Engaging stakeholders in assessing and addressing their own needs


Material Information

Engaging stakeholders in assessing and addressing their own needs planning and implementing participatory agricultural development in St. Lucia and Dominica
Physical Description:
x, 254 leaves : ill. ; 29 cm.
Sullivan, Amy Jean
Publication Date:


Subjects / Keywords:
Interdisciplinary Ecology thesis, Ph. D   ( lcsh )
Dissertations, Academic -- Interdisciplinary Ecology -- UF   ( lcsh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 2004.
Includes bibliographical references.
Statement of Responsibility:
by Amy Jean Sullivan.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 003103086
System ID:

Full Text








I would like to express my appreciation and thanks to the members of my

supervisory committee, Dr. Peter Hildebrand, chair, Dr. Anita Spring, Dr. Sandra

Russo, and Dr. Tracy Irani for their encouragement and guidance in this project.

This work would not have been possible without all of the people encountered

along the way. First and foremost I recognize my families and the communities to

which I belong. The sense of security I derive from them has allowed me the freedom

to continue exploring the world and my place in it.

Learning remains a team effort. My respect and gratitude go to my mentors,

colleagues, and friends at the University of Florida for allowing me to join their team.

Dr. Peter Hildebrand provided the continued guidance and support one can expect from

a true educator. Dr. Anita Spring made this study possible by giving me entr6e into

another side of international development. Tracy Irani exhibited the patience and

curiosity needed to turn my experiences into a work that I hope she is proud of. Dr.

Sandra Russo encouraged combining academic training and real-world experience by

placing local events on the global stage.

My colleagues from across the university who have pushed, pulled, led and

been led, I salute you and would not have enjoyed the journey nearly as much without

you, had I chosen to make it at all. In particular, Dr. Christina Gladwin, Dr. Heather

McIlvaine-Newsad, Dr. Elena Bastidas, Dr. Hannah Carter and Jim Barham, have

generously given their time, encouragement, support, and most importantly friendship

throughout my time at the University of Florida.

Special acknowledgement is due the individuals in the institutions who

combined to deliver the best development projects we knew how. In particular, the

farmers, front-line extension officers, and friends in St. Lucia and Dominica who do the

hard work, and for whom 'getting it right' is more than an academic exercise.



ACKNOWLEDGMENTS ...................................................................... ii

LIST OF TABLES ............................................................................... vii

LIST OF FIGURES ............................................................................V..iii



1 INTRODUCTION ........................................................................ 1

Participation in Development ......................................................... 5
Need for the Study ...................................................................... 6
Building a Framework for Examining Context and Process in
Participatory Agricultural Development....................................10
Research Setting .......................................................................14
The Intervention........................................................................15
Statement of the Problem ..............................................................16
Research Questions ................................................................... 17
Sum m ary ............................... ...................... .......... ................ 18

2 LITERATURE REVIEW..............................................................20

Political Ecology ...................................................................... 21
Gendered Political Ecology & MERGE ............................................25
Participation in Development .........................................................32
Agricultural Extension............................................................... 55
PRA................................................................................... 58
Summary ............................................................................. 60


3 METHODOLOGY....................................................................62

Qualitative Research ..................................................................63
Feminist Research ..................................................................... 65
Research Questions ....................................................................68
The Data .............................................................................. 69
Procedures ............................................................................ 70
Gendered Political Ecology Framework ............................................73
Summary ............................................................................. 78


St. Lucia and Dominica ................................................................80
British and French ..................................................................... 81
The Trap of Monocrops ...............................................................85
Settlement, Land Use and Tenure Patterns ........................................96
Nationalism, TNCs and Banana Wars ............................................ 107
International Response ...............................................................120
Similarities and Differences ........................................................ 122
Sum m ary ............................................. ................ ....... ....... ... 124


Project Plans for Incorporating Participation in Extension Program
Planning in St. Lucia and Dominica ..................................... 127
Stakeholders ........................................................................... 132
PRA in St. Lucia ......................................................................137
PRA in Dominica .................................................................... 160
PRA Findings: Gendered Resource Use........................................... 176
Summary ............................................................................ 189

6 FIN D IN G S ............................................................................. 191

Findings ............................................................................. 192
Sum m ary ................ .......... ....... ............................. ........... 217

7 CONCLUSIONS AND RECOMMENDATIONS ................................219

Project D esign...........................................................................220
Project Implementation...............................................................221

APPENDIX ......................................................................................233

B FAO SCHEDULE OF ACTIVITIES ................................................237
C REGIONAL REPORTING OUTLINES............................................240

LIST OF REFERENCES .....................................................................242

BIOGRAPHICAL SKETCH .................................................................254

Table page

2-1 White's framework ..........................................................................53

2-2 RRA and PRA compared ....................................................................59

3-1 Adapted MERGE framework, section I...................................................74

3-2 Adapted MERGE framework, section II .................................................76

3-3 Process analysis framework for participatory development
projects in St. Lucia and Dominica ....................................................77

4-1 Dominant land tenure types in St. Lucia and Dominica in 1995......................105

4-2 Characteristics of the Windward Banana Industry, 1992 .............................115

4-3 National economic indicators for 1995 ..................................................117

5-1 Staff participating in daily PRA activities by gender ..................................135

5-2 Livelihood production system type by household head in St. Lucia PRA ......... 179

5-3 Production activities and market arrangements in Dominica ........................ 181

5-4 Number, type, and composition of focus groups in St. Lucia ....................... 182

5-5 Constraints identified and prioritized by root crop focus group, St. Lucia ........ 183

5-6 Number, type, and composition of focus groups in Dominica .......................183

5-7 Frequency of constraints by major enterprise in Dominica .......................... 184

5-8 Input constraints and causes from farmers' perspective-Dominica ................. 186

Figure page

2-1 MERGE conceptual framework ..........................................................31

5-1 Relationship between government and farmers ........................................ 134

5-2 Timeline of activities for PRA in St. Lucia .............................................139

5-3 Toolbox delivered during training in St. Lucia .........................................141

5-4 Timeline of activities for PRA in Dominica ............................................162

6-1 Relationships among stakeholders in PRAs ........................................... 192

6-2 Relationships among direct stakeholders in training programs....................... 206

6-3 Relative stakeholder influence on project team decision making in .................211
St. Lucia and Dominica

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Amy Jean Sullivan

August 2004

Chair: Peter E. Hildebrand
Major Department: School of Natural Resources and Environment

In this dissertation, concepts and frameworks from participatory development and

gendered political ecology were used in combination with qualitative research methods

to explore factors affecting design and delivery of participatory agricultural

development projects in St. Lucia and Dominica. In order to understand the influence

of site-specific context on project planning and implementation, this study examined

the gendered political ecologies of the project settings as well as decision-making and

interactions within and among stakeholder groups.

An existing framework was used to examine how the combination of historical,

political, socioeconomic, cultural and ecological factors enhances or constrains the

potential for participatory development. Tracing the evolution of each of these factors

from European settlement of St. Lucia and Dominica through the spread of

globalization in the 1990s set the stage for the development projects analyzed in this


A project analysis framework was developed for organizing data and analyzing

decision-making within each project according to specific phases they had in common.

Data were analyzed to determine who made decisions, and how these decisions

impacted stakeholders in each project including farmers, extension officers,

government officials, representatives from the sponsoring agency, and consultants,

including the researcher. Following an overall assessment of both projects, results from

specific phases of each were compared to determine if and how different approaches to

project planning and implementation affected decision-making and stakeholder


Results suggested that project design for incorporating stakeholder participation

in agricultural development must acknowledge the limitations that affect the livelihood

choices made by target audiences. These limitations include those externally imposed

such as trade restrictions and unsuitability of foreign sourced agricultural inputs, as well

as domestic issues such as changing consumer preferences reducing demand for locally

produced crops, and unpredictable labor availability. Results also showed that failure

to explore and acknowledge stakeholder expectations, at all levels of decision-making,

can lead to misunderstanding, misuse of scarce development resources and conflict

between stakeholders. The combination of frameworks used in this interdisciplinary

study will enable policy makers, practitioners, and others with an interest in

participatory development to plan or evaluate projects that acknowledge the

interactions between context and process.


Can the world be redefined and reconstructed from the perspective of the multiple
cultural and ecological practices that continue to exist among many communities?
This is above all a political question, but one that entails serious epistemological,
cultural, and ecological considerations. Escobar, 1998, p.76.

Farmers in developing nations like St. Lucia and Dominica carve out a living

within a mosaic of biophysical and socioeconomic forces, including environments,

bureaucracies, and markets. These farmers choose among livelihood activities and

strategies heavily influenced by exogenous factors as illustrated in their historical and

present position as producers within an increasingly globalized system. The reality of

globalization has recently increased the challenges facing farmers in St. Lucia and

Dominica by inducing them to compete in the global marketplace. This dissertation

examines participatory agricultural development projects designed to help farmers face

these challenges in both countries, and questions how effectively participatory

approaches address these challenges.

After three hundred years of colonization and nearly fifty years of aid from

foreign governments, international organizations, and institutions, there are problems

and symptoms of underdevelopment-poverty, unemployment, emigration,

environmental degradation, and limited economic growth. Current thinking in

development suggests that simply integrating farmers into the global economy will


improve their situation. However, as Trouillot points out about Dominica, and which

can be said of St. Lucia, these Caribbean island nations are not "underdeveloped" due

to reliance on subsistence agriculture.

On the contrary: its present condition, including its low material standard
of living, is a byproduct of its forced integration within the world
economy. Hence, one is likely to wait in vain for the island to take off on
the capitalist path: it has been on that path for more than two centuries.
(Trouillot, 1988, p. 286)

Harcourt defines development as "a collection of theories (development as

economic growth), institutions (Bretton Woods, NGOs [Non-Governmental

Organizations]), practices (projects, community, societies), and procedures

(programmes, plans) through which individuals and communities are constituted"

(1994, p. 22). While this definition indicates elements of the process of development, it

does not illuminate the dynamic nature of the global economic web that sees the rich

getting richer, and the poor getting poorer, while there is large-scale ecological

degradation. The persistence of these conditions has led researchers and practitioners

to question theories, approaches, tools, and techniques for reaching target audiences of


Early approaches to reaching target audiences were based upon transmission of

information and technologies (Rogers, 1995). Rural communities and specific

populations (farmers) were seen as backward and in need of an external technological

fix. This model was top--down with policy makers establishing targets, researchers

identifying problems and developing technical solutions, and field level extension

agents delivering them to farmers for adoption (Pickering, 1987; Rogers, 1995). The

linear, one-way, relationship between policy makers, researchers, extension, and

farmers left little opportunity for farmers to influence policies, research priorities, and

technology delivery.

Attempts to change the top-down nature of research and extension in developing

countries spawned a number of approaches including Farming Systems Research and

Extension (FSRE), Training and Visit Extension (T & V), Farmer First (FF) and

Beyond Farmer First (BFF). These approaches altered the nature of research and

extension by changing farmers' position and role in the process, from recipient of

information to informant, collaborator, and eventually researcher (Chambers, Pacey and

Thrupp, 1989; Collinson, 2000; Waugh, Hildebrand and Andrew, 1989). They also

affected the roles of front-line extension staff and their role as the outreach arm of

ministries of agriculture. Traditional approaches to development, research, and

extension had policy makers and researchers setting targets, designing technologies,

and passing them to extension personnel for transmission to farmers.

The emergence of these more participatory approaches coincided with the

recognition of women's roles in agriculture in developing countries (see a review in

Chapter 2). Central to these approaches was a fundamental change in the role of

farmers-their participation in their own development. Ignited in the field,

participation has become a common target for development (Chambers, 1997;

Collinson, 2000; Feldstein and Poats, 1989; Hayami and Ruttan, 1985).

This research explores factors affecting the design, delivery, and outputs of two

participatory agricultural development projects meant to incorporate farmers'

perspectives into extension program planning in St. Lucia and Dominica. The

implementing agency, the Food and Agriculture Organization of the United Nations,

(FAO), planned the projects to improve service to farmers by bringing them into the

processes of assessing and addressing their needs. Front-line extension officers (EOs)

were responsible for facilitating this participation and incorporating the results into

their strategies for helping farmers. FAO sought to increase farmer participation with

extension in terms of problem identification and formulation of solutions.

In contemplating the research for this dissertation, it became apparent that due to

a confluence of factors, analysis of project activities without acknowledgement of

contextual issues was inadequate. Therefore, an approach was developed that included

a multi-disciplinary understanding of the study areas for recognizing, contextualizing,

and analyzing as many of those factors and their interrelatedness as possible. The result

was a framework that accommodated analysis of processes of project designs and

delivery, analysis of stakeholders and their participation in different phases of the

projects, and analysis of context as framed by a gendered political ecology. In the

global picture within which they operate, this includes an explication of stakeholder

expectations from the projects and stakeholder participation in decision-making. It also

includes recognition of factors influencing farmers' use of natural resources to sustain

themselves and their families within particularly complex social and ecological


Much of the literature on participation in development offers tools to use in the

field and techniques for getting target audiences to participate. Few scholars explain

how particular approaches relate or lead to particular outcomes. Fewer suggest how to

plan for, incorporate, or operationalize participation within site-specific conditions.

This study provides links among (1) theoretical approaches to participation; (2)

situation-specific economic, political, social, and ecological conditions; and (3)

institutional, practical, and procedural aspects of designing and implementing

participatory development projects.

Participation in Development

Although used since the days of Freire (1970), stakeholder participation was not

emphasized by mainstream development agencies such as FAO, World Bank, and

United States Agency for International Development (USAID) until the early 1990s.

Since then, they have encouraged greater involvement or participation of stakeholders,

namely those historically treated as recipients of foreign aid, who were mainly small-

scale, limited-resource populations (Blackburn and Holland, 1998; Chambers, 1998;

Slocum and Thomas-Slayter, 1995). Recipients' participation was expected to

incorporate their perspectives into creating solutions to their problems. Involving more

stakeholders earlier in the process was expected to improve development by generating

accurate situational analyses, identifying appropriate target audiences, delivering better

technologies, providing more sustainable results, and utilizing scarce resources more

efficiently (Chambers, 1997; Clayton, Oakley and Pratt, 1998; Guijt and Shah, 1998;

Karl, 2000).

This trend toward 'participatory development' signaled a major shift as it implied

a change in the nature of the relationship between donors and recipients. Top-down

models of development gave way to more consultative or collaborative approaches to

identifying and solving problems where roles and responsibilities could shift according

to the needs of expanded lists of stakeholders (Cornwall, 2003; Pritchett and Woolcock,


Given the persistence of rural poverty in developing countries, a great deal of

attention was focused on incorporating farmers' perspectives in the design and delivery

of development projects. However, environmental degradation, free trade, and market

liberalization had to be considered as additional constraints facing actors (large and

small producers, extension staff, and ministries, among others). This left farmers and

those institutions designed to assist them with little margin for error.

In reality, farmers in developing countries cover the spectrum from highly

diversified subsistence farmers who survive on the food they grow, to highly

specialized, mechanized, technology-rich growers. The latter have the ability and

resources to change enterprises quickly in response to national policy or market

changes. Yet until relatively recently, farmers across the spectrum were all subject to

the same top-down development model and (one size fits all) approach (Pickering,

1987; Rogers, 1995).1

Need for the Study

Integrating participation in agricultural development projects continues to

challenge donors, program planners, and practitioners. Participation, a concept that at

first glance seems straightforward, quickly becomes complicated when applied to the

entire breadth or length of any site-specific effort at planned change. For instance, who

participates, what are indicators of participation, or what indicators can be used to

measure, monitor, and evaluate participation? A common one is number of

participants, yet the number of people involved gives no indication of how they

See, for example, the failure of the Green Revolution to improve the situation of households without
access to resources such as fertilizer, herbicides, insecticides, and/or irrigation technology. Green
Revolution technologies were designed for and adopted by those resource-endowed farmers who could
fit the new production regimes within their existing schemes (Hayami and Ruttan, 1985).

participated. How people participate, perhaps indicated by a list of their activities or

their point of entry in a project, gives no indication of why they participated.

