed for cooking and drinking in relation to total water available in a
household with piped/unpiped water (based on a constant figure of 4l per capita for both households, and a
figure of 60l and 20l for piped households and unpiped households respectively).

I move now onto the issue of hygiene which encompasses notions of cleanliness and order within
the household, as well as social presentation of self.

107
Hygiene and the household

Through my observation and various focus groups I discovered that there were various ideas
about levels of hygiene. Spheres of cleanliness or hygiene. That is to say, the household domain,
being the location where household chores, agricultural work and maintenance of livestock took
place was a domain where being perceived as clean was not considered important. That is to
say, wearing dirty clothes was seen as acceptable, even unavoidable to a certain degree
considering how difficult it was to keep the environment clean at all times, especially when one is
working the whole day in dirty environments such as animal pens or the shamba (Howard
1997:194). Thus it was observed that members of the household would often wear the same
clothing - even after soiling (ibid: 20) - for many days simply because there may be no clean
ones available for them to change into (ibid: 168).



Figs. 56 57: Children wearing soiled, ragged clothing

So if we look at Fig. 58 we see that the household is represented as a black rectangle which
signifies the domain where it is most acceptable to be unclean
119
.


119
However, a caveat must be observed that when someone who is socially superior visits the household domain, then
efforts are generally made to show heshima (respect) by donning clean clothes (Howard 1997: 152), though is not always
strictly observed.

108


Fig. 58: Schematic of Chagga domains of hygiene

However, my informants were quite clear that once an individual leaves the household domain
and enters the public domain, then it is not acceptable to be seen in a dirty state. However, there
are two further domains to explain. There is a domain which is where one can be seen travelling
between two destinations for work purposes. Here, one is expected not to be dirty but not
necessarily to be wearing the best clothes. In fact, if on an errand such as fetching lumber or
fodder, then one can wear workwear clothes which are not necessarily the same as those worn in
the house, but cleaner nonetheless. So for example, in one household, all the women would put
clean kangas on top of their house-clothes when they went to deliver busa, a local brew that
they had prepared to a nearby bar.



Fig. 59: Village women wearing clothes for the road

However, the final domain is a formal social event or function. This is when it is expected that
one should not only be personally clean and well groomed, but that ones attire is pristine too.
109
Such events include going to see the doctor, visiting the priest, going to mass or any religious
ceremony, attending a wedding or funeral. On these occasions, it is regarded negatively by
villagers to be anything other than safi - clean, neat men wearing pressed suits, women
wearing Western dresses or kitenge a garment similar to the kanga worn either as a shawl or
skirt, but made of heavier material and more formal. To be other is to invite ridicule or
speculation as to ones state of mind. Furthermore, it is possible that notions of cleanliness and
public presentation are infused with notions of sexual and moral rectitude and are reinforced by
attempts to induce shame response (aibu), as revealed by the interaction between a mother and
daughter:

Why must you be so dirty? Do you imagine you can stay long with a husband if you are such a
slut? (Raum 1940: 182)

So contained within efforts to appear clean and well groomed within the public domain are ideas
about what constitutes normality and for females at least - appropriate sexuality. However,
there is also an element of impression management at stake - of the vigilant control of public
perception of status because as one informant confided to me, We might be poor, but we cant
show that.

Dirt within Chagga conceptions is not just a physical state, One can be dirty in at different levels:
physical mental and spiritual. A pastor talking about AIDS: We must re-commit ourselves to
leading clean lives. There are many kinds of filth. There is bodily filth, mental filth, and spiritual
filth. (Setel 1999: 224). However, physical dirtiness or unkemptness may therefore be read as
symbolic of impurity at other levels.




Figs. 60 61: Public event = Public display of cleanliness

Despite the restrictions imposed by water scarcity, informants indicated that they felt socially
pressured to present themselves in clean attire when participating in a public event or when
meeting socially superior individuals such as the village doctor or priest as when such attire
would be seen as a sign or heshima or respect. Those unable to do so adhere to these norms still
participated, but in a marginal capacity. On one occasion I witnessed one individual who came to
church wearing a tattered grey overcoat and mud spattered shoes that were falling apart. He
clung to the back of the wall throughout the service, and slunk away without interacting with
anyone afterwards. According to some informants, the very poor were said to avoid using clinics
because of the aibu they felt in not being able to comply with the demands for heshima made by
110
local codes of public dress because the only clothes they possessed betrayed their poverty
(Howard 1997: 194)
120


Clothes washing

Another activity which consumed large quantities of water was the washing of clothes, which
often took place twice a week - typically midweek and on Saturday, so as to have clean clothes
for both church and school the following day.

The drain on water resources is mitigated in some households by reusing dirty water for other
purposes, but it is such a major drain that sometimes a household can consolidate its washing
once a week, and often individuals will minimize the amount of washing that has to be done by
wearing the same clothes for days on end.

Water-rich households can be characterized as households that wash frequently, and in large
volume. I would often witness bedclothes hanging out to dry in my host household, whereas this
was a much rarer sight in households with limited water availability. In the worst cases,
bedclothes were never washed, and were black with the dirt and accumulated grime from the
bodies of several infrequently-washed children sleeping together.

One of the signs which I soon learnt to recognise of a household where washing was commonly
practiced was the existence of washing lines. Households with multiple washing often had a high
turnover of dirty clothing; households with one much less; households with none were
considered suspect, and the substitute practice of hanging clothes over large boulders to dry
often resulted in the immediate soiling of clean but wet clothes.



Fig. 62: Dirty clothes awaiting their destiny Fig. 63: Washed clothes. Note the multiple washing lines

Children returning from school in relatively clean school clothes, might often slip into much dirtier
house clothes, clothes that we noted had been worn for several days in succession. What
indicated that this was a norm was the fact that occasionally a household would ensure that its
children wore clean clothes during the observational period, but ad hoc surprise visits found them
wearing the same clothes they had worn before our visit, indicating some level of stage
management.


120
Stigma associated with poverty also brings with it social marginalisation (Howard 1997: 12)
111
Personal Hygiene

The use of water for cleaning domestically was relatively rare given the typical environment, mud
houses and compounds. However, some households did have the occasional wooden structure,
and some of these even had concrete patios. For most households these did not merit special
attention but for others, the minority, though, water was devoted to the patio as part of a
general behavioural pattern which strove to maintain a high level of cleanliness within the
household environment.



Fig. 64: Washing the patio

In general, order was maintained within the household compound by the mother or daughters
sweeping it clean each morning, and ensuring that poultry such as chickens, ducks and turkeys
didnt cause too much mess by confining them to their coop for extended periods or by even
tethering them to a stake in the ground, so that they couldnt wander too far.

However, within an African rural setting where humans and livestock
121
live in close proximity,
and the building materials are as basic as can be, it would be fair to say that maintaining a clean
and ordered environment is a difficult task to accomplish on a daily basis. However, a noticeable
difference that emerged between households facing similar circumstances was the degree of
effort that a household would put into such a task, no matter how thankless and ephemeral the
results. As such, it was instructive to observe that households which strove against the forces of
entropy and sought to impose order were also households which were characterized by a high
degree of structure elsewhere in their time management, organization of tasks, delegation of
responsibilities. Some households even managed to devote some of their water resources on the
beautification of their environment by growing potted plants or maintaining flower beds.


121
One of the reasons for such proximity lies in a history of cattle rustling conducted both between different Chagga
chiefdoms and from the traditional foes, the Masai (Howard 1997: 42) which led to cattle being held in stalls within one
end of a dwelling, often next to the sleeping quarters. With the institution of a legal framework and the existence of
governmental control in the form of local police, plunder of cattle on a large scale is largely a threat of the past. Taking its
place, though, is the theft of livestock which is much feared and often guarded against in wealthier households by
employing a night watchman who sits up all night listening for intruders.


112


Fig. 65: Sweeping the compound Fig. 66: Watering plants

Those that invested little energy in maintaining their environment, allowing animals to roam
freely within the compound, lack of organization and clear definitions of responsibilities were also
most at risk from an irregular supply of water and poor water management within the household.
Needless to say, such households exhibited no desire for any horticultural enhancement of their
environment.



Fig. 67: In some households not a temporary state

It was useful, therefore, to assess the level of effort expended into maintaining an ordered
environment since this provided a relatively accurate assessment as to how water was similarly
treated within a household. A high degree of structure, organization and economically-focused
activity often translated into a maximal usage of water and a minimal risk of ill-health. The
opposite, unfortunately, often served to predict poor, inefficient usage of water resources and a
household characterized by various health threats.

Water and health conceptions, well being

Water is defined in different ways, in terms of desirability and health risks

Bottom ranking is water which is perceived as being dirty maji machufu either through
particle contamination or through the presence of animal life, or even if it smells bad. This is in
binary opposition to water that is clean maji safi which is clear, devoid of animal life and with
no detectable aroma. However, such water is commonly recognised as not necessarily being safe
to drink. Thus it is clean but because of the possible presence of micro-organisms vijidudu -
represents a potential health risk, one that is only testable through means of the stomach. The
113
water least likely to compromise an individuals health possesses all the attributes of maji safi but
also lacks vijidudu that could be harmful. Such water is designated maji salama.

Clean water Properties
Maji salama Clear, devoid of particles, obvious
animal presence, micro-organisms
or identifiable smell
Maji safi As above, but potential presence
of micro-organisms
Dirty Water Properties
Maji machufu Suspended particles, animal life,
presence of micro-organisms
unknown, possible unpleasant
aroma.


Table 10: Chagga definitions of water typologies

Personal hygiene

Face washing

Whilst the majority of respondents questioned have indicated that they had received some
education about the importance of face-washing from various sources, its practice is far from
common. During household observations, we frequently noted the sounds of distress as a young
child, unused to the novelty of the experience, had its face washed by its mother. Sometimes the
children of the household would be instructed to wash en masse so as to impress the visitors, but
unannounced spot checks would find their faces dusty and marked with the traces of sleep in
their eyes, even in the afternoon. The issue of stage management was something that we
encountered consistently, and parental claims of child face-washing were treated with some
skepticism, especially when it was clear that the parent had not engaged in the behaviour
themselves!


Fig. 68: Group face-washing

During the entire period of residence in the village only one household was observed where a
child spontaneously performed face-washing behaviour, and this was a household where hygiene
education was explicit (see Fig. 71 below), and great efforts were made all round to maintain an
ordered, clean environment.

114


Fig. 69: Mother teaching her children the importance of hygienic practices (washing of hands)

Teeth-brushing

In contrast, ample evidence for teeth-brushing was observed which made for a curious contrast.
Though at the time, informants were not questioned as to why this was so. But a common idea
expressed by many was that effective face washing was water-intensive, and a secondary fear
that there wouldnt be enough water for all. When we physically demonstrated how much water
was needed, this was often accompanied by informants expressions of surprise. So it is possible
that teeth-brushing is perceived and experienced as needing minimal water, and can thus be
engaged in by the entire family.

Hand-washing

Whilst breakfast would not necessarily be accompanied by any washing of hands, lunch time
would be characterized by a formal display of hand washing usually the mother would pour water
over her husband or childrens hands before they ate which may use perhaps a quarter of a liter
of water. When questioned about this, informants would say that it was important to wash hands
because if they had been working in the fields then their hands would often be dirty and since
they had to eat with their hands (the Chaggas rarely use eating implements), it would be
important that they be clean. This notion of contamination is expressed by the greeting practice
employed when Chaggas choose to do more than exchange more than superficial, passing
ambulatory greetings, and stop instead to converse in more depth, presenting a hand to shake as
they do so. Should one of the persons being greeted have been involved in some work which has
dirtied their hands, or their hands are wet, then the wrist will be proffered instead, avoiding the
embarrassment or nuisance of the other person getting their hand dirty or wet.

Whilst it was observed that when soiled livestock was cleared out from the pens inside the
household, this would often be done manually resulting in direct contact with animal excreta and
urine, and upon completion of this task, hand-washing would seen to occur. However, whilst it
was operationally difficult to identify when adults used household latrines, a lack of positive
observation of hand-washing behaviour across a variety of households and across various points
during the day would seem to indicate that post-latrine hand-washing is not widely or frequently
practiced.

Washing of extremities

Late afternoon is often a time for washing extremities. Typically within a household compound
there is a large stone, often situated near the door of the house, which has been embedded into
the ground. This is the washing stone which can be used not only as a seat during washing but
115
as a giant pumice. This is a designated area for the washing of faces, arms and legs. These are
washed by pouring water into a large plastic bowl and then the individual can wash their
extremities whilst remaining clothed. One interesting feature is the use of old corn/maize cobs as
a natural scraper or brush during the washing procedure, which are also used to scrub dirt from
flip flops as well the soles of feet.



Fig. 70: Foot scrub

If face washing is to occur, this is the most likely time with the backs of ears and necks also
being factored into the washing procedure as well.

This kind of washing behaviour is designed to remove the accumulation of dirt and grime from
the shamba or from being on the road, and mostly occurs at the end of day. However, it can
occur at any time, particularly if an individual has somewhere to go, or some function to attend,
be it a funeral or doctors appointment etc., Then, suitably clean, they will don clean clothes and
depart from the household domain. However, in some households, it must be noted that the
regularity of washing of extremities could be held in question by the physical evidence of a dry
washing area and childrens legs which were clearly marked with familiar dirt, indicating that no
washing had occurred the preceding day, despite assertions to the contrary.

Bathing and showering

Whilst the washing of extremities often occurs at the end of a days work, showering or full body
washing takes less frequently, and is conducted out of sight of the central household area. This is
said to take place mainly last thing at night when the sun has set to preserve privacy, but was
not an activity that was observed because of time and personal safety restrictions.

Bathing order

Within Chagga households there is a clear hierarchy in who gets to wash first. Whilst informants
may publicly state that the man of the house gets first access to water for bathing, this can be
contested by both informants testimony of the reality that women bathe more often, as well as
observed practice confirming such behaviour. Children are often last in line, and often miss out
during dry periods.


