Social institutional profile of the Cayes Plain Basin

Material Information

Social institutional profile of the Cayes Plain Basin
Coffey, William J.,
Lewis, Laurence A.
Haugue, Ann B.
Jauwerysen, Herman J.
Clark University -- Institute for Development Anthropology
Place of Publication:
Worcester, Mass.
Clark University, Institute for Development Anthropology
Publication Date:


Subjects / Keywords:
Farming ( LCSH )
Agriculture ( LCSH )
Farm life ( LCSH )
Spatial Coverage:
North America -- United States of America -- Massachussetts


Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact Digital Services ( with any additional information they can provide.
Resource Identifier:
23032334 ( OCLC )

Full Text
William J. Coffey, Laurence A. Lewis Ann B. Hauge
with the assistance of
Herman J. Jauwerysen
Haiti Regional Report
Clark University/Institute for 'Development Anthropology
Cooeraiv Aremne (UeA
............ . .. .
Clark University Institute for Development Anthropology
International Development Program Suite 302, P.O. Box 818
950 Main Street 99 Collier Street
Worcester, MA.01610 Rinnh;)mtrtn Nv 1 'QOA9

William J. Coffey, Laurence A. Lewis Ann B. Hauge
With the Assistance of Herman J. Lauwerysen
Clark University/Institute for Development Anthropology Cooperative Agreement (USAID)
January 1984

In this proposed plan for coordinated rural regional
development, the two primary goals are to improve the rural quality of life and at the same time to improve the land resource base. These two objectives are only possible to attain by considering all lands within the targeted areas in regard to their pot ential capabilities for development and resource improvements. Among others, these resource improvements include possible production capabilities related to food, fuel (charcoal), and timber; physical factors including hydrology, drought likelihood, slope stability, and erosion-siltation rates; infrastructure improvement including storage facilities and transportation; economic improvements in marketing, credit, and processing; and improvements in the agricultural use of lands including seed improvements, new-crops, conservation practices and field consolidation.
The physical factors affecting production and environmental quality in a land-rural based economy are largely related to the nature of the drainage basin in which the lands are situated, including climatic, lithologic, and topographic properties. The degree to which development activities are conceptualized within the physical environmental constraints--that is, work in harmony
with nature--will determine to a large degree interactions of the human activities in maintaining or improving the area's environmental situation.
Regional development may be implemented within any boundary, but planning environmental improvements, infrastructures, and mobilization of human resources will encounter significant restrictions unless carried out within meaningful units. In a river basin all natural resources can be controlled from the highest to lowest lands through the regulation of surface waters. For this reason basin boundaries delimit a meaningful planning unit in most humid areas where agricultural expansion is the primary goal. Almost all river basins are formed through a system of tributary or smaller stream basins. Because of this structural property, a development plan can be executed gradually in a regional context sub-basin by sub-basin according to the availability of human, financial, institutional, and other development requisites. When the planning strategy is expanded from tributary to tributary basin on a systematic basis, upon completion of the planning and implementation stages of the project, the whole river basin's socio-economic and land resource base will be improved.
From the physical perspective, a planning strategy attempts to initiate improvements in the tributary basins located in the major watershed areas of the river basin, then sequentially incorporates the remaining basins in the downstream direction.

This approach reflects the natural unidirectional flow of energy and matter within river systems. However, human systems such as market and social networks needed to initiate and absorb physical changes usually have different functional relations than the upstream-downstream spatial pattern of the river network. Until the regional river basin development project is complete, the human physical domains must be integrated into a complementary planning unit to insure continuous acceptance and maintenance of completed land use and institutional improvements.
Most projects are begun and completed with various ranges of success in the short term. However, even the most successful projects often need to be restarted after a number of years. All too often with the completion of a project in the formal sense (termination of external funding), its dynamics gradually wind down, and lands revert to their original status. A number of years later the project in new trappings is reintroduced and the old scheme is recreated. With the increasing population, deteriorating hydrologic conditions, and ongoing soil degradation, a new project incurs higher costs and less degrees of freedom in meeting its expected goals. Many Haitian agricultural projects have been analogous to the Alice Through the Looking Glass phenomenon of running forward ona reverse running conveyor belt. To remain where you are, let alone to improve the situation, requires even more effort. For example, ever since colonial times when irrigation was introduced in Saint-Domingue, a vicious cycle of abandonment and rebuilding of irrigation has been set into motion. Yet with each rebuilding scheme, potential has decreased due to the chronic processes of deforestation and soil erosion that directly affect water and soil resources. Superimposed on these ongoing physical factors are a plethora of human variables such as land tenure, organizational deficiencies, and increasing population densities that diminish the likelihood of success.
Perhaps the one underlying design flaw in most previous
rural-based projects was that projects were conceptualized within the very narrow confines of the immediate project. Offsite project variables and impacts were rarely considered. Likewise, the engineering and public works components were usually stressed while the human infrastructure, including organization and maintenance components, were downplayed. It is for these reasons that this plan stresses a regional rather than sectoral approach and that the conceptualization of the rural development strategy includes both the physical and human components in a cootdinated framework. The river basin, as a whole, as opposed to specific units within it, is the focus for the development strategy since all areas within the basin have direct and indirect impacts throughout the entire system. Thus, the coordinated rural regional development strategy that is presented in the following sections of this report attempts to incorporate existing processes as they operate in rural areas. It attempts to stimulate the region's latent potential as well as to strengthen its existing capabilities.

FORWORD .....................................................
LIST OF TABLES .............................................. o ... iv
LIST OF FIGURES ........... ........... ............ v
CHAPTER 1 Introduction .... ............. o ...... o ......... 1
CHAPTER 2 An Approach to Rural Regional Development: Planning
and Implementation in the Haitian Context ... 7
CHAPTER 3 The Physical Setting: Active Processes
and the Existing Environmental Situation ...... 18
CHAPTER 4 Market Structure ........... o .................... o ... 38
CHAPTER 5 Agriculture: The Physical Economic Interface .. ..... 66 CHAPTER 6 Institutional Context .......................... 91
CHAPTER 7 A Coordinated Rural Regional Development Plan
for the Cayes Plain Basin .... ........ ........... 149
CHAPTER 8 Implementation Strategy .... ......... o ........... 168
SUMMARY AND CONCLUSIONS ....... o ...................... o ........ 189
APPENDIX A ................... o..o ...... o ....... o.. 191
BIBLIOGRAPHY ............ o ...... o ............................ o. 195

