This article was downloaded by: [University of California Santa Cruz]On: 26 October 2014, At: 22:08Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Oxford Development StudiesPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/cods20

Which Order? Whose Order? Balinese IrrigationManagement in Sulawesi, IndonesiaDik Roth aa Department of Social Sciences, Law & Governance , Wageningen University , Postbus 17,6700 AA, Wageningen, The NetherlandsPublished online: 23 Aug 2006.

To cite this article: Dik Roth (2006) Which Order? Whose Order? Balinese Irrigation Management in Sulawesi, Indonesia,Oxford Development Studies, 34:1, 31-46, DOI: 10.1080/13600810500495956

To link to this article: http://dx.doi.org/10.1080/13600810500495956

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Which Order? Whose Order? BalineseIrrigation Management in Sulawesi,Indonesia

DIK ROTH*

ABSTRACT This paper deals with irrigation management among Balinese migrant settlers inSulawesi, Indonesia. As settlers in the command area of a state-built irrigation system, they havebecome part of its blueprinted managerial structure. However, many settlers derived theirexperience from subak, the Balinese irrigators’ institution. This paper explores the technical,organizational and normative complexity hidden behind claims of order, manageability and controlof a “modern” irrigation system, examining three issues that illustrate the tension between order anddisjuncture. First, it criticizes conceptualizations of local management as cycles of degradation byfarmer neglect and rehabilitation by government attention. Second, it traces the local history ofirrigation development by putting into perspective the assumption of normative, technical andorganizational uniformity on which the management structure is based. Third, differences arediscussed between engineering approaches to management and Balinese conceptualizations ofmanagement, and their consequences for management practices.

1. Introduction

This paper deals with irrigation management among Balinese migrants in Luwu, Sulawesi.

The point of departure is the dual character of state-established water users’ associations

(WUAs): anchored in participatory ideology, but legally enforced and bureaucratically

imposed. As settlers in the command area of a state-built irrigation system, the Balinese

rice farmers in this study have become part of the blueprinted structure of a system built by

the state agency Public Works. However, many settlers were experienced irrigators who

already had knowledge of irrigation management associated with subak, the “traditional”

Balinese irrigators’ institution.

I explore the diversity and complexity hidden behind the seemingly uniform structure of

a “modern” irrigation system, focusing on the articulations between technical, normative-

legal and organizational arrangements and irrigation practices associated with the tertiary

unit (TU)/WUA structure and with subak. Rather than assuming that certain state-

engineered relationships exist between people, water and technology, I have made the

character and quality of those relationships—of local forms of resource use and

management—the object of critical inquiry.

ISSN 1360-0818 print/ISSN 1469-9966 online/06/010031-16

q 2006 International Development Centre, Oxford

DOI: 10.1080/13600810500495956

*Dik Roth, Department of Social Sciences, Law & Governance, Wageningen University, Postbus 17, 6700 AA,

Wageningen, The Netherlands.

Oxford Development Studies,Vol. 34, No. 1, March 2006

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

Three issues are discussed that are relevant for understanding “order” and “disjuncture”

in irrigation development. First, I put into perspective the prevailing image of local

irrigation management in terms of cycles of degradation (by farmer neglect) and

rehabilitation (by government attention). The historical, socio-cultural and otherwise

embedded character of irrigation management tends to be completely elided from

accounts by external experts and officials (Boelens, 1998; Mosse, 1997). Second, I trace

the local history of irrigation development by taking a critical look at the assumption of

normative, technical, organizational uniformity. More specifically, I shall analyse the role

of subak and TU/WUA, and the history of their interaction in local irrigation management.

Third, I discuss different conceptualizations of “management” between external,

engineering-based forms of irrigation development and Balinese irrigated agriculture,

and their consequences for management practices.

An irrigation system requires compatibility between rules, infrastructural technology

and organizational arrangements.1 Ideally, norms of water distribution and water rights

based on them form the foundation of, and are physically reproduced by, material

technology like division works. Similarly, norms and principles pertaining to rights,

responsibilities and conceptions of equity form the basis of organizational elements like

local irrigators’ organizations. These, in turn, reproduce such norms and principles but

are also co-determined by technical properties and requirements of the system.

The normative-legal dimensions of irrigation, then, are important connecting elements in

irrigation systems. Therefore, they should not be analytically isolated from the other

dimensions of irrigation and water control mentioned above.

A distinction can be made between endogenous and external irrigation design and

development (Boelens, 1998). In the former, technical, normative and organizational

design principles originate from system users and their socio-cultural environment. In the

latter, users are not the designers. Irrigation design and development are based on external

inputs and interventions. In situations of introduction of external design principles,

cultural identity and the socio-culturally embedded character of irrigation management

practices are crucial but often neglected. Irrigation settings cannot be reduced to their

technical and economic properties, but should be analysed as socially, culturally and

historically situated (Gelles, 1998; Mosse, 1997).