Identifying people's activities within a project gives no indication of those with whom

they collaborated or conflicted and why, and so on. All of these together may not

predict outcomes, success, or sustainability.

While donors, policy makers, researchers, and practitioners continue their search

for more effective ways to incorporate participation into development (Chambers,

1998; Pretty, 1995) and develop resources (manuals of tools, techniques, and methods)

that explain what participation looks like on the ground, little is understood about the

interactions between all players in the process. These interactions, however derived

and driven, embody differences in perceptions, decision-making authority, and mode of

transmitting and receiving information. Therefore, existing processes need to be

examined to determine how best to incorporate these differences.

As currently perceived, participation is focused almost exclusively on the

situation of those considered beneficiaries. According to Spring, "beneficiaries are

specific groups within an implementation area whose access to and control over

resources ... is actually increased or enhanced" (1993, p. 61). This target audience

should be distinguished from participants "who allocate time and/or resources to a

project on one or more occasions over its life cycle" (Spring, 1993, p. 61). This

distinction deserves exploration in participatory development projects because

beneficiaries, target audiences, and participants do not always imply the same people

with the same stake or interest in outcomes.

Agricultural development remains a needs-based (versus assets-based)

proposition, with participation seen as a means to address a particular symptom of

populations in the developing world. Donors encourage planners to include an

endogenous perspective early enough in the process to reflect local contexts and needs.

Although important, the focus on local actors limits efforts at sustainable agricultural

development because it fails to address the heterogeneity of local actors, and their

engagement or lack thereof, with other stakeholders in the process (Guijt and Shah,

1998). And until recently, minimal attention was paid to the connection between

participants and their ecological, social, political, or economic environments (Martin &

Sutherland, 2003; Rocheleau, Thomas-Slayter, and Wangari, 1996; Schmink, 1999).

As currently perceived, participation puts the burden of change on local actors without

ensuring that other stakeholder groups are willing or able to facilitate this change.

Therefore, participation must be understood and explored as a series of context-specific

interactions or processes. Within such interactions, participation is affected by power

dynamics and patterns of inclusion and exclusion illustrated by communication and

decision-making within and between stakeholder groups.

Despite the recent premium put on stakeholder participation in development,

much uncertainty remains around the term itself, its potential (why), implementation

(how), and implications (outcomes). In Paradoxes of Participation, Frances Cleaver

claims that participation has become "an act of faith," with little evidence to back up its

claims of improving livelihoods or inducing social change (1999, p. 597). Her critique

of participation covers the spectrum from who participates, (how, when, where, and

why) to who drives the process. She challenges the widely held belief that participation

by recipients of aid naturally leads to their empowerment, and implies that the term

empowerment has been co-opted or watered down by its association with participation.

Her argument suggests that true empowerment means individual as well as class action

with the goal of structural change, reconnecting empowerment with its inherently

political nature. However, after years of participation in development there are no new

peasant governments or few new or revised institutions whose priorities have been

changed to address the needs of the poor as a direct result of their participation

(Cleaver, 1999).

This critique suggests issues in need of clarification in participation and

development. It is impossible to address all of these issues at once; therefore this study

is directed toward a final point raised by Cleaver who argued that there was a need for

"analysis of 'component' communities and 'successful' participatory projects that focus

on process, on power dynamics, on patterns of inclusion and exclusion. This would

involve more process documentation and analysis of conflict, consensus building and

decision-making within communities; not just those activities related to the particular

development project at hand" (p. 609). To this critique I would add that further

analysis of these component communities must expand to include the larger, multi-

faceted realities in which they survive.

Based on the above, this study examines initial components of two specific

development projects, one in St. Lucia and one in Dominica, designed to increase

farmer participation in extension program planning and delivery. The projects analyzed

in this study were meant to change the nature of the relationship between farmers and

EOs. After examining historical economic, political, social, and ecological issues, this

study explores stakeholder participation at each step in the projects in order to assess

the appropriateness of the participation in that situation. The design and delivery of

both projects will be analyzed (using a qualitative, inductive approach) to illustrate

differences in decision-making and participation within and between stakeholder


The stakeholders involved in design and delivery of the projects were farmers,

front-line EOs, consultants, ministry officials, and representatives of the sponsoring

agency. Besides identifying them, it is necessary to situate them within the larger

political, economic, ecological, and social processes that shape conditions within these

countries (Rocheleau, Thomas-Slayter, and Wangari, 1996; Schmink and Wood, 1987).

Hence, exploration of these conditions is a part of this study.

Building a Framework for Examining Context and Process in Participatory
Agricultural Development

The theoretical framework for this study combines concepts from literature on (1)

political ecology and (2) participation in development. Political ecology is a necessary

element of this framework because it incorporates past and present relationships

between local cultural dynamics, relations of international exchange, political economy,

and the environment (Greenberg and Park, 1994; Schmink and Wood, 1987). The

political ecologies of St. Lucia and Dominica set the stage, or context, for the

development projects analyzed in this study. Participation in development literature

contributes to this framework by serving as a microscope for examining observable

processes between individuals and groups nested within a political ecology framework

(Harcourt, 1994; McDougall & Braun, 2003). The brief introduction to key issues

presented in this section is further developed in following chapters.


Political ecology is a field of study emphasizing "decision-making processes and

the social, political, and economic context that shapes environmental policies and

practices" (Rocheleau, Thomas-Slayter, and Wangari, 1996, p. 4). It expands

ecological concepts to encompass cultural and political activity "within an analysis of

ecosystems that are significantly but not always entirely socially constructed"

(Greenberg and Hank, 1994, p. 1). Political ecology explicitly ties together

sociopolitical and ecological aspects that set the stage for development. Martinez-Alier

supports this point by suggesting "the struggle for survival brings the poor to defend

access to resources, and sometimes, their conservation; therefore the ecology of the

poor has been very present, both historically and now" (1991, p. 623). Political

ecology allows for both the social and ecological questions commonly addressed by

development to be examined in relation to each other.

In expanding the scope of inquiry around human/ecological interactions, early

political ecologists made no mention of gender. Yet, ecological, economic, and

political changes at the international, national, and local levels are known to have

differential effects on women and men. Subsequent literature has incorporated both

feminism and gender in political ecology to reflect women's and men's differential

access to, relationships with, perceptions and use of natural resources and the

environment (McIlvaine-Newsad, 2000; Rocheleau et al., 1996; Schmink, 1999). By

including a gendered political ecology perspective, this study includes a focus on how

the politics of power in participation in development impact women and men.

Participation in development can be examined from multiple levels and

perspectives. Each stakeholder group-policy makers, donors, researchers,

practitioners, and subjects-conceives of participation differently depending on their

experience and expectations. Feminist theory plays a role in understanding

complexities of participation. Starting with a basic assumption of unequal access to

power, emerging feminist theory acknowledges differences based not only on gender,

but also on age, ethnicity, socioeconomic status, and culture (Connelly, Li, MacDonald

and Parpart, 2000; Cornwall, 2003; Harcourt, 1994; Rocheleau et al., 1996). This

departure from androcentric and ethnocentric theories accommodates more possible

explanations for the persistence and expansion of participation as a factor in


One existing framework for examining and presenting gender implications within

political ecology has been adapted for the purpose of this study. The Managing

Ecosystems and Resources with Gender Emphasis (MERGE) framework takes a

gendered political ecology approach (Rocheleau et al., 1996) to identify factors that

influence natural resource use and interactions between them (Schmink, 1999). In

keeping with principles of political ecology, the framework supports exploration of

interactions between socioeconomic, political, and ecological factors affecting natural

resource use, while adding to the equation gender differentiated relations to these


The MERGE framework is presented as a practical way of organizing and

exploring both the contextual and procedural aspects of conservation project planning

and delivery. The multiple scales accounted for by this framework allow researchers to

2 Androcentrism is "A term developed by feminist theorists to describe the dominant worldview that,
until recently, mostly excluded the experiences of women from its analysis. This term also refers to an
approach taken to knowledge and the production of knowledge." Ethnocentric refers to "Believing that
one's race, nation, or culture is superior to all others" (Parpart, et al., 2000, Appendix 1, Key Concepts).

incorporate levels of activity and decision-making frequently excluded from project

planning frameworks. By combining political ecology, and macro and micro-level

stakeholder analysis, as well as gender analysis, this framework includes the range of

factors that confront successful design and delivery of development projects. The

MERGE framework was developed to examine inputs and outputs of community-based

conservation efforts. Although designed for use in conservation projects, the

framework proves flexible enough to be applied to development projects.

The second part of the theoretical framework for this study concerns the rationale

behind incorporating farmer participation in development projects. According to

Norman, farmer-centered approaches to agricultural development began in the 1970s as

a response to the failure of conventional development efforts to reach those facing

severe economic and environmental constraints (1980). As a result, agricultural

research and development became more site-specific in terms of resource availability

(Collinson, 2000; Feldstein and Poats, 1989; Hildebrand, 1986; Norman, 1980). This

specificity--or tailoring--depended on farmer input, which took them from recipients

of technologies to participants in the process (Feldstein and Poats, 1989). Chapter 2

highlights recent synthesis and justification for farmer-centered development.

Concepts and ideas from political ecology and participation in development

guided the development of this study. Given the range of factors affecting the study, a

gendered political ecology framework emerged as a logical approach to organizing,

analyzing, and presenting the data available. Gendered political ecology and

participation in development are discussed in Chapter 2.

Research Setting

St. Lucia and Dominica, like many developing nations, suffer the consequences

of market liberalization and globalization associated with structural adjustment in the

1980s and 90s. The rules are now made by The International Monetary Fund (IMF),

World Bank, World Trade Organization (WTO), European Union, and United States,

among others. As a result, developing countries are susceptible to the influence of

outsiders in matters such as local government structures and policy initiatives. While

not mandatory, their participation in the global economy is tied to continued

development assistance on which they rely.

Due to historical factors (presented in Chapter 4), St. Lucia and Dominica were

particularly susceptible to the rising tide of globalization and its related goal of market

liberalization. Both countries relied heavily on monocrop export agriculture-

bananas-for employment and export earnings. A 1992 WTO ruling against a long-

standing protected market system between Britain and several of its former colonies,

including St. Lucia and Dominica, initiated the 'banana wars.' This trade dispute

(expanded in Chapter 4) had significant influence on rural livelihoods in these two

countries and their economies in general. Historical factors leading up to the ruling, the

ruling itself, and resulting changes have created serious economic hardships that

exemplify the position of St. Lucia and Dominica relative to other countries,

transnational corporations, and governing bodies.

The WTO ruling against the favorable market arrangement between Britain and

its former colonies threatened livelihood security of farmers and their families in St.

Lucia and Dominica. As a result of WTO forced market liberalization, St. Lucia and

Dominica needed development aid to combat rising unemployment, out migration, and

economic recession brought on by declining banana revenue. Currently, as these

farmers struggle to make ends meet, their governments and international development

agencies have little choice but to step in and help them find ways to survive. As Dujon

suggested, "the economy stands in transition to a crisis" (1997, p. 1538).

The Intervention

As the banana wars raged in foreign courtrooms in the mid-1990s, farm

households in Eastern Caribbean countries like St. Lucia and Dominica lost a main

source of their livelihoods (Bryan, 1997; Lewis, 2000; Sutton, 1997). The price per

pound of bananas declined to the point where many producers were losing money but

continued to produce bananas to cover household expenses.

Farmers needed assistance, and some was provided by FAO in the form of short-

term projects designed to strengthen delivery of agricultural extension to farmers. The

projects, "Strengthening Extension Delivery Services," sought to help ministries of

agriculture and farmers collaborate more productively by introducing participatory

methods and approaches. The rationale for the projects suggested that national

agricultural ministries could serve their clientele (primarily farmers) better by

increasing their understanding of farmers' realities (FAO, 1999b, 2001).

The FAO projects sought to re-orient national policy-making bodies (Ministries

of Agriculture) and their outreach arms (Extension divisions) from traditional top-down

approaches to supplying agricultural technologies to a more client-centered, needs-

based approach (FAO, 1999b, 2001). In this new approach, front-line staff would

dialogue with farmers, gain an understanding of the conditions they faced, and work


with them to develop appropriate strategies for addressing existing challenges-namely

limited opportunities for agricultural livelihood activities. There were many intended

outcomes of the projects, including strategies for each ministry, extension division,

front-line officers, and farmers to address existing situations (FAO, 2000, 2002).

Each FAO project was 18 months in duration and involved a series of component

projects designed to address issues ranging from computer training to agricultural

record-keeping and farmer group formation and function (see Appendix B for

examples). Each component of the 18-month long projects was targeted toward

building capacity of extension personnel whose role was to incorporate these new skills

into their work with farmers. The focus of this study is the initial component of the

FAO projects in St. Lucia and Dominica, referred to as Participatory Rural Appraisals

(PRAs) or "the projectss)" The PRAs were designed to provide baseline data for many

of the subsequent project components.

FAO made farmer participation in the projects a priority and suggested that

extension delivery, and therefore agricultural sector performance, could be improved by

giving greater weight to farmers' perspectives than had been done in the past. FAO's

interest in hearing farmers' voices included those not typically reached by extension.

As such, efforts were made to include a diverse audience of farmers across ranges of

socioeconomic status, agro-ecological zones, scales of production, age, ethnicity, and


Statement of the Problem

While designed to address local conditions, participatory agricultural

development efforts must acknowledge the interrelatedness of actors with each other

and their larger environments. Endogenous and exogenous factors not only set the

stage by contributing to the situation, they also influence possible solutions. Although

participatory agricultural development has become 'right and good,' focus on intended

beneficiaries-in isolation-may fail to address (1) the heterogeneity of these

stakeholders; (2) their engagement, or lack thereof, with other stakeholders in the

process; and (3) connections between stakeholders and their greater gendered political


The participatory agricultural development projects that are the focus of this

study were designed to increase livelihood security of farmers in St. Lucia and

Dominica by incorporating their participation into extension program development and

delivery. This approach suggested that livelihood insecurity could be addressed by

changing relationships between farmers and extension, and that together, these two

groups of stakeholders had the power to address the situation.

Research Questions

Concepts and theories from gendered political ecology as presented in the

MERGE framework, and participation in development guided the development of the

research questions stated in this study. Based on these concepts, gendered political

ecology provides a framework for the confluence of factors that created the present

situation while participation refers to the process-the series of activities or roles-

undertaken by stakeholders.

This study addresses the following research questions:

1. How did the confluence of political, economic, social, and ecological factors in
St. Lucia and Dominica contribute to the need for intervention?

2. Who were the stakeholders in each project and how did they participate in the
different phases of each project (project preparation and planning, training,
fieldwork, synthesis, analysis, and reporting of results)?

3. How did interactions within and between several stakeholder groups affect
outcomes of each of the participatory development projects in this study?

Objectives of this study are as follows:

1: Conduct a gendered political ecology analysis of St. Lucia and Dominica.
Identify major themes and associate their historical emergence with the current
2: Identify key stakeholders in each of the study projects including their stake
and expectations of the projects.
3: Identify and document decision-making processes undertaken in preparing,
implementing, and reporting on each phase of the two study projects. Indicate
how stakeholders did or did not participate in decision-making, including who
participated, how, when, and who facilitated their participation.
4: Determine what led to conflict and/or collaboration in each project, how it was
managed, and how that impacted outcomes.


A detailed accounting of relationships among participation, power, and gendered

political ecology contribute to understanding decision-making, consensus building, and

meeting expectations in participatory development projects. It builds on existing

political ecology and development theory, and sheds light on the importance of

designing and following a process in addition to using appropriate tools. This study

bridges a gap in the literature by examining processes involved in designing,

developing, and implementing participatory development projects within a political

ecology framework.