Tables Page
3.1 Class Characteristics and Recommended Land Use .......... 25
3.2 L'Acul River Basin Topographic Units .................... 29
3.3 Monthly Moisture Availability Index for L'Acul
River Basin ........................................... 33
3.4 Soil Classes of Upper L'Acul ............................ 35
4.1 Topologies of Haitian Markets .......................... 40
4.2 Market Days in L'Acul River Basin ....................... 43
4.3 Market Characteristics .................................. 46
4.4 L'Acul River Basin Marketing System Hierarchy ........... 49
4.5 Origin--Destination Matrix: Commodity Flows ............ 56
4.6 Market Prices: L'Acul River Basin Marketing System ..... 58
4.7 Retailer Profits .................. .................... 62
5.1 Potential Rainfed Crops for Cayes Plain Basin ........... .67
5.2 Crops Observed: November December, 1983 in
L'Acul River Basin ..................................... 69
5.3 Modes of Tenure .............................. 74
5.4 Ownership/Tenancy Arrangements by Topographical Unit .... 88
6.1 DCCH/CARITAS Affiliated Groupments in Cayes Basin ....... 98
6.2 DCCH/CARITAS Land Farmed by Groupments .................. 98
6.3 DCCH/CARITAS Groupment Investments in Cayes Basin, 1983.. 99
6.4 Irrigation Projects in the Torbeck Plain ................ 104
6.5 Irrigation Projects in the L'Acul and Cayes Plain ....... 106
6.7 Estimated Water Taxes Assessed in Les Cayes
Agricultural District .................................. 118
6.8 1982-83 Extension of Credit in Cayes Plain Basin by IDAI. 126
6.9 1981-82 BCA Loans for the Cayes Regional Office ......... 128

Figures Page
1.1 The Cayes Plain Basin Region ....................... 4
2.1 Matrix of Planning Approaches .......................... 11
2.2. Delimitation of River Basin ........................... 14
2.3 Conceptual Model of Rural Regional Development
Planning in Haiti .................................... 15
3.1 L'cu River Basin .. ............. ....... ........ 24
3.2 Sequence of Assigning Classes to Land Suitable for
Cultivation ................................26
3.3 Terrain Characteristics--L'Acul River Basin .............28
3.4 Distribution of Soils in Cayes Plain Basin ..............36
4.1 Market Locations--L'Acul Basin........................ 44
4.2 Spatial and Temporal Dimensions, L'Acul
Marketing System................. ................... 45
4.3 Market Hierarchy, L'Acul Basin Marketing System .........50
4.4 Origins of Goods Flowing to L'Acul River Basin
Markets (13 Figures) ................................ 51-54
4.5 Temporal Variations in Average Monthly Prices for
Selected Commodities ............................ 59
4.6 Les Cayes Market Price Averages ........................ 60
5.1 F~requency and Percentage Distribution of Number of
Purchased Plots by Location ..................... 76
5.2 Frequency and Percentage Distribution of Number of
Inherited Plots by Location ........................... 76
5.3 Frequency and Percentage Distribution of Number of
Owned Plots (Inherited or Bought) by Location ...........76
5.4 Frequency Distribution of Number of Plots (per farmer)
by Location.................................... 77

5.5 Frequency and Percentage Distribution of Number of
Rented Plots by Location .............................. 77
5.6 Frequency and Percentage Distribution of Number of
Sharecropped Plots by Location ........................ 77
5.7 Frequency and Percentage Distribution of Number of
Rented Plots by Size .................................. 79
5.8 Frequency and Percentage Distribution of Number of
Owned Plots by Size ................................... 79
5.9 Frequency and Percentage Distribution of Number of
Shar6cropped Plots by Size ............................ 80
5.10 Frequency and Percentage Distribution of Total Number
of Plots by Size ...................................... 80
5.11 Frequency Distribution of Farm Size Average by Location. 80
5.12 Frequency Distribution of Purchased Land Prices
(per karo) by Location ................................ 82
5.13 Frequency Distribution of Rented Land Prices
(per karo) (per year) by Location ..................... 82
Frequency Distribution of Land Prices (Purchased)
(by decades) by Location .............................. 82
5.15 Frequency and Percentage Distribution of Owned
(Inherited or Purchased) Land by Location ............. 84
5.16 Frequency and Percentage Distribution of Rented Land
by Location ........................................... 84
5.17 Frequency and Percentage Distribution of Sharecropped
Land by Location ...................................... 84
5.18 Frequency Distribution of Average Plot Sizes
by Location ........................................... 85
5.19 Frequency and Percentage of Number of Farmers by
Type of Tenure ........................................ 85
7.1 Planning Goals and Interventions ....................... 152

A Social Institutional Profile (SIP) usually is intended to
provide information on the social-economic, natural resource, and
organizational/institutional structures in a particular area, so as to facilitate the USAID in-country mission's pursuit of its goals in that
area. However, the present report extends the conventional limits of
a SIP since, in addition to furnishing useful background information
on the Cayes Plain Basin study area (in the form of "hard" data,
interview results, field observations and more impressionistic
information), it seeks to accomplish three other objectives. First,
it attempts to establish a general methodology that can be used to
analyze the physical, economic, and institutional characteristics of a
rural region and, subsequently, to construct a workable development
plan for that region. Second, it attempts to conduct the background
analysis and to specify the dimensions of such a development plan for
the Cayes Plain Basin. As indicated in the Foreword, the principal
goals of a development plan for this area are the amelioration of the
quality of rural life and the improvement of the land resource base.
Finally, this report seeks to develop an appropriate and effective
strategy for implementing the, development plan. This latter objective is especially crucial since, without a set of effective implementation mechanisms, even the' most well-conceived plan is merely a hypothetical
The focus of this extended SIP, the Cayes Plain Basin, has been
selected for analysis for a number of cogent reasons.
1. The Cayes Plain Basin is in many ways an archetype of the humid
portions of Haiti since it manifests all of the topographic units
found in these humid areas. The Platons Unit, while of minor areal
extent compared to the total area of the Cayes Plain Basin, is in many
ways a microcosm of the physical environment in the Central Plateau
and, to some extent, it is characterized by similar problems of
restricted accessibility.
2. The Basin is both a major agricultural area and one of the
country's most densely populated regions, approximately 375,000 people
live within the Cayes Plain Basin (SNEM 1982). It is an area
characterized by a range of settlement types, including the city of Les Cayes (pop. 37,000), Haiti's third largest center. The land in
the basin is intensively utilized and covers the spectrum in terms of
quality and productivity.