116


Fig. 71: Eldest daughter washes a younger sibling

UNEQUAL ACCESS OF WATER: ITS IMPACT ON HEALTH

4.9
15.2
7.9
17.8
0
5
10
15
20
Daily per
capita
water
usage
Impact of water supply on washing
and bathing
Unpiped Washing
Piped Washing
Unpiped Bathing
Piped Bathing

Fig. 72: Impact of water supply on washing and bathing
Source: Thompson 2001: 28



117


Fig. 73 - 74 Jiggas in a hand and foot Fig. 75 Ringworm

A question of neglect?

Certain children and Vulnerability

Certain children may be looked upon in a family more favourably than others within the same
family (Howard 1997: 173). Boys are much anticipated, especially the first-born since they act to
secure the family lineage, and are expected to provide financial security for the parents. Girls,
whilst not as eagerly received are still positively looked upon since they can be of great
assistance to their mother as they grow up, as well as bring bridewealth into the household when
they come of age and marry (ibid: 175).

However, not all children are viewed in a positive light, and some may receive less parental
attention than others. Illegitimate children, especially boys were regarded with some suspicion
since their spirits were thought to be able to cause harm to others around them, and this belief
may result in less care nutritional, hygiene, attention etc., being extended to such children (ibid:
177).



Fig. 76: Abandoned child plays in the dirt Fig. 77: Flies feeding off the tear duct







118
Recycling issues/practices

In a water scarce region, the difference between a household that maximizes what water it can
access, and a household that doesnt can spell the difference between a household that endures
the scarcity without too much financial or health repercussions and one that feels the strain on
both fronts.

Water recycling can be defined as the rational management of a scarce resource (Almendon
1997: 93), and within the context of the village a whole range of household recycling behaviours
was observed: from households that did not recycle to households that used the same water for
several different purposes.

Typically, recycling involved:

Water used for washing food, clothes and utensils given to livestock to drink
Water used for clothes floor cleaning

However, those that didnt were particularly vulnerable, especially during the dry season when
water becomes ever-more valuable.

Non-recycling households there seemed to be various factors at play: a generally poor level of
household activity as well as geographical and social isolation.

Case study

One household which fitted this profile was a trachoma-blinded mother with a family of five
children characterized by an absent father. The mother was generally disorganized, trying to
maintain a household in an ad hoc fashion and apparently unable to cope economically with her
responsibilities of running a household, a household characterized by extremely poor housing
conditions, sickly children with worms, trachoma, and Jiggas - sand flea infestation of the
extremities (see figs 73 74). Clothes were washed infrequently, no washing line was available
to hang clean clothes to dry; they were instead lain on dirty rocks instead. Water that had been
used for washing clothes or preparing food was simply poured into the ground instead of being
recycled in another fashion. This household was characterised by poor educational attainment,
geographical isolation (compared to the rest of the village), social isolation (mostly voluntary)
and lack of support networks (i.e. non-membership of any social organisation).

In contrast, a household possessing similar socioeconomic and socio-environmental
characteristics extracted a tremendous degree of utility from similar resources, using water
recyling as the primary vehicle. This household was characterised by an absent father, similar
numbers of children. However, the environment was very-well maintained, the household highly
productive economically, and the children regularly washed, including their faces. However, the
one factor that was significantly different was the degree of social connectedness which may
have meant that the mother of the household was more receptive to learning new ideas, and
able to implement them. This was a household, though, where the impact of recycled water was
clearly visible and points to the possibility that re-cycling of water should be actively included
within the SAFE strategy as a vehicle to leverage available limited supplies.






119
Impact of water scarcity on hygiene

Water shortages effect all aspects of daily life during the dry season, but none more so than
hygiene. But once again major difference emerge between men and women, adults and children.
Whilst women consistently stated in focus groups that their husbands always washed first, as a
mark of respect, given that the men themselves admitted to washing infrequently, perhaps once
a week, and women insisted that they washed everyday, it would seem that in actuality women
are the largest consumers of water for personal hygiene purposes. Much laughter was derived
from posing the question Are men cleaner than women?. Both men and women agreed that
women were cleaner than men. God built them that way, stated one male informant somewhat
cryptically. However, it was agreed that womens anatomy required greater personal
maintenance which could not be put off wed smell, said one female informant.

Women manage not only to maintain a daily hygiene regime which includes genitalia and
extremities and the face with a shower every couple of days. Men, on the other hand would wash
partially every two to three days with a full shower perhaps once a week.

Children, though, bear the brunt of water scarcity during the dry season. They were listed as last
in line for water access for personal hygiene purposes. Whilst it was stated that children would
be washed frequently during the dry season, in interviews with young school children, it was
stated that sometimes the very youngest children would not get washed for several days simply
because there was insufficient water.

Children informed me that the dry season often meant that households simply ran out of stored
water, and that there was no water left after other household activities had been completed for
personal hygiene. Sometimes this affected the entire family, including the youngest children.
Sometimes for up to 2-3 days.
120

DISCUSSION


If youve got it, use it

As we have seen there are multiple variables in determining water access, but a key factor that is
instrumental in determining how much water is accessed seems to be socioeconomic status. Such
households are able to use their financial resources to gain access to and store more water.
Unsurprisingly, this results in greater water usage which often includes greater overall hygiene
practices. This is a critical differential since there are indications within the literature that
socioeconomic status plays a major role in determining trachoma infection rates within a
household. Trachoma occurs more frequently and more severely the lower a household is
situated socio-economically (Marx 1989: 24), and declines as household per capita income rises
(Prost 1989: 15).

Households which possess less financial resources are usually much more restricted in the
amounts of water they can access, store and use. The dynamics of household water economy are
therefore very different to those of a wealthier household, the more difficult it is to collect , the
more valuable each liter collected becomes. This value is increased especially if distance to
source is large which often means that less water is collected. Distance to source, therefore, is a
key factor in determining the amount of water collected and the value placed upon it.

All things being equal

What such a model fails to account for is difference in water usage patterns for households with
socioeconomic and socio-environmental (identical water source, similar distance travelled to
source water, identical household numbers and living conditions) parity. That is, why two
households which may possess similar financial resources and access similar amounts of water
exhibit different water usage behaviours within the household. Such differences may be
considerable, including different usage foci, activities covered, amounts used and amounts re-
cycled all of which may have a considerable impact on the health status for household
members, especially in terms of eye disease and trachoma in particular.

Differentials in trachoma infection rates amongst such households would therefore suggest that it
is intra-household differences in water usage behaviour, particularly in reference to hygiene
behaviour - rather than the amounts sourced that is the key determinant (Marx 1989: 25).

Increasing water availability; decreasing distance to water.

One solution mooted to help reduce differentials in water access is to develop new sources of
water drilling local borehole, for example. The reasoning behind this is the key notion that lack
of water was connected to disease intensity (Marx 1989: 24), with increased distance to water
source associated with an increased risk of trachoma (Taylor 1989: 1824), and increased
distances shown to be associated with greater infection at younger ages (Prost 1989: 14).
Perhaps part of the explanation for this is that decreased distance makes more frequent water
collection more feasible (ibid: 14).

Increasing water availability is thus thought highly likely to impact trachoma prevalence within a
community. Schemes which have sought to demonstrate an increase in water consumption by
reducing water collection times to below a critical 30 minute juncture (Cairncross and Feachem
1993), have seen per capita usage of water quadruple (Prost 1989: 14). However, while a
reduction in water collection time may mean additional water is brought into the house, and
121
more water is used per head, this does not necessarily mean that additional water usage changes
existing patterns of water consumption for hygiene purposes, and may mean little improvement
in family health.

The key, therefore, may not lie in the absolute amounts of water stored in a household, but in
the perceptions of the water that is available, and the behaviours that flow from those
perceptions. In fact, some researchers argue that the main thrust of trachoma reduction depends
on working with and influencing behavioural patterns (Marx 1989: 25; Prost 1989: 15).

The effect of a planned nearby borehole on hygiene behaviour within Urauri

Plans to drill a borehole in neighbouring Kahe may result in an additional source of water being
soon made available for local communities. However, some observations need to be made about
its perceived future benefit. Firstly, whilst a proportion of Urauri villagers currently use Kahe as a
potential source of water, the same is true for some Kahe villagers who often come to Urauri on
allocation days. This would mean that the presence of an additional water source may relieve
pressure on Urauri water sources, making more water available for its residents. Secondly, the
drawback of a potentially bountiful water source in Kahe is that this would not only be massively
popular among Kahe residents but with residents from surrounding villages and would therefore
rapidly become over-subscribed, thus resulting in approximately the same level of water
availability as before. Thirdly, since Kahe lies on the border with Nessae, the benefit of such a
borehole would only be applicable to a small proportion of Urauris residents. However, the
majority of residents of Nessae subvillage would most probably still use Kenya as a principal
source since despite its drawbacks collecting water from there might still be a faster process
compared to trying to source water from an over-subscribed nearby source. Thus it is likely that
a nearby borehole would have a small impact on total water availability in the village.
Furthermore, in the Urauri households that used the Kahe borehole, this would not necessarily
translate into any notable change in patterns of water usage, especially in relation to face-
washing.

The value of water

Since water is a valuable commodity within a household economy based not just on its purchase
price but its costs in time and energy to source and transport it, households will often prioritize
its usage on its utility and value for the maintenance of a household. Generally speaking, at a
fundamental survival level water is valued and prioritised as essential for basic survival in
allowing for cooking and drinking. Thereafter, water becomes a tool for economic development
be that agricultural, the rearing of livestock and the brewing of alcohol. For most households
economic development is a concern, and water is prioritised accordingly. Hygiene usually follows
last, and is perhaps the activity that is most expendable within a households priorities since it is
neither essential for immediate survival or economic productivity. Thus the general value placed
on hygiene within a community may be a critical issue in determining whether or not increased
water supply is going to signify increased usage of water for hygiene purposes (West 1989). A
common pattern found is that increased water supply does not result in greater hygiene
practices; often additional water is channeled into economic production purposes since these are
perceived as perhaps giving the family more leverage for immediate and future survival. As one
informant said, More water means we can feed our families better. Thus intervention strategies
which focus on additional water supplies must at the same time pay attention to educational
programmes which underline the importance of hygiene and its relationship to family health.

However, such programmes must be collaborative the imposition of an externally derived
programme based on Western notions of idealised hygiene principles is unlikely to be favourably
122
received or implemented by the intended recipients unless a full account is made of restrictions
on water usage for hygiene purposes, especially face washing.

Encouraging face-washing

With reference to encouraging face-washing, various obstacles remain to the widespread uptake
of this hygiene behaviour:

1) A general lack of perception of the threat posed to children by trachoma sore,
inflamed and itchy eyes are seen as part of childhood development (Lane 1993).

2) Minimal recognition of the latter stages of trachoma as constituting part of a
continuum with earlier forms (ibid: 1993).

3) Perception of amount of water needed invariably this was grossly over-
estimated (Mccauley 1990: 1244).

4) Perception of time needed. Often parents would comment on the time face-
washing would take which might be taken away from other, more valuable
activities.

5) A lack of conviction as to the utility of pro-active behaviour to prevent something
that couldnt necessarily be seen.

6) A lack of prioritization within household water usage patterns cooking, drinking
and livestock often coming first before hygiene.

A complete success?

In houses where face-washing has been recognised as important, and incorporated into daily
schedules, the activity itself can still represent a potential vector of trachoma transmission since
children may be encouraged to share the same water to wash their faces (see Fig. 68) or a
mother may wipe their faces with a dirty kanga representing another potential transmission
vector (Collier 1999). Any face-washing educational programme must attempt to incorporate
these observations into their key messages, otherwise a sudden enthusiasm for face-washing
amongst villagers may result in greater, not lesser rates of infection.

However, educational programmes which are handled sensitively are likely to be favourably
received since there is already a conceptual basis for the transmission of diseases which is based
one Western notions of bacteria. Most individuals expressed a familiarity with the importance of
hygiene practices and mothers were often found to have sought immunization for their children.
Thus it may be that the focus of any educational programme may lie in convincing communities
of the importance of hygiene whilst taking into account pre-existing priorities for water usage.

Bathing order and gender differentials in terms of water access

One of the most valuable findings of this study was the discovery of the widespread practice of
placing young children last in the bathing order. Under normal circumstances when water is
generally available, this does not necessarily constitute a risk to health, but under conditions of
water scarcity being last in line often means that those who most at risk of a host of water-
related diseases which could be prevented by frequent washing, are in fact put in a position
where they are highly susceptible to infected or sick.

123
Studies conducted within the field of nutritional anthropology (Chen. Huq. & DSouza 1981; Engle
& Nieves 1993; Griffiths, Matthews & Hinde 2002) indicate similar processes in terms of
nutritional distribution whereby younger members of the family often receive substantially small
portions of food than adult members. Within Griffiths study such behaviour or nutritional strategy
(DeWalt 1983) was based on the perception within the household of economic productivity i.e.
small children were unable to contribute as much to a household as an adult member and
therefore merited less food. Allocation of food based on a contributions rule (Cassidy 1987)
suggests that there is a sense of competition for scarce resources, and that some individuals will
fare better than others in a household (Griffiths, Matthews & Hinde 2002: 786).

In addition, such allocation patterns may also have a gender overlay, with female children
receiving less food than male siblings (Chen. Huq. & DSouza 1981) because of gender
favouritism alone.

The contributions rule can be countered by the needs rule where individuals are allocated
according to need, rather than their potential or actual contribution to the household economy.
Unfortunately whereas a needs model can operate in conditions where there are sufficient
resources, in times of resource stress or marginality it would seem that the contributions rule
may dominate as heads of households make decisions about the long-term survival of the
household rather than any particular household member. Sometimes the consequence of this is
that decisions are made that may seem outwardly callous and uncaring, but are ultimately are
rooted in utilitarian and culturally-derived criteria.

Although householders within Urauri village did not articulate their reasoning for bathing order in
depth, the commonly reported pattern of: father, mother, children (or the actual pattern of:
mother, father and children) would seem to indicate the existence of such behaviour being rule
based, even if only unconsciously.