This dissertation is structured as follows: Chapter 2 provides a review of the

literature on two main themes: (1) gendered political ecology as an appropriate

theoretical framework for this study and (2) participation in development. Chapter 3


presents the qualitative methodology, informed by feminist theory, used to design and

conduct this study. Chapter 4 presents the two study sites within the historical and

ecological perspectives using gendered political ecology as suggested by the MERGE

framework. In addition to setting the stage for the study, Chapter 4 provides the first

step in analysis of factors affecting the projects. Chapter 5 presents project-specific

data analysis, following the MERGE framework beginning with local-level stakeholder

analysis and continuing with project analysis and gendered resource use in St. Lucia

and Dominica. Chapter 6 presents findings, and Chapter 7 presents conclusions and

recommendations of the study.


This chapter reviews literature related to gendered political ecology and

participation in development, that together frame this study. An introduction is

followed by, (1) an overview of streams of thought contributing to political ecology; (2)

gendered political ecology and the MERGE framework (Schmink, 1999); and (3)

background on the emergence, evolution, and current issues pertaining to participation

in development. The section on participation concludes with a review of agricultural

extension as a vehicle for fostering farmer participation, and Participatory Rural

Appraisals as a methodology for doing so.

While the focus of this study is on decision-making within two development

projects, it is important to recognize that real people made these decisions, and did so

within a matrix of forces that affected their decision-making. Exploring gendered

political ecology and participation facilitates recognition and analysis of these forces

from the local to the global level.

This dissertation does not simply explain current natural resource use patterns or

the political ecology in St. Lucia and Dominica. It focuses on how the combination of

factors that is political ecology shapes people's participation in development projects

designed to improve their natural resource-based livelihoods. The political ecology of

the setting ties together a history of the agricultural sector and the political and

economic goals of global, regional, and local influences. For the purpose of this study,

'agricultural sector behavior' includes national, regional, and local goals and objectives

to indicate how farmers survived in the face of changing policy, markets, and


This framework will be used to explain the historical development of natural

resource use in St. Lucia and Dominica, and how global, regional, and local concerns

have contributed to the conditions that currently exist. These conditions spawned the

study projects, helping to define their purpose and the sponsoring agency's expected


Political Ecology

With roots in physical science, social science, and political economy, political

ecology accepts that social, economic, and political forces influence how, when, and

why humans interact with their environments (Bryant, 1992; Greenberg and Park,

1994). Political ecology integrates political economy and human ecology principles in

its approach to understanding the interconnectedness between social, economic, and

political processes that influence natural resource use (Greenberg and Park, 1994;

Escobar, 1998; Schmink and Wood, 1987; Stonich, 1993). This perspective is

important to this study as it allows for examination of projects designed to improve

natural resource based livelihoods without ex-ante determination of which forces are



The first field of influence in political ecology comes from ecology. The roots

of ecology can be traced back to the 1860s, when Haeckel defined it as, "the science of

the relations of living organisms to the external world" (Greenberg and Park, 1994, p.

2). Early ecologists included Humboldt, who traveled widely studying plant geography

and Charles Darwin whose Origins of Species informed future research on evolution

and species diversity.

Two major schools of thought developed within ecology: evolutionists-who

emphasized evolution at the individual level; and ecosystem advocates-who focused

on interactions within and between communities of organisms. Evolutionists grounded

themselves in individual level competition as the driving force of change, believing that

communities were just groups of competing individuals, while ecosystems specialists

encouraged a more holistic conceptualization of action/interaction among and between

organisms. Whether considered in isolation or from a more holistic perspective,

ecologists emphasize the evolution of organisms at the individual or community level

(Greenberg and Park, 1994; Holdgate, 1994). Both of these perspectives focus

exclusively on the 'natural' world to the exclusion of the influence that humankind and

politics have on organisms.

Social Sciences

The second field of influence in political ecology comes from the social

sciences. Social sciences shifted the focus of the human/environment interaction to

consider people's interactions within ecological and environmental settings. According

to Greenberg and Park, "ecological ideas have had to be extended in line with social

science understandings of the mutual interactions between human society, human

productive activity, and the (now only slightly "natural") environment" (1994, p. 4).

Beginning in the 1950s, human and cultural ecologists within anthropology sought a

clearer understanding of relationships between human populations and their natural

environment. Among their interests was the relationship between existing ecological

systems and changing cultural practices, or cultural practices and changing ecologies.

However, establishing causality for such change proved difficult, and "although micro-

approaches had considerable success with small, rural populations, the application of

the simple ecological models to human societies soon seemed problematic" (Greenberg

and Park, 1994:4). The failure of social sciences to diagnose issues of social and

technological changes over time raised issues of scale, boundaries, space, and

institutional influences on cultural evolution, leaving room for additional factors.

Political Economy

The third stream contributing to political ecology is that of political economy.

Political economy framed the question of people's relationship to the greater world

(social and environmental) as one primarily driven by their location vis-la-vis political

institutions and markets. Deriving theory from political science and economics,

political economists couched the evolution of populations in terms of their relationships

to broader institutions and markets. Political economy encouraged examination of

integration between local populations and the larger political economies in which they

were embedded (Wolf 1982).

Rostow (1960) suggested that societies go through prescribed stages of

economic development on their way to 'modernization,' and political economy came to

the forefront with this theory. Modernization theorists expected nations to progress

from traditional to modem via their contact with global capitalism. They equated

'traditional' with labor intensive, exchange types of production systems and 'modern'

with capitalist economies.

Modernization theory was challenged by dependency theory, that suggested that

traditional societies are thus as a result of their dependence on capitalist societies, rather

than their lack of contact with them. Frank suggested that economic stratification

existed and persisted based on an extractive relationship by the metropolis of the

satellite (1966). Metropolis represented not only developed nations' domination over

underdeveloped, but urban-rural relations within nations as well. The metropolis-a

functioning capitalist economy--extracted surplus value from and profited at the

expense of the satellite or traditional economy.

The emergence of world systems theory in the mid 1970s introduced a

worldwide market that had expanded to cover the globe in the last 500 years.

Wallerstein defined states as core, semi-periphery, or periphery based on their ability to

extract value from labor, and claimed that the persistence of this system has led to

structural differences between the different types of states (1974). This version of

political ecology accounts for core states economic and political domination of

peripheral states, and peripheral states functioning to meet the requirements of the core.

Political ecology is useful for understanding and addressing

human/environmental interactions at local, regional, and global scales by examining

them in concert with political-economic forces (Grossman, 1993; Holdgate, 1994). The

resulting approach "aims at developing a new paradigm of production that incorporates,

for any given ecosystem and social group, cultural, ecological, and technoeconomic

factors into a strategy that is ecologically and culturally sustainable" (Escobar, 1998, p.

62). The focus that political ecology gives to external forces and local community

adaptation to their influences provides an adequate framework for those who work with

rural communities continually subjected to the will of global political and market forces

(McIlvaine-Newsad, 2000).

Gendered Political Ecology & MERGE

As comprehensive as political ecology appears, in its conceptualization and

early practice it did not explicitly recognize women's roles, needs, and perspectives in

relation to natural resources, or acknowledge socially constructed differences between

resource use by women and men.' The importance of women's roles in development

was brought into the discourse as Women in Development (WID) in the 1970s,

followed by Women and Development (WAD), and by Gender and Development from

the 1980s onward.

WID was predicated on the idea that both women and men must be lifted from

poverty, and that both women and men must therefore be recognized and included in

development. Closely associated with modernization theory, the WID approach

focused on improving women's welfare by incorporating them within existing

development projects. WID advocates recognized that women simply were not counted

and began addressing the need for statistical representation of women's lives, roles, and

contributions where little had existed. Once women appeared, WID pushed to include

them in education, training, employment, and credit opportunities by claiming that

these opportunities would increase women's economic efficiency (Moser, 1993;

-Reddock, 2000). Having gotten on the development agenda by focusing on women's

1Esther Boserup is credited with putting women on the agricultural development map by publicizing a
sexual division of labor that dictated their roles as food producers in Africa, Asia, the Caribbean and
Latin America, and by classifying agricultural production systems by who held major responsibility for
major production activities. Her research illustrated that most development efforts of the period ignored
women, offered training and resources to men, and often undermined women's positions within the
household (Boserup, 1970).

productive capacity, WID was unable to untangle or address the causes of their

subordination. WID increased women's visibility as productive members of

households, communities and countries but did not seek to address the root causes of

women's disadvantaged position vis-i-vis access to and control of resources.

The shift from modernization theory to dependency theory as the dominant

development paradigm in the 1960s corresponded with a shift from WID to Women

and Development (WAD). Rather than being seen as the solution to

underdevelopment, the West was seen as the cause, and rather than adding women to

projects designed for men, women needed attention in their own right. As reflected in

the names, these approaches differed in their perception of women's relationship to

development. In moving from women in development to women and development, this

approach elevated women from a bit player within the process to the focus of the


Gender and Development (GAD) emerged during the 1980s and seemed to

reframe the question from what do women need, to what are the root causes of their

needs? This approach is predicated on the belief that women's position in society is

socially constructed rather than the natural order of things. GAD, with attention to

social constructs that determined women's space in relation to men's changed the focus

from treating the symptoms to exploring the causes by "addressing women's and men's

social roles in relation to development" (Smyth, 1999, p. 9).

GAD quickly moved from a theoretical framework to a series of analytical

frameworks used today. In addition to women's material condition and class position, /

GAD explores and analyzes patriarchal ideas and structures that reinforce women's

disadvantaged position. Molyneux (1985) introduced the concept of women's interests

as either practical or strategic, and Moser (1993) adapted these terms for planning to

meet women's needs.

-- Practical gender interests/needs: If these needs were met, the lives of women
(or ment) would be improved without changing the existing gender division of
labour or challenging women's subordinate position in society. Meeting practical
interests/needs is a response to an immediate perceived necessity; interventions
which do this are typically concerned with inadequacies in living conditions such
as water provision, health care and employment.

.. Strategic gender interests/needs: If these were met, the existing relationship of
unequal power between men and women would be transformed. These
interests/needs relate to gender division of labour, power, and control. Those
identified by women may include issues such as legal rights, domestic violence,
equal wages, and women's control over their bodies (Smyth, 1999, p. 20).

Having evolved from the early imperative to count women, gender analysis

helped researchers and practitioners recognize women's roles as decision makers in

cultivating, collecting, and care giving while acknowledging social and structural

barriers to women's empowerment. However flexible this approach appeared, its

primary use was within the agricultural sector where much development was focused.

That is not to suggest that researchers focused solely on women's production activities;

to the contrary, women's triple roles-production, reproduction and community

management activities---were recognized and accounted for with corresponding

analytical frameworks (Moser, 1993).2

While women's association with natural resources was embedded in gender and

development thought and practice, less attention was given to women's relationship /

with their broader ecological environments. Whereas women's roles in agriculture

2 See March, Smyth and Mukhopadhyay, 1999 for a comprehensive guide to gender analysis

became clearer to the development world in the 1970s and 1980s, their roles as

managers of ecosystems lagged behind until into the 1990s. Closer examination

suggested that,

There is also a gender dimension to ecological inequality .Women's role in
provisioning and care of the household leads to a special concern with such issues
as scarcity and pollution of water and lack of firewood. Women often have a
smaller share of private property and depend more heavily on common property
resources. Also, women often have specific traditional knowledge in agriculture
and medicine, which is devalued by intrusion of market resource exploitation or
state control. (Martinez-Alier, 1998, p. 23)

Feminism, Gender, and Political Ecology

The following discussion of feminism and gender is not meant as a

comprehensive or comparative review of the subjects. Rather, it is meant to illuminate

linkages between them, and how together they contribute to a political ecology that

explicitly includes women. Feminist thought has remained behind-the-scenes in the

wider discourse on gender and development. Gender, a less charged term, has been

used to indicate a female-centered perspective even when couched in a social construct

that analyzes women's position relative to men. Somewhat conspicuous by its absence

in development practice, feminist theory has informed and impacted much of the work

done within gender and development. While gender claims that women's reality in a

given situation is largely determined by local expectations, norms, laws, etc, rather than

dictated by biology, the de facto focus of gender research is inequality between men

and women. Recognition of this inequality is but a short step from the core of feminist

theory-the subordination of women.

A notable contribution of feminist thought to development has been feminist

critique of science with a call to question and challenge dominant paradigms, models,

and approaches (Harcourt, 1994; Harding, 1987; Hartman and Messer-Davidown,

1991; Naples, 2003; Rocheleau, et al., 1996; Rosser, 2000). Feminist theorists claim

that objective, positivist science is not value free, promotes a particular agenda, and is

by nature disadvantageous to women.3 It must be noted that feminist critiques of

science and dominant development paradigms are based not only on sex, but find their

center in questions of power. While gender and development acknowledged power as

'access to and control of' resources, questions of power at other levels and in other

spheres remained the domain of feminist theorists. Their concern raised the debate

about women's situations from forests and fields to questioning what is legitimate

knowledge, who produces it and who can claim it.

This researcher suggests in this way, feminism is an inherently political

orientation with gender as its application in the real world of development practice.

Feminist political ecology serves as both a methodology and a political stance

(Harcourt, 1994), while gendered political ecology suggests methods for recognizing

and incorporating women's roles, responsibilities and realities in natural resource use.

Women emerged in political ecology during the 1990s, both explicitly

(Rocheleau, et. al, 1996; Schmink, 1999), and implicitly (Agarwal, 1994; Agarwal,

1997; Harcourt, 1994). Like other political ecologists, Rocheleau et. al, (1996)

examined environmental practices and policies as a series of decision making processes

within particular social, political, and economic contexts, acknowledging uneven access

to, control, and distribution of natural resources. However, they claim a distinctly

3 Regardless of the type of feminism, feminists promote challenging dominant paradigms wherever they
exist. For a comprehensive review of feminisms and their critiques of science, see Rosser, 2000.

feminist perspective on political ecology by treating "gender as a critical variable in

shaping resource access and control, interacting with class, caste, race, culture, and

ethnicity to shape processes of ecological change" (1996, p. 4). They justify their

approach as feminist on two fronts: by examining the constructed definition of

environment through the lens of feminist critique of science; and the emergence of

feminist environmental activism.

Another critical association of women with political ecology is Schmink's

MERGE conceptual framework (1999). I argue that while not explicitly feminist, this

framework relies heavily on feminist methods of self-reflection, positionality and

women and their lived experiences as valid sources of knowing. Emerging from

applied research and community based conservation experiences, the MERGE project

sought to "address the need to strengthen the understanding of gender issues in natural

resource management for both academic researchers and local technicians responsible

for the implementation of natural resource management projects" (Schmink, 1999).4 In

keeping with political ecology, MERGE accounts for historical, ecological, cultural,

socioeconomic and political factors affecting natural resource use. In addition, the

MERGE framework puts a premium on local actors, their patterns of use, and their

perceptions. Recognizing and acknowledging local actors' roles in creating and

maintaining sustainable natural resource use strategies ties together participation and

political ecology. Inherent in this conceptual framework is the recognition that

empowering local actors in isolation from those intervening factors, inadequately

4 For an account of the evolution of the MERGE project with the Tropical Conservation and
Development Program at the University of Florida, and MERGE conceptual framework, see

addresses issues of power and competing interests. Only by mapping out influential

factors, in relation to each other, can researchers, practitioners, and community

members accurately assess and address the situation.

1. How is the potential for community-based conservation projects constrained or
enhanced by historical, ecological, cultural, socioeconomic and political factors at
diverse scales?
[Political Ecology analysis]
2. Who are the multiple stakeholder groups involved in direct or indirect negotiation
for resources? In what ways are their interests complementary and/or in conflict? How
do their different levels of power and resources affect the outcomes of negotiations?
[Stakeholder analysis]
3. Under what conditions does participation by local communities contribute to goals
of achieving conservation with improved livelihoods?
[Stakeholder analysis within the community]
4. In what ways do gender relations differentiate people's connections with natural
resources and ecological systems, including knowledge of, use of, access to, control of,
and impact on natural resources, and attitudes towards resources and conservation?
[Gender relations and resource analysis]
5. Does stakeholder participation in participatory learning with a gender focus improve
the ability of local groups to negotiate their interests in conservation? [Project analysis]
6. How are changes in resource use and management by local communities linked to
biodiversity and conservation?
[Sustainability analysis]
7. How can stakeholder learning contribute to conservation success in the long run?
How can it be incorporated into a broader strategy for institutional change and
partnership that provides continuity in research, exchange, technical assistance and
other participatory activities with local communities?
[Institutional analysis]
Figure 2-1. MERGE Conceptual Framework. Source: Schmink, 1999.