3. The Basin is an area having untapped water resources which hold high potential for the enhancement of agriculture. Further, like other portions of Haiti, it contains active irrigation works which are characterized by a continual need for rehabilitation.
4. Not only does the Basin appear to be an operationally valid planning region, from both human and physical points of view, but it also appears to be a manageable planning unit. The latter denotes that the Basin is sufficiently compact (compared, for example, with the Central Plateau) such that investments committed will not be spread too thinly; and that logical sub-regional units exist that may be employed for the implementation of a development plan. That is to say, from a temporal perspective it is impossible to simultaneously implement all elements of a plan in all portions of the planning region. Implementation must be phased, and this requires the delineation of sub-regions which possess an internal functional integrity in terms of both human and physical criteria. The Cayes Plain Basin is completely occupied by three such units: The drainage areas of a set of small rivers that emerge from groundwaters on the lowlands (e.g., the Picot and Bordonne Rivers), the Rivi~re de la Grande Ravine du Sud, and L'Acul River Basin.
For reasons of logistics and human resources it has not been
possible for the Study Team to examine each of these three sub-regions in equal depth. For the following reasons L'Acul River Basin has been selected as the detailed case-study of this report.
a. Just as the Cayes Plain Basin is a model for the humid portions of Haiti, L'Acul River Basin is a model which manifests the general characteristics of the larger Cayes Plain region. In addition, L'Acul Basin contains the Platons highlands which, as noted above, are unique in the Cayes Plain Basin, but are typical in the broader national context.
b. Due to L'Acul Watershed Project conducted by the U.S. Soil Conservation Service, there exists a useful set of background data which can be used as a point of reference for L'Acul sub-region. The dearth of consistent and reliable data continues to be a major constraint to development planning in all regions of Haiti.
c. A number of AID projects and interventions have been instituted in L'Acul Basin in recent years. It is hoped that the program outlined here can complement and build upon these initiatives in such a manner that the individual impacts of all activities will be multiplicative.
d. Just as the Cayes Plain Basin is identified as a valid and manageable planning region, at a reduced spatial scale the same characteristics are attributed to L'Acul River Basin. L'Acul Basin is a relatively self-contained unit from both a human and a physical perspective. Further, it is composed of four distinct sub-regional units which can be effectively utilized for the construction and implementation of a coordinated rural regional development plan.

Located in the southwestern portion of the Southern Peninsula,
the Cayes Plain Basin comprises approximately 720 square kilometers (Figure 1.1). Two large river systems, L'Acul and Grande Ravine du Sud, as well as ten smaller permanent river systems are found within the basin. From a perspective of rural development, a wide spectrum
of ecologic conditions exist in the area. These range from high-risk,
low potential, severely degraded lands to low-risk, high 'potential
agricultural lands (DATPE 1982).
Estimates of population vary widely for the Cayes Plain Basin.
According to the 1982 census, approximately 375,000 people live within the area. Of this total, 10 percent live in Les Cayes. Excluding Les
Cayes, and to a far lesser extent Camp Perrin, almost all the
remaining inhabitants earn their livelihood from occupations directly
connected to agricultural pursuits. Accurate figures on the age
structure of the population are elusive, but a casual, non-scientific
sample undertaken by the Study Team indicates that approximately 40
percent of the population is under fifteen years. As this population ages and enters the reproductive stage, either ever increasing demands will be placed on the immediate lands within the basin or outmigration
will dramatically increase. Already population densities range from
over 475/km2 on the lowlands to a "low" of 125/km2 in the poorer
agricultural areas within the slope zones. No place within the whole
Cayes Plain Basin can be classified as underpopulated.
As would be expected, the economic structure within the Cayes
Plain is dominated by the agricultural activities of-its inhabitants.
No secondary industries exist within the basin. Additionally, none of
the primary industries utilize materials that are imported into the
region. Thus the industrial sector that exists within the area is
directly linked to the land resource. This is obvious from the
following list of active industries. Within the Cayes Plain Basin
there are a few small clairin distilleries, one small sugar centrale,
one small dairy plant, a few vetiver plants, and some livestock slaughtering facilities; these represent the gamut of industrial
activity. A small tomato processing plant functionning in 1982-1983
appears at this time to be out of business. This paucity of
industrial development is reflected in the economic structure of the
family unit for the overwhelming majority of the Cayes Plain Basin
inhabitants. The rural family unit is largely a self-sufficient
entity that lacks any real specialization. Even the irrigated rice
farmers grow a variety of crops. While this self-sufficient structure
maximizes the family unit's ability to survive under present
environmental conditions, it must result in a decreased level of
economic efficiency and output. This is reflected in the low
agricultural yields when the Cayes Plain Basin is compared to other
areas in the world.

HAITI ~$$U AAtlantic Ocean
- International boundary 1ort-dePallx
Departement boundary Nord-Oluest Cap-Haftien
o National capital
o Departement capitals Nord
0 Cities
0 10 2.0 30. miles Gonarve Ennery 0 Saint-RapbailJ
0 10 20 30 Kilometers Golt. do )a
________________ Gonave ['Artibonite
0_Port-au-Prince U
Port-5 u Caribbean Sea

The crops grown in the area are dominated by varieties that are largely consumed locally, as will be shown in Chapter 5. This is particularly true for the rainfed agricultural areas. The two dominant irrigated crops, tobacco and rice, clearly enter the national market. Throughout the Cayes Plain Basin a well articulated market system exists. The flows of commodities through this market system reach the most distant points of the basin with surprising facility. Yet, like the industrial composition of the region, the market's goods are dominated by the agrarian economy of the area. The structure and functions of the Cayes Plain Basin marketing system are developed in Chapter 4.
The institutional composition of the basin in many ways reflects both the patterns of population densities and the ease of accessibility within the Cayes Plain Basin. By far, the greatest concentration of institutions is located on the Cayes Plain. Except for the church, government security, conseils communautaires, a few PVOs and a number of informal local community-based organizations, the upland areas of the basin have been largely ignored. It is on the lowlands that the majority of public and private, both formal and informal organizations are located. Most government ministries are represented in the basin, specifically in Les Cayes. However reflected by the magnitude of their presence and the level of the functionaries, this region appears to have been a relatively low priority for the national government.
Given this low level of involvement, the government has chosen to emphasize the lowlands in the 'majority of its spheres of work. For example, while limited government credit (BCA) has just become available (1983) for the small farmers, the BCI\ efforts have only been on the lowlands. This is likewise true for most church activities with the exception of some missionary work. Since most PVOs in the Cayes area are related to church activities, these too have been proportionally biased to the lowlands. One slight exception to this is in the Camp Perrin area, a location immediately juxtaposed to the lowlands. Even USAID has emphasized the lowlands. This concentration of activities on the lowlands, by almost all organizations, has resulted in an unequal allocation of resources within the basin to the detriment of all areas in the region including the lowlands. The role and nature of organizations in the context of co-operative rural development for the Cayes Plain Basin, is discussed in detail in Chapter 6.