Whilst the identification of such a bathing hierarchy within a rural village setting may lead to a
desire to commence a campaign which attempts to educate householders about the importance
of washing infants and young children first in order to reduce trachoma prevalence, such
educational efforts may flounder if they fail to recognise the dynamics which drive such hygiene
behaviour in the first place. However, the likelihood of such a hygiene education campaign
succeeding may be much higher if it is combined with an increased and accessible supply of
water. This would alleviate the pressure to allocate a hitherto valuable resource according to
perceived household contribution - actual or potential, and more according to perceived need.
Combining hygiene education and increased water supply at the same time would probably make
for a more appropriate intervention strategy, focused at the right junctures to encourage change
than the implement of just one element alone.

This perspective concurs with the conclusion of the second Drawers of Water survey and that of
earlier findings (Cairncross 1988) which found that the greatest difference to the health of a
community came not from improvements in quality per se, but in greater quantities being made
available. However, such structural improvements need to simultaneously be accompanied by
educational programmes to capitalize on the increased availability of water which encourage the
emergence of health-promoting hygiene behaviour (Thompson 2001). Failure to do so may result
in greater usage of water, but no significant changes in hygiene behaviour, or an educational
campaign which may be transmitted but not received (Yoder 1997).






124
Flaws in the gem: trachoma education

The history of international biomedical educational efforts is a patchy one. The success of a
campaign is often measured by whether the target audience has been deemed to have complied
with the message and altered their behaviour accordingly, but when a health education message
has not been heard or acted upon this is often labeled as non-compliance, which places the onus
for the success of the communication (within terms defined by those implementing the
campaign) on the audience targeted. However, blaming the audience for the non-reception of a
message deflects responsibility for the messages sent away from the originator(s). George
Foster, a pioneering medical anthropologist, was instrumental in helping to refocus responsibility
on those who sought to influence and change current behaviours, arguing that the very culture
of such agencies was an integral element that needed to be factored in when presenting and
promulgating health-promotional behaviour (Foster 1987). Failure to recognise the influence that
the culture of the promoting agency might have on the diffusion of critical messages might
continue to blame the receivers rather than grasping the fact that cultural assumptions and
dynamics might play a major part in why a message failed to be acted upon.

My perception of previous trachoma interventions and educational efforts by outside agencies
within the village was that most people had not received much, if any education. Most average
informants did not know what it was, did not understand how something that was invisible could
represent a problem. Furthermore, the balozi leaders who had received trachoma sensitization
training claimed they had received information about the disease were somewhat vague about its
exact transmission and how to prevent it: whilst most recognised it had something to do with
environmental conditions latrines, refuse etc., and that an important element in preventing it
was the use of water, the majority failed to remember that face-washing was a key element, if
not a corner stone of preventative efforts.

This puzzled me greatly until I pieced together what had been happening within the village.
Firstly, over the past few years there had been a variety of organisations working within the
village. On the one hand, the International Trachoma Initiative (ITI) was based in Mkuu and
occasionally came to the village to give talks on what trachoma was and how it could be
prevented. Then there was the Rombo Trachoma Research Project (RTRP) which was also based
in Mkuu at Haruma Hospital (next door to ITI) which was responsible with evaluating the
prevalence of trachoma in the region, rolling out various interventions such as the controlled
distribution of the anti-trachoma antibiotic, anthromyzin, while at the same time performing an
educational service. Then there was the Christoffel Blinden Mission (CBM) which visited the
village on an irregular basis and performed surgical interventions on those with trichiasis of
trachoma. Little wonder then that villagers expressed confusion as to who was who or what they
were meant to know and remember since the different agencies were interfacing with the village
in an uncoordinated and seemingly confused disorganized way. So part of the communication
problem lay in the seemingly haphazardness of how these agencies interacted with the village.

Another problem was how information was diffused within the village. ITI, the agency most
involved in trachoma education, predominantly used the Village Office to hold trachoma
education seminars. The problem with this venue as it soon became apparent through informant
interviews is that whilst this was convenient for the ITI staff travelling from Mkuu, they failed to
recognised that villagers had very specific and often gendered relationships with this venue. So
whilst in theory everyone was invited to attend, women rarely did so because it was widely
perceived as a male space. This resulted in an ironic situation where those least involved in day-
to-day child care were briefed on how to identify the causes and best ways to prevent trachoma
whilst those who most needed this information, the main caregivers and water usage decision
makers, would have acted upon it were it related to them by their husbands (who rarely did),
and unfortunately remaining largely uninformed. ITI meetings were therefore characterised by a
125
majority of men and a handful of women attending, most village women remaining within their
households, convinced that the meetings werent relevant to them since they occurred in a space
where they were rarely welcomed.

Another problem emerged in how the meetings were promoted within the village. On one
occasion my fieldworker and I were travelling to a household at 6 o clock one morning when we
heard the village crier announcing an ITI meeting for 10 oclock that very morning. Whether ITI
had failed to inform the village well beforehand, or whether the village administration had
belatedly realised that the meeting was imminent will never be known, but later that morning
when ITI turned up, instead of a finding a crowd of interested people waiting to learn about
trachoma prevention, they found a mere twelve villagers had turned up, instead of a quorum
which should have measured a couple of hundred. This particular meeting was deemed a failure,
and rescheduled for two weeks later. Two weeks later, though, the rescheduled meeting had
been extensively promoted within the village and considerably more people turned up for the
trachoma sensitization meeting, milling around for hours before realizing that it was a no-show,
and venting their anger at ITI and the village administration having wasted their time which
could have been better used being economically productive. Whoever is responsible for the poor
handling of both occasions is neither here nor there; the important point is that ITI was
simultaneously alienating its target audience whilst failing to communicate its core message.

Compounding the problem is the fact that many of the balozi leaders who managed to make it to
previous meetings often failed to communicate the key learnings and messages on to the
households they represented. It may be possible, though, that they didnt understand everything,
a likely possibility given the variation of answers told me when asked about causes and
preventative mechanisms. It would appear, therefore, that there are some systemic difficulties in
the communication of basic trachoma facts which need to be addressed to ensure that villagers
are at least provided with accurate information so that they have a choice whether to act on this
or not.

ITI also have a representative within the village (the Village Health Officer) who is provided with
a bicycle and whose job it is to go out once a week to reinforce trachoma sensitization and
ensure that people understand the various components of the SAFE strategy with visits to
households within the village. However, whilst this is undoubtedly important role, the ITI village
representative revealed that they had formulated no real plan or strategy as to how this was
going to be done or received no direction and guidance from ITI as to what to do. Households
were therefore visited on an ad-hoc rather than systematic basis which meant that coverage
within the village was patchy and progress slow.

Reaching the target audience

Efforts must be expended in the area of health education as it relates to preventing and reducing
trachoma. However, according to informants accounts and my own observations, such
educational efforts as have occurred are sporadic, unstructured, unfocused and generally to date
largely unsuccessful. Given the current limitations (above), what might be the best way forward
to ensuring that the various message(s) that trachoma agencies wish to convey reach their
audience?

1) Perhaps agencies could co-ordinate more with each other so that a more unified and
coherent front is presented, reducing confusion amongst end users as to their presence
and purpose.

2) Re-evaluate the role that the agency itself plays in poor uptake of intended messages

126
3) Reconsider using alternate mechanisms and venues which act to maximize attendance
and participation. Possible ideas include:

a. decentralizing meetings so that they are held on a smaller scale, say at a
subvillage rather than village level. This would help boost attendance by reducing
distances that individuals have to travel, thereby making it less inconvenient to
attend and resulting in less time lost for productive purposes.

b. Learnings from subvillage meetings could be consolidated further by balozi-level
meetings where key trachoma messages are reinforced and individual
understanding easily verified.

c. Ensure that meetings are regularly scheduled, that the schedule is reliable and
that they are advertised well in advance.

d. In addition to official political structures, a variety of other structures exists, the
main one being those created by the Church such as the numerous religious
groups which could be used to diffuse trachoma educational messages. Using the
Church would be advantageous because of the size of its membership virtually
the entire village, the regularity of contact, and the expressed willingness of the
priests to diffuse health messages.

e. In order to counter the perceived male bias in using conventional political
structures, more attention should be paid in cultivating and encouraging the
nascent womens groups to act as a vehicle for trachoma education.

f. There have been some efforts made within primary schools to teach the
principles of trachoma prevention. These seem to have petered away perhaps
because of a lack of water for demonstration purposes. Thinking long-term these
may be a valuable juncture to start changing behaviours by educating future
adults as to the benefits of the F and E components of the SAFE strategy.
However, a commitment to repeated reinforcement of such messages must be
made so that with subsequent exposure, these messages become more familiar,
less alien
122
.

4) The ITI field representative is potentially a highly valuable asset. However, in order for
that potential to be fully realised, a more coherent household-visiting strategy needs to
be created which needs to be more actively managed in order to ensure systematic
coverage of the various households within the village.

5) So far trachoma education efforts have all been top-down affairs, and have imparted
little sense of partnership or co-ownership of trachoma educational messages. Efforts
should be made to explore how best to encourage villagers to form grassroots self-
education groups.
123124
However, there is a precedent for the positive influence of peer

122
Lynchs 1994 study indicates that repeated exposure of face-washing messages do result in improved face-washing
practices, including thoroughness, and a decrease in flies observed on childrens faces: a key factor in tackling the
transmission of trachoma (Lynch 1994: 516)

123
Studies how that external programmes that dont meet local defined needs often suffer from poor uptake (Egovil 1993:
995), the key being that the educational goals have to be seen as practical and relevant (Coulibaly 2002) as worthwhile to
strive for by the general community (Nurse 1985).

124
Such groups, called Care Groups by Sutter have been shown to be popular and effective in transmitting knowledge
and instrumental in changing attitudes within the community (Sutter 1983: 1815 - 1816).
127
approval leading to a significant uptake of health-seeking behaviour. Mothers who have
had positive child immunization programmes encourage other mothers to get their
children immunized. The result is that there is widespread compliance with child
immunization programmes, with possibly 70-80% of mothers possessing immunization
records, an indication that enlisting the support of the villagers themselves might be a
highly effective route for diffusing key trachoma prevention messages.

If the SAFE strategy is going to maximally adopted by those most at risk of trachoma infection,
then greater creativity and flexibility needs to be implemented by responsible agencies in
ensuring that this message is heard in as many different ways and at as many different levels as
possible, and in ways that are most acceptable to the target audience.

Getting the message across

The fact that the SAFE strategy messages werent getting through was found in the common
perception expressed of a lack of connection between initial stages of trachoma and latter stages.
Even individuals who had become blind through repeated trachoma infection and finally corneal
abrasions failed to make a connection between their blindness and the various stages of
trachoma, using whatever explanatory model that would fit their cosmology and provide meaning
for their experience
125
. However, once we explained the concept of repeated infection and
physically demonstrated the physical process of inversion and abrasion that develops, individuals
seemed to be able to make a connection between their own experience and the physically based
explanation we had provided, and seemed genuinely astonished that this was the developmental
trajectory of trachoma. They were even more astonished, it seemed, because although they had
children that had been identified with trachoma infection, they had not connected their condition
with their blindness. Once they had grasped this, they seemed more than willing to listen to our
advice about trachoma prevention. Furthermore, when asked if they would be willing to
participate in any trachoma education programmes within the village, they indicated that they
would.

The implications of this is that perhaps a full review of how the concept of trachoma is presented
to rural villagers of varying educational attainment and possessing a range of various health-
seeking behaviours. It may be that new ways of presenting what may seem rather abstract and
mysterious medical concepts to most villagers could be found that would present these ideas in
simple, graspable and understandable terms. Attempts to present complex information in terms
and language that most villagers would understand might go someway to helping bridge the
communication barriers that are often experienced between medically trained personnel and the
lay public. It is possible that introducing physical models of the eye using everyday objects like
fruit and something to represent the eyelashes scraping across the cornea might be help to
represent a constellation of ideas in one impressive visual example. Furthermore, enlisting the
assistance of individuals who have been blinded through trachomatic processes who may be able
to relate their own experiences of going blind and of subsequently learning about how they could
have saved their sight if only. The physical presence of such individuals who are known to the
local villagers might help forcibly convince the doubtful that there are consequences to not being
pro-active and that trachoma therefore deserved to be taken more seriously. This might result in
the message being more readily accepted than some city slicker health worker with more often
than not a somewhat disdainful attitude to the poor and the sick (Van Etten 1976: 84) trying to
convince them in perhaps arcane and abstract ways about the dangers of trachoma.


125
One individual had reasoned that her blindness was caused by God who used the Devil as the agent to punish her for
her parents attempts to cure her trichiasis using traditional medicine.
128
CONCLUSION AND RECOMMENDATIONS


Overview

Across many parts of the world householders are confronted with the daily need to source water,
nowhere more so than in Africa (see Appendix 2) where this challenge to daily survival has been
observed to be increasing across the continent, particularly amongst East African households
lacking piped water supply (White 1972; Thompson 2001).

This study has attempted to present a comprehensive picture of patterns of water access and
water usage within one small rural community nestling between Tanzania and Kenya which has
revealed great differences between different socioeconomic groups as well as within similar
socioeconomic divisions in terms of water access. Furthermore, it has highlighted patterns of
water usage which may have a bearing on the prevalence of trachoma.

Recommendations

Based on informants accounts and previous research findings, increasing water availability within
the study site is unlikely to result in a decrease in the prevalence of trachoma unless new
supplies of water are accompanied by health education programmes which focus on reinforcing
pre-existing notions as well as presenting a case for the importance of hygiene for family well
being.

Such programmes should not necessarily aim to elevate the importance of hygiene
behaviours by devaluing existing water usage priorities, rather they should aim to build
on and extend existing hygiene behaviours.

A key aim of such campaigns should be a deliberate attempt to connect childhood forms
of trachoma with the later stages of the disease. New and creative ways of presenting
such information should be considered, including using willing individuals suffering from
TT within health education efforts.

Greater co-ordination is needed in delivering health messages so that target audiences
are presented with coherent and reliable information rather than the fragmented
messages they are receiving at present.

A variety of alternative vehicles for information dissemination should be considered:
reliance on official structures alone are unlikely to reach the most influential
demographic.

Efforts to disseminate information should include community perspectives and
involvement as well as be characterised by a high degree of structure and management
with long-term follow-up.