2. The Setting

2.1 Transmigration in Luwu

Luwu is a largely mountain-covered area in South Sulawesi (Figure 1). North Luwu

became a major centre of exploitation of land and water resources. The rivers flowing into

the North Luwu Plain make the area very suitable for irrigation development. From the

1930s, it became a destination for the Dutch “colonization” programme through which

Javanese (and, on other islands, Balinese) were resettled.2 The main objectives were

poverty alleviation, reduction of population density on Java and Bali, and economic

development of the “outer regions” of the Dutch colony. A key element was the creation of

rural settlements based on irrigated agriculture, using the experience of Javanese and

Balinese farmers. After independence, the Indonesian transmigration programme

(transmigrasi) continued this policy. From the late 1960s thousands of farmer families,

especially from Java and Bali, were resettled in Luwu.3

32 D. Roth

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

2.2 Irrigation Development and Water Users’ Associations

In the 1960s, foreign donor funding for transmigration, irrigation and infrastructure

development became available. Indonesian irrigation development under the Suharto

regime was a project of modernization, the exclusive business of engineers. Preoccupation

with centralized control, suppression of local initiatives, and construction interests have

Figure 1. Luwu and the Kalaena irrigation system

Balinese Irrigation Management 33

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

not been conducive to the development of strong local organizations and initiatives outside

the technical and organizational frameworks of the irrigation bureaucracy. Low performance,

operation and maintenance problems, degrading infrastructure, and dependence on

rehabilitation projects characterize most systems (Booth, 1977a, b; Bruns, 2004).

Systems built, rehabilitated or expanded under the responsibility of the Ministry of

Public Works and based on engineering technology have a similar physical-technical,

organizational and operational set-up.4 They typically consist of a weir and a hierarchy of

primary, secondary, tertiary and quaternary canals. Off-take structures and devices for

measurement and control form the division points between primary and secondary, and

between secondary and tertiary canals. The system section below the division gate to the

tertiary canal is called the TU. Inside the TU, water division boxes with flap gates regulate

water flows to quaternary canals, from which water is divided to the individual plots.

From the 1970s, water users were organized in WUAs responsible for operation and

maintenance of the TU.5 Later, training projects for “participatory irrigation management”

were introduced. A 1984 presidential decree made WUAs in the TUs of Public Works

systems compulsory. While the main system (weir, primary and secondary canals,

including the tertiary gate) remained under the responsibility of Public Works, the WUA

was given a key role in tertiary irrigation management. Ideally, the WUAs created through

such policies become legal entities.6 TUs cover between 50 and 150 ha, and are defined by

principles for design and use, internal organization, and formal procedures. After the

introduction of WUAs, problems remained with system performance. In 1987, the

Irrigation Operation and Maintenance Policy (IOMP) was formulated in response to donor

concerns. This included the introduction of cost recovery principles and an irrigation

service fee (ISF). In the late 1990s, political changes and the growing awareness of

problems in the irrigation sector led to new reforms to enable further devolution of rights

and responsibilities for irrigation management, and to empower WUAs to become viable

and autonomous institutions (Bruns, 2004; Oad, 2001).

The Indonesian WUA model has been criticized for its focus on formal organizational

arrangements, routines and procedures rather than local decision-making and management

capacities and effectiveness in coping with management problems. It has created formal

organizations that are often incapable of mobilizing water users for management needs.

By stressing formal structures and creating bureaucratic procedures for even the smallest

problems, WUAs mainly raise transaction costs for water users (Bruns, 2004).

Another criticism concerns the scope of WUA rights and responsibilities. These are

usually presented as (near) “total” rights, giving farmers a “sense of ownership”.

In practice, few rights are delegated to WUAs. Turnover was a top-down devolution of

management tasks and responsibilities to farmers rather than a participatory process

involving the establishment of strong local rights. Water and irrigation infrastructure are

exclusively owned by the state. WUAs have the right to use a state-owned natural resource

delivered to the TUs through a public irrigation system.

By this lack of attention to property rights, the complex character of jointly managed

irrigation systems and their property dimensions was disregarded. Hardly any attention

was paid to the interface between the main system and its public property characteristics,

and the WUAs with their common property characteristics; nor to the interface between

the latter and the private property dimensions of the resource as it enters individual fields.

In addition, in many irrigation systems in Indonesia, the transfer part of the establishment

of TUs as legal entities has not or has only partly been realized. Even under the reform,

34 D. Roth

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

WUA development remained a target-driven transfer of responsibilities to WUAs rather

than their empowerment with a more complete bundle of responsibilities and rights

(Bruns, 2004; Oad, 2001).

2.3 Kertoraharjo: A Balinese Village

The Balinese village of Kertoraharjo forms the northern part of the former transmigration

settlement Kertoraharjo I. It is located in the command area of the Kalaena irrigation

system in North Luwu (Figures 1 and 2). In 1972–73, 500 transmigrant families from Bali

and Java were resettled here. They received 2 ha of forest-covered land, to be developed

into home yards, irrigated fields and rain-fed land. From the 1980s, when the system

reached Kertoraharjo, it became a relatively thriving village. Almost all the land, including

the land initially planned for rain-fed agriculture, can be irrigated, and yields two rice

harvests a year. Many Balinese now also cultivate cocoa on land bought from the local

population.

Balinese migration and settlement entailed a recreation and reinvention of Balinese

culture, identity and social organization. Alongside the state administrative arrangements

at village level, a domain of Balinese customary institutional and administrative

arrangements was established. Key Balinese institutions are the customary village, the

customary hamlet, temple groups and various voluntary associations. In Kertoraharjo,

the gods are still “real social partners” (Guermonprez, 1990, p. 62; Warren, 1993).

Last but not least, there is the subak, the irrigators’ association. The gods are real social

partners not only in the village but also in the rice fields. For the Balinese, the initially

forested land was not just a resource to be economically exploited by transforming the

forest into agricultural land. These processes are deeply embedded in cultural meaning.

