WATER CONTROL IN SOCIOTECHNICAL SYSTEMS: A CONCEPTUAL

FRAMEWORK FOR INTERDISCIPLINARY IRRIGATION STUDIES

Lecture Notes for Irrigation and Development

Peter P. Mollinga (draft, 1997), revised by Philippus Wester (2010), shortened by Gert Jan
Veldwisch (2014)

The central question of these lecture notes is:

What are the elements of a conceptual

framework for interdisciplinary analysis of
irrigation in the context of rural development?

This chapter discusses the concepts that can be employed to investigate and explain the
interrelation of technical, organisational or institutional and socio-economic and political
aspects of irrigation. Together these concepts constitute a conceptual framework or an approach
that can be used for the analysis of different irrigation situations. Before embarking on
conceptual explanation, it is useful to indicate where the approach that will be presented below
comes from, and what the biases within it are.

The main bias of the framework is its focus on irrigation systems in the sense of canal
infrastructure. This reflects the professional irrigation engineering background of the author and
his colleagues. The irrigation engineering discipline worldwide is mainly occupied with dams,
canals, structures, and how these are to be combined into viable, working infrastructural
networks. In addition to this focus on the system level, Wageningen University has always had a
strong emphasis on the field and farm level. There the issue is the agronomical aspects of
irrigation, the farming system and field irrigation technologies. The third level at which
irrigation can be studied is the watershed or regional level. Irrigation is then studied in the
context of water resources development more generally. One of the questions becomes the share
of irrigation in total water use, which also comprises drinking water, industrial water, etc. We
believe that the general features of the approach can be equally applied to all three levels, but for
the field and farm and for the watershed level additional work is required.

The origins of the framework lie in a critique of the professional irrigation literature. With
professional irrigation literature I refer to three bodies of intellectual work.
1) The irrigation engineering literature. This can be found in journals like the Journal of
Irrigation and Drainage Engineering, Agricultural Water Management, Irrigation
Science, and many others, and also in the proceedings of the International Commission
on Irrigation and Drainage (ICID).
2) The irrigation economics literature. Economics, particularly neo-classical economics,
has always been very important in government-sponsored and government-managed
irrigation. In colonial times it was used to calculate the financial returns of government
investment in irrigation (through land revenue and taxes). It is now similarly used for
calculating benefit/cost ratios and internal rates of return to evaluate the feasibility of
irrigation projects. More recently the discipline of economics has become more
important in the field of irrigation studies by using economic models (game-theory, rent
seeking analysis) for understanding institutional processes. To my knowledge there is no
special journal on irrigation economics.

-1-
3) The irrigation management literature. This body of literature has emerged since the
1970s as a response to the disappointing performance of the irrigation systems newly
constructed in the 1950s and 1960s in a massive wave of irrigation investment. In the
1970s and early 1980s most attention went to the organisation of water users (in water
users associations) and user participation more generally. Two influential authors are
Robert Chambers and Norman Uphoff. In 1984 the International Irrigation Management
Institute (IIMI) was established, renamed the International Water Management Institute
(IWMI) in 2000, which is now the main international institution doing research on
irrigation and water management. Main system management and the irrigation
bureaucracies have also become the focus of research. Performance assessment is the
main theme, linked to irrigation management transfer or decentralisation of water
management, and the introduction of market principles (volumetric pricing) in
management. Publications from this stream of literature can be found in a journal like
Irrigation and Drainage Systems, and in publications by IWMI, IFPRI (the International
Food Policy Research Institute) and ICID.

These three literatures are very closely linked to irrigation practice and intervention funded by
international donors like the World Bank and national governments. In addition there is a more
academic literature on irrigation, much less linked to irrigation intervention practices. These are
historical, geographical, social-anthropological, political-economic and other studies on
irrigation, aiming to understand ongoing processes more than directly desiring to change them.
Institutionally these studies are usually based in universities. Though there are definitely
linkages between the ‘professional’ and the ‘academic’ literatures on irrigation (both in persons
and in ideas), to a large extent they have remained two separate worlds.

This is not the place to review the professional irrigation literature and develop a critique of it,
but the approach presented below is based on such an exercise. Summarised at the most general
level, the critique consists of three points (to be clarified below and in the remainder of the
course).
1) The treatment of technology as a black-box. This means that engineers as well as social
scientist tend to consider technology as something to be used and abused, but inherently
neutral. In contrast we will emphasise the social dimensions of technologies themselves.
2) A limited concept of human agency. With this I mean the basic idea about what
motivates peoples’ behaviour, and how that behaviour should be analysed. In the
economics and management literature for example the concept of human agency is
usually that of the utility maximizer or optimiser and the rational decision maker. In
contrast to this we suggest that human behaviour is a much more complex affair, which
requires less idealised and simplified models.
3) The absence of the social relations of power in professional analytical frameworks. In
contrast we argue that social power is a crucial element in understanding relations
among people (and between people and material things for that matter).

In much of the ‘academic’ irrigation literature points 2) and 3) are quite common place, and this
literature can therefore be usefully employed to enrich the professional irrigation literature, and
we will do so. Point 1) on technology is the more problematic one. In most academic studies of
irrigation, technology is equally treated as a black box. We will devote considerable space to
arguing that this is unjustified for a comprehensive understanding of irrigation.

After this introduction on the background of the approach, I can now outline the structure of
these lecture notes. The two most general concepts of the framework are that irrigation systems

-2-
are sociotechnical systems, in which water control is the central activity. The first concept,
sociotechnical system, is as it were the ‘static’ part of the framework. It describes what kind of
structure an irrigation system is, what its different elements are, and how these are connected.
The second concept, water control, provides the ‘dynamic’ element. It refers to the activities and
processes that take place in irrigation systems, who are involved in this, how these persons
interact, which resources and strategies they employ in these interactions, and what the
outcomes are.

The concepts sociotechnical system and water control are discussed in two separate sections
(2.1 and 2.2). After this we operationalize the concept by identifying the water control processes
within irrigation systems (section 2.3) and the different impacts of irrigation on wider processes
of rural transformation (section 2.4). The discussion in these lecture notes moves at a general
level. The framework presented is applied and elaborated in subsequent sessions of the
Irrigation and Development course.

2.1 IRRIGATION SYSTEMS AS SOCIOTECHNICAL SYSTEMS

Irrigation systems are complex systems. Complex can mean two different things. The first is
‘difficult to understand’ or ‘complicated’. Sometimes it is indeed difficult to grasp how an
irrigation system works. Complex also means ‘consisting of different elements, related in
diverse ways’. It is this second meaning of complexity that is relevant for developing the
conceptual framework. Figure 2.1 illustrates the second meaning of complexity.