In its original form, the MERGE framework was directed toward community

based conservation efforts.5 The progressions from political ecology analysis through

local stakeholder and institutional analysis "provide a systematic understanding of the

interplay of socio-environmental factors that lead to the observed patterns of resource

use," and can result in location and situation specific policy interventions (Schmink,

5 For the purpose of this study, I have adapted the MERGE conceptual framework to be applicable to the
projects that are the subject of this research. The adapted version is presented in the Chapter 3.

1999). In that context, it is grounded in community level participation as a key to

equitable and sustainable development with "emphasis on understanding the

opportunities and constraints, the incentives and disincentives that influence the

decisions that are made by individual actors and groups" (Schmink, 1999).

Participation in Development

Development assumes some basic equality between developed and
underdeveloped nations within the present world. That assumption has been
questioned since at least the 1950s, and continues to be challenged with
increasing sophistication. Peripheral societies differ from the industrialized West
in terms of their internal structures, their integration within the world economy,
and their immediate historical and cultural heritage. (Trouillot, 1988, p. 286)

This review of participation in development is meant to acquaint the reader with

the progression from expert centered, information-based efforts to the current focus on

client-centered empowerment. This review has not treated participation within specific

fields, for example health, education, or conservation, but presents an account of the

change of mentality within development over the last fifty years. As implied, a certain

amount of uncertainty accompanies discussions and defenses of participation.

Since its arrival in the 1970s, participation has emerged as a dominant paradigm

in international development (White, 2000). Funding proposals must often prove to

donors that their project considers all stakeholders (Deshler and Sock, 1985; Dichter,

1989). Policy makers want to know what 'the people' think. Governments in the

"North" and development agencies are promoting decentralization and democratization.

As participation spread and the number of practitioners increased, so did the need

for understanding effective tools, techniques and methods. Information was passed on

through networks of practitioners and participation became its own innovation process

(Collinson, 2000). As participation took hold, there arose a need to standardize

processes and approaches. This had advantages and disadvantages. The advantage was

that the idea was reaching an ever-growing audience. The disadvantage was that in

some cases it was being standardized and diluted (Chambers, 1998; Guijt and Shah,

1998,Pritchett and Woolcock, 2004).

Standardization of tools for participation led to criticisms of the approach. It had

lost its flexibility and dynamism. Rather than being responsive and context-specific,

participation became a necessity in any 'good' project. These criticisms caused

practitioners and researchers to clarify meanings of participation. This clarification

came mainly in the form of typologies of participation. Currently several typologies of

participation are recognized, and are typically presented as a scale or continuum by

which to gauge the nature of participation of the intended recipients in development

projects (Waisbord, 2002; White, 2000). These typologies suggest that not all

participation is created equal (Deshler and Sock, 1985; Lilja and Ashby, 1999; Pretty,

1995; White, 2000). Connections have been established between approaches

(typologies) to participation and expected outcomes; however they come complete with

the assumption that success is as simple as picking the correct approach to

incorporating participation (Guijt and Shah, 1998; van Veldhuizen, Waters-Bayer, and

Zeeuw, 1997).

History of Participation in Development

That participation has become part of the development agenda at all is no mean

feat and its current emphasis signals a grand paradigmatic shift in the last five decades.

Although observers might suspect that participation is a recent innovation in

development practice, it is more accurate to say that there have always been those who

approached this work with the belief that answers lie within individuals, households,

and communities, rather than within donor organizations and policy making bodies

(Freire, 1970; St. Anne, 1999).

Political science and economics dominated 20th century development theory after

World War II. The primary goals were to transform institutions and markets in the

developing world into versions that more closely resembled those of the developed

world (Hancock, 1989; Pritchett and Woolcock, 2004; Rostow, 1960). Development

was seen as the duty to close the gap between the haves and the have-nots by giving the

latter group a road map to prosperity, based largely on the paths followed by the former

(Rogers, 1995, Rostow, 1960). The underlying assumptions) was that conditions in

the developing world could, and should be improved by focusing on political systems,

markets, and education (Tomich, Kilby, and Johnston, 1995). As Pritchett and

Woodcock claim, economists "focused their tools on the question of what governments

should do, with relatively less attention to the economics and politics of how to

accomplish the 'what'" (2004, p. 192).

According to Waisbord (2002) who examined the role of development

communication, the last 50 years of practice and discourse were dominated by two

basic assumptions about the causality of underdevelopment, (1) lack of appropriate

information (such as technologies); and (2) unequal power among populations. The

first of these assumptions reflects modernization theorists' belief that adequate

information (and education) could change the behavior of populations and carry them

toward development (Rostow, 1960). The second assumption was more in line with

dependency theorists who believed that the underdevelopment of some was a result of

development of others (Frank, 1966).

Proponents of modernization theory addressed complicated issues by attempting

to change behavior by providing adequate information. In their view, appropriate

information would lead to behavioral change, putting populations on the right track

toward modernization. It was suggested that underdeveloped nations suffered

bottlenecks at certain points on the path due to their traditional cultural beliefs, and that

these kept them from moving up the ladder of consumption. These bottlenecks were to

be addressed by providing new ideas-to overcome the old-that would lead to desired

behavior, progress, and ultimately development. This approach to development was

based on the belief that individual, group, or societal behavior was the problem (cause

of underdevelopment) and could be solved by introducing better ways of behaving

(Rostow, 1960, Waisbord, 2002).

Modernization theory manifests itself in efforts to diffuse information,

technologies, ideas, or practices with little or no attention to the heterogeneity of the

intended audience and their realities: political, economic, environmental, ecological,

social, or religious. Nor was it deemed necessary to engage recipients in meaningful

dialogue when one-way communication seemed sufficient to deliver the message. It

was thought that these non-technical factors should not impact an audience's ability to

adopt new or improved technologies. In other words, neither these, nor other

exogenous factors, were expected to affect people's capacity to change (Rogers, 1995).

Governments sought to skip the steps involved in slowly establishing and maintaining

governance structures necessary for meeting their citizens' needs. Instead, they

mimicked the organizational forms of existing, 'developed' nations, and made their

own situations worse (Pritchett and Woolcock, 2004). This shortcoming traditional

approaches-which dominated development from the 1950s into the 1980s---opened

the door for examination of the reasons behind the failures, and an examination of

underlying assumptions made by decision-makers. One of the most costly assumptions

was that of a need that could be met with a technical solution, delivered by an

impersonal, rules-driven provider (Pritchett and Woolcock, 2004).

Modernization theory was eventually challenged by dependency theory, which

supported the second of the major assumptions presented by Waisbord: that unequal

power relations were responsible for underdevelopment. Rather than putting sole

blame for their plight on underdeveloped nations, dependency theory suggested that

external factors were responsible, at least in part, for the disparity between the haves

and the have-nots. Underdevelopment was due to a combination of factors including

faulty social processes and the practice of powerful nations extracting assets from the

weaker, peripheral states and so on down the line (Wallerstein, 1974).

As this theory gained momentum, the focus shifted to interconnectedness, power

differentials, and competing interests. Underdevelopment could no longer be blamed

on lack of information, but rather on social and political factors. It followed then that

better information leading to individual behavioral change was not the solution. The

answers lay in political and social change. Dependency theory shifted the burden from

individual behavior to the function of institutions and power differentials. Nonetheless,

development during this era still followed the "need as the problem, supply as the

solution, civil service as the instrument" model for institutions and organizations

(Pritchett and Woolcock, 2004, p. 192).

Although casting underdevelopment in a different light, dependency theory failed

to change significantly the way development-as planned change-was managed by

practitioners, researchers, or policy makers. Answers, and solutions were still derived

and implemented by outsiders rather than those most affected by the situation. Little

attention was paid to the individuals on whose backs and by whose labor development

was to happen. Women and men, young and old, Muslim and Christian, relatively

wealthy and relatively poor people were all treated the same by policy makers,

researchers, planners, and practitioners. It was as if there was an average rural resident

in a developing country, and everyone else knew what was best for her or him.

According to Karl (2000), long-term failure of conventional development efforts,

as evidenced by increasing poverty, caused a paradigmatic shift away from

modernization and dependency theory. The emphasis on information and institutions

had failed to create substantive or sustainable change in the well being of the poorest of

the poor. Upon reflection by examiners of development history, it became clear that for

several decades, projects and programs had been plannedfor and delivered to people,

rather than with them.

Alternatives to the dominant, top-down approaches to development did exist. In

fact, as early as the mid 1970's, groups of researchers and practitioners working across

the globe established contact with each other and compared notes on emerging

approaches to development (Chambers, 1997). With more differences than similarities

between the various pockets of activity, some common themes emerged, (1) a client-

centered approach whereby individuals were consulted and their input helped shape

subsequent research agendas; (2) an increased reliance on social scientists and social

science rather than just biophysical scientists and science; and (3) a systems, or holistic

approach to examination and analysis. Focus shifted to interactions between

households, their members, their activities, resources, requirements, and constraints in

concert with plants, water, soil, or animals.

Early practitioners and advocates of participation in development were field staff

who experimented with new ways of communicating with farmers, making research

more germane to their individual needs. The idea and practice of involving farmers-

allowing them to participate-gained a following as it became clear that not all farmers

were benefiting from the traditional, 'top-down' agricultural development model, and

that not all farmers were the same (Chambers, 1997; Collinson, 2000; Escobar, 2000;

Hildebrand, 1986; Whyte, 1981).

Participation grew into something 'good and right,' almost regardless of its form

or function (Cleaver, 1999; Guijt and Shah, 1998). Participation included political

scientists promoting democratization; economists seeking ways to integrate people into

markets; sociologists focusing on community action; and others seeking richer data that

emerged from engaging stakeholders (White, 2000). With no single rationale,

participation played many roles and served many purposes. Much like gender sensitive

and eco-friendly approaches to development, participation was credited with everything

from increasing efficiency of projects to empowering the traditionally powerless. And

much like gender analysis and 'green' approaches, participation is open to

interpretation by those responsible for its institutionalization, explanation, and

implementation (White, 2000).

Participation is now promoted as a key to ensuring appropriate technology

development, dissemination, and sustainability (Chambers, 1997; Guijt and Shah, 1998;

Karl, 2000). Current views suggest that rural populations with limited resources in the

developing world, especially women, must have a greater voice in setting development

agendas that influence their livelihood systems and dictate their survival strategies

(Harcourt, 1994; Rocheleau, et. al, 1996; Reddock, 2000; Saito and Spurling, 1992).

The evolution of development theory from information centered, modernization

approaches toward power centered, participatory approaches has provided new lenses

through which to view the idea of planned change in the developing world.

As a result, participation, became part of mainstream development strategies for

addressing the failures of the past. According to Karl, "Development agencies began

to introduce concepts of participation in projects and programmes in the late 1970s and

early 1980s after lack of beneficiary participation was identified as a reason for the

failure of many development efforts" (2000, p. 85)

Guijt and Shah (1998) agree that although not a new idea, participation emerged

on the global stage of development as early as the 1970s. Its major proponents were

field workers and practitioners who recognized the value of the existing knowledge of

farmers and sought to formalize it. This system of trial and error seemed in direct

response to decades of top-down directives from researchers and policy makers with

little understanding of the reality on the ground.6 Early experiences with participation

led to research that was more tailored to specific audiences and their needs.

The 1980s saw an expansion of the use of participatory approaches, and

corresponding confusion over terminology, methods, and practice. Those promoting

participation based their approaches on respecting local knowledge and experimenting

with people, rather than for them. As more work was done with individuals and

communities, greater emphasis was placed on increasing their capacity to solve their

own problems.

From the early 1990s, participation has been a dominant paradigm in

development (Guijt and Shah, 1998; Peters, 1996; Webber and Ison, 1995; White,

2000). The evidence is the emphasis put on 'participation' by donors as a condition for

funding projects and programs. Participation came to be characterized as an

appropriate approach as well as a means of ensuring sustainability of projects and

programs (Cleaver, 1999). Policy makers and practitioners sensed that participation

was a positive innovation and the word spread. Practitioners wrote about their

experiences, sharing them with others in an effort to multiply positive outcomes (for

examples see: Ascroft, and Masilela, 1994; Guijt and Shah, 1998; Swantz, 1994;

Wacker, 1994).

High expectations of participation are reflected in vague definitions and laundry

lists of likely outcomes cited by practitioners, researchers, development agencies and

policy makers (Karl, 2000; Guijt and Shah, 1998; Jacobson, 1994; White, 1994).

6 Much of the theoretical foundation of people helping themselves while learning from each other as a
process is attributable to Paulo Freire's Pedagogy of the Oppressed (1970).

Participation is more likely to be explained by what it can do, than by what it is or

cannot do. This is not to say that participation cannot meet these objectives, rather that

without clear definitions and expectations, participation might just as easily miss as hit

its targets. It is interesting to note lack of delineation between definitions, rationale,

expectations, or techniques for implementing participation (Karl, 2000). Given a lack

of precise definitions of participation, skeptics such as Cleaver (1999) and Dichter

(1989) question the validity and utility of participation in development.

There is a wide range of interpretations that can be reduced to a few common

themes. It is difficult to discern one definition of participation. Interpretation suggests

that lack of a common definition reflects the complex nature of the term and its

practice. In the end, participation is context specific, and could look different from one

situation to the next depending on factors such as those identified within gendered

political ecology. According to Jan Servaes, writing in the field of development


Participation is a term used to refer to a number of social and planning processes
occurring in many different places and in many different contexts. To some,
participation is a means to reach a certain goal; to others, participation is an end
in itself. (1996, p. 15)

Bordenave suggests, "In order to avoid banalization of the term, participation should be

referred to as the joint efforts of people for achieving a common, important objective

previously defined by them" (1994, p. 46).

Whereas concise definitions of participation are limited, there are many

justification and expectations. The underlying assumption suggests that stakeholder

inclusion leads to more accurate situational analysis, more sustainable results, and more

effective investment of resources. There seems to be no single rationale for

incorporating participation in development efforts, but justification is based on the

belief that increased participation leads to increased effectiveness (Chambers, 1997;

Cleaver, 1999; Guijt and Shah, 1998; White, 1994). John Thomas, a political scientist


The principal reason for involving the public in decision-making is to increase
acceptance of the decision, especially when implementation hinges on that
acceptance. Involvement increases the likelihood of public acceptance and of
successful implementation by nurturing citizen "ownership" of the decision.
(1995, p. 48)

In the same vein, White argues genuine participation occurs when "the development

bureaucracy, the local elite, and the people are working cooperatively throughout the

decision-making process and when the people are empowered to control the action to

be taken" (1994, p. 17).

According to Karl, "Some of the common objectives and expected benefits of

participation in development are improving the efficiency, effectiveness, sustainability

and coverage of projects and programmes and promoting stakeholders capacity, self-

reliance and empowerment" (2000, p. 23).

Clayton, Oakley, and Pratt, summarize widespread hypotheses of the benefits of

participation in rural development projects and programmes. It is expected that

participation can:

"Increase the efficiency of development activities by involving local resources
and skills and thereby make better use of external costs.
* Increase the effectiveness of activities, by ensuring that they are based upon local
knowledge and understanding and are more relevant to local needs.
* Build local capacities and develop the ability of local people to manage and
negotiate development activities.
* Increase coverage and help expand the range of activities.
* Better target beneficiaries to the poor through the identification of key
stakeholders who are most affected by the activities.

Help ensure the sustainability of activities as the beneficiaries assume ownership.
Improve the status of women by providing the opportunity for them to play a part
in development work" (1998).