As noted at the beginning of this chapter, the present document
is designed to accomplish four objectives:
1. The provision of a range of specific data and general
information concerning the study area;
2. The establishment of a planning methodology which can'be
generally employed throughout Haiti;
3. The specification of a coordinated rural regional development
plan for the Cayes Plain Basin and, in particular, L'Acul River Basin;
4. The recommendation of an appropriate and effective set of
mechanisms for the implementation of the development plan.
Chapters 2 through 6 deal with the first two objectives.
Chapter 2 specifies the conceptual model of the planning process that the Study Team utilizes subsequently in this report. Chapters 3 and 4
are detailed descriptions and analyses of the physical setting and marketing structure of L'Acul River Basin; Chapter 5 deals with the
interface of these two elements in L'Acul system: agricultural
activity. These latter three chapters are relatively lengthy as they are intended not only to present specific information about the study
area but also to inform the reader concerning the relevent factors and variables that need to be taken into consideration, and concerning the
appropriate methods of analyses to be utilized. To the reader
concerned strictly with policy formulation and policy implementation
these technical chapters may seem painfully detailed. It is the hope
of the Study Team, however, that this document can serve as a useful
model for the entire planning process--technical assessment as well as
policy formulation and implementation..
Chapter 6 presents a description and analysis of the
institutional context of the study area and considers a range of
formal and informal, governmental and non-governmental organizations which have significant implications not only for the formulation of a
development plan but also, and most particularly, for the
implementation of a plan.
The third and fourth objectives cited above--the specification
of a coordinated rural regional development plan for L'Acul River
Basin, and the recommendation of an effective set of implementation mechanisms--are the concern of the final sections, Chapters 7 and 8.
The material contained in these two chapters is the logical
culmination of the approaches developed and the analyses conducted in
Chapters 2 through 6.

The purpose of this chapter is to make explicit the view of the
planning process that the study team will utilize subsequently in this
report. After some general comments on major issues in development
planning, the elements of the coordinated regional approach employed
in this exercise are outlined and justified. A conceptual model of
the development planning process based upon these elements is then
Over the past thirty years, during which period regional
development, in general, and rural development, in particular, have occupied the attention of governments and planners in the developed
and developing worlds, a number of identifiable approaches to the
stimulation of economic growth and social well-being have been
adopted. The nature of these individual approaches is, in turn, a
function of the view of the causes of underdevelopment that is explicitly or implicitly held by decision-makers and planners.
There are essentially six views of the roots of
underdevelopment; naturally, these views are not mutually exclusive
and there is considerable overlap between them. The first view is based upon a combined balance of trade/economic base framework and holds that exports are the principal growth stimulus for a region.
Export-producing activities ("basic activities", in the terminology of
economic base theory) create "injections" into the local (regional)
economy which, through the multiplier mechanism and the circular flow
of income, not only generate wealth but also stimulate local-serving
or "non-basic" activities (e.g., retail stores, services,
entertainment facilities etc.). A related concept is that of import
substitution. By reducing the "leakages" from the local economy that
occur when imported goods and services are purchased--that is, by
substituting local products for the imports--an effect similar to that
resulting from the stimulation of exports can be achieved. Regional
development policies which are based upon this view stress the
creation and expansion of export markets and opportunities for the
substitution of local goods and services for imports. A related
concept is the staples theory, which attributes the lack of growth or the decline of the economies of certain regions to a natural resource base for which there is a low demand, i.e., for which there is little
export potential.

A second view concerns the role of migration as an economic adjustment mechanism. According to this line of neo-classical reasoning, per capita income is the sole acceptable measure of development; per capita income is a function of both the total population of a region and its total earned income. An increase in per capita income may be achieved by a reduction in the divisor, i.e. by the out-migration of unemployed and underemployed portions of the population to areas where employment opportunities are more plentiful. It is believed that this economic adjustment will generally occur as a result of the normal functioning of the labor market mechanism. The major constraint to this approach is policies (e.g. social assistance programs) which attempt to interfere with the market mechanism by creating a better standard of living for the population 'in situ, thus reducing the motivation for migration. Policies based upon this view emphasize the reduction in the social "safety net", the encouragement of the free operation of the labor market, and the provision of incentives to migration.
A third view stresses the role of physical capital and
infrastructure in development. The argument is essentially a simple one based on the notions that industry requires a suitable environment in order to develop and to flourish and that improving both intraregional and interregional accessibility will stimulate economic transactions. The construction of roads and sewers, the extension of electrical networks, and the establishment of serviced industrial parks are among the policies associated with this approach.
A fourth view stresses the importance of a region's industrial structure. The argument here is that in order for economic growth to occur a region requires both certain strategic industries (for example, those that can create value added from the region's resource bases) and a sufficiently well-developed set of linkages between industries. The latter point refers to the complementarity achieved when two or more firms create functional connections between their products in the form of forward or backward linkages; for example, a firm producing automobile engines has backward linkages to the firm producing carburators and forward linkages to the firm that assembles the finished automobile. Policies based upon this view involve the analysis of industrial structure, generally using input-output or inter-industry analysis, and attempts to alleviate bottlenecks and to fill-in "missing-links" in the industrial structure. A policy that overlaps both the physical capital/infrastructure and industrial structure viewpoints involves the establishment of growth centers. A growth center represents an attempt to maximize a region's connectivity with the rest of the national economic system, particularly with the nation's urban network, so that "growth impulses" generated nationally will diffuse throughout the region via "spread" or "trickle down" effects. The designated growth center is generally an existing urban node into which various types of economic activity are concentrated in order to induce regional economic growth and to create the potential for facilitating inter-industry linkages.