Figure 2.1 Domains and linkages of a canal irrigation system

Source: Chambers (1988:44, figure 2.7)

In the chapter on ‘how to think about irrigation’ in his 1988 book Managing Canal Irrigation,
Robert Chambers discusses the many different elements that constitute an irrigation system.
There is water in different forms: rain, water in storage, water in canals, and water in the soil.
There are people with different roles: farmers or water users, labourers working in agricultural

-3-
production, and government irrigation officials at different levels of the bureaucratic hierarchy.
There are soils, crops and animals, there is money as cost and benefit, and less tangible factors
like sunlight, temperature and relative humidity. And finally there are technologies: canals,
outlet structures and other devices necessary for bringing water from A to B.

The arrows in the figure indicate that this heterogeneous set of elements hangs together through
a large number of relationships. These relationships are of different kinds. The relation between
irrigation staff and farmers may be a formal, administrative one, but may also involve illegal
payments for water (corruption). It may be cooperative or conflictive. Farmers relate to
labourers by paying them in cash or kind and may be combined with giving credit. The climate
relates to crops through the process of photosynthesis and evapotranspiration. Water users and
governments officials relate to the irrigation technologies through their skills and forms of
organisation. And so forth.
These elements and relations jointly constitute Box 2.1
the irrigation system. The complexity or
heterogeneity of the system is summarised in Socio-technical systems and processes
the term sociotechnical system. (See Box 2.1
“Both human and physical aspects interact
for quotations of authors with a similar view.) continually and profoundly in irrigation
There are social and technical elements, and enterprises, so a hyphenated construct of irrigation
there are social and technical relationships. No as a socio-technical process seems appropriate.”
irrigation system is exclusively technical or (Uphoff, 1986:4)
exclusively social in nature. Real irrigation
“Irrigation systems are socio-technical systems,
systems can only be both at the same time.
i.e. they embrace both social and technical system
components and subsystems. The essential
This then raises the question how to think attributes of socio-technical systems include:
systematically about the sociotechnical - close interrelationships between structural, social
characteristics of irrigation systems. and technological features;
Approaches that attempt to analyse irrigation - openness of the systems to their system
environments;
in its full complexity are rare. Most
- an emphasis on conversion processes in which
approaches are disciplinary in nature. A imputs imported from the system environment are
particular aspect is focussed upon in a transformed in a conversion process (throughput)
disciplinary approach, at the cost of other and exported to the system environment as
factors. The focus may be on the hydraulic, outputs.” (Huppert, 1989:27)
the agronomic, the engineering, the
managerial, the socio-economic, the legal, the
cultural, or any other aspect. There is no doubt that disciplinary approaches are of great value
and produce very important pieces of knowledge. But what we aim for here is a conceptual
framework that allows the integration of these different disciplinary foci into a single approach.

2.1.1 A labour process perspective

Our starting point lies outside the domain of irrigation. More than a century ago Karl Marx was
interested in a similar problem as we are: the relation between the social and technical aspects of
production. He wanted to analyse how the ‘forces of production’ and the ‘relations of
production’ were connected to each other in capitalist manufacturing and industry. He
developed a general concept of the labour process in which both the technical and the social
elements of production found a place. He proposed that every labour process, that is every
human activity aiming to produce a useful product, consisted of three elements:
1) the personal activity of man, i.e. work itself,

-4-
2) the subject of that work, and
3) its instruments. (Marx, Capital Volume 1, chapter 7).

Figure 2.2: Marx's concept of the labour process

Labour

Raw Materials Technology

When we regard irrigation as a labour process that aims to bring water from A to B in order to
grow agricultural crops, the figure can be filled in as follows.1)

Figure 2.3: Elements of irrigation as a labour process

Individual & Collective Action

Water Irrigation infrastructure

This figure shows that in irrigation a farmer's work is to bring water to the fields in which s/he
grows his or her crops. A farmer can do this individually, but often s/he does it together with
other persons. These other persons may be other farmers but also government officials from an
irrigation agency that is responsible for the management of the irrigation system. These joint
activities give rise to particular forms of organisation. All this would be impossible without the
aid of irrigation technologies. These are the instruments to guide water while it is on its way
from the point of capture to the farmer's field.

1
For reviews of labour process theory, see Thompson (1989), Knights and Wilmott (1990) and Burawoy (1985).
For applications in agriculture, see van der Ploeg (1991) and Friedland, Barton and Thomas (1981). Please note that
the figure has been filled in for the system level. At farm/field level the subject of work will also include crops, and
perhaps the soil, and other technologies will be involved. At the watershed level the subject of work remains water
when looking at water resources management in general, but the technologies involved expand: flood control works,
drinking water pipeline systems, watershed protection technologies, etc. There is an important difference between
the type of labour process Marx was primarily thinking of, capitalist manufacturing, and our concern, irrigation.
Manufacturing is a process of transformation. In this process raw materials are materially changed, transformed into
an end product. Irrigation is not a process of transformation but of regulation, or eco-regulation as Benton (1984)
puts it. Benton’s example is agriculture in general. The growing of crops is not adequately described as transforming
a raw material (a seed) into an end product (the crop or animal), but it is a case of guiding (providing favourable
conditions for) the process of biological growth. Similarly, irrigation is the activity of capturing, storing and guiding
water from one place to another by manipulating the forces of gravity and hydraulic pressure. The significance of
this difference in the nature of the labour process will become clear below (particularly see 2.5.1).

-5-
2.1.2 Irrigation activities

The triangle of people, technologies and water summarises in a single formula all activities that
take place in irrigation. Further operationalisation is possible by looking in more detail at
exactly which activities are part of irrigation. Norman Uphoff has provided a useful and
comprehensive description of irrigation activities with his well-known cubic matrix.

He distinguishes three types of activities: water use activities, control structure activities and
organisational activities. Under each type there are four activities, which he describes as follows
(Uphoff, 1986: 38-40).

WATER USE ACTIVITIES

1. ACQUISITION of water from surface or subsurface sources, either by creating and operating physical
structures like dams, weirs or wells, or by actions to obtain for users some share of an existing supply.
2. ALLOCATION of water by assigning rights to users, thereby determining who shall have access to water.
3. DISTRIBUTION of water brought from the source among users at certain places, in certain amounts, and at
certain times.
4. DRAINAGE of water, where this is necessary to remove any excess supply.

CONTROL STRUCTURE ACTIVITIES

1. DESIGN of structures such as dams or wells to acquire water, channels and gates to distribute it, and drains
to remove it.
2. CONSTRUCTION of such structures to be able to acquire, distribute and remove water.
3. OPERATION of these structures to acquire, distribute and remove water according to some determined
plan of allocation.
4. MAINTENANCE of these structures in order to have continued and efficient acquisition, distribution and
removal of water.