The forest had to be physically and ritually transformed into irrigated fields. Sawahs

(irrigated rice fields) are managed ritually and ceremonially to maintain the balance

between the godly world, human beings and the resources involved (field, water and

crops). For Balinese, turning forest into agricultural land is more than a mere application

of human labour to a neutral natural environment. Other “stakeholders” are involved, in

particular the various spirits which, if not treated with awe and care, may become a threat

to people and crops. Forest clearing, then, is accompanied by ceremonies, rituals and

offerings (Charras, 1982).

The relationship between the spiritual world, humans and resources does not stop once

the forest has become agricultural land. In Bali, the rice cycle, from preparatory activities

to post-harvest offerings, forms a chain of highly ordered and interrelated activities.

The Balinese conceive of agriculture as a human activity intimately bound up with the

godly world. Sari, the essence of rice, originates from the body of the rice goddess, Devi

Sri, and should be returned to her after harvest to ensure the continuation of the cycle.

Gods subsist on the essence of rice and are the ultimate source of its substance in the

human world. Hence, its cultivation is susceptible to threats of transgression of ceremonial

rules, pollution and disturbance of the delicate relationship between humans and the rice

goddess. Therefore, the cultivation cycle is accompanied by a ritual–ceremonial cycle,

which suggests an analogy to the human life cycle (Howe, 1991).

In Bali, the subak performs many functions related to irrigated agriculture in the

broadest sense, including construction, repairs, operation and maintenance, agricultural

planning and pest control, conflict resolution, the organization of ritual and ceremony,

Balinese Irrigation Management 35

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

temple construction and maintenance, maintenance of religious purity, collection of tax

and fines, creation and enforcement of subak legal regulations, and application of

sanctions. A subak typically includes a complex of between tens and hundreds of hectares

of (mainly) irrigated fields. Physical boundaries, and hydrological and socio-political

Figure 2. Kertoraharjo village, tertiary units and approximate subak areas

36 D. Roth

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

factors play an important role in determining the subak area. Subaks may be subdivided

into smaller units or be part of larger complexes. The subak head (klian subak, pekaseh) is

assisted by other functionaries.7 Membership is associated with rights and duties with

regard to water, ritual, agricultural planning, organization and management. Subak

technology for water division is based on fixed proportional division of continuous flows

through wooden or stone overflow weirs (temuku). An advantage is that the water flow is

divided in a direction parallel to the current. The only variable being the width of the

proportional openings in the structure, water division is relatively transparent and easily

controllable by farmers (Geertz, 1980; Jha, 2002; Horst, 1996; Sutawan, 1987).

3 Two Short Histories of Irrigation Development

3.1 Cycles of Degradation and Rehabilitation

Many TUs in Kertoraharjo look rather degraded. Water control is low, there are serious

maintenance problems, and traces of tampering with infrastructure. Locks have been

destroyed, quaternary gates fixed with wire, sawn off or removed, thresholds broken

away, gate openings filled, and division boxes smashed to pieces. Eighty per cent of the

boxes can no longer perform their function: rotational water division in the TUs. For

officials, the issue is simple: apart from infrastructure, all you need is a hierarchical

command structure, a top-down decision-making routine for determining opening and

closure dates of the system, and WUAs for the TUs to take over formal management

responsibilities and enforce rule-conformity upon their members. If working things

deteriorate, degradation is attributed to the farmers’ “unruliness” or “stupidity”. As an

official put it:

Farmers are not yet aware of the need for a more active attitude. The government has

pampered them. They have been given land and an irrigation system, so that they can

have two rice harvests in a year. They have their WUAs, through which the

government has tried to give them a sense of ownership. We organize WUA courses

to make farmers aware of their duties and to teach them how to run their WUAs.

What else can we do? But still, they do not turn up at WUA meetings or perform

collective labour, and let their WUAs degrade until another rehabilitation

programme for their tertiary units is necessary.

Wayan, a Balinese farmer and WUA chairman, knows the problems mentioned by the

official. However, the course he attended was more about formal rules and procedures than

about the daily reality of irrigation management:

Wayan has just returned from a course on tertiary water management for WUA staff,

organized by Public Works. The main objective is the revitalization or

re-establishment of WUAs. The construction section of Public Works organized

the course; the Irrigation Service provided lists with the names of the trainees. For

some years, Wayan has wanted to step back down from his function. After ten years

he is fed up with it. His TU is among the most degraded, and its WUA largely

inactive. Members usually do not join collective labour or meetings, and do not pay

fines. Formally, a new board is elected or the sitting one re-elected every three years,

Balinese Irrigation Management 37

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

but every time elections were planned, Wayan and his staff were forced to continue

for lack of any other candidate, or because there were not enough members present

to elect a new head. Therefore, he had to attend this course. Wayan: “we learned

about WUA organization, about government regulations pertaining to TUs, about

rice cultivation, post-harvest practices, and plans to reach three harvests per year.

We also learned how to operate the tertiary system using the division boxes for water

rotation.” “But”, I ask him, “you don’t use those boxes at all, do you? Most boxes

have decayed or been smashed to pieces, haven’t they? And water isn’t rotated at all,

is it?” “Well”, Wayan replied, “that is what I tried to ask them about, our TUs,

during the course. But they don’t want to hear all that. They told me that we have to

learn how to use them. “One day”, they said, “all division boxes will be rehabilitated

so you can rotate the water as you should”.’