A fifth view assigns the causality for underdevelopment not to intrinsic factors within the region itself but to external exploitation; this is the neo-marxist approach of dependency theory. The reasoning here is that the economic and political "center" of a country is able to dominate the "periphery" (i.e., the lesser developed regions) and to extract surplus economic value from its natural resource base, its relatively weak industrial structure, and its labor force. Dependency theory thus builds a model of polarized economic development in which the center continues to grow both as a result of its own attributes and as a result of its exploitation of the periphery, and in which the periphery remains at a low level of development. Policies associated with this approach involve selective regional closure, that is, attempts to reduce the linkages between the center and the periphery, thus curtailing the domination by the former of the latter.
Finally, there are a number of related views which attribute underdevelopment to certain barriers preventing the indigenous regional population from reaching its full economic and societal potential. Lack of information, knowledge, and know-how, as weii as insufficiently developed levels of animation, motivation, and entrepreneurial spirit retard the capacity of the population for raising its level of social and economic well-being. This approach is based upon the notion that the source of development iies in the local. population rather than in the region's natural resource endowment or in external factors. Policies stemming from this viewpoint stress the need for better knowledge and skills, better access to information, and a higher degree of both social animation and entrepreneurial activity. To a large extent, these types of policies, stressing as they do the indigenous population and its potential, are the inverse of those mentioned above which stress external linkages, physical capital, infrastructure, and industrial structure.
In the discussion above two general issues are explicit: that of increasing the degree of integration of a region into the national economic structure versus regional closure; and that of emphasizing the development of human resources versus emphasizing physical capital and infrastructure. Related to these two issues, and implicit in the above discussion, are two others. First, there is the question of the goal of regional development: achieving national economic efficiency (i.e. maximizing total economic output) or promoting equity between individuals and/or regions. The mutual exclusion of these two goals is one of the classic dilemmas of development theory, one which appears incapable of resolution. Second, there is the question of adopting a regional (spatial) approach versus a pectoral approach. This dichotomy is discussed in detail in the following section. However, it must be stressed that there is no single answer to the complex problem of regional development. All of the viewpoints that have been presented contain within them elements of truth.

In this section, the Study Team's basic planning philosophy is
made explicit. This document is neither philosophical nor theoretical
in purpose, but the nature of the approach adopted must be specified from the outset. The most effective manner in which to present this
approach is to situate it in the context of a simple typology of
planning alternatives (Figure 2.1). This typology is constructed
using two dimensions: spatial orientation and functional
orientation. Type I approaches are not only sectoral (rather than
regional) in orientation, but also focus upon a single activity. This
is often referred to as a "development project" strategy and may be
exemplified by the construction of a dam or of an irrigation project.
A Type II strategy involves a single activity but is concerned with its distribution over a certain region; for example, an attempt to
increase tree planting in the Artibonite River Basin. Type III
approaches are non-regional but do attempt to coordinate sectoral
activity. One may cite as an example a general national health
program having the following components: the provision of better
sanitation facilities, the provision of potable water, increased capacity for the treatment of malaria, and the suppression of the
mosquito population. Finally, a Type IV strategy is both regional and coordinated in nature. This is the approach adopted by the Study Team
and is now explored in more detail.
A coordinated approach to planning signifies one in which
various economic sectors (e.g., marketing, transportation,
agriculture, industry), various components of the physical environment (e.g., rainfall, soil depth and quality, geomorphology and lithology),
and various elements of the socio-cultural milieu (e.g., attitudes,
values, and beliefs associated with farming practices and land tenure)
are considered. Further, this approach recognizes that these three
broad categories are themselves functionally interrelated. Tnus, one
would take into consideration not only agricultural, marketing, and
processing activities but also the relationship of these to both
physical and socio-cultural factors. From a systems perspective the planning environment can be regarded as a set of interacting systems,
each of which is composed of interacting sub-systems. Obviously there
are degrees of coordination between the extremes of the single
activity approach and the "perfectly" coordinated approach. While
"perfect" coordination is unlikely to ever be achieved, one can strive
to identify and to coordinate a set of key interacting elements. One
important implication of a coordinated approach to planning is that
its successful operationalization requires an interdisciplinary effort.
A regional approach is one in which the spatial dimensions of
both human social-economic activity and the physical environment are incorporated into the planning process. In contrast to the sectoral
view implicitly held by the classical economist, who treats the world
as if it were located on the head of a pin and as if location in space
is of no import, the regional view acknowledges two important facts:

first, that all human and physical phenomena are located (in both absolute, and relative terms) somewhere in terrestrial space; and,
second, that the specific location at which a phenomenon is found can not only exert a profound influence upon the nature of that phenomenon
but can also affect the degree to which that phenomenon is capable of influencing other phenomena. For example, the economic viability of an aluminum smelter may range from very poor to very good, depending upon how it is situated relative to sources of hydroelectric energy.
Spatial Orientation
Sectoral Approach Regional Approach
Activity I II
Approach IIIV

Similarly, its location will have potentially strong effects upon ooth the employment levels and the environmental quality of the surrounding area. Thus, economic geographers, regional planners, and regional economists concern themselves not only with such concepts as distance decay functions, spillover and neighborhood effects, and distribution over space, but also with the manner in which individual subregions interact with one another over space through flows of people, goods, information, and physical material. From an entirely pragmatic viewpoint the regional approach is eminently sensible since it encourages the concentration of efforts within a well-defined area rather than allowing them to be spread too thinly over a wide area. Having discussed the elements of the regional approach, it is necessary to define more precisely the concept of "region" and then to specify the regional unit employed in this study. A region is an intellectual construct having no existence beyond the minds of the planners and policy-makers who establish it for a specific purpose using various human and physical criteria that will best permit planning goals to be achieved. That is to say, regions do not exist naturally but are delineated as needed. There is, however, a logic to the delineation of a region, and this generally involves the consideration of both the homogeneity of attributes (e.g., per capita income, soil and crop types) and the flow patterns of people, products, information, money, energy and physical matter (e.g., soil particles, water). The basic rule of thumb in delineating a region is that the variation in the criteria utilized should be lower within the region than that between the region and an external area. For ex ,ample, one would expect that the per capita income level (if that were one of the criteria utilized) at two points within the region would vary less than between one point within the region and one external point.
In the approach utilized here the river basin is the regional unit of analysis. There are two reasons for this decision. First, many of the variables and attributes upon which this rural regional development planning methodology is based (e.g., pedological, biological, and hydrological factors) clearly function within the boundaries of major river basins. Second, field investigation by the Study Team has shown that many manifestations of human social and economic activity in Haiti also tend to function most strongly within river basin boundaries. That is to say, there are generally more social and economic flows within river basins than between basins. (See, for example, the discussion of marketing in chapter 4). This is in contrast to developed countries characterized by better transportation capabilities and more mobile populations, where the river basin is generally not, from a human perspective, an appropriate regional unit since it does not effectively delimit the boundaries of economic and social flows. In the Haitian context there is also a third consideration. Due to the general unavailability of socio-economic data for units below the national level, and in the