ORGANISATIONAL (MANAGEMENT) ACTIVITIES

1. DECISION-MAKING: This applies to acquisition, allocation, distribution or drainage of water; to design,
construction, operation or maintenance of structures; or to the organisation which deals with these
activities. PLANNING is one major form of decision-making.
2. RESOURCE MOBILISATION AND MANAGEMENT: This involves the marshalling as well as application
of funds, manpower, materials, information or any other inputs needed for the above activities, or for any
general organisational tasks.
3. COMMUNICATION: This concerns the needs and problems in any of the activity areas noted above,
conveying information about decisions made, about resource mobilisation, about conflicts to be resolved,
etc. to farmers or any other persons involved in irrigation. One purpose of communication may be
COORDINATION.
4. CONFLICT MANAGEMENT: This must deal with differences of interest that arise from activities of
acquisition, allocation, distribution, drainage, design, construction, operation or maintenance, or from
organisational activities generally.

A 4x4x4 matrix can be constructed that looks as shown in figure 2.4.

Figure 2.4: Uphoff's matrix of irrigation activities

-6-
(Source: Uphoff, 1986:42)

The matrix once more illustrates the sociotechnical nature of irrigation. Other classifications of
activities and processes in irrigation have been made (see for example Chambers, 1988;
Huppert, 1989). The exact classification one favours depends on the purpose of the analysis in
which it is used, and the overall approach to irrigation. The point here is that irrigation consists
of many different activities that need to be executed in combination.

2.1.3 Levels

A further element of the ‘static’ conceptualisation of irrigation is that irrigation systems consist
of a number of levels. Irrigation as a sociotechnical system has different sociotechnical levels
with different sociotechnical connections. Hydraulically, irrigation systems are made up of
different levels of canals, which are connected by hydraulic structures, notably division
structures. Figure 2.5 shows differences among irrigation systems in terms of their number of
levels.

The levels-model can be extended upward by seeing irrigation systems as parts of a watershed,
and watersheds as parts of regional, continental and world wide agro-ecological systems.
Downward the model can be extended by including the different levels of the drainage system,
from field drain to ocean. In short, the whole hydrological cycle can be seen as a system with
many levels.

The point is that these levels are not only hydraulic or ecological levels, but also social levels.
At each of the levels there are different institutions and forms of organisation in relation to
water flow at that level. These range from national and international policy making and
legislation, to collective action at canal system level, to intra-household cooperation in field
irrigation. The social levels are organisationally related just like hydraulic levels are technically
and physically related. For example, the connection of a secondary and a tertiary canal is not
only a division structure with gates and locks, or other technical features, but also the person of

-7-
the water distributor (the canalero in Mexico, the niiruganti in India, etcetera) with different
institutional attributes: appointed by, paid by and accountable to farmers or the government,
etcetera.

Figure 2.5: Levels in irrigation systems

(Source: Uphoff, 1991:33)

2.1.4 Irrigation activities in context

Irrigation activities are not self-contained, isolated activities, but they are part of wider
processes. Irrigation activities as we encounter them in practice have a number of conditions of
possibility. With conditions of possibility we mean the circumstances or context that make it
possible that irrigation activities are conducted as they are. There are material and social
conditions of possibility.

The rainfall pattern and soil fertility are examples of material conditions of possibility. We
notice this when rain does not fall and rivers and groundwater aquifers dry up, or when soils
become saline or suffer from erosion. Other examples are the existence of passable roads in the
irrigation system in order to make gate adjustment and other management activities possible,
and working telephones to make communication between different administrative divisions in
the system possible. When the right material conditions are not fulfilled, irrigation as usual can
no longer take place.

There are also many social conditions of possibility.

-8-
- A farmer that uses a pumpset to lift groundwater to his fields can only do this when he
has a right to put a pumpset in his field, and extract water from the groundwater aquifer
below it. This right is part of the legal system of property rights in a society.
- Another condition of possibility is the price of water. When water is very expensive
farmers may use it judiciously, while they may not be very particular about water use
efficiency when it comes free of cost. The price of water depends on several factors,
among them government water pricing policy.
- When farmers that happen to have land in the same irrigation system need to cooperate
to distribute water among themselves, this may be easier or more difficult depending on
differences in class, caste, gender, ethnic and other social relations.

The different conditions of possibility for irrigation activities can be summarised in three
categories2):
1) the agro-ecological system and technological infrastructure (climate, weather,
vegetation, soil, topography, technologies other than the irrigation system itself),
2) the agrarian structure (markets for labour, land, technology, credit, inputs and outputs,
and social relations of class, gender, ethnicity, religion, caste, kinship, etcetera at
household, village/community and other levels), and
3) the state and the institutions of civil society (government line agencies like the Irrigation
Department, the legal system, policy making institutions, development NGOs, social
movements, education and training institutes, international donor and lending agencies,
local government institutions, etc).

On the basis of this discussion we can now draw the following figure.

Figure 2.6: Irrigation activities in context

2
The idea of embeddedness presented here is derived from Burawoy’s (1985) notion of ‘production regime’ or
‘political and ideological apparatuses of production’, and Benvenuti’s (1982) TATE concept (the Technological and
Administrative Task Environment).

-9-
The figure shows that irrigation activities in which water, technology and people interact, are
embedded in a set of wider material and social structures or processes. When we express this in
the computer-speak of the 1990s, we could say that irrigation activities consist of a hardware
(the technologies and water) and a software component (people's form of organisation), and that
these activities need an operating environment (the three different conditions of possibility) to
be able to take place.

We have now developed a description of the elements that together form ‘irrigation activities in
context’. In figure 2.3 these elements are related by arrows, all of which point both ways. This
indicates that all the elements are interrelated, and that changing one would lead to changes in
the others. This is why it is a system.

The next question is: what is the content of these arrows? How does this system hang together?
What are these relationships between all the different elements made off? In this chapter we
answer this question in a very general manner. We leave more specific answers to later chapters.

2.1.5 Networks and intermediaries; social power and resources

It follows from the discussion above that irrigation can be described as a network consisting of
interrelated heterogeneous elements. People are the most crucial part of the irrigation network,
because they are the network builders. People have particular objectives (like acquiring
sufficient water for their crops) and actively pursue these (by buying a pumpset, by engaging in
forms of organisation for water distribution, etcetera), and in this way they give shape to the
relationships of the irrigation network. People are network builders, trying to put all the
elements in the right place.

People make and consolidate linkages between different elements by means of what Callon
(1992) calls intermediaries. Intermediaries are the things that a person puts between him/herself
and the objects and other persons s/he wants to relate to. For physical relationships this is easy
to understand. When an irrigator wants to regulate a water flow in a stream or canal or on
his/her field, s/he puts artefacts like weirs, drops, division structures, pipes, siphons, pumps and
sprinkler installations between him/herself and the water. The relationship between people and
water is mediated by technologies.