Is it possible to write another history of irrigation development in these TUs? A history

that is not exclusively based on a mono-causal and external view of farmers’

behaviour?

3.2 An Alternative History of Irrigation Development

In a TU with a part Balinese, part Toraja8 farming population, I met a group of Balinese

preparing for the agricultural season. They were improving the farm road, repairing

canals and working on a construction to provide tail end fields with more water. Earlier

negotiations about water had resulted in a plan for water distribution and the reuse of

drain water within this small group of 11 farmers, who were constructing a system of

canals, PVC pipes and crossings. This group had been active for 3 years. They could no

longer stand the malfunctioning water division boxes of the Public Works system, and

decided to replace the remains of their box with a Balinese temuku, a proportional

division structure, and to become more active as a group. They created a small group

fund from which members could borrow money, and legal regulations (awig-awig)

stipulating rights, obligations and fines. Further, they had introduced into their group

ways of organizing collective labour on the canal infrastructure in proportion to the area

of land owned.

These farmers do not really fit the picture sketched by the official. Though a relatively

successful group, it is certainly not exceptional. It shows that the history of local irrigation

management is more than just a cycle of degradation caused by farmers and rehabilitation

by the government. Irrigation development in Indonesia was, until the late 1990s, largely

determined by foreign donor funding. While the irrigation bureaucracy converted donor

funding into construction packages and training courses, farmers were readapting the

resulting infrastructure to their priorities, wishes and values, and developing their own

irrigation and organizing practices. This was also the case in Kalaena. Initially TUs were

the outcome of the application of standardized design criteria. According to farmers, parts

of the TUs were functioning but many were not. Most TUs required considerable

adaptation and change by the farmers. Thus, farmers have been engaged in a dual role of

adapting to the Public Works infrastructure, rule system and organizational arrangements,

and changing the physical infrastructure and social organization of irrigation management

to their needs, priorities and interests.

38 D. Roth

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

4. Irrigation Management between Subak and WUA

Irrigation development has been based on a number of basic tenets of modernization

thinking: the superiority of new technology and regulatory arrangements over existing

“traditional” ones, the unproblematic “transfer” of such external elements into a “target”

population, and their untransformed acceptance and internalization by water users

“participating” in WUAs. However, the management practices that have developed—the

construction of proportional water division structures, the creation of group regulations

and forms of organization—derive primarily from subak. Any analysis of local irrigation

practices should, therefore, pay attention to the history of interaction between the subak

and the TU/WUA, and the impact of their articulation upon farmer behaviour and

irrigation practices.

Although much land was still forested and agriculture rain-fed, the first settler groups

had established a subak upon arrival. After some years, when all four Balinese settler

groups had arrived and made progress in developing their land, this subak was split into

four separate subaks. The subaks were spatially defined by the boundaries of the blocks of

land allocated to these groups, and are still known among the population by reference to

the settler groups that initially formed their membership: subak 150KK, 100KK, 50KK

and 50KK Tampaksiring.9 Each of the four subaks was headed by a klian, assisted by other

functionaries. Together, these formed a pekaseh, headed by a functionary with the same

name. Subak regulations taken from Bali were adapted to the local situation. These

included religious-ritual issues as well as those related to irrigated agriculture.

At this time, the construction to expand the Kalaena system beyond the Dutch works

had not even started. Why, then, were subaks established at all? Their early formation

points to the central importance of their religious-ritual functions, next to irrigation

management functions in a more restricted sense. Former subak functionaries and farmers

stressed the need for organizational arrangements to stage agricultural rituals. The forested

area was full of destructive forces and evil spirits, which in Balinese eyes can only

effectively be combated through the appropriate rituals. Further, the subak was essential to

determining a propitious day for planting and, thus, in disciplining cultivation practices.

It was also associated with early attempts to control small water sources.

Thus, contrary to the subaks in Bali, those in Kertoraharjo were not first defined by

physical and hydrological boundaries. The pattern of land allocation to settler groups

rather than water flows determined subak membership and boundaries. Definition of subak

in terms of land allocation by the state became an important “resource” in conflicts over

the boundaries of legitimate subak authority, especially the right to collect subak tax (sarin

tahun) and determine the day on which Balinese are allowed to start transplanting rice.

Those who define “subak” in terms of the boundaries of state-allocated land tend to reject

subak authority and subak regulation pertaining to land resources outside these areas, such

as on land bought in other villages. However, the early developments make clear that the

subaks could display the same wide variety of functions as in Bali, covering ritual,

agricultural practices and decision-making, and construction, as well as irrigation

management in the sense used by government agencies.

Irrigation water reached Kertoraharjo around 1983. Introduction of the system had

serious consequences for the subaks. As explained above, the organizational arrangements

for tertiary irrigation management—in the WUAs—were delivered as a “package” with the

physical TU infrastructure. Layout and construction of TUs and establishment of WUAs

Balinese Irrigation Management 39

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

were based fully on irrigation-technical criteria. The TU boundaries (based on design

criteria) cut across the pre-existing subaks, which, based on earlier land allocation, had

never been linked to irrigation. With the completion of the infrastructure, the cross-cutting

boundaries of TUs and subak areas and the obligatorily established WUAs, the subaks lost

their potential relevance as organizations for irrigation management. Nor could the

TU/WUA complexes, with their sometimes ethnically heterogeneous farmer population,

fulfil that other important subak function: organizing and carrying out Balinese rituals for

irrigated rice cultivation.