129

The short time frame precluded in-depth exploration of the following areas which could benefit
from further anthropological research to establish whether they merit inclusion in future
strategies to improve water access and usage:

The role of problematic drinking in limiting access to water supplies, its impact on
trachoma prevalence, and how best to tackle such issues and factor them within
trachoma control programmes.

The importance of water-recycling within households of equal socio-economic and socio-
environmental status in determining the prevalence of trachoma.

Greater exploration of differentials in bathing order and possible impact on trachoma
prevalence.

Further researching into determining factors of household water usage priorities, and
how best to ensure greater water availability can be translated into greater usage of
water for hygiene purposes.



































130
BIBLIOGRAPHY


Abel-Smith, N. (1986) 'The World Economic Crisis - Repercussions on Health', Health Policy
and Planning, Vol. 1 # 3.

Ajewole, J. et al.(2001) Understanding the community perspectives of trachoma: the Gambia as
a case study, Ophthalmic Epidemiology, 8 (2-3): pp. 163 180.

Almendon, A. (1997) Hygiene evaluation procedures: Approaches and methods for
Blumenthal, U. assessing water and sensitization-related hygiene practices, INFDC,
Henderson, L. London.

Anonymous (1973) The Chagga, In: The Family of Man, Vol. 2, Part 18, Marshall
Cavendish, London.

Arens, W. (1979) On the Frontier of Change: Mto Wo Nbu, Tanzania, University of
Michigan Press, Ann Arbor.

Baer, H. & (1982) Why not have a Critical Medical Anthropology?, Paper presented
Singer, M. at the Annual Meeting of the American Anthropological Association, Washington,
D.C.

Bailey, R. et al. (1991) Trachoma and water use: A case control study in a Gambian village,
Transactions of the Royal Society of Tropical Medicine and Hygiene, 85:
824 828.

Bailey, R. (2001) The SAFE strategy for the elimination of trachoma by 2020: Will it
work?, Bulletin World Health Organisation: 233 236.

Barlow, M & (2003) Blue Gold: The Fight to stop the Corporate Theft of the Worlds
Clarke, T. Water Resources, New Press. New York.

Bjorndalen, J. (1992) Tanzanias Vanishing Rain-Forests Assessment of Nature Conservation
Values, Biodiversity and Importance for Water Catchment, Agriculture
Ecosystems & Environment, 40 (1-4): 313-334.

Brewin, D.R. (1964) Kilimanjaro Agriculture, Tanganyika Notes and Records, 64: 115
117.

Cairncross, S. & (1987) Water use and health in Mueda, Mozambique, Transactions of the
Cliff, J. Royal Society of Tropical Medicine and Hygiene, 81: 51 54.

Cairncross, S. (1988) Domestic Water Supply in Rural Africa, In: Rural Transformation in
Tropical Africa, D. Rimmer ed., Bethune, London.

Cairncross, S. (1989) Elasticity of demand for water in Khartoum, Sudan, Social Science &
Medicine, Vol. 34 # 2: 183 189.

Cairncross. S. & (1993) Environmental Health and Engineering in the Tropics, 2
nd
edition.
Feachem, R. Wiley, Chichester.

Clark, C.O. & (2003) Interdecadal Variability of the Relationship between the Indian Ocean
131
Webster, P.J. & Zonal Mode and East African Coastal Rainfall Anomalies, Journal Of Climate,16
Cole, J.E. (3): 548-554.

Caplan, P. (1997) African Voices, African Lives: Personal Narratives from a Swahili
Village, Routledge, London.

Cassidy, C. (1987) World-View Conflict and Toddler Malnutrition: Change Agent Dilemmas
In: Child Survival, ed. Nancy Scheper-Hughes, Dordrecht, Holland, Reidel,

Chen, L. (1981) Sex-biased behaviour in the intra-family allocation of food and the use of
Huq, E. & healthcare services in rural Bangladesh Population and Development
DSouza Review, 7(1) 55-70.

Collier, L. (1999) Impact of face washing on trachoma', The Lancet, 345: (8950): pp.
650 - 651.

Congdon, N. et al (1993) Exposure to children and risk of active trachoma in Tanzanian women,
American Journal of Epidemiology, 137 (3), pp. 366 372.

Coulibaly, H. (2002) Mothers Perception of Quality of Growth, Ecology of Food and
Nutrition, Vol. 41 # 6: 937 947.

DeWalt, K. (1983) Nutritional Strategies and Agricultural Change in a Mexican
Community, Ann Arbor, UMI Research Press.

Dolin, P. et al. (1997) Reduction of trachoma in a sub-Saharan village in absence of a disease
control programme, The Lancet, Vol. 349, May 24.

Drangert, J.O. (1993) Who Cares About Water? Household Water Development in
Sukumaland, Tanzania., Linkping University Press, Linkping.

Egeno, B. & (1973) Population growth in the population of Tanzania, Dar es Salaam:
Henin, R. Bureau of Resource Assessment and Land Use Planning, University of Dar es
Salaam and the Bureau of Statistics. Dar es Salaam.

Egovil, M. (1993) Nutritional programs, Social Science & Medicine, Vol. 37 # 8: 995
1003.

Engle, P. & (1993) Intra-Household Food Distribution Among Guatemalan Families in a
Nieves, I. Supplementary Feeding Program: Behaviour Patterns, Social Science &
Medicine, Vol. 36 # 12: 1605 1612.

Foster, G. (1987) Bureaucratic Aspects of International Health Agencies, Social Science
& Medicine, 25: 1039 1048.

Gillingham, M. (1999) Gaining access to water: Formal and Working Rules of Indigenous
Irrigation Management on Mount Kilimanjaro, Tanzania, Natural Resources
Journal , 39 (3): 419-441.

Greenway, P.J. (1964) The Vegetation and Flora of Mt. Kilimanjaro, Tanganyika Notes and
Records, 64: 97 98.

Griffiths, P. & (2002) Gender, Family and the Nutritional Status of Children in three Culturally
132
Matthews, Z. & Contrasting States of India, Social Science & Medicine, Vol. 55 # 5: 775
Hinde, A. 790.

Glaser, B. & (1967) The Discovery of Grounded Theory, Aldine, Chicago.
Strauss, A.

Grove, A. (1993) Water Use by the Chagga on Kilimanjaro, African Affairs, 92: 431448.

Hastenrath, S. (1997) Glacier recession on Kilimanjaro, East Africa, 1912-89, Journal of
Glaciology 43 (145): 455-459.

Heald, S. (1989) Controlling Anger, The Sociology of Gisu Violence (Manchester/New
York.

Helman, C. (2000) Culture, Health and Illness (4
th
ed), Arnold, London.

Hendry, J. (1994) Drinking and Gender in Japan, In: McDonald, M. (Ed.) Gender, Drink
and Drugs, Berg, Oxford.

Howard, M. & (1997) Hunger and Shame: Child Malnutrition and poverty on Mount
Millard, A.V. Kilimanjaro, Routledge, London.

Hull, R.W. (1976) African cities and towns before the European Conquest, W.W.
Norton, New York.

Huntingdon, S. (1996) The Clash of Civilisations, Touchstone Books, New York.

Ikenberry, G.J. (2002) Americas Imperial Ambition, Foreign Affairs, Sep/Oct 2002

Irion, R. ( 2001) The Melting Snows of Kilimanjaro Science, 291 (5509): 1690-1691.

Jones, B. (1975) The prevention of blindness from trachoma, Trans. Ophthal. Soc. UK.
Vol. 95, 16.

Kabanda, T. & (1999) Inter-annual variability of short rains over northern Tanzania
Jury M.R. Climate Research, 13 (3): 231-241 Dec.

Kimambo, I. (1996) Environmental control and Hunger, In: Custodians of the Land
Maddox, G. Giblin, J., Kimambo I. eds. James Currey, London.

Kolb, D.A. (1984) Experiential learning: Experience as a Source of Learning and
Development, Prentice-Hall, New York.

Koponen J. (1996) Population: A dependent variable, In: Custodians of the Land
Maddox, G. Giblin, J.l Kimambo I. eds. James Currey, London.

Korzybinski, A. (1933) Science and Sanity, Institute of General Semantics, New York.

Lane, S. et al, (1993) Sociocultural Aspects of Blindness in an Egyptian Delta Hamlet: Visual
Impairment vs. Visual Disability, Medical Anthropology, Vol. 15 : 245 260

133
Lema, T. (1999) Chagga Religion & Missionary Christianity on Kilimanjaro: The initial
phase 1893 1916, In: East African Expressions of Christianity, Spear, T.
& Kimambo I. eds. James Currey, London.

Lynch, M. (1994) Testing a participatory strategy to change hygiene behaviour - face
washing in central Tanzania', Trans. R. Soc Trop Med, 88 (5): pp. 513 - 517.

Malinowski, B. (1922) Argonauts of the Western Pacific, Routledge, London.

Marealle, P. (1963) Chagga Customs, Beliefs and Traditions, Tanganyika Notes and
Records 64: 56-61.

Maro, P.S. (1988) Agricultural Land Management under Population Pressure - The
Kilimanjaro Experience, Tanzania Mountain Research And Development
8 (4): 273-282 Nov 1988.

Marx, R. (1989) Social factors and trachoma: A review of the literature, Social Science
& Medicine, Vol. 29 # 1: 22 34.

Masao, F.T. (1974) The Irrigation System in Uchagga: An Ethno-Historical Approach,
Tanzania Notes and Records, 75, 1-8.

Mathur, G & (1970) Influence of some socio-economic factors on the prevalence of
Ramewshwar, S. trachoma, IndianJournal of Medical Sciences, Vol. 24: 325 333.

Mdoe, N. (1997) Returns to Smallholder Dairying in the Kilimanjaro Region, Tanzania,
Agricultural Economics, 17 (1): 75-87 OCT 1997.

Mccauley, A. (1990) Changing water-use patterns in a water poor area - lessons for a
et al trachoma intervention Project', Social Science & Medicine, 31 (11): pp. 1233
- 1238.

Mccauley, A. & (1992) Household decisions among the Gogo people of Tanzania: Determining
West, S. & the roles of men, women and the community in implementing a trachoma
Lynch, M. prevention program Social Science & Medicine, 34 (7): pp. 817 824.

Mercer, C. (2002) The Discourse of Maendeleo and the Politics of Women's Participation on
Mount Kilimanjaro Development And Change, 33 (1): 101-127 Jan.

Moore, H. (1996) Space, Text and Gender: An Anthropological Study of the
Marakwet in Kenya, Guilford Press, NY.

Moore, S.F. (1986) Social Facts and Fabrications Customary Law on Kilimanjaro,
1880 1980, CUP, Cambridge.

Moore, S. (1996) Post-Socialist micro-politics: Kilimanjaro, 1993 Africa 66 (4): 587 - 606

Msuya, S.E. (2002) Female Genital Cutting in Kilimanjaro, Tanzania: Changing Attitudes?
et al. Tropical Medicine & International Health, 7 (2): 159-165 Feb.

Munoz, B. & (1997) Trachoma: the forgotten cause of blindness, Epidemiological
West, S. Reviews, 19: 205 217.

134
New, C. (1873) Life, Wanderings and Labours in Eastern Africa, Hodder and
Stoughton, London.

Nurse, D. (1979) Classification of Chaga Dialects: Language and History on
Kilimanjaro, the Taita Hills and the Pare Mountains, Buske, Hamburg.

Nurse, G.T. (1985) Malnutrition, Ignorance and Poverty, British Medical Journal, Vol.
291, 14
th
September: 738.

O Connor, I. (1975) Social and Cultural Factors Influencing Drinking Behaviour Irish Journal
of Medical Science (suppl.) June, pp. 65-71.

Pakenham, T. (1991) The Scramble for Africa: White Mans Conquest of the Dark
Continent from 1876 1912, Avon Books, New York.

Peace, A. (1992) No fishing without drinking: The construction of social identity in rural
Ireland, In Gefou Madianou, D. (ed.) Alcohol, Gender and Culture,
Routledge, London: 167 180.

Pike, A.G. (1964) Kilimanjaro and the Furrow System, Tanganyika Notes and Records,
64: 95 96.

Pilger, J. (2002) The New Rulers of the World, Verso, London.

Polack, S. (2002) Evaluating the effect on the prevalence of active trachoma of
improving the water supply in Rombo District, Tanzania, LSHTM, London
(unpublished)

Prost, A. (1989) Water, trachoma and conjunctivitis', B. World Health Organ, 67 (1):
pp. 9 - 18.

Pruss A, & (2000) Preventing Trachoma Through Environmental Sanitation: A Review of the
Mariotti, S. Evidence Base, Bulletin of the World Health Organisation, 78 # 2: 258.

Pruss, A. (2002) Estimating the burden of disease from water, Sanitation, And Hygiene At
A Global Level, Environmental Health Perspectives, 110 (5): 537-542.

Ramsey, J.C. (1964) Kilimanjaro: Water Supplies, Tanganyika Notes and Records, 64: 92
94.

Raum. O.F. (1940) Chaga Childhood: A Description of Indigenous Education in an
East African Tribe, OUP, London.

Richardson, L. (1990) Narrative and Sociology, Journal of Contemporary Ethnography,
19: 116-35.

Rodney, W. (1981) How Europe Underdeveloped Africa, Bogle Louverture Publications,
London.

Rosegint, M. & (2002) Dealing with Scarcity, IFPRI, Washington. Cai X.

Scheper-Hughes(1987) Culture, Scarcity and Maternal Thinking: Mother Love and Child Death in
N. northeast Brazil, In: Child Survival, ed. Scheper-Hughes, N. Reidel, Dordrecht.
135

Setel, P. (1996) AIDS as a Paradox of Manhood and Development in Kilimanjaro,
Tanzania, Social Science & Medicine, 43 (8): 1169-1178.

Setel, P. (1999) A Plague of Paradoxes: Aids, Culture and Demography in
Northern Tanzania, University of Chicago Press, Chicago.

Singer, M. (1986a) Developing a Critical Perspective in Medical Anthropology, Medical
Anthropology Quarterly, Vol. 17 (9)5: 128-129.

Singer, M. (1986b) Toward a Political Economy of Alcoholism: The Missing Link in the
Anthropology of Drinking, Social Science & Medicine 23(2): 113-130.