As summarized by Karl, more specific objectives of participation include

"Efficiency: to promote agreement, cooperation and interaction to achieve a
smoother flow of project services and minimize overall costs.
Effectiveness: to help achieve the project objectives through the involvement of
beneficiaries in project design and implementation.
Beneficiary capacity: to build the capacity of beneficiaries to share in and take
responsibilities for the tasks of projects.
Self-reliance: to break the mentality of dependence, to promote self-awareness
and confidence of rural people and help them learn how to plan and implement so
that they have greater control over their lives.
Empowerment: to achieve a more equitable sharing of power and a higher level
of political awareness and strength for the disadvantaged.
Sustainability: to help ensure that local people maintain the project's dynamic.
Coverage: to increase the numbers of rural people who potentially benefit from
development" (Oakley 1991) (2000).

One likely explanation for uncertainty around participation can be attributed to

the fact that at its very core, it advocates bringing more stakeholders to the table, and

acknowledging their points of view. Although simplistic, this interpretation highlights

one of the most sensitive issues surrounding the concept of participation: that of power.

Taken to its logical end, participation directly impacts the balance of power within and

between all levels of stakeholders in the development industry.

Genuine participation directly addresses power and its distribution in society.
Participation involves the more equitable sharing of both political [democratic]
and economic power, which often decreases the advantage of certain groups.
Structural changes involve the redistribution of power. (Servaes, 1996, p. 16)

Critiques of Participation

Criticisms of participation emerged as the approach gained popularity. These

criticisms ranged from maintaining 'ugly' fields in the name of on-farm research to a

willing suspension of disbelief on the part of practitioners to trust farmers (Whyte,

1981; Dichter, 1989). Current critiques target the inherent uncertainty around

participation ranging from terminology and implementation to impact and scale.

Researchers question the role of participation in 'good' science as it may compromise

reliability and validity of studies (Gladwin, Peterson and Mwale, 2002). In their broad

critique of the inadequate treatment of gender in participation, Guijt and Shah (1998)

question whether or not participation has evolved into a simplified recipe suitable for

broad adaptability. This concern implies that participation was meant to be adapted to

the context within which it was utilized. They also suggested that participation had

become a popular method of conducting needs assessment or data collection but was

rarely applied to analyzing or addressing underlying causes of existing situations, i.e.,

power relations.

Chambers (1997) often cited perspective on participation claimed that previous

development efforts failed to realize local complexity, diversity, dynamism, and

unpredictability, and incorporate it into development planning. Although few would

disagree with the need to recognize these characteristics, Chambers was vague about

strategies to follow up on this recognition. It was as though solutions would present

themselves after recognition of the nature of people, systems, and problems. Guijt and

Shah expand upon this point by suggesting that participation

S. is more than inviting people to partake in a needs assessment or a decision-
making process. Offering the marginalized opportunities for consultation,
without following this through with analysis about causes of oppression and
feasible action to redress the causes, is unlikely to be empowering. (1998, p. 11)

Like gender, participation has been advocated for a range of reasons from a more

efficient use of resources, to a means for empowering the less powerful. Like gender,

participation as an ideal has been advocated and accepted by development institutions

without consensus on goals, indicators, expectations, or understanding of how best to

operationalize it. As such, different organizations, and often individuals with the same

organization, may conceptualize and practice participation differently.

Participation has been used to describe target populations providing researchers

with data, as well as those same beneficiaries planning for, implementing, monitoring

and evaluating their own development efforts (Karl, 2000; Pretty in Veldhuizen et al.,

1997). Given the range of possibilities, advocates of participation would be hard

pressed to articulate 'how much' participation was called for, or what type of

participation was best for a particular situation.

Much remains to be understood about participation. Political science offers

insight into one of the issues of participation: what makes people get involved. The

following passage highlights some of the shortcomings of participation in development.

In thinking about why some people are active while others are not, we find it
helpful to invert the usual question and to ask instead why people do not take part
in politics. Three answers immediately suggest themselves: because they can't
(resources); because they don't want to (engagement); or because nobody asked
(recruitment). (Verba, Schlozman, and Brady, 1995, p. 15)

White (2000) offers a critique of participation by suggesting that focusing on who

participates and at what level are not enough. As she explains, "sharing through

participation does not necessarily mean sharing in power" (p. 143). Her point resonates

in an era where people's participation may mean showing up at the table to listen or

respond to experts' questions. By acknowledging differential access to power, White

has shifted focus from quantity of participation to quality. Referring to types of

participation where people are simply consulted, she suggests,

At their best, such measures can only facilitate fuller participation, they cannot
deliver it. More critically, framing the problem in these terms ties us to

observing the mechanisms for participation; it gives us no mean of assessing its
content." (p. 144)

Typologies of Participation

Early explorations of participation resulted from practitioners' examination of

their own work in search of lessons learned. Practitioners themselves recognized some

of the difficulty of operationalizing participation, and tried to provide a roadmap for

subsequent efforts (Chambers, 1997). Typologies emerged that sought to clarify some

of the issues around the term and practice. Researchers such as Lilja and Ashby (1999),

Pretty (1997), and White (1994), identified different types of participation and their

associated power, location, communication, inputs, and outputs. They suggested that

particular types of participation could lead to particular types of outcomes, and which

types of participation were more appropriate for which particular settings.

Based on this work, it is now recognized that there is a continuum of participation

and frameworks showing various levels of stakeholder involvement. These frameworks

do not always indicate who participates and the roles they play, but suggest where

various approaches to participation fall relative to each other, and possible outcomes.

Guijt and Shah suggest a useful distinction between participation as a means

versus participation as an end in and of itself. While not prescriptive, this basic

categorization is a useful.

An instrumental approach views participation as a means to achieving better cost-
effectiveness, living standards, etc., while an empowerment approach values the
process of increasing participation as an important end in itself. (1998, p. 9)

Other researchers have addressed the ends versus the means issue. Campbell and

Salagrama echo this view when they offer three categories of participation: functional,

empowering, and philosophical. According to the authors, functional participation

addresses issues of efficiency and effectiveness; empowering participation

encompasses independence, awareness, and increased capacity; and philosophical

participation allows for the expression and recognition of alternative world-views (Karl,


The World Bank suggests three categories of roles to be played by rural poor,

based on their level of empowerment. They are as follows:

Beneficiaries: recipients of services, resources and development interventions
through such things as community organizing, training and one-way flows of
Clients: capable of demanding and paying for goods and services from
government and private sector agencies. World Bank-supported development
interventions at this stage focus more on building sustainable market-based
financial systems, decentralizing authority and resources, and strengthening local
Owners and managers of their assets and activities: the highest stage in terms of
the intensity of participation involved (World Bank 1996).

Deshler and Sock (1985) reviewed the development participation literature and

found that, although support and advocacy for participation were high,

few studies are available with a rigorous analysis of the phenomenon. Concepts,
measures and indicators of rural development participation are lacking, as are
theories, definitions and conceptual frameworks for empirical validation or
formulation of hypotheses (1985, p. 1).

They used concept mapping to identify a key theme, "'extent or control of power'"

exercised by participants in various conceptualizations" (p. 4). They created four

different types or categories of participation: "domestication, assistencialism,

cooperation, and empowerment" (p. 5). Deshler and Sock then grouped these four

types of participation into two major categories: pseudo-participation, that includes

domestication and assistencialism; and genuine participation that includes cooperation

and empowerment.

Domestication: involves informing, therapy, and manipulation
Assistencialism: includes placation, and consultation.
Genuine participation
Cooperation: refers to partnership and delegation of power
Citizen control: which means empowerment.

Pseudo-participation occurs in development when decision-making power and

control of the project rests with planners, administrators, and community elites.

Genuine participation occurs when the development bureaucracy, local elites, and the

people work cooperatively throughout the decision-making process, and when the

people have the power to control the action to be taken (White, 1994).

Jules Pretty (1997) presented a seven-step typology of participation ranging from

passive participation to self-mobilization. This particular typology is useful in its

presentation of a detailed range of possibilities for participation; however these

distinctions seem rather theoretical and fail to acknowledge the dynamic nature of

people and situations. Pretty's language indicates that he intends this typology for

consumption by researchers and administrators rather than those expected to


He characterizes passive participation as that of individuals whose role is to

listen to external professionals explain what will, or already has happened. At this

level, individuals have no input into the process, and are only receivers of information.

The next level, participation in information giving, shows people participating by

providing researchers data determined to be of importance by researchers. At this level,

people have no power over how the investigations are conducted and no relation to the

outputs. The third level, participation by consultation, still reflects dominance in

decision making by outsiders. Insiders may be consulted about their views, which

researchers may take into account, but have no authority to impact proceedings. Level

four, participation for material incentives, describes nominal participation by insiders

who are expected to contribute (land, labor, inputs) in return for goods or services

designed by outsiders to help them. Pretty points out that this type of participation

often ends when the outside incentives dry up. Level five is functional participation in

which he describes individuals adopting collective action-forming groups-to meet

some existing objectives. These groups are not internally derived or driven, but created

and managed by outside facilitators. Level six is interactive participation that includes

insiders in the situation analysis stage of the project cycle and allows for their planning

and implementation of initiatives. In this stage locally induced institutions or groups

exercise authority in determining priorities and decisions. The seventh type of

participation identified by Pretty is self-mobilization. At this stage, insiders take

initiative to address issues that they deem important. They reach out to known

institutions rather than waiting for change agents to approach them. Pretty

acknowledges an important point in his description of this type of participation, that

even self-mobilization may not recognize or address existing inequities (Veldhuizen et

al., 1997, p. 42).

Lilja and Ashby's (1999) typology identifies levels of participation by accounting

for decision-making authority and communication between scientists and farmers in

research and innovation processes. They suggest that "organized communication"

plays a role as a "well-defined methodology for carrying out a procedure (informal

surveys, transect walk, etc. as well as formal surveys)." It is treated as either one-way,

where farmers respond to researcher driven inquiries, or two-way where either side may

initiate, and follow-up, on inquiry in an effort at mutual understanding.

They suggest five possibilities for interaction between and within these two

categories of participants. Grounded firmly in a research context, their typology ranges

from scientist dominated approaches to those where ultimate decision making authority

rests with farmers. Their first category of participation, on-farm research, refers to

scientist-controlled processes "without organized communication with farmers." The

second category is considered consultative whereby decision-making authority rests

with scientists but farmer input is solicited. In this type of participation, scientists have

an understanding of farmers' perspectives, but may not incorporate this information

into their decision-making. The third category of participation is collaborative where

farmers and scientists share decision-making power. In this case, neither side makes

unilateral decisions, or can renege on those jointly made decisions. Joint decision-

making is based on mutual understanding grounded in two-way communication of

opinions, preferences, and priorities. Type four is collegial participation. As the name

implies, farmers and scientists in this type of process are envisioned as colleagues.

Farmers are aware of scientists' perspectives and may or may not incorporate those

views or concerns into their decision-making. Scientists in this process often act as

facilitators or resource persons for farmers who control the situation and can renege on

decisions as they choose. The fifth type of participation,farmer experimentation,

suggests a process whereby farmers have sole decision making authority, with no

organized communication with scientists.

Although useful, typologies themselves may have become part of the problem. In

suggesting that participation is losing the edge it may have had as an effective and

flexible approach, Guijt and Shah identify pitfalls of existing typologies.

1. "Wrong assumption of a static picture." In reality change is constant as will be
all resulting participation.
2. "Simplifying difference in terms of 'insider and outsider.'" This distinction helps
hide differences within either of these groups, and is particularly dangerous when
all insiders are considered the same.
3. "Normative assumption of an 'ideal' form of participation." The implication that
some participation is better than others leads to a best type, towards which every
project must work.
4. "Ignoring diversity." When participation is categorized, artificial boundaries are
created between types. In all likelihood, the quality of participation-however
measured-will change throughout the lifecycle of projects or interventions.
(1998, p. 9-10).

Reviewing typologies suggests a tendency toward best practices for a limited

number of stakeholders in the process, namely practitioners and recipients. Typologies

are rarely flexible enough to cover the range of stakeholders who enter and exit

throughout the processes of project design, delivery, implementation, and evaluation.

In addition, in presenting a continuum of options, few typologies suggest which

particular approach to use in a given situation to achieve a particular result.

Sarah White (2000) created a framework that associates approaches to

incorporating participation with a particular function or rationale for doing so. She

offers two perspectives for each approach, top-down, where power resides in the hands

of institutions or agencies, and bottom-up where power resides in the hands of those

historically considered recipients. By incorporating multiple approaches, multiple

perspectives, and power differentials, her framework acknowledges that participation is

political and based on the power to decide. Like Chambers (1997), Cleaver, (1999),

and Guijt and Shah (1998), White recognizes that participation is inherently political,

with the more powerful stakeholders typically dictating conditions of participation to

the less powerful. While not always visible or malevolent in nature, questions of power

are never far from participatory development projects. An invitation to participate is

rarely an invitation to participate equally, benefit equally, or share equally in the risks

associated with the project. White acknowledges that while an invitation to participate

offers entr6e, it does not guarantee a shift in the balance of power, nor is always meant

to. White distinguishes between approaches in terms of their function or rationale

while indicating the diversity of interests and power inherent in participation.

As shown in the first column of Table 2-1, White's framework distinguishes four

forms of participation. The second and third columns delineate varying interests in

participation, with column two highlighting the interest of those who design and

implement a development intervention (top-down), and the third column highlighting

the interests of the participants (bottom-up). These columns indicate how stakeholders

perceive their participation and what they expect to get out it. Interest of both the top

and bottom can be likened to an agenda-what their expectations and goals are for

participating in any given development intervention. The fourth column shows the

overall function or rationale for each type of participation. White specifies a range of

approaches to treating participation and indicates interests of the top (more powerful)

and interests of the bottom (less powerful) according to desired outcomes legitimationn,

efficiency, sustainability, and empowerment).

Table 2-1. White's Framework
Nominal Legitimation Inclusion Display
Instrumental Efficiency Cost Means
Representative Sustainability Leverage Voice
Transformative Empowerment Empowerment Means to an end

Source: White, 2000, p. 144.

Nominal participation can be equated with passive participation. The top uses

this approach to legitimize or support itself by solving problems for the less powerful.

Efforts are taken to convince target audiences (the bottom) that their needs are being

met. Decision-making authority rests solely in the hands of a power structure and the

target audiences remain passive recipients throughout the development process. For

example, asking farmers to identify their own needs, then implementing a pre-existing

program that may or may not address them is nominal participation by farmers.

An instrumental approach to participation is based on achieving efficiency by

way of better-targeted interventions and more appropriate use of technologies. Those

within the power structure seek information or input from the target audience in an

effort to better understand their needs, and thus design an intervention that attempts to

address these needs. An instrumental approach can be considered consultative in that

target audiences may be seen as initial contributors to the process even though they

remain peripheral to most decision-making. This situation is common in agricultural

research where researchers design and administer a survey reflecting their own

priorities and incorporate farmers' input into their research.

Representative participation promotes a collaborative relationship between the

power structure and target audience in hopes that the latter will carry on the process

after planning and initiation by the former. This approach differs from the previous two

in that some decision-making responsibility, and thus power, is transferred to the target

audience throughout the process. Nonetheless, power still resides with the outsiders

who drive the process and expect audience members to take advantage of the

opportunity to contribute to their own well being, the opportunity to participate.

Successful agricultural research projects have used this approach resulting in farmer

centered and led research groups (Ashby, 2003).

Transformative participation is equated with empowerment. Those previously

treated as audience members control and own decision-making processes in this

approach. The relationship between insiders and outsiders is considered collegial with

insiders ultimately driving the process while developing and recognizing their capacity

to address further needs and situations as they arise. The obvious paradox found in the

far right spectrum of this empowerment approach power structures instituting a

process to change the power structures means that few development interventions are

designed with a truly transformative approach in mind.

While White's framework fills a void in participation theory and literature, it

treats stakeholders as two dimensional (top or bottom), and fails to address project

phases, stages, or activities that correspond to different approaches. In reality,

participatory development projects have multiple stakeholders who function between

top and bottom, and whose position may shift depending on the sequence of activities

to be undertaken. This study explores two such projects.