Consequently, upon establishment of TUs and WUAs, the complex of agricultural,

irrigation-managerial and religious-ritual activities of irrigated agriculture was torn apart.

This separation arose from divergent perceptions of irrigated agriculture between the

government-administrative world and the life world of Balinese farmers. An engineering

and administrative understanding of “management” (Ind. manajemen) includes routine

operation and maintenance tasks like canal cleaning and repairs, but excludes the

agronomic and religious-ritual dimensions of irrigated agriculture. While the former is

mainly the responsibility of PERTANIAN (Agricultural Service), the latter is classified as

belonging to the domain of agama (religion). Neither of the two is associated with

irrigation management.

In Balinese irrigated agriculture, such distinctions are not relevant. Balinese do not

commonly use the term “management” to refer to this complex field, but tend to use

persubakan to refer to all activities of rice agriculture. However, it was “irrigation

management”, defined as the operation and maintenance of tertiary infrastructure, which

became the formal responsibility of the WUAs. Agricultural planning belongs to the

government domain as well (Agricultural Service and district Irrigation Committee).10

Hence, the subaks as organizations were forced to retreat to their religious-ritual function

and refrain from interference in irrigation management as defined by the government

agencies. A different story can be told for the subak as an institution—regularized patterns

of behaviour between individuals and groups (Leach et al., 1999; Meinzen-Dick &

Pradhan, 2001). Even where subak organizational authority was formally reduced to the

religious-ritual sphere, there remained many settings in which subak and pekaseh

continued to influence matters for which government agencies are formally responsible.

After WUA establishment and separation of religious-ritual and “management”

functions, the subaks continued to play a role. The Balinese had difficulties in making

WUAs function without recourse to elements of the subak. Soon, organizational

arrangements deriving from the subak emerged in the WUA domain. The Balinese created

an organizational structure in which the pekaseh became “WUA co-ordinator” of

Kertoraharjo.11 According to the interviewed farmers, subak functionaries and a Public

Works functionary involved in those days, the subak-derived regulations were applied

quite successfully. TUs and WUAs showed a rapid development. Farmers invested capital

and labour in their TUs, and the WUA co-ordinator-pekaseh maintained regular contacts

with a functionary of the Irrigation Service. The former pekaseh explains:

To make WUAs function better, I wanted to use subak regulations for all activities in

WUAs with Balinese farmers. From the transplanting ritual, farmers were given two

weeks to plant. Late planters were fined, and the fine went to the WUA. We also had

a maintenance programme for canals and drains. Every season the farmers had to

clean part of the drain. After the plan had been approved by the Kertoraharjo village

40 D. Roth

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

administration, subak regulations were put on paper and applied to the TUs. We also

made improvements to the system, the water of which did not reach all farmers.

Funds were raised by collecting eight kilograms of unhulled rice from each farmer.

We have placed many culverts in the WUAs, where the constructors had not

provided crossings of canals, drains, and farm roads.

The plans were also supported by the Javanese branch head for the Kalaena right bank

branch of the Irrigation Service. Again, in the words of the former pekaseh:

I formally asked permission for all plans and changes in the TUs, for instance

placing culverts or improving a canal. He said, “it is all up to you down there what

you want to do with the TUs and the WUAs, as long as what you do makes the

organization function more smoothly without disturbing other farmers”. Indeed,

things ran smoothly in those days. There were regular contacts with the Irrigation

Service.

The branch head, who was actively involved in irrigation development and establishment

of WUAs, confirms this view:

The WUA co-ordinator was an initiative of the Kertoraharjo farmers, meant to make

the relationship between the Irrigation Service and local organizations more efficient.

The co-ordinator could create understanding about government policy among

farmers in their own language. The system functioned well under the co-ordinator.

If there was a meeting or other activity, he was always there, so that the relations with

the Irrigation Service were smooth. The WUAs in Kertoraharjo were among the

better functioning ones, with an active farmer population. If there were activities like

collective labour on canals, all Kertoraharjo farmers joined. As the WUAs were

involved, all farmers, including the Javanese, accepted the structure.

This happy marriage between subak and WUA was short-lived. After a conflict between

the WUA co-ordinator-pekaseh and the (then Javanese) administrative village head about

the financial resources controlled by the WUAs, subak regulations were confiscated and

application to the WUAs forbidden. The function of WUA co-ordinator was discontinued,

and the pekaseh-co-ordinator was summoned to the police to account for “undermining

WUA regulations”. After this incident, TUs and WUAs are said to have degraded.

Maintenance was neglected, infrastructure disappeared and regulations for water

distribution were no longer followed. Contacts between farmers (through their WUAs) and

the Irrigation Service diminished to the point where today there is no contact at all between

farmers and the agency.

Thus, the functional separation between subak and WUA was reasserted. The subak

regulations adapted for the WUAs disappeared after confiscation, and were gradually

replaced by new subak regulations specifically made for each of the four subaks and

restricted to issues for which the subaks were allowed to be responsible. The pekaseh was

no longer actively involved in tertiary irrigation management and, since then, the subaks

and the pekaseh have organized rice rituals, guarded ritual purity, collected the seasonal

subak tax (on a per hectare basis), determined the agricultural calendar (primarily the start

of transplanting) and provided cash loans to their members. However, as formal

Balinese Irrigation Management 41

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

organizations the subaks are no longer involved in “water issues”: the latter belong to the

WUA domain.