Singer, M. (1989) The Coming of Age of Critical Medical Anthropology, Social Science &
Medicine, 28(11): 1193 - 1204.

Singer, M. (1995) Beyond the Ivory Tower: Critical Praxis in Medical Anthropology,
Medical Anthropology Quarterly, 9(10: 80-106.

Singer, M et al. (1998) Why does Juan Garcia have a drinking problem? The perspective of
Critical Medical Anthropology, In: Understanding and Applying Medical
Anthropology, Brown, P. (ed.), Mayfield Publishing, Mountain View, California.

Spear,T. (1996) Struggles For The Land: The Political And Moral Economies of Land on
Mount Meru, In: Custodians of the Land Maddox, G. Giblin, J.l. Kimambo I.
eds. James Currey, London.

Stahl, K.M. (1965) Outline of Chagga History, Tanganikya Notes and Records: 35 49.

Stambach, A. (2000) Lessons from Mount Kilimanjaro: Schooling, Community and
Gender in East Africa, Routledge, London.

Sutter, E. & (1983) Community participation in the control of trachoma in Gazankulu, Social
Science & Medicine, 17 (22): pp. 1813 1817.

Swantz, M.J. & (1975) Socioeconomic causes of malnutrition in Moshi district. Research
Henricson, U.S. paper # 38, Bureau of Resource Assessment and Land Use Planning ,
Zalla, M. University of Dar es Salaam, Dar es Salaam.

Taylor, H. et al (1989) Hygiene Factors and increased risk of trachoma in central Tanzania,
Archives of Ophthalmology, 107, 1821 1825.

Therkildsen, O. (1988) Watering White Elephants?: Lessons from Donor-Funded
Planning And Implementation of Rural Water Supplies in Tanzania,
Scandinavian Institute of African Studies, Uppsala.

Thompson, J. (2001) Drawers of Water 2: Summary: 30 years of change in domestic
et al. water use and environmental health in East Africa. IIIED, London.

Thylefors, B. (1990) The World Health Organisations programme for the prevention of
blindness. International Ophthalmology, 14: 211 219.

136
Tumwine, J. (2002) Diarrhoea And Effects of Different Water Sources, Sanitation And
Hygiene Behaviour in East Africa Tropical Medicine & International Health,
7 (9): 750-756.

UNOPS (2002) Aerial Survey to Mount Kilimanjaro, UNOPS, Geneva.

Van Etten, G. (1976) Rural Health Development in Tanzania, Amsterdam, Van Gorcum.

Van Maanen, J. (1988) Tales of the Field: On Writing Ethnography, University of Chicago
Press, Chicago.

Vavrus, F. (2002) Making distinctions: privatisation and the (un) educated girl on Mount
Kilimanjaro, Tanzania International Journal of Educational Development
22 (5): 527-547.

Von Clemm, F. (1962) The Interdependence of Political and Economic Activity amongst
the Chagga in Tanganyika, D.Phil. Thesis, Oxford University, Oxford.

West , S. (1989) Water availability and trachoma, Bulletin of the World Health
Organisation, 67: 71 75.

West, S. et al. (1995) Impact of face-washing on trachoma in Kongwa, Tanzania, The Lancet,
345: 155 8.

West, S. et al. (1991) Facial cleanliness and the risk of trachoma in families, Archives of
Ophthalmology, 109: 855 857.

West, S. et al. (1996) Risk factors for constant, severe trachoma among pre-school children in
Kongwa, Tanzania, American Journal of Epidemiology, 143: 73 78.

White, G. & (1972) Drawers Of Water: Domestic Water Use in East Africa, University
of Bradley, J. Chicago Press, Chicago.

WHO (1995) Community Water, Supply and Sanitation: Needs, Challenges and Health
Objectives, 48
th
World Health Assembly A48/INF.DOC/2,28 April 1995 Geneva,
Switzerland.

WHO (1997) Report of the First Meeting of the WHO Alliance for the Global
Elimination of Trachoma, Geneva, Switzerland,, 30 June 1 July 1997,
Available at: http://wholibdoc.who.int/hq/1997/WHOPBLGET97.1pdf

WHO & UNICEF (2000) Global Water Supply and Sanitation Assessment 2000 Report.
WHO Geneva and UNICEF, London.

Willis, J. (2000) Potent Brews: A Social History of Alcohol in East Africa, James
Currey, London.

Yoder, P.S. (1997) Negotiating Relevance: Belief, Knowledge and Practice in International
Health Projects Medical Anthropology Quarterly, Vol. 11 # 2: 131 141.

Zorita, E. & (2002) Rainfall Variability in Northern Tanzania in the March-May Season (Long
Tilya, F.F. Rains) and its Links to Large-Scale Climate Forcing, Climate Research, 20 (1):
31-40 Feb.
137
APPENDIX 1



The worlds water: Source UNEP



Global access to water and sanitation: Source WHO



Predicted global water stress and scarcity in 2025: Source UNEP
138
APPENDIX 2






139
APPENDIX 3

WHO trachoma grading card



140







141
APPENDIX 4

ITI (Tanzania) Trachoma Prevention Programme Promotional Leaflet





142
APPENDIX 5

Briefing leaflet for village participants

RESEARCH PROJECT: WATER COLLECTION AND USAGE IN A CHAGGA VILLAGE COMMUNITY

Researchers: Robert Th & Mosoi Balthazar

Information Sheet

We are conducting research into how water is obtained and used within your community. This sheet
explains why we are here, and how you can help us.

Why are we doing this research?

We are trying to prevent and control an eye disease called Trachoma. Perhaps you have heard of it since it
is found in many countries throughout Africa, as well as Tanzania. Trachoma is often found in areas with
little or no available water. It is an infection that affects the eyes, specifically behind the upper eyelids.
When there is little water available, faces and fingers may be washed less often and if just one person has a
Trachoma infection, it becomes very easy for other people to then get it until many people in the
community are affected.

Trachoma is often found in children under ten who may be re-infected many times as they grow uip. With
each additional infection, the eyelid becomes more and more damaged inside. Over time, a scar can form
which pulls and shortens the eyelid so that the eyelashes rub against the surface of the eye. This causes a
great deal of pain, damages the eye and often leads to blindness in adulthood. You may know or have
heard of someone who has experienced this.

Why Urauri?

The Rombo Trachoma Research Project has found that there are many people in Urauri who have been
infected with Trachoma. We have also found that water can be used to prevent Trachoma spreading within
a community. However, since water is not always easily or commonly available within Urauri village, we
would like to work with the community to find a way to best use the limited water it has so that its
members, especially its children, can be kept free of Trachoma infection and not become blind as they get
older.

What we want to know

To do this, we need to know exactly how the community obtains and uses its water. In order to do that, we
need to be based in the village so that we can get to know the villagers well and spend time observing
exactly what they do with the water. In addition, we would like to ask them questions about their
experiences with water.

What will this mean?

In order for our research to paint a true picture of the villages experience, we need to be able to spend
time with a wide range of village families. We will observe how water is used in different ways within each
household, and will try and minimise our presence as much as possible.

Your co-operation

We have not come here to judge or criticise, but to learn about how things are done in the village. If
anyone feels they want to ask additional questions about our research before agreeing to answer our
questions or be observed, then we would be happy to answer them. Lastly, even if an individual or
household agrees to participate in assisting with our research, they may freely withdraw at any stage.

Original English text by Robert The
143
Briefing leaflet for village participants Kiswahili translation

MRADI WA UTAFITI: MAJI NA MATUMIZI YAKE KATIKA JAMII YA KIJIJI CHA UCHAGANI

Watafiti: Robert Th na Mosoi Baltazar

Karatasi ya maelezo nakala ya wanakijiji

Tanaganya utafiti juu ya jinsi maji yanavyopatikana na matumizi yake katika jamii. Karatasi hii
inaeleza ni kwanini tupo hapa na namna utakavyotusidia.

Kwanini tunafanya huu utafiti?

Tunajaribu kuuzuia na kuuthibiti huu ugonjwa wa macho wa kikope (trachoma). Labda
umeshawahi kuusikia tangu utokee katika nchi nyingi za Afrika na pia hapa Tanzania. Ugonjwa
wa kikope (trachoma) hunapatikana has sehemu nyingi zenye maji haba ua hamna maji
kabisa. Ugonjwa huu hushika sehemu ya kifuniko cha juu cha jicho kwa ndani. Mahali penye
uhaba wa maji, usafi wa uso na mikono ni vigumu kufanyika na hufanyika mara chache na
kama mtu mmojo akiwa na trachoma inakuwa rahisi kwa watu kuambukizwa na hii huenaea
kijiji kzima.

Trachoma inapatikana hasa kwa watoto wadogo wenye umri chini ya wa miaka kumi ambao
huambukizwa mara nyingi kadri wanawyokuwa. Kilk wanapoambukizwa kifuniko cha jicho
huaribika kutokana na kuvo, hukunjika na kusababisha kope za jicho kuvutwa na kuangalia
ndani na kukwaruza kioo cha jicho unapopepesa (kufunga na kufungua) mach. Hii husababisha
maumivu makli huharibu kioo cha jicho na kuleta upofu ukubwani. Labda umeshawahi kusika
au kumwaona mtu mwenye shida hii.

Kwanini tumekuja Urauri?

Wataftiti wa ungonjwa huu wa kikope (Rombo Trachoma Research Project) walipata watu
wengi sana walioathirika na ugonjwa huu katika cha Urauri walipofanya utafiti mwaka jana.
Tumegundua kuwa maji yanaweza kuzuia ueneaji wa ugonjwa huu katika jamii. Hata hivyo
majihayapatikani kwa wingi katika kijiji Urauri, tungependa kufanya kazi na jamii katafuta njia
rahisi uya utumiaji wa maji ili watoto waweze kuzuia uambukizwaji wa ungonjwa wa kikope
na kuepuka upofu ukubwani.

Tunachotaka kujua

Kufanya haya, tunahitaji kujua uhakika wa jamii inavyopata maji na matumizi yake. Tukitaka
kufahamu haya inatubidi tuishi kijijini na wanavijiji tuwaelewe wanakijiji kukaa nao na
kuangalia wanavyoyatumia maji. Paia tutauliza maswali kuhusu uzoefu wa matumizi ya maji.

Hii ina maana gani?

II utafiti wetu uwe na picha halisi ya hali ya uzoefu wa maji kijijini, tunahitaji kutukia muda
mwingi na wanakaya wengi. Tutaangalia jinzi maji yanavyotumika kwa nia nyingi kwa kila
kaya pia tutajaribu kupunguza madu wa kukaa nanyinyi ikiwezekana.

Ushirikiano wenu

Hatukuja hapa kuamua au kukosoa balu kujifunza jinsi mambo yanavyofanyika kijijini. Mu
yeyote anayetaka kuuliza swali lolote kuhusu tuafiti huu kabla ya kukubali kujibu maswali yetu
au kuchunguzwa, tutafurahi kuyajibu maswali yake.

Mwisho mtu yeyote anayeshiriki au kusaidia katika utafiti huu yukp huru kujitoa wakati
wowote.

Kiswahili translation by Patrick Massae
144
APPENDIX 6

Research Schedule

Week 1 (2
nd
8
th
June 2003) Engagement

Arrival in Rombo district, meeting KCCO and RTRP staff, Meeting key local political figures to
obtain permission to conduct study, arranging accommodation within village

Week 2 (9
th
June 15
th
June) Taking initial field bearings: Semi structured interviews with
village political officers

Monday 9
th
June Water chairman (WC), Village Health Officer (VHO)
Tuesday 10
th
June Subvillage leaders
Wednesday 11
th
June Reha Balozi leaders + Funeral ceremony
Thursday 12
th
June Mhaka Balozi leaders
Friday 13
th
June Nesae Balozi leaders
Saturday 14
th
June Village Chairman (VC), WC and VHO History of water issue &
village history Travel to Mkuu to update and review fieldnotes,
write field report for both Brunel and LSHTM, and download
digital pictures.
Sunday 15
th
June Return to Urauri Village

Week 3 (16
th
June 22
nd
June) Exploring local experiences. focus groups and initial mapping

Monday 16
th
June Reha Focus Group Discussion (FGD) 5 men and 5 women
Tuesday 17
th
June GPS activity identifying main sources of water
Wednesday 18
th
June Mhaka FGD 5 men and 5 women
Thursday 19
th
June Nesae FGD 5 men and 5 women
Friday 20
th
June Reha and Mhaka Primary School FGD
Saturday 21
st
June VC, WC and WHO interview # 2. Visit Mkuu to update and
review field notes, write field report, and download digital
pictures.
Sunday 22
nd
June Return to Urauri

Week 4 (23
rd
June 29
th
June) Household Observations 1 (Mhaka) Grass roots experience,
establishing baseline households

Monday 23rd June Household # 1
Tuesday 24th June Household # 1
Wednesday 25th June Household # 1
Thursday 26
th
June Household # 2
Friday 27
th
June Household # 2
Saturday 28
th
June Visit Moshi to meet with Paul Courtright and Susan Llewellyn at
KCCO, write brief field report for both Brunel and LSHTM,
download digital pictures
Sunday 29
th
June Return to Urauri






145
Week 5 (30
th
June 6
th
July) Household Observations 2 (Nesae) Exploring trachoma-positive
households in Nesae Subvillage

Monday 30
th
June Household # 3
Tuesday 1
st
July Household # 3
Wednesday 2
nd
July Household # 4
Thursday 3
rd
July Household # 4
Friday 4
th
July Gabriel Toktas Wedding (ceremony and reception) Witness
the interrogation of a suspect thief
Saturday 5
th
July Visit Mkuu to update and review fieldnotes, write brief field
report for both Brunel and LSHTM. Return to Urauri village
Sunday 6
th
July Attend church services. Wake at Household # 1

Week 6 (7
th
July 13
th
July) Household Observations (Nesae) Exploring trachoma-negative
households in Nesae Subvillage

Monday 7
th
July Household # 5 No TF/TT
Tuesday 8
th
July Re-interview Householders # 3 and # 4
Wednesday 9
th
July Household # 6 Nesae Balozi leader
Thursday 10
th
July Household # 7. GPS readings of Village roads
Friday 11
th
July Visit Moshi to meet KCCO staff, update and review fieldnotes,
submit field report and exchange information with fellow Brunel
medical anthropologist in the field
Saturday 12
th
July Return to Urauri
Sunday 13
th
July Semi-structured Interviews with two traditional healers

Week 7 (14
th
July 20
th
July) Household observations (Mhaka) exploring trachoma-positive
households, Focus groups on household management and
parenting issues.