This review of typologies suggests the range of possible criteria for categorizing

participation. Some typologies are more specific than others, and most are directed at

specific audiences. They are thus written in a particular language. Nonetheless, taken

together they represent countless hours of experience, reflection and analysis of

participatory processes within the field of development. And more importantly, they

inform policy makers, programmers, and practitioners of the possibilities and pitfalls of


Agricultural Extension

Agricultural extension systems serve farmers by making "scientific knowledge

available to all who need it" (Seever, Graham, Gamon, and Conklin, 1997, p. 3), and

exist in virtually every developed and developing country (Rivera and Schram, 1987).

Agricultural extension began in the United States in 1914, and throughout its history

"has encouraged invention, experimentation and education" (Seevers et al., 1997, p.

15). The U.S. system is characterized as university-based because statewide research

and extension activities are coordinated by each state's Land Grant University (Seevers

et al., 1997; Warner and Christenson, 1984). Extension in the U. S. is administered

locally at the county level.

In contrast to the U. S. system, ministry-based extension dominates agricultural

outreach in the rest of the world. According to FAO, this system provided more than

90 percent of all extension services in 1993 (FAO, 1993). This system considers

agricultural extension as a service of the national government, and aims to improve

agricultural productivity for the benefit of both producers and consumers (Nagel, 1997;

Pickering, 1987). Ideally, such public extension is developed to serve the needs of all

types of farmers in a country (FAO, 1993).

In most countries, extension and research exist as separate divisions under the

Ministry of Agriculture and often have weak linkages with each other (Cernea, Coulter

and Russell, 1985; McDermott and Andrew, 1999; Pickering, 1987). Information that

gets transferred is often outdated or not pertinent to the farmer's situation (Nagel,

1997). EOs are often responsible for a large geographic area encompassing a wide-

range of farmers in terms of agro-ecosystems, socioeconomic status, and livelihood

systems. In addition, EOs often enter the field with only a few years of technical

training and little or no understanding of participatory techniques for incorporating

farmer input. This may lead to a limited understanding of farmers' problems, and

difficulty prioritizing those issues for research (Collinson, 1985; Gomez, 1985). Large-

scale commercial farmers often have more interaction and influence with government

research bodies, but small-scale farmers are virtually unseen by researchers (Baxter,

1987; Jain, 1985).

Traditional top-down extension can be a constraint to equitable distribution of

resources (Berger, DeLancey, and Mellencamp, 1984; Saito and Spurling 1992). The

biased perspective of policy makers and practitioners alike reinforces a research-

extension continuum that limits the value of extension to women farmers (Dwyer and

Bruce, 1988; Weidemann, 1987). By misdiagnosing the contributions of female

farmers, policymakers, researchers and extension have failed to understand and address

their unique challenges. Research and extension have perceived the household as a

cooperative unit with consensus on goals and objectives as articulated by a male head.

This generalization has caused the system to miss the differences that exist between

members of a household, or between various household types (Dwyer and Bruce,


Several factors limit women's access to extension including systemic bias and the

realities of women's lives. Extension often fails to address issues of concern to female

farmers, or assumes that their concerns mirror those of their male counterparts.

However, in Honduras, Colverson (1996) found a discrepancy between the training

offered and the training that women indicated they desired. Research and extension

typically devote a majority of resources to increasing cash crop production, often at the

expense of food crops that women produce for household consumption or local sale

(Saito and Spurling, 1992).

Agricultural extension in St. Lucia and Dominica are ministry-based divisions

following the model of most British Commonwealth countries (FAO, 1993). St. Lucia

and Dominica each have a Director of Extension who is responsible for oversight of all

frontline staff, organized into regional teams. Extension in each country relies on

limited research from within their own MOA, as well as from other institutions such as

the University of he West Indies (UWI), the Caribbean Agricultural Research and

Development Institute (CARDI) and other international agricultural research centers.

Their extension approach has been categorized as top-down with limited effectiveness

in meeting farmers' needs (FAO, 1999b, 2000).

Given extension's role as the outreach arm of the MOA in St. Lucia and

Dominica, FAO developed projects to reorient their approach to serving farmers.

FAO's project plans (discussed in detail in Chapter 5) sought to train EOs to facilitate

greater farmer participation in extension program planning and delivery. The first step

of this training for reorientation was familiarizing EOs with PRA (FAO, 1999b, 2000).

As development policies and priorities shifted during the last forty years, so did

approaches to understanding local realities. According to Ellis, "the shift is from the

general to the particular, from seeking solutions with widespread application to

addressing specific problems in a limited context, from implementing solutions from

above to permitting solutions to be generated from below (2000, p. 27). Structural

adjustment and market liberalization policies were accompanied by the advent of an

'actor-oriented,' or participatory philosophy of working with intended beneficiaries

(Chambers, 1997; Ellis, 2000).

Rapid Rural Appraisal (RRA) emerged in the 1970s as a methodology for

collecting accurate, location-specific information, more quickly and less expensively

than formal survey methods (Chambers, 1997; Ellis, 2000). RRA did not use structured

questionnaires or surveys, and adopted a systems-oriented approach to understanding

realities of rural populations. RRA helped researchers recognize and retain site-

specificity in terms of agro-ecosystems, seasonality, gender issues, and diversity among

households and individuals.

By the mid-1980s, RRA was evolving into PRA. This shift signaled an

expansion from data collection to encouraging local actors to participate in diagnosing

and addressing their own needs (Blackburn and Holland, 1998; Chambers, 1997; Ellis,

2000). PRA has been defined as "a family of approaches that enable people to express

and analyze the realities of their lives and conditions, to plan themselves what action to

take, and to monitor and evaluate the results" (Chambers and Blackburn, 1996, p. 1).

Table 2-2 compares key aspects of RRA and PRA, indicating common elements of


Table 2-2. RRA and PRA compared
Major development Late 1970s, 1980s Late 1980s, 1990s
Major innovators in Universities NGOs
Main users Aid agencies, Universities NGOs, Government field
Key resource earlier Local people's knowledge Local people's capabilities
Main innovation Methods Behavior
Outsider's mode Eliciting Facilitating
Objectives Data collection Empowerment
Main actors Outsiders Local people
Longer-term outcomes Plans, projects, Sustainable local action
publications and institutions
Source: Chambers (1997, p. 115).

PRA is associated with a diverse set of field methods designed to elicit

information and involve community members in setting priorities for projects and

policies. Methods and tools associated with PRA include (1) group discussions; (2)

wealth ranking; (3) maps; (4) transect walks; (5) timelines and trend analysis; (6)

seasonal calendars; (7) matrix scoring; and (8) Venn diagrams (Chambers, 1997; Ellis,

2000). PRA is a collaborative learning experience based on reversals of power. Local

perspectives replace outsiders' as drivers of the process. Chambers refers to this

reversal as 'turning over the stick' and allowing intended beneficiaries to lead processes

of change (1997). "The emphasis of these methods is on active involvement by

respondents, the outsider as learner rather than teacher, and qualitative prioritizing or

ordinal ranking of variables and options, instead of quantitative measurement of them"

(Ellis, 2000, p. 193).

PRA has been widely used in Africa, Asia, and Latin America by NGOs, aid

agencies, government ministries, and local communities (Chambers, 1997; Ellis, 2000;

Humble, 1998; Paul, 1998; Thomas-Slayter, 1995; Wordofa, 1998). It has been used to

diagnose and address issues as diverse as rural health care planning and delivery in

Ethiopia (Wordofa, 1998); revitalizing farmers' organizations in Estonia in the early

1990s (Thompson, 1998); and reorienting agricultural extension services to be more

responsive to farmers' needs in Grenada (Eckman, 1990).

As a more participatory form of development, PRA calls for outsiders to act as

catalysts for local people to decide what they want to do with the information generated

(Blackburn and Holland, 1998). This requires outsiders engaged in the process to treat

local people as subjects, rather than objects, of their own development, and "pay

particular attention to the way they behave when interacting with local people"

(Blackburn and Holland, 1998, p. 7). One of the strengths of PRA is its flexibility as a

methodology (Chambers, 1997). Rather than using a prescribed set of tools and

techniques, PRA is designed to be responsive to findings as they emerge (Slocum and

Thomas-Slayter, 1995). This requires skilled facilitators and collaborators who treat

participants as partners rather than subordinates.

This review of participation literature explored elements of its emergence as a

challenge to the status quo in development principles, policy, programs, and practice. It

included justifications, shortcomings and typologies to illustrate that participation

means different things to different people. These differences in interpretation, and their

implications, are central to this research.

The combination of gendered political ecology and participation literature created

a theoretical framework for this research that justifies a multidisciplinary perspective on

agricultural development projects. The framework gives legitimacy to examining

people's participation in relation to economic, social, political, and ecological factors

without giving a priori weight to any one of them.

A basic accounting of participation is necessary as this dissertation proceeds

through the methodology used for examining participation in this study (in Chapter 3),

establishing the political ecological context for that examination (in Chapter 4), as well

as conceptualizing and dissecting participation as a process (in Chapter 5).

This chapter establishes the methodology used in this research. After a brief

introduction, it presents principles and justification for combining qualitative and

feminist research approaches. Description of the data and specific research procedures

are followed by gendered political ecology and process analysis frameworks adapted

for this study. This is followed by an indication of the data utilized to address the

research questions and objectives presented in Chapter 1.

While a great deal of research has been done on participation in agricultural

development, the majority of existing resources is centered on tools and approaches for

facilitating farmer participation (e.g., Ashby, 2003; Chambers, 1997, 1998; Pretty,

1995). Although literature exists on the rationale for incorporating participation in

development (see Chapter 2), few studies have taken a holistic approach to exploring

how these efforts are planned and delivered. In particular, interactions-determining

target audiences, setting objects, staffing, and decision-making-between groups of

stakeholders within the hierarchy of development (policy makers, donors,

administrators, planners, practitioners and target audiences) have received little

attention. Even less attention has been paid to relationships among these processes of

incorporating participation in agricultural development within specific gendered

political ecologies like St. Lucia or Dominica, and for good reason. "One of the

challenges facing an ex-poste study of innovation is that there is no evidence left of the


critical steps in the innovation process, and that people have forgotten about them"

(Douthwaite, Keatinge and Park, 2002, p. 110). This study gathered and analyzed just

such evidence.

Qualitative Research
This study examines stakeholder interactions in two participatory agricultural

development projects against the backdrop of gendered political ecology in St. Lucia

and Dominica. This study provides a link between procedural aspects of incorporating

stakeholder participation in agricultural development projects and the specific gendered

political ecology context within which these processes took place. As such, it is

holistic, qualitative and descriptive in nature. Qualitative approaches allow researchers

to begin with general research questions rather than with specific hypotheses, compile

large amounts of observation, narratives, and interview data from participants, examine

these data in order to organize them into coherent forms, and tell the story that explains

the phenomena they have studied (Leedy and Ormrod, 2001). This study has been

undertaken according to this process.

Qualitative research is typically used to "answer questions about the complex

nature of phenomena, often with the purpose of describing and understanding the

phenomena from the participants' point of view," as is the case in this study (Leedy and

Ormrod, 2001, p. 101). It "implies a direct concern with experience as it is 'lived' or

'felt' or 'undergone,'" and seeks to transport the reader to the scene by telling an

evocative story (Sherman and Webb, 1988, p. 7). Qualitative research listens for the

voices of those studied, within the context(s) in which they exist, rather than controlling

for or excluding influencing factors. Qualitative researchers seek to discover new

insights into phenomena in their natural settings.

Patton (1990) identifies several types of qualitative research including basic

research, designed to contribute to the existing body of knowledge, and applied

research for illuminating societal concerns. In this tradition, this study leans toward

applied research because in addition to understanding and explaining a phenomenon, it

aims to generate potential solutions for application to real-world problems and


Patton's perspective on qualitative research is considered applied-he uses the

term evaluative-in that it is designed to "inform action, enhance decision making, and

apply knowledge to solve human and societal problems" (1990, p. 12). According to

Patton, questions in applied research emerge from the problems and concerns

experienced by people. Accordingly, the challenge of evaluative research is to get the

best possible information to the people who need it in a way that they can use it to

address real world issues. This research is designed for just that purpose.

The need for this study to stay grounded in the empirical world led to qualitative

research methods and this researcher's roles led to a theoretical orientation in heuristic

inquiry. This tradition is appropriate to this study because it brings to the front the

personal experience and insights of the researcher. "Heuristic inquiry asks: What is my

experience of this phenomenon and the essential experience of others who also

experience this phenomenon intensely?" (Patton, 1990, p. 71). This orientation

emphasizes meanings and knowing through personal experience, and treats the

researcher as the primary instrument in qualitative inquiry, demanding that the

researcher "have personal experience with and intense interest in the phenomenon

under study" (Patton, 1990, p. 71). Although focused on the experience of the

researcher, heuristic inquiry requires that others whose actions or reflections are among

the phenomenon being observed share the experience. Patton claims "rigor of heuristic

inquiry comes from the systematic observation of and dialogues with self and others"

(1990, p. 72).

Feminist Research

This research is grounded in-and built on-my experiences and reflections on

basic social structures and processes. The intersection of heuristic- and feminist

theories recognizes experience and education as sources of knowing (Harding, 1991;

Patton, 1990). Diverging from the inductive nature of qualitative research, heuristic

inquiry can be oriented toward a particular perspective such as feminism. In doing so,

Patton claims, "orientational qualitative inquiry begins with an explicit theoretical or

ideological perspective that determines what variables and concepts are most important

and how the findings will be interpreted" (1990, p. 86). A feminist perspective in

heuristic inquiry assumes that power differences exist within social structures and /

manifest themselves in social processes and interactions (Glesne, 1999).

Feminist approaches to research directly challenge the dominant, positivist

scientific paradigm by placing women's 'problems' at the center of the research i

process, designing the research to address these problems, and "placing the researcher

in the same critical plane as the research subjects" (Harding, 1987, p. 181). Feminist

research assumes that gender is a key dimension of human societies and cultures.

Research that ignores gender or treats it in the context of family or sex roles, for

example, "do[es] serious disservice to our understanding of social and cultural

dynamics" (Gailey, 1998, p. 205). Feminist research is grounded in women's everyday

experiences, addresses ethical questions in research such as power differentials, places

the researcher within the research in order to recognize the influences of cultural beliefs

and behaviors, and challenges dominant scientific philosophy and practice along the


Feminist research does not exclude men, nor does it start with men and

extrapolate to women. Feminist research seeks to reflect diversities of both male and

female realities. A concern with gender does not mean segregating women and men. It

means being able to focus on men and masculinity, in order to research the powerful as

well as the powerless. Feminist researchers accomplish this by hearing voice of their

subjects, acknowledging their own positionality, and practicing critical reflection and

self-reflection (Harcourt, 1994; Hartman, 1991; Naples, 2003; Rosser, 2000).

This is an applied research study that seeks to find solutions to issues around

incorporating participation in agricultural development identified in Chapters 1 and 2.

This is an in-depth study of: (1) the actions and interactions of individuals comprising

stakeholder groups; (2) decision-making within and between key stakeholder groups;

(3) the processes that resulted from these decisions; and (4) the influence on these

interactions and decisions of gendered political ecology of St. Lucia and Dominica.

Each of the two projects is an individual case, consisting of the same phases or

components, identified by FAO and undertaken under by contractors under FAO

supervision. Quantitative data on the processes being studied do not exist; therefore

quantitative analysis representing stakeholder interactions and decision-making was not


Issues of reliability and validity in this research must be addressed. While

heuristic and feminist inquiries allow the researcher to be part of the phenomenon

studied, efforts must be made to ensure that others can judge the outputs with some

certainty. This study includes several verification procedures to help ensure validity.

These procedures include, (1) prolonged and intense engagement as expressed in

heuristic inquiry (Patton, 1990); (2) clarification of researcher subjectivity according to

feminist inquiry (Gailey, 1998; Harding, 1991); (3) rich, thick description of the

phenomenon being examined; (4) member checking by sharing analytical thoughts and

drafts with other participants; and (5) triangulation by using multiple data collection

methods, sources, and theoretical perspectives.