absence of specific fieldwork, insufficient human activity information
exists for attempting to delineate regions. Thus, by default
exclusively physical criteria must often be utilized. In summary, when higher agricultural productivity and the preservation of the
environment in a particular area of Haiti are the goals of the
planning exercise, as they are here, both physical and human evidence
suggests that the river basin is an appropriate regional unit.
By definition a river basin comprises all of the area above the
mouth of a river that contributes to the discharge of the river system. A basin includes both the area that contributes to the
surface (runoff) waters as well as the zones contributing groundwater,
since the discharge of a permanent river is determined by both the
former and the latter. In practice, however, most river basin
boundaries are determined solely by topographic features that produce
the surface flows into the river network (Figure 2.2).
In a river basin all energy and matter flow in a unidirectional
manner from the upper reaches to the mouth of the river. It is this
property that makes the river basin a logical unit for regional development concerned with agriculture and environment since two
crucial variables affecting agriculture, water and soil, are removed and transported from the land via the river system. In addition, the
flow characteristics of these two variables as well as the
interactions of numerous human interactions (e.g., crop treatment)
with these variables are determined by the energy attributes
(altitude, slope) of the basin. According to Donner (1975: 44-45),
because of Haiti's rugged topography, only when the river basin is the
basic planning unit can agricultural development occur.
Having explored the nature of the terms "coordinated" and
"regional", the conceptual model of the coordinated rural regional
development process that is utilized by the Study Team is now
presented (Figure 2.3). It must be stressed that this model is not
simply a theoretical creation but is, rather, one designed to achieve
results. It is for this reason that, in the planning process, the
implementation of the plan is given equal weight with the development
of a logical plan.
Rural development in Haiti is ultimately a question of being
able to improve its agricultural system. This statement is not meant
to imply that other activities such as primary and secondary
processing and urban-based growth center functions do not play a role; it is simply a recognition of the fact that the overwhelming majority
of the rural population earns its livelihood from worRing the soil.

Delimination of River Basin
River Basin Divide
Cross -sector along line A- B D = basin divide S = stream channel
Figure 2 .2

Technical Assessment Filter 1. Development Filter 2.
Natural Attitudes, Potentials and Existing and Potential
Resource Values, Constraints: Development Mechanisms
System Beliefs, and Organizations
Perceptions Rural Regional
of Population Development (Public and Private
Concerning Sectors, PVOs)
Agricultural System Natural,
Market Systems
Market System
Social-Cultural- Design, Execution,
Institutional Monitoring, EvaluaContext tion, Phasing,
Investment Level
Implementation Strategy

Agricultural production is subject to the constraints and potentials of the natural resource system: rainfall, climate, soil type and depth. Similarly, in an economy such as Haiti's where produce is not grown exclusively for on-site consumption, the nature of the marketing system has important implications for the creation and retention of surplus value by individual households. Thus, the first step in the planning process is a technical assessment of these three interrelated systems.
An "objective" technical assessment, however, does not
necessarily reflect the "real" world. It is an axiom of social science that the use of the natural environment by man and the manner in which he conducts social and economic activities will be a function of his attitudes, values, beliefs, and perceptions: not only those that relate directly to the systems in 'uestion but also those that relate to ancilliary phenomena and processes, e.g., religious practices, the ownership of land and animals, the role of the extended family. It is for this reason that the technical assessment must be passed through a first social-cultural-institutional filter. At this point certain directions and initiatives suggested by the technical assessment will be shown to be inappropriate while other, new considerations may emerge.
Having developed a thorough understanding of the three critical systems and of the social-cultural-institutional context in which they are situated, it becomes possible to weigh the regions's potentials and constraints and to arrive at a development plan which explicitly considers the interaction of the various economic sectors, components of the physical environment, and elements of the socio-cultural milieu, all within a regional framework. Depending upon the specific nature of the planning exercise and upon the specific goals established, the regional scope and substantive breadth will vary.
However, the most well-conceived plan is merely an intellectual curiosity until it is implemented. The implementation stage is thus the most critical element of the planning process and has two components: the identification or establishment of suitable implementation mechanisms and the formulation of an implementation strategy. The former may involve governmental or quasi-governmental authorities and organizations, actors from the private sector, private voluntary organizations and other organizations and institutions in the not-for-profit sector. Once again, in order to arrive at the appropriate implementation mechanisms, "objective" reality must be passed through a social-cultural-institutional filter which may eliminate certain alternatives and suggest others. Once this has been accomplished the latter concern, the implementation strategy, may then be addressed. As indicated, this includes considerations of the design of individual plan components as well as execution of the overall strategy; the on-going monitoring and evaluation of plan implementation; the phasing of the specific components of the plan implementation strategy; and requisite levels of investment.

Having specified the nature of the coordinated rural regional development approach adopted, the following chapters examine the individual elements of the'planning model. Chapters 3, 4, and 5 comprise the technical assessment of the physical, marketing, and agricultural systems. Chapter 6 explores the social-culturalinstitutional context that underlies the model's two filters, focusing specifically upon institutions and organizations--not only describing those that exist but also citing what will be required in order to successfully implement a coordinated rural regional development plan. The final two chapters outline the specific development plan and implementation strategy that are indicated by the factors and conditions reviewed in Chapters 3 to 6.