Monday 14
th
July ITI scheduled sensitisation meeting but no-show & angry
villagers remonstrate with village leadership
Tuesday 15
th
July Semi-structured interview with a rainmaker (AM) and FGD with
female informants about household management, childrearing
and parenting (PM)
Wednesday 16
th
July 2 FGDs on household management, childcare, gender and
hygiene
Thursday 17
th
July Household observation # 8
Friday 18
th
July Household observation # 9
Saturday 19
th
July Visit Mkuu to update and review fieldnotes, submit field report
Sunday 20
th
July Meeting and discussion of fieldwork with Hannah Kuper and
Sarah Polack (LSHTM). Return to Urauri

Week 8 (21
st
July 27
th
July) Conclusion and departure from field. Final household
observations trachoma-negative household in Mhaka subvillage. Final focus groups and
interviews.

Monday 21
st
July Household observation # 10
Tuesday 22
nd
July FGDs with village political leadership (about village
political issues) and group of 5 female informants
Wednesday 23
rd
July Mkuu GPS technical assistance from Sarah Polack, Interviews with ITI
and Kiliwater (KW) management. Final departure to Moshi followed by
departure for Dar es Salaam and the UK.
146
APPENDIX 7

Standard Question Schedule for Informants &
Household Observation Participants
____________________

Name of household: Household #

Status Age Educational Level Literacy












Land Ownership (Since)
Size
Agricultural Produce
Avocadoes
Bananas
Beans
Cassava
Coffee
Fruit
Maize
Millet
Other
Peas
Trees (for wood)

Livestock Number
Chickens
Cows
Ducks
Goat
Pig
Turkey


WHO WORKS THE SHAMBA?

***
Income Source:

147
Income
Weekly
Monthly



Wealth indicators

Bicycle
Motorbike
Car
Electricity
Radio
Television
One wooden house or more
Permanent House

***
Household Water Status

Private tap
Cement Storage
Rain Harvesting

Water Sourcing

Source
Why do they go there?
Distance
Who Collects
How much do they collect
How often
What time do they leave, return?

How much water do they use per day? Week/ How much does it cost?

Water usage

Who decides?
What is water used for?


Who uses the most water? Why?
Who uses the least water? Why?
Do they prioritise water usage in the rainy
season? How





Do they recycle water? How?
148


Are women cleaner than men? Why?
Is there any order in personal washing?


Who washes the smallest children?

***
Knowledge and experience of trachoma education/treatment programmes

Have you been to any trachoma information meeting?
Has anyone come to you before to tell you about trachoma?
Have you or anyone in your family been examined for trachoma?
Have you or anyone in your family received any pamphlets about trachoma?

If you have heard of trachoma what can you say about it?
What causes it?
What prevents it?

***

Alcohol consumption in the house

Mother
Father

***

Membership of support organisation

Type of
organisation
Frequency of
meeting
Number of
members
Membership
dues
Kiarano
Clan
Religious groups

***











149
APPENDIX 8

Indicative questions for Focus Group Discussions

General Questions

Why am I here (Why do you think the mazungo is here in the village)?
Why is there a water problem in the village?
Whose responsibility is it to solve the problem?
What have the village council done so far to solve it?
To what extent have their efforts been successful/unsuccessful? How?

Village political leadership

What do you think of the village council leadership in general?
How effective are they at leading the village? What are its shortcomings?
What do you think of the balozis? To what extent do they do a good job? Are their
opinions important?

Health Issues

What are the major health problems in the village?
Why do people get sick?
When people are sick, what do they do to get better (Health-seeking behaviour)
What do you think is the best way for the village to solve its own health problems?
What are the difficulties involved in tackling village health problems?
To what extent should the village solve its own health problems?
When should outsiders become involved with issues of health?
What is the best way for outside people to assist the village?
What efforts have been made to tackle eye problems/illnesses in Urauri? Do you feel you
have been kept informed or meetings/treatments?

Hygiene issues

We have noticed that people change their clothes when they leave the house, why is
this?
When is it important to wear clean clothes?
Why is it OK to wear clothes that are not so clean within the household?
We have noticed that women make more effort to appear clean (personal washing and
cleaner clothes) than men, why do you think this is so?
How often (ideally) should people wash (including children)?
Which parts of the body are the most important to keep clean?

Household management

How can you tell when a household is well-run (or organised) what might such a
household look like?
How can you tell when a house is poorly-run (disorganized) what might such a
household look like?
Why do you think some households are better organised than others?
What are the consequences of a household being managed well economically, and in
terms of family health?
150
What are the consequences of a household being managed badly economically, and in
terms of family health?
Are there any other consequences of a well or badly-run household?
To what extent do you think that alcohol consumption has an effect on how a household
is run?

The consumption of alcohol within the village

What do you consider is a reasonable amount of alcohol to drink in one sitting/per day?
Why do some people drink beyond this amount?
Is excessive drinking a significant problem in Urauri?
If yes, how big a problem do you think it is?

Being a parent/childcare issues

What are the responsibilities of a husband/wife in Chagga culture?
What are the attributes of a good father?
What are the attributes of a bad father?
What are the attributes of a good mother?
What are the attributes of a bad mother?
Why do you think there are good and bad fathers?
Are these reasons the same for good and bad mothers?
What are the consequences for the children if the parenting is good, if the parenting is
bad?
We have noticed that some children are left alone in the house for long periods of time,
and the parents are absent, why do you think this occurs?
Can this be avoided?
How would you describe this as parenting behaviour. How good or bad is this as
parenting behaviour?
If unavoidable, how old should the eldest child be to look after the youngest child?
We have also noticed that many children wear dirty clothes in the household, why do
think this is?
Some children are also sometimes naked. What kind of parents do these children belong
to?
We have also noticed that some children have very dirty faces, why might this happen?
What do you think of the flies that are everywhere? To what extent are they a problem?
In what way?












151
APPENDIX 9

Tanzanian Political Organisation and Political Offices


Ten household cell
Nyumba Kumi
Ten Household cell
Representative
Balozi wa Nyumba
K i
Village
Kijiji
Village Executive Officer
Afisa Mtendaji wa Kijiji
Village
Kijiji
Village Chairman
Munyakiti wa
Ward
Kata
Ward Secretary
Katibu Kata
District
Tarafa
District Commissioner
Katibu Tarafa
Region
Mkoa
Regional Commissioner
Mkuu wa Mkoa
Nation State
Taifa
President
Rais
Village
Kijiji
Village Chairman
Mwenyakiti wa
Subvillage
Kitongoji
Subvillage Chairman
Mwenyakiti wa
Kitongoji
152
APPENDIX 10

Political organisation of Rombo District









Mengwe
Division
Mkuu
Division
Kikelelwa
Village
Mbomai
Village
Motamburu Kitendeni
Ward
Kibaoni
Village
Mhaka
Subvillage
23 Balozis
Reha
Subvillage
26 Balozis
Nesae
Subvillage
20 Balozis
Urauri
Village
Msangai
Village
Nanjara Reha
Ward
Kilombero
Village
Endoneti
Village
Motamburu
Ward
Tarakea
Division
Mashati
Division
Usseri
Division
Rombo
District
153
APPENDIX 11 - Threats To Mt. Kilimanjaro Forests
(Source UNOPS 2002 Aerial Survey to Mount Kilimanjaro)

















154
APPENDIX 12

Kiliwater Contract for provision of standing pipe

(Kiswahili and English)


KILIWATER COMPANY LIMITED

HATI MKATABA WA UWAKALA WA KUKSANYA MALIP YA MATUMIZI YA MAJI KATIKA VITUO VYA
KUCHOTEA MAJI. MKATABA HUU UMEFANYIKA KATI YA KAMPUNI YA MAJI KILIWATER
COMPANY LIMITED NA:

BW/BIBI

WA ENEO LA MATUMIZI YA MAJI NAMBA

KATIKA KANDA YA

KWA MKATABA HUU, BW/BIBI/

ATAKUWA MKUSANYAJI WA MALIPO YA MATUMIZI YA MAJI KATIKA KILULA NA KILICHO
KATIKA BOMBA/TAWI LA

LILILOPO KAYIKA KIJIJI CHA

Mkataba huu masharti yafuatayo:

1 Kwamba wakala atawajibika kwa kampuni ya KILIWATER chini ya uongozi wa Afisa
wa Kanda.

2 Kwamba Kilucha hicho kitafungwa mita ya kupima wingi wa maji yali yotumika kwa
gharama ya kampuni.

3 Kwamba Kilucha hicho kitawekewa koki na gate valve ya kufunga maji wakati
hayatumiki kwa gharama ya kampuni.

4 Kwamba mita hiyo itajengwa na itasomawa na kufanyiwa matengenezo na Kampuni.

5 Kwamba makusanyo ya malipo ni lazima yalingane na kiasi cha maji yaliyotumika
kama mita itakavyosoma.

6 Kwamba kila debe moja (au lita 20) litalipiwa Tsch . tu.

7 Kwamba kati ya hizo Tsh .. kwa kila debe moja (lita 20) la maji wakala
ataliowa posho ya Tsh . Na Kamati ya maji italipwa Tsh. . na
bakaa ya Tsh . italipwa kwa kampuni.

8 Kwamba afisa wa kanda atasoma mita hiyo kila wiki mara moja na kuchukua kiasi cha fedha
zilizopatikana pamoja na posho ya kamati ya maji ambapo kamati ya maji itachukua posho
yake kutoko kwa afisa wa kanda.

9 Kwamba usafi wa kilula na mazingira yanayokizunguka ni jukumu la wakala.

155
10 Kwamba usalama wa kilula na mita kwa ujumla ni jukumu la wakala.

11 Kwamaba kukitokea uharibufu wa makusudi unaorudia mara tatu mfululizo kwa kipindi cha
mwezi mmoka, itabidi huduma ifungwe katika kilula hicho.

12 Kwamba wakala atapendekezwa na kamati ya maji, na li lazima awe na wadhamini wawili.

13 Kwamba endapo wakala atatoroka na fedha taslimu, wadhamini wake watawajibika kulipa kiasi
wakala alichotoroka nacho, na wanaweza kumchukulia hatua za kisheria, na kampuni itatoa
ushahidi.

Mkataba huu umefanyika leo tarehe 19..

USHUHUDA:

i) Wakala au Mkusanyaji

Jina Sahihi .

Majina matatu (Jina la Kwanzaa, jina la Baba, Jina la ukoo

ii) Wadhamini

Jina kazi Sahini

Makazi/kijiji

Jina kazi Sahini

Makazi/kijiji

iii) Mashahidi

Jina kazi Sahini

Makazi/kijiji

Jina kazi Sahini

Makazi/kijiji

iv) Kwa niaba ya Kiliwater Company Ltd
(Jina na sahihi ya afisa wa kanda)

Jina Sahini







156
Kiliwater Contract

KILIWATER COMPANY LIMITED

AN AGREEMENT OF AN AGENT FOR WATER BILL COLLECTION FROM WATER STANDING PIPE
THE AGREEMENT IS UNDER KILIWATER COMPANY LIMITED AND:

MR/MISS.
THE AREA OF WATER USAGE NO: .
ZONE ..
THE AGREEMENT OF MR/MISS .
WILL BE THE COLLECTOR OF WATER BILL READINGS TO STANDPIPE No
IN THE MAIN LINE PIPE/BRANCH.....
IN THE VILLAGE

Regulations of the agreement are as follows:

1. The agent will be answerable to the Zone supervisor.

2. There will be an installation of meter reading to the standing water pipe and the expenses will be
borne by Kiliwater.

3. There will be a cock and gate valve for controlling water when not in use which will be installed at
Kiliwaters expense.

4. Installation of meter reading and repair of the meter will be the responsibility of Kiliwater.

5. Water bills will equal the amount of water used.

6. The cost of a single bucket of water (20 liters) will be Tsh only.

7. From this Tsh . amount, for every bucket of water (20 liters) of water, an agent will be
paid an allowance of Tsh .and water committee will be paid Tsh . The rest
of Tsh will for the company.

8. The zone officer will take meter readings every week and collect monies due weekly. This includes
the allowance of the water committee, which will be given to the committee by the Zone officer.

9. The agent will be responsible for the upkeep of the standing pipe and its immediate environment.

10. The security of the standing pipe and meter reading will be the responsibility of the agent.

11. If any equipment installed by the company are vandalized or destroyed three times within a single
month, the service will be immediately terminated.

12. The agent must supply two referees and be nominated by the water committee.

13. Should an agent disappear with monies collected, the referees will be liable for refunding said
monies, and in addition may be liable to prosecution.



Agreement date19.




157
Witnessed:

i) Agent / water collector

Name . Signature..

Three names: (first name, fathers name, clan name)

ii) Referees

Name Occupation Signature
Resident / village .

Name . Occupation . Signature ..
Resident / village

iii) Witness

Name Occupation Signature
Resident / village .

Name . Occupation . Signature ..
Resident / village

iv) On behalf of Kiliwater Company Ltd

(Name and signature of the Zonal co-ordinator)

Name .. Signature.



