5. Different Conceptualizations of “Management” and Water as a Sanctioning

Instrument

The rigid separation between WUA and subak was based on diverging perceptions of

irrigated agriculture between government officials and the Balinese system users. In their

perception, classification of interrelated domains in terms of a distinction between

“management” and “religion” does not make sense. The way in which qualifications for

leadership functions in the subak domain are defined conveys the impression that the

spheres of WUA and subak are regarded as one “wet” sphere, requiring the same type of

knowledge, leadership qualities and characteristics. As the former pekaseh said about the

qualities of a subak leader or pekaseh:

He must be an all-round person. He should have knowledge of the weather, position

of the stars and the Balinese calendar, but also of leadership and administration,

water distribution, agricultural inputs and the characteristics of the rice varieties

planted. He should also know subak ceremonies. He should have field knowledge

and be diligent in keeping an eye on field conditions. He has to be conscious and

patient, a person who does not do damage to other persons and is capable of

composure. If it were about planting only, being pekaseh would be easy.

What does this separation mean for irrigation practices? According to many Balinese, the

separation of subaks/pekaseh and WUAs has had important consequences for both. WUAs

are weak organizations, lacking authority to enforce rules and sanction transgressions.

The small farmers’ groups that have emerged in several WUAs are more effective, and

the authority of their leaders more respected. The subaks have not fared much better.

Since the 1980s, Balinese landownership outside the initial subak areas has rapidly

expanded. This has led to problems of definition of Subak land and boundaries, and

contestation of Subak authority.

The segmentation of crucial linkages between subak functions has influenced irrigation

management. Engineers and officials, with their instrumental perception of irrigated

agriculture, classify this Balinese “wet” world into categories such as “water management”,

“agriculture” and “religion”. Thus, crucial relationships between interrelated dimensions

of subak have been lost, as the remarks by a farmer and subak functionary comparing the

Kertoraharjo subaks with those in his place of origin illustrate:

Why has irrigation management so deteriorated? The major cause is the separation

of subaks and WUAs, of water and collective labour for repair and maintenance

from the offerings. Take, for instance, mapag toya (ceremony for welcoming the

season’s first irrigation water). What does it mean here? It has lost its meaning

and importance for irrigation. In Bali, the canals are cleaned and repaired before the

agricultural season starts. Once the canals are ready for use, the ceremony is held

and the water can be used. In Bali, these are very important occasions. Usually

people are rather massively present. Here it is different: mapag toya belongs to the

subak, and canal cleaning is a task for the WUA. Often, when mapag toya is

42 D. Roth

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

held, the canals have not even been cleaned and repaired. If offerings are brought

to the places where water enters the subaks at all, hardly anybody is present. Often,

offerings are made in the temple only.12

A village elder made a similar point, with sharp observations on the situation in

Kertoraharjo:

The subak has only retained its ritual function. It does no longer have a relationship

to irrigation. The government determines the planting schedule, only slightly

adapted to the Balinese calendar by the subak. It also determines the system

schedule. WUA regulations also derive from the government. Because the TUs and

WUAs are not completely Balinese and others do not recognize subak, its

regulations cannot be applied in the WUAs. In Bali, farmers join subak on the basis

of their water source. There is a close connection between collective labour

for repairing and cleaning canals, and ritual activities accompanying each stage.

The subak regulates everything: water, planting, seed choice, labour and offerings.

As a consequence, work is always executed in time. All members are present and do

their share. Here, decision-making and tasks related to irrigation and planting have

moved to the government. What remains for the subaks is ritual. Those functions. . .

can also be executed by the customary village. If my proposal were accepted, subak

and pekaseh would actually no longer exist. An important subak task, irrigation, was

taken away from it long ago anyway. If the new system is accepted, the customary

village leader will collect tax for all irrigated land owned by Balinese, irrespective of

its location and origin. If we continue using the old subak system, the problems will

never be solved. Land outside the subak areas could never be taxed, and payment of

tax would be dependent on the willingness of owners to pay. There is another

important difference with Bali here: if somebody refuses to pay tax or perform

labour in Bali, he will be fined by the subak. If he does not pay, his sawah is given a

field mark and he will no longer receive water. Here, water cannot be used as a

sanctioning instrument. The subak has no control over water; authority over water

allocation is in the hands of the WUAs. WUAs have the right to close off people

from water, but the subak does not. Many people refuse to pay tax or do not even

want to become a subak member, but the subak cannot effectively force them to

observe all subak rules. In the banjar and customary village, people fear sanctions.

In the subak, there is no such thing. Like in the WUA, which is also not feared

because it has no effective sanctions against farmers who breach the rules.

In this view, the forced separation of “water” and “religion” has seriously weakened both

subaks and WUAs, with negative consequences for local irrigation management.

6. Which order? Whose order? Towards a Contextualized Understanding of LocalIrrigation Management

I have discussed local responses to state attempts to create managerial order in a public

irrigation system. Considerable diversity in irrigation management practices exists within

the seemingly uniform technical, legal and organizational order of this system. I have

identified specifically Balinese forms of management of irrigated agriculture, mainly with

Balinese Irrigation Management 43

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

the subak, as the main source of such diversity. New, context-specific articulations have

emerged between “engineering” approaches, and conceptualizations of rights, norms,

forms of regulation and organization, technology and practices associated with subak.