This mediation also exists in the case of social relationships. For example when a farmer wants
to communicate with a government irrigation official to improve water supply, s/he may talk to
the official personally, speak with him/her on the telephone, offer to pay the official a bribe,
write the official a letter or petition, file a case against the official/department at the local court,
or ask a local politician to put pressure on the official.

In these different instances of communication and interaction the intermediaries are different.
When the farmers talk to the official personally the intermediary is language: the spoken word.
When they talk on the telephone the technology of the telephone system is the additional
intermediary. In the case of bribe payment the intermediaries are language and money. When
writing a letter, written text and the postal service are the intermediaries. The court case brings
in several people as intermediaries: police officers, advocates and judges. The politician who
can influence the behaviour of government officials is another example of a person acting as an
intermediary. The intermediary in the relationship between the farmer and the politician may be
the vote at election time, and that between the politician and the official, the former's power

- 10 -
over the latter's transfer. The intermediaries in the transfer relation are texts (administrative rules
for example) and money (payments by officials to politicians to influence their transfer).3)

Callon (1992) classifies the different types of intermediaries in four categories:

1) texts (language, policy documents, administrative forms, manuals and textbooks, laws,
ordinances, etcetera)
2) artefacts (material objects, including technologies),
3) people, and
4) money.

The point about intermediaries is that they are the resources that people use in pursuing their
objectives, and that not everyone has equal access to them. For example, a poor farmer may not
be able to purchase plastic siphons for smooth and regular irrigation of his/her furrows, or
acquire a sprinkler installation for the irrigation of high-value horticultural crops. As a result
his/her field irrigation may be less efficient and crop yields reduced. The poor farmer is not able
to exert the control (or power) over the water flow that s/he would like to exert, because s/he
can't get hold of the right intermediaries/resources.

In the discussion of the farmer and the irrigation official on water supply there is also an issue of
differential control and power. The official may, for example, try to intimidate, impress or
convince a farmer by arguing that the farmer's approach is `unscientific'. The official may argue
that according to his/her hydraulic calculations the farmer is treated fairly. When the farmer is
unfamiliar with the language of hydraulic engineering, s/he is unable to contradict the official's
argument. In this way, a particular type of language, based on hydraulic engineering texts,
becomes a means to exert power.

Another -obvious- example of differential access to resources is that rich farmers are able to pay
higher bribes to officials than poor farmers, and thus may get more water. Also access to
politicians is not the same for everyone. Politicians may only listen to local leaders who
mobilise votes for them, and not to any odd farmer with a complaint.

The degree and way in which different people are able to mobilise different
intermediaries/resources and structure the irrigation network according to their own objectives,
express the social relations of power in that society. Social relationships, mediated by different
resources, are social relations of power.

Three conclusions follow from this discussion.

1) Irrigation can be understood as a network consisting of a set of heterogeneous elements,
kept together by a set of diverse relationships. These relationships can be concretely
studied by looking at the intermediaries or resources that people mobilise to link the
different elements and consolidate these linkages.
2) The different ways and degrees in which people are able to mobilise intermediaries or
resources for their own purposes, express the different ways and degrees of control or
power that people are able to exert over the physical and social processes that occur in
irrigation systems.
3) There are no such things as ‘purely’ physical and ‘purely’ social processes in irrigation
systems. To a greater or lesser degree people intervene in physical processes like water

3
The farmers-politicians-officials example is based on Wade (1982).

- 11 -
 flow, soil degradation and plant growth. The social regulation of these physical
processes is exactly the objective of irrigation. And to all social relationships there are
physical aspects. Even the simplest person-to-person communication assumes the
physical properties of the human body (for example the ability to speak clearly, not to
mention the importance of ‘body language’). In many social interactions physical objects
like telephones, bullock carts, motorcars, the pieces of paper that we call money, and
computers are crucial elements. The physical and the social are not the same thing, but
they are intimately related, and need to be understood by looking at the way they shape
each other.

We have now, in general terms, described the elements and relationships that constitute an
irrigation system as a sociotechnical system. This is an outline of the `static' dimension of
irrigation systems. We now move on to the `dynamic' dimension: how the day-to-day
functioning of irrigation systems can be understood.

2.2 WATER CONTROL

Water control for agricultural production as a general concept includes a) irrigation, b) drainage,
and c) flood control. Water conservation and other water-related techniques used in rainfed
agriculture, are not usually referred to as water control. In this book we understand irrigation
systems to include the water supply infrastructure that brings water to agricultural fields to add
to available soil moisture, as well as the drainage infrastructure to evacuate excess water from
the irrigated area (see Uphoff's water use activities discussed above). We do not discuss flood
control systems, that is infrastructure to keep water out of (agricultural) land.4)

When we want to understand what happens in and with irrigation systems, we first have to
describe their central purpose. This has already been mentioned in the discussion above.
Irrigation systems aim to control the flow of water for agricultural crop production. By means of
irrigation systems people manipulate part of the hydrological cycle to enhance the growth of
crops (see chapter 1). This may not be the only purpose of an irrigation system, but it is the
defining one.

The concept of water control can be usefully employed to analyse the processes within irrigation
systems.

2.2.1 Three dimensions of water control

4
We will also abstract from the flood control functions of irrigation systems. The approach developed in this book
could mutatis mutandis be applied to other water control system than irrigation, such as flood control systems,
drinking water systems and hydropower systems. It could probably also be extended to water control in rainfed
agriculture (water conservation, erosion control, etcetera), but a major difference would be that there is not one or a
limited number of sources of (concentrated) water supply in such situations, but a diffuse natural supply, that is
rainfall. We haven’t thought through the implications of this for the conceptual framework. Where rainfed
agriculture involves organised drainage (as in polders), the applicability seems more straightforward.