158
APPENDIX 13

Glossary

Swahili English

Aibu Shame
Asante: Thank you
Balozi: Representative leader of ten households
Bongo Flavour: Tanzanian hip hop
Busa: Kind of alcohol
Bwana
mkubwa: A big man
Chai Spicy tea
Daktari: Doctor
Duka: Shop may refer to a Western style shop or a roadside stall
Elimu: Education
Gongo: Illegally distilled spirit: made from molasses
Heshima: Respect
Hohehahe: The destitute
Jiggas: Sand fleas
Kanga: Traditional shawl/skirt
Kiarano: Self-help group
Kibarua: Day labouring
Kichagga: Language of the Chagga people spoken on Kilimanjaro
Kihamba Prime agricultural belt in the highlands of Kilimanjaro
Kiswahili: National language of Tanzania, originates with the Swahili people on the coast of
East Africa
Kitenge: Similar to the Kanga but more formal and made of heavier material
Kubwa Big
Kuvimba: Swelling of the stomach Kwashiorkor
Maendeleo Progress
Maji Water
Maji safi Clean water (looks clean)
Maji salama Clean and healthy water (no micro-organisms)
Maji machufu Dirty water
maji ya mbua Rain harvesting
Mbege: Kind of alcohol
Mbuzi: Goat
Mangi: Traditional Chagga chief
Maskini: Poor person
Masumba: Wealthy man plural: wasumba
Mazungo: White man
Mchagga: Kiswahili noun for Chagga person plural: wachagga
Mfongo: Traditional Chagga furrows and irrigation system plural: wafongo
Mfuruwanj: Traditional Kilimanjaro tree
Mfumu: Traditional Kilimanjaro tree
Mizimwi: Ancestors
Mkuyu: Traditional Kilimanjaro tree
Mlevi: Problematic or alcoholic drinking behaviour
Morani: Masai Warrior
Mtembwe: Traditional Kilimanjaro tree
Mtoto wa maji: Water child offspring from rape which occurs while collecting water
159
Mungo: Lord - God (Christian conception)
Ndasoe: Sacrifice
Nginenyi Chagga word for the sky dwelling place of Ruwa
Nyati: Buffalo
Nyika Plains
Nyumba: Village household
Piki piki: Motorbike
Pombe: Kind of alcohol
Rungu: Masai Club
Ruwa: Chagga word for supreme being
Safi: Clean
Shamba: Traditionally lower grazing land, but now means arable land farmland
Simba: Lion
Tambiko: Ceremony ritual
Tembo: Elephant
Ugali Staple food of Tanzania, maize and cassava based
Uji: Watery porridge consumed for breakfast
Ujama: Socialist policy of collectivization implemented in Tanzania in the 1970s
Ukoo: Clan
Vijana wa
mitarni: Street boys
Vijana wa
Sururaji: Street boys, hoodlums, layabout
Vijidudu: Microorganisms
Vikao: Groups of men
Wafadhili Donors
Warumu: Ancestors
Watu wa
porini: Backward people of the plains
Yesu: Jesus

160
APPENDIX 14
Water Conflict Chronology
126



Date Parties
Involved
Basis of
Conflict
Violent
Conflict?
Description Sources
1503 Florence and Pisa
warring states
Military tool Yes Leonardo da Vinci and Machievelli
plan to divert Arno River away from
Pisa during conflict between Pisa and
Florence.
Honan 1996
1642 China; Ming
Dynasty
Military tool Yes The Huang He's dikes have been
breached for military purposes. In
1642, "toward the end of the Ming
dynasty (1368-1644), General Gao
Mingheng used the tactic near
Kaifeng in an attempt to suppress a
peasant uprising."
Hillel 1991
1863 United States Civil
War
Military tool Yes General U.S. Grant, during the Civil
War campaign against Vicksburg, cut
levees in the battle against the
Confederates.
Grant1885,
Barry 1997
1898 Egypt; France;
Britain
Military and
political tool,
Control of water
resources
Military
maneuvers
Military conflict nearly ensues
between Britain and France in 1898
when a French expedition attempted
to gain control of the headwaters of
the White Nile. While the parties
ultimately negotiates a settlement of
the dispute, the incident has been
characterized as having "dramatized
Egypt's vulnerable dependence on the
Nile, and fixed the attitude of
Egyptian policy-makers ever since."
Moorhead 1960
1924 Owens Valley, Los
Angeles, California
Political tool,
Control of water
resources,
Terrorism, and
Development
dispute
Yes The Los Angeles Valley
aqueduct/pipeline suffers repeated
bombings in an effort to prevent
diversions of water from the Owens
Valley to Los Angeles.
Reisner 1986,
1993
1935 California, Arizona Political tool,
development
dispute
Military
maneuvers
Arizona calls out the National Guard
and militia units to the border with
California to protest the construction
of Parker Dam and diversions from
the Colorado River; dispute ultimately
is settled in court.
Reisner 1986,
1993
1938 China and Japan Military tool,
Military target
Yes Chiang Kai-shek orders the
destruction of flood-control dikes of
the Huayuankou section of the Huang
He (Yellow) river to flood areas
threatened by the Japanese army.
West of Kaifeng dikes are destroyed
with dynamite, spilling water across
the flat plain. The flood destroyed
part of the invading army and its
heavy equipment was mired in thick
mud, though Wuhan, the
headquarters of the Nationalist
government was taken in October.
Hillel 1991,
Yang Lang
1989, 1994

126
Compiled by: Peter Gleick, Pacific Institute for Studies in Development, Environment, and Security.


Available at: http://www.worldwater.org/conflict.htm
161
The waters flooded an area variously
estimated as between 3,000 and
50,000 square kilometers, and killed
Chinese estimated in numbers
between "tens of thousands" and
"one million."
1940-
1945
Multiple parties Military target Yes Hydroelectric dams routinely bombed
as strategic targets during World War
II.
Gleick 1993
1943 Britain, Germany Military target Yes British Royal Air Force bombed dams
on the Mohne, Sorpe, and Eder
Rivers, Germany (May 16, 17). Mohne
Dam breech killed 1,200, destroyed
all downstream dams for 50 km.
Kirschner 1949
1944 Germany, Italy,
Britain, United
States
Military tool Yes German forces used waters from the
Isoletta Dam (Liri River) in January
and February to successfully destroy
British assault forces crossing the
Garigliano River (downstream of Liri
River). The German Army then
dammed the Rapido River, flooding a
valley occupied by the American
Army.
Corps of
Engineers 1953
1944 Germany, Italy,
Britain, United
States
Military tool Yes German Army flooded the Pontine
Marches by destroying drainage
pumps to contain the Anzio
beachhead established by the Allied
landings in 1944. Over 40 square
miles of land were flooded; a 30-mile
stretch of landing beaches was
rendered unusable for amphibious
support forces.
Corps of
Engineers 1953
1944 Germany, Allied
forces
Military tool Yes Germans flooded the Ay River, France
(July) creating a lake two meters
deep and several kilometers wide,
slowing an advance on Saint Lo, a
German communications center in
Normandy.
Corps of
Engineers 1953
1944 Germany, Allied
forces
Military tool Yes Germans flooded the Ill River Valley
during the Battle of the Bulge (winter
1944-45) creating a lake 16
kilometers long, 3-6 kilometers wide
greatly delaying the American Armys
advance toward the Rhine.
Corps of
Engineers 1953
1947
onwards
Bangladesh, India Development
disputes, Control
of water
resources
No Partition divides the Ganges River
between Bangladesh and India;
construction of the Farakka barrage
by India, beginning in 1962, increases
tension; short-term agreements settle
dispute in 1977-82, 1982-84, and
1985-88, and thirty-year treaty is
signed in 1996.
Butts 1997,
Samson &
Charrier 1997
1947-
1960s
India, Pakistan Development
disputes, Control
of water
resources, and
Political tool
No Partition leaves Indus basin divided
between India and Pakistan; disputes
over irrigation water ensue, during
which India stems flow of water into
irrigation canals in Pakistan; Indus
Waters Agreement reached in 1960
after 12 years of World Bank-led
negotiations.
Bingham et al.
1994, Wolf
1997
162
1948 Arabs, Israelis Military tool Yes Arab forces cut of West Jerusalems
water supply in first Arab-Israeli war.
Wolf 1995,
1997
1950s Korea, United
States, others
Military target Yes Centralized dams on the Yalu River
serving North Korea and China are
attacked during Korean War.
Gleick 1993
1951 Korea, United
Nations
Military tool and
Military target
Yes North Korea released flood waves
from the Hwachon Dam damaging
floating bridges operated by UN
troops in the Pukhan Valley. U.S.
Navy plans were then sent to destroy
spillway crest gates.
Corps of
Engineers 1953
1951 Israel, Jordan,
Syria
Political tool,
Military tool,
Development
disputes
Yes Jordan makes public its plans to
irrigate the Jordan Valley by tapping
the Yarmouk River; Israel responds
by commencing drainage of the Huleh
swamps located in the demilitarized
zone between Israel and Syria; border
skirmishes ensue between Israel and
Syria.
Wolf 1997,
Samson &
Charrier 1997
1953 Israel, Jordan,
Syria
Development
dispute, Military
target, Political
tool
Yes Israel begins construction of its
National Water Carrier to transfer
water from the north of the Sea of
Galilee out of the Jordan basin to the
Negev Desert for irrigation. Syrian
military actions along the border and
international disapproval lead Israel
to move its intake to the Sea of
Galilee.
Samson &
Charrier 1997
1958 Egypt, Sudan Military tool,
Political tool,
Control of water
resources
Yes Egypt sends an unsuccessful military
expedition into disputed territory
amidst pending negotiations over the
Nile waters, Sudanese general
elections, and an Egyptian vote on
Sudan-Egypt unification; Nile Water
Treaty signed when pro-Egyptian
government elected in Sudan.
Wolf 1997
1960s North Vietnam,
United States
Military target Yes Irrigation water supply systems in
North Vietnam are bombed during
Vietnam War. 661 sections of dikes
damaged or destroyed.
Gleick 1993,
Zemmali 1995
1962 to
1967
Brazil; Paraguay Military tool,
Political tool,
Control of water
resources
Military
maneuvers
Negotiations between Brazil and
Paraguay over the development of
the Paran River are interrupted by a
unilateral show of military force by
Brazil in 1962, which invades the area
and claims control over the Guaira
Falls site. Military forces were
withdrawn in 1967 following an
agreement for a joint commission to
examine development in the region.
Murphy and
Sabadell 1986
1963-
1964
Ethiopia, Somalia Development
dispute, Military
tool, Political tool
Yes Creation of boundaries in 1948 leaves
Somali nomads under Ethiopian rule;
border skirmishes occur over disputed
territory in Ogaden desert where
critical water and oil resources are
located; cease-fire is negotiated only
after several hundred are killed.
Wolf 1997
1965-
1966
Israel, Syria Military tool,
Political tool,
Control of water
Yes Fire is exchanged over "all-Arab" plan
to divert the Jordan River headwaters
and presumably preempt Israeli
Wolf 1995,
1997
163
resources,
Development
dispute
National Water Carrier; Syria halts
construction of its diversion in July
1966.
1966-
1972
Vietnam, US Military tool Yes U.S. tries cloud-seeding in Indochina
to stop flow of materiel along Ho Chi
Minh trail.
Plant 1995
1967 Israel, Syria Military target
and tool
Yes Israel destroys the Arab diversion
works on the Jordan River
headwaters. During Arab-Israeli War
Israel occupies Golan Heights, with
Banias tributary to the Jordan; Israel
occupies West Bank.
Gleick 1993,
Wolf 1995,
1997,
Wallenstein &
Swain 1997
1969 Israel, Jordan Military target
and tool
Yes Israel, suspicious that Jordan is
overdiverting the Yarmouk, leads two
raids to destroy the newly-built East
Ghor Canal; secret negotiations,
mediated by the US, lead to an
agreement in 1970.
Samson &
Charrier 1997
1970s Argentina, Brazil,
Paraguay
Political goal,
Development
dispute
No Brazil and Paraguay announce plans
to construct a dam at Itaipu on the
Paran River, causing Argentina
concern about downstream
environmental repercussions and the
efficacy of their own planned dam
project downstream. Argentina
demands to be consulted during the
planning of Itaipu but Brazil refuses.
An agreement is reached in 1979 that
provides for the construction of both
Brazil and Paraguays dam at Itaipu
and Argentinas Yacyreta dam.
Wallenstein &
Swain 1997
1974 Iraq, Syria Military target,
Military tool,
Political tool,
Development
dispute
Military
maneuvers
Iraq threatens to bomb the al-Thawra
dam in Syria and massed troops along
the border, alleging that the dam had
reduced the flow of Euphrates River
water to Iraq.
Gleick 1994
1975 Iraq, Syria Development
dispute, Military
tool, Political tool
Military
maneuvers
As upstream dams are filled during a
low-flow year on the Euphrates, Iraqis
claim that flow reaching its territory is
"intolerable" and asks the Arab
League to intervene. Syrians claim
they are receiving less than half the
rivers normal flow and pull out of an
Arab League technical committee
formed to mediate the conflict. In
May Syria closes its airspace to Iraqi
flights and both Syrian and Iraq
reportedly transfer troops to their
mutual border. Saudi Arabia
successfully mediates the conflict.
Gleick 1993,
1994, Wolf
1997
1975 Angola, South
Africa
Military goal Yes South African troops move into
Angola to occupy and defend the
Ruacana hydropower complex,
including the Gov Dam on the
Kunene River. Goal is to take
possession of and defend the water
resources of southwestern Africa and
Namibia.
Meissner 2000
1978-
onwards
Egypt, Ethiopia Development
dispute, Political
No Long standing tensions over the Nile,
especially the Blue Nile, originating in
Gleick 1991,
1994
164
tool Ethiopia. Ethiopias proposed
construction of dams on the
headwaters of the Blue Nile leads
Egypt to repeatedly declare the vital
importance of water. "The only
matter that could take Egypt to war
again is water" (Anwar Sadat-1979).
"The next war in our region will be
over the waters of the Nile, not
politics" (Boutrous Ghali-1988).
1981 Iran, Iraq Military target
and tool
Yes Iran claims to have bombed a
hydroelectric facility in Kurdistan,
thereby blacking out large portions of
Iraq, during the Iran-Iraq War.
Gleick 1993
1980-
1988
Iran, Iraq Military tool Yes Iran diverts water to flood Iraqi
defense positions.
Plant 1995
1988 Angola, South
Africa, Cuba
Military goal,
military target
Yes Cuban and Angolan forces launch an
attack on Calueque Dam via land and
then air. Considerable damage
inflicted on dam wall; power supply to
dam cut. Water pipeline to
Owamboland cut and destroyed.
Meissner 2000
1982 Israel, Lebanon,
Syria
Military tool Yes Israel cuts off the water supply of
Beirut during siege.
Wolf 1997
1986 North Korea,
South Korea
Military tool No North Koreas announcement of its
plans to build the Kumgansan
hydroelectric dam on a tributary of
the Han River upstream of Seoul
raises concerns in South Korea that
the dam could be used as a tool for
ecological destruction or war.
Gleick 1993
1986 Lesotho, South
Africa
Military goal;
Control of water
resources
Yes South Africa supports coup in Lesotho
over support for ANC and anti-
apartheid, and water. New
government in Lesotho then quickly
signs Lesotho Highlands water
agreement.
American
University
2000b
1990 South Africa Development
dispute, Control
of water
resources
No Pro-apartheid council cuts off water to
the Wesselton township of 50,000
blacks following protests over
miserable sanitation and living
conditions.
Gleick 1993
1990 Iraq, Syria, Turkey Development
dispute, Military
tool, Political tool
No The flow of the Euphrates is
interrupted for a month as Turkey
finishes construction of the Ataturk
Dam, part of the Grand Anatolia
Project. Syria and Iraq protest that
Turkey now has a weapon of war. In
mid-1990 Turkish president Turgut
Ozal threatens to restrict water flow
to Syria to force it to withdraw
support for Kurdish rebels operating
in southern Turkey.
Gleick 1993 &
1995
1991-
present
Karnataka, Tamil
Nadu (India)
Development
dispute, Control
of water
resources
Yes Violence erupts when Karnataka
rejects an Interim Order handed
down by the Cauvery Waters
Tribunal, empaneled by the Indian
Supreme Court. The Tribunal was
established in 1990 to settle two
decades of dispute between
Gleick 1993,
Butts 1997,
American
University
2000a
165
Karnataka and Tamil Nadu over
irrigation rights to the Cauvery River.
1991 Iraq, Kuwait, US Military target Yes During the Gulf War, Iraq destroys
much of Kuwaits desalination
capacity during retreat.
Gleick 1993
1991 Iraq, Turkey,
United Nations
Military tool Yes Discussions are held at the United
Nations about using the Ataturk Dam
in Turkey to cut off flows of the
Euphrates to Iraq.
Gleick 1993
1991 Iraq, Kuwait, US Military target Yes Baghdads modern water supply and
sanitation system are intentionally
targeted by Allied coalition.
Gleick 1993
1992 Czechoslovakia,
Hungary
Political tool,
Development
dispute
Military
maneuvers
Hungary abrogates a 1977 treaty with
Czechoslovakia concerning
construction of the
Gabcikovo/Nagymaros project based
on environmental concerns. Slovakia
continues construction unilaterally,
completes the dam, and diverts the
Danube into a canal inside the
Slovakian republic. Massive public
protest and movement of military to
the border ensue; issue taken to the
International Court of Justice.
Gleick 1993
1992 Bosnia, Bosnian
Serbs
Military tool Yes The Serbian siege of Sarajevo, Bosnia
and Herzegovina, includes a cutoff of
all electrical power and the water
feeding the city from the surrounding
mountains. The lack of power cuts
the two main pumping stations inside
the city despite pledges from Serbian
nationalist leaders to United Nations
officials that they would not use their
control of Sarajevo's utilities as a
weapon. Bosnian Serbs take control
of water valves regulating flow from
wells that provide more than 80
percent of water to Sarajevo; reduced
water flow to city is used to smoke
out Bosnians.
Burns 1992,
Husarska 1995
1993-
present
Iraq Military tool No To quell opposition to his
government, Saddam Hussein
reportedly poisons and drains the
water supplies of southern Shiite
Muslims, the Ma'dan. The European
Parliament and UN Human Rights
Commission deplore use of water as
weapon in region.
Gleick 1993,
American
University
2000c
1993 Yugoslavia Military target
and tool
Yes Peruca Dam intentionally destroyed
during war.
Gleick 1993
1995 Ecuador, Peru Military and
political tool
Yes Armed skirmishes arise in part
because of disagreement over the
control of the headwaters of Cenepa
River. Wolf argues that this is
primarily a border dispute simply
coinciding with location of a water
resource.
Samson &
Charrier 1997,
Wolf 1997
1997 Singapore,
Malaysia
Political tool No Malaysia supplies about half of
Singapores water and in 1997
threatened to cut off that supply in
Zachary 1997
166
retribution for criticisms by Singapore
of policy in Malaysia.
1998 Tajikistan Terrorism,
Political tool
Potential On November 6, a guerrilla
commander threatened to blow up a
dam on the Kairakkhum channel if
political demands are not met. Col.
Makhmud Khudoberdyev made the
threat, reported by the ITAR-Tass
News Agency.
WRR 1998
1999 Lusaka, Zambia Terrorism,
Political tool
Yes Bomb blast destroyed the main water
pipeline, cutting off water for the city
of Lusaka, population 3 million.
FTGWR 1999
1999 Yugoslavia Military target Yes Belgrade reported that NATO planes
had targeted a hydroelectric plant
during the Kosovo campaign.
Reuters 1999a
1999 Bangladesh Development
dispute, Political
tool
Yes 50 hurt during strikes called to protest
power and water shortages. Protest
led by former Prime Minister Begum
Khaleda Zia over deterioration of
public services and in law and order.