This difficult engagement between externally imposed arrangements and locally existing

and developing forms of knowledge, norms and rule systems, and organizing practices

produces tensions, contradictions and disjunctures rather than a predetermined regulatory

and managerial order.

I have given three examples. First, I have contrasted “external” explanations of

degradation in terms of farmers’ shortcomings leading to cycles of degradation and

rehabilitation with an alternative explanation. Balinese farmers do try to manage their

TUs. However, they do so by making use of specifically Balinese subak technology, norms

and rules, and organizational arrangements. Second, I have traced this history of

articulation of elements of subak and TU/WUA in the development of local irrigation

management. “Reinvented” in a non-Balinese environment, subaks were defined in terms

of land allocated to the settlers. When irrigation reached the village, the blueprinted

structure of the system demanded that farmers organized in WUAs. Gradually, the subaks

as organizations were marginalized; their role in irrigation management diminished.

Major functions of the current subaks and pekaseh are religious-ritual. However,

important domains of interaction with the broader field of irrigated agriculture continue to

exist: subak tax collection and determination of a transplanting date. Elements deriving

from the subak as institution have entered the domain of TU and WUA. Third, I have given

examples of the disjunctures created by the differing perceptions of irrigated agriculture

among external state agency officials and Balinese farmers. I have contrasted narrow

definitions of “irrigation management” with Balinese “persubakan”, which also includes

agricultural planning and ritual. Separation of these domains along the lines of WUAs and

subaks/pekaseh has weakened both.

Recent years have seen the emergence of a development policy interest in institutions,

the “rules-in-use” that guide human behaviour in institutional domains. Largely based on

neo-institutional economics, this interest in institutions derives its appeal mainly from the

basic assumptions and promises of manageability it holds. “Efficient” institutions can be

“crafted” by keeping transaction costs low and optimizing incentive structures (Ostrom,

1992; Schlager & Ostrom, 1992).

This social engineering approach has become rather hegemonic, often at the expense

of a more critical understanding of the complexities of intervention. I have shown here

how certain developmental norms about irrigation management have been translated

into highly routinized and blueprinted forms of irrigation development, in which farmer

adaptation is simply taken for granted. Rather than assuming that farmers will behave in

accordance with the externally introduced rules, I have started “on the ground” from

irrigation management practices that have developed among Balinese (Benda-

Beckmann et al., 1996; Spiertz, 2000). I have analysed TUs and WUAs in their

normative-legal, organizational and technical dimensions of regulation. Local irrigation

management is complex in all these dimensions. There is not just one orderly system of

rules, organizational arrangements and technology “in use”. There is uncertainty,

contestation of rules, interaction and tension between state and subak regulation, and

merging into new local forms of regulation. As to the relationship between regulation

and behaviour, it can be concluded that the Balinese farmers are guided in their

44 D. Roth

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

managerial practices by elements of the institution of subak rather than by those of

TU/WUA.

Understanding local irrigation management requires an analysis that takes into account

the locally specific and embedded character of resource use and management. This

requires some distancing from policy-driven agendas that treat institutions as

unproblematic and uncontested rule systems. The outcome of planned changes through

civil and socio-legal engineering is basically uncertain and contingent. It is only by

recognizing this fact of life that we can better understand the dynamics of development

interventions and the forms of order and disjuncture they create in real-life settings.

Notes

1 This chapter was inspired theoretically by two insights. First, “legal pluralism” or “legal complexity”

refers to the existence and interaction of different (often state and non-state) legal orders in the same

socio-political space (F. von Benda-Beckmann et al., 1996; Bruns, Meinzen-Dick, 2000; Meinzen-

Dick & Pradhan, 2001; Spiertz, 2000). Second, irrigation systems can be analysed as “socio-technical

systems”, complexes of physical-technical, organizational and normative-legal dimensions of water

control that develop in a wider agro-ecological, politico-economic and socio-cultural context

(Mollinga, 2003; Vincent, 2001).2 “Colonization” means pioneer land settlement, not colonial rule.3 State-sponsored transmigration resettled Javanese and Balinese farmers on islands with a relatively

low population density, such as Sumatra, Kalimantan and Sulawesi.4 In 2000, Public Works became the Ministry of Settlement and Regional Infrastructure (Oad, 2001).5 P3A (Perkumpulan Petani Pemakai Air; Water Users’ Association).6 The WUA legal framework generally contains: a national basic law (“enabling law”) authorizing

establishment of WUAs as legal entities and providing regulations pertaining to water fees, etc.;

“bylaws” for specific WUAs, containing data on these WUAs and further specifications of

organization, rights and obligations; and “transfer agreements” in which transfer is arranged in greater

detail (Salman, 1997).7 Also used for the higher level encompassing various subaks, and its head. See this case.8 Toraja: ethnic group from highland South Sulawesi. Many Toraja have migrated to Luwu in search of land.9 KK (Kepala Keluarga) means “household head”, and refers to the number of initial Balinese settler

families (totalling 350). The second group of 50 households (from Tampaksiring) is often referred to as

“Tampaksiring”, to distinguish it from the other group of 50 settler families.10 PERTANIAN and Panitia Irigasi, respectively.11 In Bali, a distinction is maintained between the “dry” (customary) village and the “wet” village (subak)

(Geertz, 1972). In government administration, the village head is responsible for WUAs.12 This is corroborated by my own observations: 1 week after the ceremony, cleaning was still going on.