According to Glesne, "multiple-data collection methods is the most common

form of triangulation in qualitative research" (1999, p. 31). This is undertaken to relate

different kinds of data "to counteract the threats to validity identified in each" (1999, p.

31). In addition to multiple data collection methods used in this study-participant

observation (field observations), interviewing participants, and document collection-

multiple kinds of data sources and multiple theoretical perspectives were used. Sources

of data were written records of field observations of decision-making among

stakeholder groups during the projects, documentation of these processes produced for

FAO, and secondary data pertaining to St. Lucia and Dominica. Theoretical

perspectives employed in this study include gendered political ecology and its effect on

implementing participatory development, and stakeholder participation.

The researcher's role as a participant in the study projects raises questions about

objectivity. However, as Patton (1990), Naples (2003) and Wolcott (1990) suggest, the

best way to address this issue is by being explicit about the researcher's role in the

study. The researcher was a graduate student contracted by FAO and the University of

Florida as a consultant on both study projects, as a project team member in St. Lucia

and as project leader in Dominica. As presented by Patton, evaluator effect pertains to

the impact of the presence of an evaluator on a process being evaluated; how does their

being there as a researcher affect the process. Given that evaluation of the study

projects did not begin until after completion of the second project, evaluator effect is

likely to not be relevant.

Research Questions
This study addresses the following research questions:

1. How did the confluence of political, economic, social and ecological factors in
St. Lucia and Dominica contribute to the need for intervention?

2. Who were the stakeholders in each project and how did they participate in the
different phases of each project?

3. How did interactions within and between several stakeholder groups affect
outcomes of each of the participatory development projects in this study?

Objectives of this study are

* Objective 1: Conduct a gendered political ecology analysis of St. Lucia and
Dominica. Identify major themes, and associate their historical emergence with
the current situation.
* Objective 2: Identify key stakeholders in each of the study projects including
their stake and expectations of the projects.
* Objective 3: Identify and document decision-making processes undertaken in
preparing for, implementing and reporting on each phase of the two study
projects. Indicate how stakeholders did or did not participate in decision making
throughout both projects, including who participated, how, when, and who
facilitated their participation.

Objective 4: Determine what led to conflict and/or collaboration in each project,
how it was managed, and how that impacted outcomes.

The Data

The data for this study come from two development projects conducted for FAO

and have been obtained according to Patton's requirements for qualitative data

collection: "(1) in-depth, open-ended interviews; (2) direct observation; and (3) written

documents" (1990, p. 10). The data are derived from fieldwork, from the perspective

of a participant observer who engaged in the activities under examination. The data

include direct quotations from people about their experiences, detailed descriptions of

people's activities, behaviors and actions, and "the full range of interpersonal

interactions and organizational processes that are part of observable human experience"

(Patton, 1990, p. 10).

Data compiled during the study projects are listed in Appendix A and consist of

project plans and terms of agreement from FAO, e-mail correspondence, field notes,

calendars of activities, meeting notes, and numerous written reports detailing activities,

behaviors, and actions of participants in St. Lucia and Dominica. Additional data

resulted from fieldwork with EOs. These include farmer demographic, socioeconomic,

agro-ecological, and livelihood data. These data tell the procedural story of the projects

in St. Lucia and Dominica and are augmented by secondary data to supply the gendered

political ecology contexts.


The data collected and analyzed for this research were derived from project

planning and implementation in St. Lucia, January through August 2000, and similar

processes for Dominica from December 2001 through April 2002. Secondary data

sufficient for establishing gendered political ecologies of St. Lucia and Dominica were

compiled beginning in January 2003.

The first research question addressed by this study-How the confluence of

political, economic, social and ecological factors in St. Lucia and Dominica contributed

to the need for intervention-is informed by secondary data sources and field notes

derived from both projects. Secondary data sources included literature on historical,

economic, political, ecological, and social factors in St. Lucia and Dominica. The data

were based on ethnographies, census, and historical accounts. The MERGE framework

was used to identify and organize relevant data into three dominant themes that explain

the situation in St. Lucia and Dominica at the time of the study projects. These themes

are presented in Chapter 4 and meet Research Objective One.

The second and third research questions in this study-Who were the

stakeholders in each project and how did they participate in the different phases of each

project? How did interactions within and between several stakeholder groups affect

outcomes of each of the participatory development projects in this study? -are

addressed by analysis of project data. The MERGE framework was used to analyze

and organize project data according to local level stakeholders, project activities, and

gendered natural resource use patterns derived from fieldwork.

All available project data (see Appendix A) were sorted by country and date.

Documents pertaining to St. Lucia were reviewed in search of a logical delineation of

project activities. The emergent list was compared to documents from Dominica and

resulted in a chronological list of activities common to both projects. They are as


1. Pre-departure research and planning
This phase covered activities such as identification of project team members, proposal
preparation and submission, background research on current and historical conditions in
the target country, identification of available resources, and listing necessary
information not available through U.S. based research. This phase lasted nearly three
2. Training design and delivery
This segment of the project involved interpreting and synthesizing expectations of FAO
into a five-day training design, creating session plans, and delivering the training. This
phase lasted approximately six weeks.
3. Field work, interviews and focus groups
This phase was comprised of fieldwork, individual farmer interviews, and focus groups.
The project teams and personnel from collaborating agencies in St. Lucia and Dominica
undertook these activities.
4. Data synthesis and analysis
This phase encompassed finalizing field-data entry, synthesis, and analysis by project
teams and EOs. It included planning and delivery of a final workshop for presenting
findings of the fieldwork to the FAO representative and MOA personnel.
5. Final Wrap-up
This phase represented production and dissemination of final outputs to project partners
according to contractual obligations, and lasted nearly three weeks.

Each of these phases was further divided into sub-activities according to project

documents, project plans and calendars from both projects (see Table 3-3 for the

resulting project analysis framework to be used in Chapter 5).

Organization of project phases and activities was followed by identification of

stakeholder groups represented within the data. Once stakeholder groups were

identified and delineated, data pertaining to each were examined for evidence of their

expectations and interests in the overall projects, and for each project phase and

activity. The results of this step responded to Research Objective Two.

All project data from each country were identified and indexed to indicate which

phase and specific activity to which they pertained. The resulting groups of data were

examined for distinct procedural decisions within each activity, and associated with the

stakeholder groups that made them, which other stakeholder groups they impacted, and

how. These decisions, who made them, with whom, affecting whom, and resulting

outcomes met Research Objective Three.

Data were further sorted according to stakeholder group, project phase, activity,

and decision making to discern examples of power sharing, collaboration, expectations,

or conflict among stakeholder groups within the project analysis framework. This step

met Research Objective Four.

Similarities between the countries, the projects and the personnel involved led to

hypothesizing about similarities and differences between the two projects, and what

accounted for them. A review of literature on participation in development along with

further critical reflection with other project personnel produced anecdotal accounts and

explanations for the variation within and between the two projects. While explanations

remained elusive, ideological, contextual and procedural differences between the two

projects began to emerge and highlighted some potentially important issues.

White's (2000) framework of approaches to incorporating participation

illuminated a number of the ideological issues that had emerged. Her typology of

participation was useful for examining procedural aspects of incorporating stakeholder

participation in development. Her framework presents a value-neutral continuum of

participation, thereby eliminating the need to determine good, better or best practices.

The framework also accounts for stakeholders at many levels, and allows for analysis

of interactions among all levels of decision-making authority.

Contextual issues pertaining to the study projects fit well within political ecology

and specifically within the MERGE framework (Schmink, 1999). Gendered political

ecology allowed for incorporating political interests and struggles, and unequal

relations of power in mediating access, control and conflict in resource use, which were

at the heart of this study. The procedural differences proved very interesting given the

contractual and staffing similarities between the two projects and proved more difficult

to disentangle.

Gendered Political Ecology Framework

As indicated in both the introduction and literature review, this study used a

theoretical framework within which to examine stakeholder decision-making and

interactions based on gendered political ecology and participation. In addition to being

a conceptual framework as presented in Chapter 2, the MERGE framework serves as an

analytical tool when applied to conservation projects. However, because the

framework was designed for community based conservation and not agricultural

development, it needed adaptation to be a useful analytical tool for this study. The

structural integrity of the framework was maintained in terms of the progression of

levels of analysis beginning with political ecology analysis, and continuing with macro

level stakeholder analysis, stakeholder analysis within the community, gender relations

and resource analysis, and project analysis. However, the specific questions posed

within each of these levels of analysis needed to reflect the site-specific nature of the

study projects: participatory agricultural development, rather than community-based

conservation. As such, each level of analysis now includes research questions that

reflect planning and delivering participatory development projects in St. Lucia and


Table 3-1. Adapted MERGE Framework, Section I.
Type of Analysis Research Questions
Political Ecology Historical
Analysis: What are the key historical periods that have shaped
current socioeconomic and agro-ecological conditions?
The potential for Political
participatory development How are these periods distinguished by changing
is constrained or government policies?
enhanced by historical, Socioeconomic
ecological, cultural, What are the connections to international, national,
socioeconomic and regional and local markets for local resources?
political factors at diverse Which groups have been involved with these markets
scales. historically, and what was their relationship?
How have patterns of land use and resource use changed
during different historical periods?
What are the key agro-ecosystems in this setting? How
are they being used and how is that use changing?
What kinds of production areas or zones exist and how are
they managed?
Macro-level Stakeholder Who are the stakeholders in the agricultural sector?
Analysis: How are their interests defined?
How do they conflict?
Multiple stakeholder What are the possible bases for cooperation or
groups are involved in complementarity?
direct or indirect How were they involved in historical agricultural practices
negotiation for input into and policy?
agricultural program What state and non-governmental organizations are
planning. Their interests involved in the area?
may be complementary What community organizations exist?
and/or in conflict. What kinds of property regimes and resource management
institutions currently exist?
For which groups do they regulate access/control to key
Source: adapted from Schmink, 1999.

In addition to setting the political ecological stage for this study, the adapted

MERGE framework served as an analytical tool as well. Given the importance of

context and process to this study, I divided the framework into two discrete sections,

the first of which relates to context while the second relates to process. The first

section of the analysis, presented as Chapter 4 combines (1) gendered political ecology

analysis of the study countries and (2) macro-level stakeholder analysis. Presenting

these levels of analysis together is justifiable because they explain the combination of

events, decisions, and decision makers who are responsible for the current reality in St.

Lucia and Dominica. By following the adapted MERGE analytical framework, Chapter

4 allows for critical examination of factors that affect current resource use strategies.

The second section of the adapted MERGE framework, focusing on process, is

presented in Chapter 5 and is used to analyze site specific conditions within the

agricultural sectors of each country as well as activities of and interactions between the

subjects of this study. This section begins with local-level stakeholder analysis and

proceeds with project analysis, and gender relations and resource analysis (see Table 3-

2). The first two levels of analysis in this section were informed by data derived from

the study projects through participant observation. The third level, gendered resource

use analysis, is based on survey and focus group data collected and analyzed during the

projects in St. Lucia and Dominica. By design, Chapter 5 analyzes decision making by

stakeholders actually present during the projects including representatives from the

sponsoring agency, the project team, MOA personnel, EOs, and farmers.

Table 3-2. Adapted MERGE Framework, Section II.

Type of Analysis

Micro-level Stakeholder

Participation by different
groups within local
communities can sustain
or contribute to improved
agricultural livelihoods.

Project Analysis:

Mutual learning
processes improve the
ability of local actors to
negotiate their interests in
agricultural livelihood

Gender Relations,
Resource Analysis:

Gender roles, identities
and relations differentiate
people's connections with
natural resources and
agro-ecological systems.

Research Questions

What scales are involved in community-based agricultural
development efforts?
In what ways does each community participate?
Within the community, who participates and how?
Who are the relevant stakeholder groups within
heterogeneous communities?
Who represents them?
Which local groups have been empowered?
What kind of support or benefits do they receive?
How are their activities affected?
In which decisions have they participated?
How long has local knowledge been recognized and
incorporated in planning?

How has agricultural development planning been done in
the past?
Who were the key actors (outsiders/local) and how did
they interact?
What were the defined objectives?
How was the project implemented?
What problems arose and how did they affect the project?
What kinds of training experiences have been offered to
To whom: #s, types and representation of participants?
For what purpose?
In what way was a focus on gender and community
participation incorporated?
What were the results of these training experiences?

What are the patterns of livelihood strategies practiced by
different groups of households?
How do gender relations differentiate links with key
natural resources and agro-ecological systems, as well as
attitudes toward development?
What are the key groups differentiated by gender and other
key social dimensions?
How do these gender differences affect resource use and
agro-ecosystem function?

Source: adapted from Schmink, 1999.

Process analysis framework

The adapted MERGE framework offers the multi-faceted perspective necessary

for examining participation in development projects but leaves little room for the


temporal aspects of interactions, or the dynamic nature in which actors move in and out

of the action. Therefore I present a process analysis framework useful for organizing

and analyzing stakeholder interactions and decision-making throughout the projects.

Table 3-3. Process analysis framework for participatory development projects in St.
Lucia and Dominica.
Phase Decision makers) Decision
Pre-departure research &
1. identification of project
team members
2. proposal preparation
3. background research
Training design and delivery:
1. contract analysis
2. training objectives &
3. session plans
4. session delivery
Field work:
1. structure
2. data needs
3. schedule
4. daily processing
Data synthesis and analysis:
1. data entry
2. analysis
3. workshop planning &
4. presentation of results
1. report writing

The process analysis framework corresponds to five distinct phases of the

participatory development projects. Each phase is analyzed in terms of who was

responsible for the decision, what decisions were made, and the nature of the decision

making process. Analysis of these decisions and actions by stakeholders answer the

research questions posed in this study and are presented in narrative form.

This study is limited by the researchers' short-term involvement in relatively

long-term projects. In each case, the researcher only has data on the PRA component

of the 18-month long projects. This prohibits analysis of training and implementation

on all subsequent project phases.


This study is descriptive, inductive and qualitative research. The choice of

feminist theory reflects commitment to research as part of a process of change that

brings women (as farmers, extension officers, administrators, policy makers,

development practitioners, and researchers) to the forefront of development efforts.

This research examines two projects designed to incorporate farmer participation

into agricultural development projects in St. Lucia and Dominica. Both the context of

the projects and the processes of designing and implementing them are analyzed.

Chapter 4 presents gendered political ecologies of St. Lucia and Dominica. Chapter 5

analyzes stakeholder participation in designing and delivering the PRAs.


This chapter presents a gendered political ecology analysis of St. Lucia and

Dominica as the setting of the study projects. According to the MERGE framework,

analysis of key historical periods and corresponding natural resource use patterns can

serve as a road map for understanding factors that constrain or enhance efforts at

participatory development (Schmink, 1999). This multi-faceted assessment is a

precursor to exploring and explaining stakeholder interactions within the specific

development projects in this study. While these projects were designed and undertaken

against a backdrop of banana wars brought on by globalization, they cannot be fully

understood independently of the historical events leading up to the current situation.

In keeping with the objectives of this study, this chapter establishes the context

within which the study projects took place, and is both descriptive and analytical. It

blends the first two levels of analysis suggested in the MERGE framework; political

ecology and macro-level stakeholder analysis (Schmink, 1999). This chapter is divided

into three sections, each of which includes historical, political, economic, social and

ecological dimensions. After a brief introduction and general history of the islands, the

first section follows the justification, emergence, and evolution of monocrop agriculture

in St. Lucia and Dominica. The second section introduces land settlement and use

patterns, with family land as a dominant land tenure system. The third section

examines key events in St. Lucia and Dominica from the 1940s until present day

including nationalism, the role of Transnational Corporations (TNCs) and the banana


Rather than presenting macro-level stakeholder analysis as a separate section, it is

woven within each of the three sections presented here. Through the stories in this

chapter, those stakeholders influential within the agricultural sector, the nature of their

interests, and how their influence affected agricultural in St. Lucia and Dominica

clearly emerge. The remaining levels of the MERGE framework, including gendered

resource analysis and project analysis will be addressed in Chapter 5.