- 12 -
In the irrigation literature water control is a very commonly used term. In the engineering
literature water control refers to the physical control of water flow by means of irrigation
technology. A classic publication by the World Bank is called ‘Modern water control in
irrigation’, and it describes the different irrigation technologies available for equipping an
irrigation system (Plusquellec, Burt and Wolter, 1994). (See Box 2.2 for a short extract.) The
use of water control in the sense of water
control technology is also common in the Box 2.2
social science literature (see for example
Burns (1993) and Mitchell and Guillet Water control methods (technical)
Hydraulic equipment for canal systems:
(1993)).
1) hand or motor-operated gates (overflow gates or
undershot (orifice-type) gates as cross regulators,
Water control in the technical sense is the power supply, electric controllers, motors)
core of many interventions in irrigation 2) automatic water level control (self-regulating
systems. Examples are projects for canal gates, automatic motor operated gates)
rehabilitation or system modernisation, and 3) flow rate control (proportional, manual,
automatic)
projects to improve the physical 4) measuring equipment (gate position sensors,
infrastructure at farmer/field level (field water level sensors, flow measuring systems,
channel construction, levelling, etcetera). interfaces)
Hydraulic equipment for pipe systems:
In the irrigation management literature water 5) isolating valves
control is used in a broader sense than its 6) on-line flow or pressure control (manual or
automatic valves)
technical meaning alone. The irrigation 7) discharge or water level control at free surface
management literature acknowledges that the head breakers or distribution structures (manual or
operation of water control technologies is automatic valves)
done by people. It further says that in systems 8) discharge control at farm outlets (discharge
with more than one user, some form of limiting device, (constant discharge) valves)
organisation or collective action is needed to 9) safety equipment (pressure relief, air exhaust and
air inlet, and automatically closing valves)
manage the system. Water control also refers 10) measuring equipment (valve position and
to managerial control of the water pressure sensors, flow measurement devices)
distribution process, and other organisational
processes in the irrigation system (see Box Source: Plusquellec, Burt and Wolter (1994:36-38)
2.3 for quotations).

In fact, control and management are almost synonyms. Webster's New Dictionary of Synonyms
gives the following description of the two words.

“Conduct, manage, control, direct are comparable when they mean to use one’s skill, authority,
or other powers in order to lead, guide, command, or dominate persons or things. (...) Manage
usually implies the handling, manipulating, or manoeuvring of a person or persons or a thing or
things so as to bring about a response or submission to one's wishes or attempts to use, guide,
lead, or command. (...) Control stresses the idea of authoritative guidance and suggests a
keeping within set or desired bounds (as of accuracy, efficiency, propriety, or discipline); it
implies a regulating or restraining often by getting or keeping the upper hand.” (1984: lemma
‘conduct’)

Irrigation management, or water control in the organisational sense, thus is about the regulation
and control of human behaviour, particularly with regard to the forms of cooperation or
collective action necessary for making irrigation systems function.

- 13 -
Box 2.3
Water control (managerial)
"(...) by organizational control I mean that the
group of farmers are in control of distribution and
conflict resolution within their group, and that the
group has the right and the ability to negotiate with
other entities over the delivery of water to their
group." (Hunt, 1990:144)
Huppert (1989:35) defines four basic management
functions: planning, organising, leading and
controlling. “Controlling is defined as the "the
continuous monitoring and adjustment of all
activities of an organisation in line with pre-
determined plans and standards.”
"On the 22nd August 1973, the learned Advocate
General of Andhra Pradesh conceded that this
Tribunal has no power to direct the vesting of the
control and administration of the Rajolibunda
headworks and the common canals within Mysore
State limits in the Tungabhadra Board. However,
he prayed that it should make suitable
recommendations for vesting control and
administration of the aforesaid works in a joint
control body." (From: Government of India,
Krishna Water Disputes Tribunal (1973:54); this is
the report of a national committee that judged
conflicts over the allocation of water in the Krishna
river to different States in India).
Management or managerial control has
become a central topic in the irrigation
literature since the 1970s. This was a period
in which the awareness of the poor
performance of many irrigation systems grew.
This problem particularly occurred in
government built and managed large-scale
canal irrigation systems. These problems are
discussed in detail in later chapters. Here the
point to be made is that the organisational or
institutional dimension of water control is
now a central topic in irrigation studies
(Sampath and Young, 1990 is a recent
overview).
Like water control in the technical sense,
water control in the organisational sense has
translated into intervention programmes.
These aim to bring about organisational and
institutional change. In many countries in the
world there are or have been programmes to
establish Water Users Associations at
different levels of irrigation systems, and
more recently programmes for bureaucratic
reform of government irrigation agencies
have been started (see box 2.4 for an example
of the latter).

"Water control as used in this paper is defined asA further extension of the scope of irrigation
the ability of farmers to plan adequately and in studies lies in the attention for water control
time for cropping decisions by having the requiredin the socio-economic and political sense.
volumes of water available at the appropriate timeTechnical and managerial control refer to
and places for crop needs plus increments irrigation practices very directly. They deal
sufficient for leaching requirements and
evaporation losses. As defined, proper water
with the technologies and people directly
involved in water distribution, maintenance
control ensures that end users will have a relatively
and other necessary irrigation activities. Not
high degree of predictability of water supplies for
making cropping decisions and meeting crop yet addressed are what in the previous section
needs. (...) Effective water control is a function of
we called the `conditions of possibility' of
several complex sets of vatriables. Along with thetechnical and managerial water control. Box
physical, technical and economic factors, complex
2.5 gives examples from India and
social and institutional factors are also worthy of
serious study." (Lowdermilk, 1990:155) Bangladesh where the socio-economic
differentiation of farmers prevents the
emergence of effective forms of cooperation
among water users. The inequalities in the agrarian structure are part of `water control' because
they shape the forms of organisation in water management. An example of water control in the
political sense is also given in Box 2.5. The case is Mexico, where, as in many other parts of the
world, irrigation served as a means for the state government to politically and economically
control peasant farmers.

- 14 -
Box 2.4

Bureaucratic reform in Egypt

"With about 1000 m3 per person [per year] currently available, Egypt is right at the threshold of scarcity. The
available water per capita declines by 50% from 1960 to 1990, primarily as a result of rapid population
growth and a fixed water supply (...). But the paradox is this: Egypt currently manages its water resources as if
they were unlimited (...). All the trend lines point in the same direction, to what I call a creeping crisis: the
question is not if but when will Egypt begin to feel the effects of increasingly severe shortfalls of water supply
compared to demand. At present, Egyptian water resourse management institutions are poorly equipped to
respond to such a crisis." (Merrey, 1995a:1-3)

In 1995 the International Irrigation Management Institute (IIMI) conducted an institutional analysis of the
water resources sector in Egypt. The conclusion was as follows.
"Effective use of past and future technological investments will require improving the overall management
processes and institutional framework for water resource management, especially within the MPWWR
[Ministry of Public Works and Water Resources], as well as continued improvement in the human resources
of the Ministry. (Merrey, 1995a:18)

"The main product [of the institutional analysis] is an Action Plan for Strengthening Water Resource
Management in Egypt (...). This plan has three components:
1) a `Management Strengthening' component that will involve training in management skills for senior policy
makers, and a structured process to review water resource policies, and the the institutional changes needed to
address future challenges;
2) a component to begin introducing cost sharing by all water users including farmers; and
3) an Irrigation Operation and Maintenance Plan that is intended to use a participatory action research process
to introduce institutional and technical innovations for irrigation management." (Merrey, 1995b:6)

From the examples given in Box 2.5 it can be concluded that the relation between technical and
managerial control on one hand, and socio-economic and political control on the other, is two-
sided. Socio-economic and political conditions influence what happens inside irrigation
systems, and irrigation practices within irrigation systems have implications for the evolution of
these socio-economic and political conditions.