References

Benda-Beckmann, F. von, von Benda-Beckmann, K. & Spiertz, J. (1996) Water rights and policy, in: J. Spiertz &

M. G. Wiber (Eds) The Role of Law in Natural Resource Management (’s Gravenhage: VUGA.).

Boelens, R. (1998) Collective management and social construction of peasant irrigation systems: a conceptual

introduction, in: R. Boelens & G. Davila (Eds) Searching for Equity. Conceptions of Justice and Equity in

Peasant Irrigation (Assen: Van Gorcum).

Booth, A. (1977a) Irrigation in Indonesia, part I, Bulletin of Indonesian Economic Studies, 13, pp. 33–74.

Booth, A. (1977b) Irrigation in Indonesia, part II, Bulletin of Indonesian Economic Studies, 13, pp. 45–77.

Bruns, B. (2004) From voice to empowerment: rerouting irrigation reform in Indonesia, in: P. Mollinga &

A. Bolding (Eds) The Politics of Irrigation Reform. Contested Policy Formulation and Implementation in

Asia, Africa and Latin America, Global Environmental Governance Series (Aldershot: Ashgate).

Bruns, B. & Meinzen-Dick, R. (Eds) (2000) Negotiating Water Rights (New Delhi: IFPRI).

Charras, M. (1982) De la Foret Malefique a l’Herbe Divine. La Transmigration en Indonesie: Les Balinais a

Sulawesi (Paris: Editions de la Maison des Sciences de l’Homme Paris).

Balinese Irrigation Management 45

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014

Geertz, C. (1972) The wet and the dry: traditional irrigation in Bali and Morocco, Human Ecology, 1, pp. 23–29.

Geertz, C. (1980) Organization of the Balinese subak, in: E. W. Coward Jr (Ed.) Irrigation and Agricultural

Development in Asia. Perspectives from the Social Sciences (Ithaca and London: Cornell University Press).

Gelles, P. (1998) Competing cultural logics: state and “indigenous” models in conflict, in: R. Boelens & G. Davila

(Eds) Searching for Equity. Conceptions of Justice and Equity in Peasant Irrigation (Assen: Van Gorcum).

Guermonprez, J. F. (1990) On the elusive Balinese village; hierarchy and values versus political models, Review

of Indonesian and Malaysian Affairs, 24, pp. 55–89.

Horst, L. (1996) Intervention in irrigation water division in Bali, Indonesia. A case of farmers’ circumvention of

modern technology, in: G. Diemer & F. Huibers (Eds) Crops, People and Irrigation. Water Allocation

Practices of Farmers and Engineers (London: Intermediate Technology Publications).

Howe, L. (1991) Rice, ideology, and the legitimation of hierarchy in Bali, Man (N.S.), 26, pp. 445–467.

Jha, N. (2002) The bifurcate subak: the social organisation of a Balinese irrigation community, Dissertation,

Brandeis University.

Leach, M., Mearns, R. & Scoones, I. (1999) Environmental entitlements: dynamics and institutions in

community-based natural resource management, World Development, 27, pp. 225–247.

Meinzen-Dick, R. S. & Pradhan, R. (2001) Implications of legal pluralism for natural resource management, IDS

Bulletin, 32, pp. 10–17.

Mollinga, P. P. (2003) On the Waterfront: Water Distribution, Technology and Agrarian Change in a South

Indian Canal Irrigation System (New Delhi: Orient Longman).

Mosse, D. (1997) The symbolic making of a common property resource: history, ecology and locality in a tank-

irrigated landscape in South India, Development and Change, 28, pp. 467–504.

Oad, R. (2001) Policy reforms for sustainable irrigation management—a case study of Indonesia, Irrigation and

Drainage, 50, pp. 279–294.

Ostrom, E. (1992) Crafting Institutions for Self-governing Irrigation Systems (San Francisco, CA: ICS Press).

Salman, M. A. (1997) The Legal Framework for Water Users’ Associations: A Comparative Study, World Bank

Technical Paper 360 (Washington, DC: World Bank).

Schlager, E. & Ostrom, E. (1992) Property-rights regimes and natural resources: a conceptual analysis, Land

Economics, 68, pp. 249–262.

Spiertz, H. L. J. (2000) Water rights and legal pluralism: some basics of a legal anthropological approach,

in: B. R. Bruns & R. S. Meinzen-Dick (Eds) Negotiating Water Rights (New Delhi: IFPRI).

Sutawan, N. (1987) Farmer-managed irrigation systems and the impact of government assistance: a note from

Bali, Indonesia, in: Public Intervention in Farmer-managed Irrigation Systems (Digana Village, Sri Lanka:

International Irrigation Management Institute).

Vincent, L. (2001) Struggles at the social interface: developing sociotechnical research in irrigation and water

management, in: P. Hebinck & G. Verschoor (Eds) Resonances and Dissonances in Development. Actors,

Networks and Cultural Repertoires (Assen: Van Gorcum).

Warren, C. (1993) Adat and Dinas. Balinese Communities in the Indonesian State (Kuala Lumpur: Oxford

University Press).

46 D. Roth

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

08 2

6 O

ctob

er 2

014