St. Lucia and Dominica

Dominica and St. Lucia belong to the Windward Islands chain of the Caribbean.

In addition to their geographic proximity, both islands share basic geological and

ecological features. They are heavily forested volcanic islands characterized by sharply

rising peaks, steep slopes and fertile river valleys. They share an essentially north-

south orientation with the Caribbean Sea to the west and the Atlantic Ocean to the east.

As a result, both islands have cliff-ridden east coasts exposed to ocean currents. Major

ports in both countries, centers of trade and commerce, are on the western, or leeward,

coast. Population centers of these islands are typically along the coast, with scattered

communities farther inland where topography and roads permit.

Biophysical factors limit natural resource based livelihoods in St. Lucia and

Dominica. They are small countries, outward expansion is impossible, their natural

resource based is finite and proximal to population centers, their volcanic origin leaves

them with only a thin layer of topsoil, and they are susceptible to ecological

disturbances such as hurricanes that compromise natural resources even further

(Kowalewski, 1981, p. 46). St. Lucia supports over 160,000 people on 616 square


miles, much of it steeply sloping mountainside, while Dominica supports nearly 70,000

people on 754 square miles of even more rugged land (CIA, 2003).

Socioeconomic factors also limit people's ability to improve their livelihoods in

these countries. Population pressure, unequal distribution of wealth and access to land,

and limited market access for buying and selling goods, curtail options available to

residents. Per capital Gross Domestic Product for each country was estimated at $5,400

in 2002, and the agricultural sector employed the largest percentage of the labor force:

22% in St. Lucia; and 40% in Dominica (CIA, 2003).

These are just a few of the biophysical and socioeconomic factors facing farmers

in St. Lucia and Dominica today. And when combined with historical, political and

cultural factors, restrict the options available to residents trying to sustain, let alone

improve, their livelihoods. This chapter tells the story of the confluence of factors that

have generated the current situation, and shape future options.

British and French

St. Lucia and Dominica were 'named' by European explorers within a few years

of each other in the late fifteenth and early sixteenth centuries. The first European

arrivals found heavily forested islands with lush vegetation, ample water, and groups of

Caribs, descendants of native South Americans. The first one hundred years of

European contact were intermittent and frequently ended in violence between the Carib

and Europeans. The first European settlement on either island began in the 1638 when

a British councilor and some three hundred men arrived in St. Lucia and established a

base. This settlement enjoyed 18 months of peace before a Carib war party arrived,

seeking revenge for trouble on another island. The few settlers who survived the attack

fled, effectively ending British interest in St. Lucia for another 20 years. The French

appeared in 1651 when two representatives of the French West India Company laid

claim to the island. Eight years later, ownership disputes between the French and

British ignited hostilities that would endure throughout the Caribbean region for 150

years. Few efforts at settling Dominica during this time were recorded. In addition to

the native Carib, arrivals faced physically daunting cliffs and few coves suitable for


Between the mid seventeenth and the early nineteenth centuries, the British and

French fought for control of St. Lucia nearly 15 times. Dominica fared only slightly

better due to less hospitable conditions. The British took St. Lucia from the French for

the last time in 1803. Dominica was granted to Britain by treaty in 1805, and was the

last Caribbean island to be settled. Although characterized by history as a period of

fighting, Europeans had managed to establish settlements during the periods of relative

peace. According to Trouillot, "between 1660 and 1805 there were long interludes of

political "stability" and economic growth" (1988, p. 53). He suggests that change of

administrators did not necessarily mean abandonment of settlements. For example, the

French established the first permanent settlement in St. Lucia, Soufriere, in 1746, and

within forty years there were 12 settlements and a large number of sugar cane

(Saccharum officinarium) plantations throughout the island. During the American

Revolutionary War, French troops drove the British out of Dominica, and found that

"Caribs and runaway slaves made raids upon the English settlements for plantains,

bananas, ground provisions, and stock" (Harrison, 1935, p. 69) indicating existing

settlements. The prevailing economic system and natural resource use patterns

survived shifting control of the territories.


British control of both islands preceded an additional thirty years of slave-based

plantation production based on sugar in both territories and coffee in Dominica. In

addition to estates there were numerous smallholders carving out a living from

agriculture and timber extraction further in the interior during this time. Following the

end of slavery in the islands in 1834, production systems and resource use changed

considerably, most notably in terms of labor and land distribution. Despite those

changes, residents of St. Lucia and Dominica continued reliance on an export economy

based on a series of monocrops; a situation that persists.

The hundred years of British post-slavery rule in St. Lucia and Dominica (1840s

to 1940s) saw shifting demographics, natural resources use, and economic development

patterns. Nonetheless, it was a period of relative stability, with guaranteed, if not

shifting, markets for produce, and efforts to apply technological advances to production

efforts. The relative stability of this period eroded with the worldwide economic

downturn of the 1930s and the beginning of World War II. The 1940s saw the rise of

global nationalist ideals to which Caribbean territories were not immune, as well as

Britain's need to divest itself of economic responsibility for its colonies. The 1950s can

be characterized by the rise of multi-national companies, bananas, and budding regional

cooperation agreements. In the 1960s, both territories were granted the right of self-

governance, maintained economic and political links with regional and global partners,

and continued reliance on monocrop, export agriculture. In the midst of a global

economic downturn, Dominica became an independent country in 1978 with St. Lucia

following gaining independence the next year.


Regional political unrest (Grenada), structural adjustment policies, and a series of

devastating hurricanes tested the resilience of the people, institutions, and ecologies of

the Eastern Caribbean in the 1980s and 1990s. As the weaker cousins of the regions'

more powerful countries-Jamaica, Trinidad and Tobago, and Barbados-St. Lucia

and Dominica adopted the same institutional and economic belt tightening prescribed

by the International Monetary Fund (IMF). On the road to structural recovery in 1992,

St. Lucia, Dominica, and other Commonwealth countries received bad news: U. S.

based TNCs had filed suit with the World Trade Organization (WTO) to dismantle their

preferential trade agreements with the British. The suit was filed in order to open the

British banana market to produce from Latin American countries, brokered by Dole,

Del Monte and Chiquita.

Bananas did not create the current economic crisis in St. Lucia and Dominica;

they were just the last in a long line of decisions that sealed the fate of these two

countries. The story is complex with socioeconomic, political, and ecological origins.

Current socioeconomic and agro-ecological conditions in St Lucia and Dominica can be

easily traced to European settlement patterns. Their role as agricultural islands was

determined as early as the late 1600s, and common threads run through the intervening

three hundred years.

Three themes deserve particular attention within a political ecology examination

of St. Lucia and Dominica. The first is the history of reliance on monocrops beginning

with plantation systems and continuing through peasant systems after the abolition of

slavery. The second theme is settlement patterns, land use and tenure schemes over the

last four centuries. The third thematic area incorporates the rise of nationalism, TNCs

and the banana wars within the last sixty years. Each of these sections contains

germane historical, political, socioeconomic and ecological issues, and was selected for

further investigation based on its relevance to current conditions.

The Trap of Monocrops

"The oddity of being a massive agricultural exporter while at the same time being

a net importer of basic foodstuffs was structured into the Caribbean economy from the

colonial epoch and continues to be a major problem today" (Kowalewski, 1981, p. 46).

Having disposed of the Carib in St. Lucia, and reaching an uneasy peace with them in

Dominica, European colonizers set about establishing moneymaking estates in the early

1700s. The importance of Caribbean territories to British and French economies cannot

be overstated. "Caribbean colonies did not enter into international trade through the

accidental discovery of new opportunities. They were created for trade and surplus,

according to the European mercantilist design" (Trouillot, 1988, p. 22). Kowalewski

suggests Caribbean islands were subjects of the earliest relatives of modem TNCs, the

West Indies Companies (1981). Therefore their development and distribution of wealth

has always been affected by global economic factors. Estate owners in the Caribbean

were spokes in the European capitalist wheel. They bought slaves, mainly from Africa,

and used them to convert raw materials into commodities needed to fuel the industrial

revolution and the increasingly consumer oriented population in Europe. It was, as the

saying goes, a buyers market, and the buyers lived in Europe.

Slave labor was used in the Eastern Caribbean from the early 1700s until

emancipation in 1834. Women and men were purchased by estate owners, housed on

the property, and used to clear land, produce and process crops, and feed and replace

themselves. Cash was typically in short supply and owners could not afford to

purchase food that slaves could produce themselves. Both male and female slaves

were seen as productive laborers and owners rarely differentiated between male and

female labor domains. The stereotypical gender division of labor, men cultivating cash

crops while women cultivate food, did not apply to plantation labor in the Eastern

Caribbean (Momsen, 1993), and does not apply today (Grossman, 2000). Slave owners

discouraged legal unions among slaves leading to women and their children as most

common social unit in the islands. Remnants of this system are visible today in the

frequency of female-headed households (36 % region-wide, United Nations, 2000, p.

42), high percentage of women and men, (22% and 19% respectively), not married by

age 45 (UN, 2000, p. 26), and women's primary roles in organizing and sustaining the

family (UN, 2000, p. 42).

The demand for slaves remained high in those islands under British control,

where not coincidentally, plantation production centered on one or two crops-sugar in

St. Lucia, and sugar and coffee in Dominica. Sugar was the first crop in St. Lucia and

Dominica to turn substantial profits for plantation owners (Baker, 1994; Trouillot,

1988) and required considerable land and labor to produce. Ecological conditions in St.

Lucia were more conducive to sugar production but owners in Dominica managed

sizeable production on less suitable land. "Sugar came to be seen as synonymous with

economic growth and the only logical investment" in the Eastern Caribbean in the mid

1700s (Baker, 1994, p. 60). This realization was bad news for laborers and the land.

Sugar cane production was best done on large tracks of land, and processing sugar was

labor intensive.

Planters made no attempts to improve techniques for sugar production or to
experiment with alternatives. Their focus was narrowly avaricious and rapacious

and can be summed up as: Get as much as possible out of the land and the slave
for nothing. The results were increasing soil erosion, dropping yields, and a
horrendous slaughter of slaves from neglect, overwork, and punishment. (Baker,
1994, p. 60)

While sugar dominated plantation production in St. Lucia from the early 1700s

well into the twentieth century, Dominica followed a slightly different path. As British

settlers had claimed the land available for sugar production, French arrivals sought an

alternative cash crop and introduced coffee in Dominica in 1740. Coffee, and not

sugar, dominated exports from Dominica until the 1830s when coffee blight struck the

plantations, coffee prices dropped and sugar prices rose. Having entered into global

capitalism with both sugar and coffee, plantation owners in Dominica were able to

reconfigure their land and labor to accommodate sugar.

From Slavery to Peasantry

Island economies continued to rely on estate-based export agriculture following

the abolition of slavery, produced now by peasant laborers working for wages.

Emancipation in the Eastern Caribbean in 1834 was followed by a four-year period of

apprenticeship whereby all former slaves received a nominal salary for their labor, 0.5

acres and Saturdays off to produce their own food. Upon emancipation, former-slaves

tended to establish themselves outside the estate boundary, on government property

(Crown Lands), and began producing provisions such as sweet potatoes (Ipomoea

batatas), tannia (Eanthosoma sagittifolium), plantains (Musa sp.), and beans (Phaseolus

vulgaris) for sale and their own consumption. As a newly formed peasantry they

acquired or appropriated small plots, enlisted the labor of family and neighbors for help

producing provisions, and eventually produced exportable agricultural commodities.

According to Trouillot "Once slavery ended in 1834 and the threat of physical

bondage disappeared, the Dominican cultivators sought control of the units of

production, including those that did not produce for export" (1988, p. 67). His

suggestion that former slaves wanted to work with and for former slaves is borne out

when the end of apprenticeship brought a sudden decline in numbers of resident

laborers on estates. The implications of this labor arrangement are far reaching. In

both St. Lucia and Dominica the phrase "coup de main" still refers to exchange labor

arrangements, called on by farmers during periods of intense labor such as land

preparation or weeding.

Emancipation in 1834 signaled a shift from slavery to peasantry that deserves

exploration. For the purpose of this exploration slavery is considered to have ended

after the four-year period of apprenticeship. Unlike populations in other developing

nations, peasants in St. Lucia and Dominica have long operated within local, regional,

and trans-continental market economies. In fact, ancestors of present day farmers in St.

Lucia and Dominica arrived after the first European colonizers. Therefore their arrival

and growth occurred after "the region was fully incorporated within the modem world

economy" (Trouillot, 1988, p. 21). Nor were their ancestors blessed with indigenous

knowledge and generations of experience on the land; other than maize, manioc and

sweet potatoes, "they depended principally upon plants and animals brought into the

region" by Europeans (Trouillot, 1988, p. 21).

There was certainly a transition period when newly freed slaves in both countries

developed survival strategies. Some continued working on estates as wage laborers

while producing their own food on the margins of the estate. Others left the estates

completely and while rarely able to establish large-scale production of the dominant

cash crop, produced and marketed other financially viable crops such as provisions-

cassava (Manihot esculenta), yams (Dioscorea), breadfruit (Artocarpus altilis), sweet

potatoes, and plantains-and cash crops like cocoa, cotton (Gossypium spp.), indigo

(Indigofera hirsute), ginger (Zingiber officinale) or nutmeg (Myristicafragrans).

Production and especially marketing of non-dominant crops soon became the

domain of women in Eastern Caribbean. Owing to their historical roles in production,

women in the region were never far removed from the economic aspects of agriculture.

There was no neat dividing line between their public and private sphere activities

(Besson, 1993; Momsen, 1993; Ho, 1999). Women were not relegated to household

subsistence production activities but were economically active producers and

marketers. While not suggesting that peasant farmers in St. Lucia and Dominica ever

marketed everything they grew, they have always operated in close proximity to

production systems that put a premium on particular crops for export. Peasants in St.

Lucia and Dominica always knew that by producing these crops they could generate


Because sugar remained the dominant crop in St. Lucia from its establishment in

the early 1700s until the 1950s, this suggests that while moving off the plantations,

peasants continued working on the estates for wages. Large river valleys suited its

agronomic needs as well as the waterpower crucial to processing. Former slaves, their

children, grandchildren and subsequent generations settled in the growing cities, on the

slopes above the estates, and on unoccupied Crown Land deeper in the interior. There

they produced crops such as plantains, dasheen (Colocasia escuelenta), tannia, sweet

potatoes, mangoes (Mangifera indica), breadfruit, yams and bananas both for market

and home consumption. They cultivated peanuts (Arachis hypogaea), rice (Oryza

sativa), citrus (Citrus aurantifolia, sinensis, and paradise), and pineapples (Ananas

comosus) for sale in city markets and for export to other islands. Women became the

main conduits from field to market. Women's economic roles included producing as

own account farmers or wage laborers, selling in local markets, buying produce at the

farm gate and selling it to wholesalers for export, and exporting to other islands in the


While St. Lucia's crop of reliance on sugar persisted for nearly 250 years, the

situation in Dominica was quite different. As indicated by the coffee/sugar overlap

from the mid 1700s until the mid and late 1800s respectively, neither Dominica's estate

owners nor its peasants enjoyed long periods of economic stability. The decline of

sugar production in Dominica after 1860 was contrasted by the rise in limes as the new

commodity of choice for both growers and buyers.

Lime production and exports from Dominica began in earnest in the 1860s and

within a decade, lime-producing estates were making more money than sugar

producers. Available land was planted in limes in hopes the market would last another

six or eight years until the trees bore fruit, and it did. By the 1890s, lime was the

dominant crop in Dominica and Dominica became the world's largest producer of

limes. Limes had several advantages over other crops: (1) the trees grew on most

island soils; (2) large tracks of flat land were not necessary; (3) farmers could plant few

trees and still recognize a profit; and (4) it could be sold raw, as concentrated juice, as

essential oil, pickled, as whole fruit, or as citric acid. For eighty years, lime production