In humid tropical areas erosion is one of the most serious
limiting factors in agricultural and livestock production. Geologic erosion is a natural phenomenon. In most areas where vegetation has
been unaltered by human activities a general balance exists between
erosion and soil formation. With the introduction of cultivation, the
soil is broken and the natural vegetation is removed. Both of these changes increase the potential for soil erosion. If measures are not
simultaneously introduced to counter these changes accelerated soil
erosion almost always follows. In Haiti, in response to both an
increasing population and limited access to the better agricultural lands, peasants have-continually needed to search for new lands to
cultivate. Most of the lands available for peasants were and are in
physical settings that from an agricultural perspective have a low
potential and high erosional risk if brought under continuous
cultivation. In past years a system of shifting cultivation and
fallowing of lands was a response to the degradation of the soil as a
result of agricultural activities.
Today, with the extreme scarcity of agricultural lands,
intensification of cultivation is necessary. There is not enough land
available for the individual farmers to continue practicing tile
previous rotation of crops interdispersed with periods of fallow.
Almost no land remains fallow for a long enough period to restore soil
fertility to adequate levels. This results in lower yields and the
need to increase the proportion of land under cultivation just to meet previous agricultural outputs. Also with lower yields, the vegetation
cover of the crops is reduced and the soil is less protected from
rainfall. Thus, with the continuous increase in the use of marginal lands, erosion has been in an accelerated phase for a long period of
time. Today, the evidence indicates that the process continues
unabated. According to Ahmad (1977), Haiti is a classic example of
continuing severe land degradation, perhaps more than any other country in the world. In the following sections we explore the
problems and processes of land degradation in general and
specifically, within the Haitian context, in the Cayes Plain Basin.
In the following sections basic concepts and methods of
measurements (assessment) are introduced. All the constructs and
indicies presented in this chapter contribute to an understanding of
the environmental situation as it exists, or could exist under
different conditions. The generation of numerical values for the

crucial variables of soil-loss, land capability, and moisture
availability provide the necessary background information to permit
rational planning decisions to be made. To meet the demands of this
SIP it was necessary to develop some primary data in order to
determine the viable options that exist for agriculture in the Cayes
Plain Basin. Even to meet the preliminary needs of this study, the data base was so poor that it was necessary to estimate the primary
data in many cases. In the next phase of a coordinated rural regional
development project more information for constructing these indicies
will need to be provided if planners are to have sufficient
information available to make valid planning decisions.
This report only considers soil erosion resulting from water
since in Haiti it is clearly the dominant erosional agent. Erosion is
caused by the impact of raindrops on exposed soil and by runoff
causing sheet, nill, and gully erosion. In restricted Haitian
environmental settings, stream bank and sub-surface flow erosion are
On cultivated lands six factors have been identified as the
crucial variables affecting the erosional rates on given parcels of
land. These are rainfall intensity and quantity, runoff, soil
erodibility, slope steepness, plant cover, and conservation
practices. Whether accelerated soil erosion occurs or not depends on
the balance between the energy, resistance, and protection factors.
Energy factors include the energy potential of rainfall and runoff to
initiate erosion (erosivity), the slope steepness, and the length Of tine field. The resistance group is comprised of those properties of
these soil that resist erosion.
The resistance of the soil to erosion, called erodibility, is
determined by its physical, chemical, and organic composition. Soil
properties increasing water infiltration (decreasing runoff) decrease
erodibility; conversely factors decreasing infiltration such as compaction, increase erodibility. Protection variables include
vegetation cover and conservation practices.
Soil erosion is a two-phase process. First, individual
particles of the soil need to be detached from the soil mass.
Raindrop impact on the soil is the most important agent initiating soil detachment. It is only effective when bare soil is exposed to
the direct impact of the raindrop. Land clearing always increases the
effectiveness of this process. Once soil is detached, the second
process of transport in running water is required. Overland (runoff)
flow is the agent that transports the soil downslope and eventually to
the river system. When sufficient energy (velocity) is no longer present to transport the soil particles in the runoff, deposition
occurs. From an agricultural perspective both excessive erosion and

deposition create situations that result in lower crop yields. In Haiti, in areas under irrigation excessive deposition has historically been a major problem.
With the clearing of the natural vegetation from most of Haiti's land and the substitution of vegetation that does not protect the soil to the same degree from raindrop impact, widespread accelerated erosion exists. Change in vegetation cover, alone, has resulted in major soil-loss on most agricultural lands. With erosion occurring over a long period, the soil moisture retention as well as the soil infiltration rates have decreased. This has increased the overall proportion of rainfall that flows as runoff and concomittantly has decreased ground water recharge. This results in larger storm flows which increases the capacity of existing streams to transport materials. Thus the initiation of accelerated erosion without immediate counter efforts sets into motion a positive feedback mechanism that makes eroding landscapes even more favorable for further erosion.
Under these conditions erosion causes disastrous floods and
pseudo-droughts, those resulting from pedological not climatological change, are created through the reduction of the water holding capacity of soils, and the drying up of springs and river flow. This leads to a continuing lowering of agricultural, grazing, and silvacultural potential of the affected area. Because of the deterioration in the environment resulting from erosion, it is easier to prevent erosion than to reverse it once it has reached a major intensity. Unfortunately in Haiti already a large proportion of its lands are experiencing accelerated soil loss. However, in tnose few upland areas where soil erosion has not become excessive, such as Les Platons, immediate preventive actions will be less costly than waiting until these areas reach the stage of the majority of today's Haitian lands.
The five types of erosion are sheet and rill erosion, gully erosion, streambank erosion and sub-surface flow erosion. Sheet erosion is the removal of soil via unconcentrated overland flow. It is the dominant form of soil-loss on gentle slopes (less than 4 percent).
Rill erosion begins to occur when the overland flow becomes
concentrated. This movement of water in flow streams results in the formation of ephemeral local channels known as rills. In Western Nigeria rills began to form on slopes greater than 4 percent, they became the dominant erosional process on 15 percent slopes (Lewis, 1980). When sheet and rill erosion occur and land is cultivated for the next planting season, manifestations of the erosion are not readily visible. One intense storm or runoff from saturated soils may remove 2 to 5 mm of soil in a few hours. On moderate or gentle slopes, sheet and rill erosion remove the fine soil particles