Also this third dimension of water control has translated into specific intervention programmes.
The clearest example is the implications of structural adjustment policies and other economic
and political reforms at the national and international level. These programmes aim to reduce or
change the government's role in irrigation and other sectors, and to make irrigation, and other
sectors, more financially viable. This means decentralising or `turning over' management to
users, and collecting higher irrigation service fees from farmers. For further discussion, see the
chapter on state-water users relations.

- 15 -
Box 2.5

Error! Bookmark not defined.Water control (socio-economic and political)

North India
"The relevant power structures engaged in controlling and directing the supplies of water were typically those
of the cultivating body within the village itself, and the picture which emerges from the enquiry into water-
course management conducted in the early 1870s (...) is one of petty elites often able to reinforce their multi-
faceted advantages within the village through their control over the allocation of scarce water. These village
elites were not unduly affected as government control extended down the hierarchy of distribution channels
and progressively confined local control to within the village boundaries." (Stone, 1984: 202)
For other studies of water control in South Indian systems see Wade (1987) and Mollinga (1996).

West Bengal (India) and Bangladesh

"[I]rrigation, or, more broadly, water control, constitutes the key technological constraint to agricultural
growth, or in Ishikawa's terminology is the `leading input'. This hypothesis is confirmed by strong relationships
between water control variables and various aspects of agricultural performance. (...) The irrigation-yield
correlations tend to rise through time, implying that irrigation’s leading role has been strengthened with the
increase in fertiliser use and the introduction of HYVs. (...) In the light of this, it is remarkable that water
control in West Bengal and Bangladesh remains so little developed. [T]he agrarian structure in Bangladesh
and West Bengal has impeded water control development, by adversely affecting the possibilities for
resolution of the associated public goods problems (...). The alternative to costly and often ineffective attempts
at bureaucratic control of water allocation would be control by organisations of the irrigators themselves, who
are in a much better position to monitor and enforce agreements and have strong incentive to do so. One
possible reason for the general failure of such institutions to emerge at the local level is that inequalities among
water users make it more difficult to achieve social control. It may be easy to enforce compliance from
relatively small and powerless cultivators, but effective limitation of water use by richer and more powerful
individuals is another matter." (Boyce, 1987:198-199, 229, 233)
For water control development in Bangladesh see Wood et al. (1990), and van Koppen and Mahmud (1995).

Mexico
"In Mexico the control and management of irrigation has a long history. Beginning as early as 1500 B.C. the
people in both the highland and the humid tropics developed a sophisticated system of water control that
included elaborate canal networks, dikes, dams, and aqueducts. After the Conquest, the Spanish, with their
own extensive experience with irrigation brought new irrigation technologies to the New World; from
Zacatecas to Oaxaca they dammed rivers, tapped groundwater, and built extensive canals and monumental
aqueducts. From the time of Mexico's independence in 1821 until the despotic reign of Porfirio Díaz began in
1880, political instability in the countryside impeded the construction of large irrigation systems. Expansion
began under the dictatorship of Díaz, who (...) gave concessions to companies, many of them foreign, to
develop agricultural lands that required irrigation. (...) The overthrow of Porfirio Díaz [in the 1910 Revolution]
(...) left the countryside torn by conflicting parties [and] made the expansion of irrigation systems impossible.
Although land reform was the rallying cry of the Revolution, the importance of water was not forgotten by the
framers of the Revolutionary Constitution of 1917, who included as Article 17 the nationalization of water. (...)
The irrigation law of 1926 was pushed through by President Plutarco Elías Calles. This law nationalised
private irrigation systems, and established the National Irrigation Commission (CNI). (...) Starting in 1930, it
created a series of irrigation districts. (...) From its foundation the CNI was highly political. The creation of
new irrigation projects was politically motivated, as the central government, located in Mexico City,
maneuvered to tighten its control over regions that had resisted national control. (...) Projects along the border
with the United States were initiated to strengthen Mexico's position in bilateral conflicts over water and even
territorial sovereignty. (...) The trend towards increased state control was slowed by a law passed in 1947 that
allowed for private holdings of irrigated land. The new law sanctioned the expansion of private agriculture in
irrigation districts, as well as the transfer of water rights when land was confiscated by the state for
redistribution. (...) This law allowed the private sector greater opportunity to control irrigated land and water
resources. (...) Today in Mexico, irrigation districts are one of the most important units of administration of
agriculture and water resources. (…) The control of irrigation districts continues to be dictated from Mexico
City. (Enge and Whiteford, 1989:5-7)

- 16 -
The three dimensions of water control are summarised in table 2.1.

Table 2.1: Three dimensions of water control

Dimension Association/meaning
Technical control Guiding-manipulating-
mastering of physical
processes
Organisational control Commanding-managing of
people's behaviour
Socio-economic and political Domination of people('s
control labour)
Regulation of social processes
Disciplines
(Civil) engineering, soil
mechanics, hydraulics,
hydrology, agronomy,
meteorology, agro-ecology
Management science,
extension science, public
administration, organisation
sociology
Political economy,
economics, rural sociology,
political science,
anthropology, gender studies,
agrarian history

2.2.2 An example of the relation of the three dimensions of water control

With this discussion we have outlined the different meanings that are attached to water control.
Related to different meanings are different scientific disciplines that analyse a specific type of
water control. The three different fields of disciplines identified in table 2.1 have very little
interaction. The concept of ‘control’ is used in all three domains of scientific discourse, which
shows the flexibility of the concept, but in neither of the three domains `control' has become an
object of theoretical reflection across the boundaries of these domains. Work on each type of
water control is done without much reference to the other types of water control. We regard this
as a missed opportunity.

Before discussing control as an analytical concept in more detail, we summarise an example that
shows that the three dimensions of water control are related intimately in reality. The example is
a historical one. Box 2.6 discusses irrigation development in Maharashtra, one of the Indian
states, in the colonial period. The colonial government attempted to introduce a fundamental
change in irrigation practices in the Nira Left Bank Canal in the early 20th century. The
government's modernisation plan had a technical element (new outlet structures), an
organisational element (new rules for distribution) and a socio-economic element (the
introduction of volumetric pricing of water). Changes in all three elements were necessary to
achieve the objectives of the plan. Because the desired changes could not be effectuated in all
three elements, the outcomes were very different than anticipated.