Ahmed 1999
1999 Yugoslavia Military target Yes NATO targets utilities and shuts down
water supplies in Belgrade. NATO
bombs bridges on Danube, disrupting
navigation.
Reuters 1999b
1999 Yugoslavia Political tool Yes Yugoslavia refuses to clear war debris
on Danube (downed bridges) unless
financial aid for reconstruction is
provided; European countries on
Danube fear flooding due to winter
ice dams will result. Diplomats decry
environmental blackmail.
Simons 1999
1999 Kosovo Political tool Yes Serbian engineers shut down water
system in Pristina prior to occupation
by NATO.
Reuters 1999c
1999 Angola Terrorism/
Political tool
Yes 100 bodies were found in four
drinking water wells in central Angola.
International
Herald Tribune
1999
1999 Puerto Rico, U.S. Political tool No Protesters blocked water intake to
Roosevelt Roads Navy Base in
opposition to U.S. military presence
and Navys use of the Blanco River,
following chronic water shortages in
neighboring towns.
New York Times
1999
1999 East Timor Military tool,
Political tool,
terrorism
Yes Militia opposing East Timor
independence kill pro-independence
supporters and throw bodies in water
well.
BBC 1999
1999 Kosovo Terrorism/
Political tool
Yes Contamination of water supplies/wells
by Serbs disposing of bodies of
Kosovar Albanians in local wells.




CNN 1999

167
Sources

Ahmed, A. 1999. "Fifty hurt in Bangladesh strike violence." Reuters News Service, Dhaka, April 18, 1999
http://biz.yahoo.com/rf/990418/3.html.

American University (Inventory of Conflict and the Environment ICE). 2000a. Cauvery River Dispute.
http://www.american.edu/projects/mandala/TED/ice/CAUVERY.HTM.

American University (Inventory of Conflict and the Environment ICE). 2000b. Lesotho "Water Coup."
http://www.american.edu/projects/mandala/TED/ice/LESWATER.HTM.

American University (Inventory of Conflict and the Environment ICE). 2000c. Marsh Arabs and Iraq.
http://www.american.edu/projects/mandala/TED/ice/MARSH.HTM.

Barry J.M. 1997. Rising Tide: The Great Mississippi Flood of 1927 and How it Changed America." Simon and Schuster,
New York. p. 67.

Bingham, G., A. Wolf, and T. Wohlegenant. 1994. "Resolving water disputes: Conflict and cooperation in the United
States, the Near East, and Asia." US Agency for International Development (USAID). Bureau for Asia and the Near East.
Washington DC.

BBC 1999. "World: Asia-Pacific Timor atrocities unearthed." September 22, 1999.
http://news.bbc.co.uk/hi/english/world/asia-pacific/newsid_455000/455030.stm

Burns, J.F. 1992. "Tactics of the Sarajevo Siege: Cut Off the Power and Water," New York Times, September 25, 1992.
p.A1.

Butts, K., ed. 1997. Environmental Change and Regional Security. Carlisle, PA: Asia-Pacific Center for Security Studies,
Center for Strategic Leadership, US Army War College.

Cable News Network (CNN). 1999. "U.S.: Serbs destroying bodies of Kosovo victims." May 5.
http://www.cnn.com/WORLD/europe/9905/05/kosovo.bodies

Corps of Engineers. 1953. "Application sof Hydrology in Military Planning and Operations and Subject Classification Index
for Military Hydrology Data." Military Hydrology R&D Branch, Engineering Division, Corps of Engineers, Department of the
Army, Washington.

Drower, M.S. 1954, "Water-supply, irrigation, and agriculture." In C. Singer, E.J. Holmyard, and A.R. Hall (ed.) A History
of Technology. Oxford University Press, New York.

Financial Times Global Water Report. 1999. "Zambia: Water Cutoff." FTGWR Issue 68, p. 15 (March 19, 1999).

Gleick, P.H. 1991. "Environment and security: The clear connections." Bulletin of the Atomic Scientists. April:17-21.

Gleick, P.H. 1993. "Water and conflict: Fresh water resources and international security." International Security 18, Vol. 1,
pp. 79-112.

Gleick, P.H. 1994. "Water, war, and peace in the Middle East." Environment, Vol. 36, No. 3, pp.6-on. Heldref Publishers,
Washington.

Gleick, P.H. 1995. "Water and Conflict: Critical Issues." Presented to the 45th Pugwash Conference on Science and World
Affairs. Hiroshima, Japan: 23-29 July.

Gleick, P.H. 1998. "Water and conflict." In The Worlds Water 1998-1999. Island Press, Washington.

Grant, U.S. 1885. Personal Memoirs of U.S. Grant. C.L. Webster, New York. ["On the second of February, [1863] this
dam, or levee, was cut,The river being high the rush of water through the cut was so great that in a very short time the
entire obstruction was washed away As a consequence the country was covered with water."

Green Cross International. The Conflict Prevention Atlas. http://dns.gci.ch/water/atlas.

Hillel, D. 1991. "Lash of the Dragon." Natural History (August), pp. 28-37.

Honan, W.H. 1996. "Scholar sees Leonardos influence on Machiavelli." The New York Times (December 8), p. 18.

Husarska, A. 1995. "Running dry in Sarajevo: Water fight." The New Republic. July 17 & 24.
168

International Court of Justice. 1999. International Court of Justice Press Communiqu 99/53, Kasikili Island/Sedudu Island
(Botswana/Namibia). The Hague, Holland 13 December 1999, p. 2 (http://www.icj-cij.org).

International Herald Tribune. 1999. "100 bodies found in well." International Herald Tribune, August 14-15. p.4.

Kirschner, O. 1949. "Destruction and Protection of Dams and Levees." Military Hydrology, Research and Development
Branch, U.S. Corps of Engineers, Department of the Army, Washington District. From Schweizerische Bauzeitung 14
March 1949, Translated by H.E. Schwarz, Washington.

Meissner, R. 2000. "Hydropolitical hotspots in Southern Africa: Will there be a water war? The case of the Kunene river."
In H. Solomon and A. Turton (editors). Water Wars: Enduring Myth or Impending Reality? Africa Dialogue Monograph
Series No. 2. Accord, Creda Communications, KwaZulu-Natal, South Africa, pp. 103-131.

Moorehead, A. 1960. The White Nile. Penguin Books, England.

Murphy, I.L. and J. E. Sabadell. 1986. "International river basins: A policy model for conflict resolution." Resources Policy.
Vol. 12, No. 1, pp. 133-144. Butterworth and Co. Ltd., United Kingdom.

New York Times. 1999. "Puerto Ricans protest Navys use of water." The New York Times, October 31, p. 30.

Plant, G. 1995. "Water as a weapon in war." Water and War, Symposium on Water in Armed Conflicts, Montreux 21-23
November 1994, Geneva, ICRC.

Reisner, M. 1986, 1993 ed. Cadillac Desert: The American West and its Disappearing Water. Penguin Books, New York.

Reuters. 1999a. "Serbs Say NATO Hit Refugee Convoys." April 14, 1999.
http://dailynews.yahoo.com/headlines/ts/story.html?s=v/nm/19990414/ts/yugoslavia_192.html

Reuters 1999b. "NATO Keeps Up Strikes But Belgrade Quiet." June 5, 1999.
http://dailynews.yahoo.com/headlines/wl/story.html?s=v/nm/19990605/wl/yugoslavia_strikes_129.html

Reuters 1999c. "NATO Builds Evidence Of Kosovo Atrocities." June 17, 1999.
http://dailynews.yahoo.com/headlines/ts/story.html?s=v/nm/19990617/ts/yugoslavia_leadall_171.html

Samson, P. and B. Charrier. 1997. "International freshwater conflict: Issues and prevention strategies." Green Cross
International.
http://www.dns.gci.ch/water/gcwater/study.html

Simons, M. 1999. "Serbs refuse to clear bomb-littered river." New York Times, October 24, 1999.

Wallenstein, P. and A. Swain. 1997. "International freshwater resources - Conflict or cooperation?" Comprehensive
Assessment of the Freshwater Resources of the World: Stockholm Environment Institute.

Yang Lang. 1989/1994. "High Dam: The Sword of Damocles." In Dai Qing (ed.), Yangtze! Yangtze! Probe International,
Earthscan Publications, London, United Kingdom. pp. 229-240.

Wolf, A.T. 1995. Hydropolitics along the Jordan River: Scarce Water and its Impact on the Arab-Israeli Conflict. United
Nations University Press, Tokyo, Japan.

Wolf, A. T. 1997. "Water wars and water reality: Conflict and cooperation along international waterways." NATO
Advanced Research Workshop on Environmental Change, Adaptation, and Human Security. Budapest, Hungary. 9-12
October.

World Rivers Review (WRR). 1998. "Dangerous Dams: Tajikistan" Volume 13, No. 6, p. 13 (December).

Zachary G.P. 1997. "Water pressure: Nations scramble to defuse fights over supplies." Wall Street Journal, December 4,
p. A17.

Zemmali, H. 1995. "International humanitarian law and protection of water." Water and War, Symposium on Water in
Armed Conflicts, Montreux 21-23 November 1994, Geneva, ICRC.





169