- 17 -
Box 2.6
Modules for modernisation

The Nira Left Bank Canal is a large scale canal irrigation system in South India. In the colonial period it was
located in Bombay Presidency; it is now part of Maharashtra State. This 54,000 hectares sized system was
designed as a protective irrigation system. Located in a drought prone area it was intended to spread available
water thinly over a large area to protect crops against crop failure and farmers against famine. The supplementary
irrigation of traditional food crops was unattractive to farmers, because the investments to make land suitable for
irrigation were considerable, while the crops were not very responsive to irrigation. In years of normal rainfall the
moisture retentive heavy clay soils could store sufficient water to bring the crop to maturity. Only when rains
totally failed farmers were interested in irrigation. There was a problem of unwanted water; the system was
underutilized and yielded very low financial returns.
In the period 1900-1940 the British colonial government tried to introduce a modernization plan that consisted
of three steps.
1) Concentrate irrigated areas in blocks, fix the demand for irrigation water of these blocks and promote the
production of a remunerative crop like sugarcane on part of the area.
2) Control the distribution of water and avoid corrupt practices and cultivator interference by the invention of
a machine that discharges fixed quantities of the expensive irrigation water.
3) Sell water by volume to avoid waste of water and see to economic use of it according to market principles.

With the block system the government sought to achieve managerial control. It hoped to establish working and
manageable institutional arrangements and procedures for water delivery to groups of farmers. The second step,
the water delivery machine was the element of technical control. It was the technical translation of the desire to
deliver quantifiable amounts of water to farmers. The third step, the introduction of the price mechanism to achieve
efficiency of water use, was believed to be an effective economic control mechanism for achieving the objective of
a modern type of agricultural production. This was an effort to control the conditions that facilitate the unrolling of
a more dynamic agrarian development process.
The block system was quickly introduced, though with an important amendment: the land consolidation needed
to create contiguous blocks was impossible to implement. This made control of the sugarcane area very difficult.
The sugarcane area grew explosively. Its high water consumption deprived a large part of the system of water. The
increase in sugarcane cultivation made the block system a financial success because the taxes collected by the
government increased. The government turned a blind eye to the indiscriminate growth of the sugarcane area.
The design of a working modular outlet (or module) that discharged fixed quantities of water independent of
upstream and downstream canal water levels, proved very difficult. The most problematic design criterion was the
desire to make the device `tamper proof'. After several decades of research and trials the effort was given up.
Because of the failure on the technical front, the volumetric pricing of water never came in sight. The
modernization plan did not achieve its objectives, an issue on which debate continues to the present day. The table
below summarizes what was intended and what happened.

Before modernization Modernization plan Actual result

Technical:
- non-modular pipe outlets (do not allow - modular outlets - non-modular pipe outlets
quantitative regulation) - irrigation in blocks with `wet' and `dry' - sugarcane irrigation concentrated in head
- widely spread command area crops mixed end of the canal system

Managerial:
- government supplying water to individual - government supplying water to - head end/affluent farmers influence block
farmers blocks/groups of farmers system and appropriate excess water
- group management inside blocks

Socio-economic/political:
- colonial government prescribes cropping
pattern, prohibiting water consuming crops
like rice and sugarcane
- colonial government mainly concerned
about limiting social unrest in case of famine,
and tax collection
- colonial government allows mix of `wet'
and `dry' crops in fixed proportion
- colonial government seeks to combine
economic growth with anti-famine
objectives
- colonial government wants to increase
tax collection
- dominant farmers grow what they like
- post-independence government dominated
by sugarcane lobby
- overall level of tax collection decreases
(populist farmer oriented policies)

Source: Bolding, Mollinga and van Straaten (1995)

- 18 -
2.2.3 Water control: a political process of contested resource use

We can now look more closely at the analytical meaning of the concept of water control. The
Nira Left Bank Canal example in Box 2.6 and some of the other examples given above, show
several things.

First, water control is about the use of a natural resource, water, for irrigated agricultural
production. Irrigated agricultural production is part of the wider process of rural development,
which in its turn has implications for the development of society and ecology in the widest
sense.

Second, the use of the resource is contested. Around the use of water there are conflicts,
struggle, negotiations, and other confrontations of the different stakeholders with their different
interests and strategies.

This means, thirdly, that water control is a political process in the broad sense of that term.5)
Kerkvliet gives the following definition of politics in the broad sense. It is "the debates,
conflicts, decisions, and cooperation among individuals, groups and organisations regarding the
control, allocation, and use of resources and the values and ideas underlying these activities."
(Kerkvliet, 1990:11) This is exactly what is involved in the use of water in irrigation systems.
Water control can thus be defined as follows.

Water control = a political process of contested resource use.

The description of water control as a case of contested resource use, places the study of water
control in a current of interdisciplinary studies on natural resources that tries to transcend the
dichotomy between human culture and behaviour and the physical environment which sustains
them (Bennett, 1990:435; for these approaches see for example Dahlberg and Bennett, 1986;
Moran, 1990; Smith and Reeves, 1989). This literature speaks of socionatural systems, where
we speak of sociotechnical systems. This difference in terminology illustrates the greater
emphasis on the technological mediation of (wo)man's interaction with the natural world in our
sociotechnical approach than in the broader socionatural approach. Our sociotechnical approach
has a strong political economy flavour, while socionatural approaches of resource studies have
more of a social anthropological hue.

References

Anderson, Perry, 1980, ‘Agency’. In: Arguments within English marxism. New Left Books, London, pp.16-58

Bennett, John W., 1990, `Ecosystems, environmentalism, resource conservation, and anthropological research'. In: Emilio F.
Moran (ed.), pp.435-457

Benton, Ted, 1989, Marx and natural limits: An ecological critique and reconstruction. New Left Review No.178, pp.51-86

Benvenuti, B., 1982, `De technologisch-administratieve taakomgeving (TATE) van landbouwbedrijven'. Marquetalia No.5,
pp.111-136

5
The narrow sense is politics as formal party and state politics. Kerkvliet (1990) uses the term, ‘everyday politics’
for the wider meaning of politics.

- 19 -
Bhaskar, Roy, 1979, `On the possibility of social scientific knowledge and the limits of naturalism'. In: John Mepham and D-H.
Ruben (eds.) Issues in marxist philosophy. Vol. III Epistemology, Science, Ideology. Harvester, Brighton, pp.107-139

Bolding, Alex, Peter P. Mollinga and Kees van Straaten (1995) `Modules for modernisation. Colonial irrigation in India and the
technological dimension of agrarian change. Journal of Development Studies Vol.31, No.6, pp.805-844

Bourdieu, Pierre, 1977, Outline of a theory of practice, Cambridge University Press, Cambridge (orig. 1972).

Boyce, James K., 1987, Agrarian Impasse in Bengal. Institutional Constraints to Technological Change. Oxford University
Press, New York.

Burawoy, Michael (1985) The politics of production. Factory regimes under capitalism and socialism. Verso, London

Burns, Robert E. (1993) `Irrigated rice culture in monsoon Asia: the search for an effective water control technology', World
Development, Vol.21, No.5, pp.771-789

Callon, Michel, 1992, `The dynamics of techno-economic networks'. In: R. Coombs, P. Saviotti and V. Walsh (eds.)
Technological change and company strategies: economic and sociological perspectives. Academic Press, London, pp.72-102

Carlstein, Tommy, 1982, `Irrigation agriculture'. In: Time resources, society and ecology. Volume 1. Preindustrial societies.
George Allen and Unwin, London, pp.257-300

Chambers, Robert, 1988, Managing Canal Irrigation. Practical Analysis from South Asia, Oxford & IBH, New Delhi, India.

Dahlberg, Kenneth A. and John W. Bennett (eds.) Natural resources and people. Conceptual issues in interdisciplinary
research. Westview Special Studies in Natural Resources and Energy Management. Westview, Boulder and Colorado

Dhawan, B.D., 1988, Irrigation in India's Agricultural Development. Productivity, Stability, Equity, Sage, New Delhi.

Enge, Kjell I. and Scott Whiteford, 1989, The keepers of water and earth. Mexican social organisation and irrigation.
University of Texas Press, Austin

FAO, 1977, Crop water requirements. FAO Irrigation and Drainage Paper 24. Rome

Friedland, William H., Amy E. Baron and Robert J. Thomas, 1981, Manufacturing green gold. Capital, labor and technology in
the lettuce industry. Cambridge University Press, Cambridge

Giddens, Anthony, 1984, The constitution of society. An outline of the theory of structuration, Polity Press, Cambridge.

Government of India, Krishna Water Disputes Tribunal, 1973, The report of the Krishna Water Disputes Tribunal with the
decision. New Delhi

Guijt, Irene and John Thompson, 1994, `Landscapes and livelihoods. Environmental and socioeconomic dimensions of small
scale irrigation'. Land Use Policy Vol.11, No.4, pp.294-308

Hunt, Robert C., 1990, `Organizational control over water: the positive identification of a social constraint on farmer
participation'. In: R.K. Sampath and Robert A. Young (eds.), 1990, Social, economic, and institutional issues in third world
irrigation management. Studies in Water Policy and Management, No.15. Westview, Boulder, pp. 141-154

Huppert, Walter, 1989, Situation conformity and service orientation in irrigation management. GTZ, Eschborn

Kerkvliet, Benedict J. Tria (1990) Everyday politics in the Philippines. Class and status relations in a Central Luzon village,
University of California Press, Berkeley

Knights, David and Hugh Wilmott (eds.), 1990, Labour process theory. Macmillan, Houndmills and London

Koppen, Barbara van and Simeen Mahmud, 1995, Women and water pumps in Bangladesh: The impact of particpation in
irrigation groups on women's status. Bangladesh In stitute of Development Studies/Department of Irrigation and Soil and Water
Conservation, Wageningen Agricultural University

Krol, Marjon, 1994, `Irrigatie is mannenwerk'. Genderverhoudingen in een kleinschalig irrigatieprojekt in de Ecuadoriaanse
Andes. Wageningen (unpublished M.Sc. thesis)

- 20 -
Lowdermilk, Max K., 1990, `Irrigation water control and anarch'. In: R.K. Sampath and Robert A. Young (eds.), pp.155-173

Merrey, Douglas J., 1995a, Governance and institutional arrangements for managing water resources in Egypt. Paper prepared
for a public lecture on 2 November 1995, Wageningen Agricultural University, Wageningen

Merrey, Douglas J., 1995b, Particpatory action research: experience from Sri Lanka and application to Egypt . Paper prepared
for a seminar, 23 October 1995, Wageningen Agricultural University, Wageningen

Mitchell, William P. and David Guillet (eds.), 1993, Irrigation at high altitudes: The social organization of water control
systems in the Andes. Volume 12. Society for Latin American Anthropology Publication Series.

Mollinga, Peter P. (1996) On the waterfront. Water distribution, technology and agrarian change in a South Indian irrigation
system, Wageningen (Ph.D. thesis) (forthcoming)

Moran, Emilio F. (ed.), 1990, The ecosystem approach in anthropology. From concept to practice. The University of Michigan
Press, Ann Arbor

Ostrom, Elinor, 1990, Governing the commons. The evolution of institutions for collective action. Cambridge University Press,
Cambridge

Ostrom, Elinor, Larry Schroeder and Susan Wynne, 1993, Institutional incentives and sustainable development. Infrastructure
policies in perspective. Wedstview Press, Boulder

Ploeg, Jan Douwe van der, 1991, Landouw als mensenwerk. Arbeid en technologie in de agrarische ontwikkeling. Coutinho,
Muiderberg

Plusquellec, Hervé, Charles Burt and Hans W. Wolter (1994) Modern water control in irrigation. Concepts, issues, and
applications. World Bank Technical Paper 246, World Bank, Washington DC

Sampath, R.K. and Robert A. Young (eds.), 1990, Social, economic, and institutional issues in third world irrigation
management. Studies in Water Policy and Management, No.15. Westview, Boulder

Smith, Sheldon and Ed Reeves (eds.), 1989, Human systems ecology. Studies in the integration of political economy,
adaptation, and socionatural regions. Westview Special Studies in Applied Anthropology. Westview, Boulder

Stone, Ian, 1984, Canal Irrigation in British India. Perspectives on Technological Change in a Peasant Economy, Cambridge
University Press, Cambridge.

Thompson, Paul, 1989, The nature of work. An introduction to debates on the labour process. MacMillan, London (orog. 1983)

Ubels, Jan and Lucas Horst (eds.), 1993, Irrigation design in Africa. Towards an interactive method. CTA, Wageningen

Uphoff, Norman, 1986, Improving international irrigation management with farmer participation. Getting the process right.
Studies in Water Policy and Management, No.11. Westview, Boulder and London

Uphoff, Norman (with Priti Ramamurthy and Roy Steiner), 1991, Managing irrigation. Analyzing and improving the
performance of bureaucracies. Sage, New Delhi

Villarreal, Magdalena, 1994, Wielding and yielding: Power, subordination and gender identity on the context of a Mexican
development project. Wageningen (Ph.D. thesis)

Wade, Robert, 1982, `The System of Administrative and Political Corruption: Canal Irrigation in South India', Journal of
Development Studies, Vol.18, No.3, pp.287-327.

WADE, Robert (1987) "The social response to irrigation". In: Village republics: economic conditions for collective action in
South India. Cambridge University Press, Cambridge, pp.72-95

Wood, Geof D., R. Palmer-Jones, Q.F. Ahmed, M.A.S. Mandal and S.C. Dutta, 1990, The water sellers. A cooperative venture
by the rural poor. Kumarian, Connecticut USA

- 21 -