resulting in a coarser soil texture. This residual soil material has
lower fertility and productive capacity. Worldwide, sheet and rill
erosion are the major form of erosion on agricultural lands.
Gully erosion is the next important form of erosion on a
worldwide basis, but where it exists, it is the most destructive phase
of water erosion. It is caused by the concentration of runoff from
extended areas. 'When both the volume and velocity of the water becomes high enough, it gradually removes the soil material and
excavates a ditch. Unlike the rill, gullies are permanent steep-sided
channels. Like the rill, they only have discharges during and
immediately following a rain event. Almost always their presence
indicates accelerated erosion. When no engineering structures are
present, a gully always forms in the lower portion of a slope and
precedes upslope. Pathways in many rural settings result in the
initiation of gully formation. Today road building and associated
drainage diversion result in excessive runoff at specific points.
Under these conditions the gully is initiated at the road and rapidly
moves downslope. In Haiti where lands over 30 degrees are under cultivation, gully erosion is a major widespread process that is
removing productive lands into the completely non-productive category.
Stream bank erosion occurs when riverbanks-cave in due to the
undercutting of the river channel. It is a natural process associated
with the normal dynamics of river channel migration. However, it can be associated with accelerated erosion in areas where the flow regime
of the river has been altered due to changes in the balance between
overland and groundwater flows. Where the overland flow component has
increased, the river channel must handle a larger discharge than
previously for the same flow frequency. This requires the river to
scour a larger channel resulting in riverbank collapse. In the
process, cropland juxtaposed to the river channel is lost.
Finally, subsurface erosion is important 'in areas where earth
materials permit concentrated flow in natural tunnels or pipes.
Limestone areas are one example of an environmental situation where
this can occur. In addition plant nutrients especially those added by
fertilizers, are removed by this process. This removal of soluble
minerals reduces a soil's resistance to erosion by reducing plant
growth and thereby reducing plant cover. Rarely is sub-surface
erosion accelerated by human interventions. When excessive fertilizer
losses occur they afe usually the result of increased overland flows
not sub-surface erosion.
The erosion rate for specific locations within a river basin is
determined by the interaction and magnitude of the numerous physical
and management variables at the sites. The Universal Soil Loss
Equation (USLE) is the most widely used model to evaluate erosional

conditions. It was originally developed to predict the long term average soil losses resulting from sheet and rill erosion from specific fields under specified cropping and management systems (Wischimeier & Smith 1978). Recently it has been extended to be applicable to noncrop land use as well as to estimating river basin erosion and sedimentation delivery rates. The USLE is defined as:
Am=R KL S CP where
Am =Soil loss in metric tons per hectare per year
R =Runoff and Rainfall factor
K =Soil erodibility factor
L =Slope length factor
S =Slope factor an evaluation of the interaction between slope length and steepness.
C =Crop and management factor
P =Conservation practices
Only the R factor is completely independent of human
interactions with the land resource. It is solely defined by the precipitation characteristics of the area. The K factor, while largely determined by the soil's structure and texture, can be altered through interventions affecting the organic content of the soil. Additionally agricultural practices such as the use of fire and irrigation can alter the K value. Field length and slope angles can be modified by boundary changes and morphology changes such as terracing. Both the C and P factors for any parcel of land are completely determined by human intervention. It can be seen clearly from the variables identified as crucial in the USLE that soil erosion rates can be strongly altered by the land use patterns established in a given region.
One of the problems in applying the USLE to tropical areas is
that it is an empirically derived equation. It was formulated in the U.S.A. where climate, soil types, and agricultural systems differ markedly from most third world countries. Recently, research in some tropical areas, mostly in Africa, have begun to develop numerical values for many tropical settings (Roose 1977). For Haiti, as a whole, little information is available at the required degree of precision to apply the USLE. In particular rainfall records are too sparse in areal coverage to determine the erosivity value, a crucial input in determining the R value. Fortunately the p2/p ratio can be calibrated and substituted for erosivity (Fournier 1960) where p is the highest mean monthly rainfall and P is the annual mean rainfall.

Only in the upper portion of L'Acul River Basin does specific data exist permitting the USLE to be applied (USDA SCSI 1983) (Figure
3.1) in Haiti without making major assumptions.
No widely acceptable methodologies comparable to the USLE exist for predicting gully, streambank, and subsurface erosion. To accurately assess these forms of erosion the need still exists to measure them in the field or from large scale remotely se 'nsed information. For example, for planning purposes the erosional hazard for gullying can be estimated directly from aerial photographs by calculating simple indices such as gully density. Yet even this simple method requires large scale and recent photographs. Unlike the USLE, this approach cannot be used to simulate the gully hazard if changes in the vegetation and conservation practices would be planned. But it does permit the identification of hazard prone areas if the frequency of gullies can be correlated to specific phenomena such as slope steepness thresholds, paths, and road culverts.
The Land Capability Classification's (LCC) (Klingebiel and
Montgomery 1966) primary objective is to regionalize land units into units of similar rural land use capability. When land is utilized inappropriately either it is underutilized which results in less than maximum economic return; or if it is overutilized, land degradation will occur. In its assignment of land capability the LCC gives heavy weight to the role of slope angle. Furthermore, it does not specify the suitability of the land within each of its land classes for particular crops. However, this can be inferred if the erosional protection of the crops are known. The LCC delimits all lands into one of eight classes sequentially ordered from Class I to Class VIII. Class I includes all lands with little or no risk of soil erosion when placed into cultivation. At the opposite end of the spectrum, Class VIII can only be safely "utilized" in its natural state (Table 3.1). Only classes I through IV are from an environmental perspective suitable for cultivation. In Haiti, all classes nave been used for cultivation at one time or another. This accounts for the rampant erosion throughout the nation, which at its extreme results in exposed bedrock. This misuse of the land, specifically placing lands in cultivation that fall within higher classes (V-VIII), resulting from land scarcity for the small farmer, is one major contributing factor to land degradation.
In addition to the prime designation of land into eight major classes, capability subclasses are formed by adding an additional e, W, or s to the Roman Numerals. The letter e indicates that erosion risk is the major limiting land use factor; w indicates poor drainage or a high water table affects plant growth; and s specifies that the shallow nature of the soil is the prime limiting -factor. In Haiti, the s sub-category has been increasing due to its ongoing erosional problem. As a result, the Haitian land resource is continuously decreasing in terms of its production potential and lands previously in the lower classes have today moved into higher categories.

Cayes Plain Basin *
L'Acul River Basin
-- L'Acul Divide Subsurface
River 'q
... USDA/SCS L'Acul Watershed

I Deep, productive soils, easily worked, on nearly level
land; not subject to overland flow; no or slight risk of
damage when cultivated; use of fertilizers an 'd lime, cover
crops, crop rotations required to maintain soil fertility
and soil structure.
II Productive soils on gentle slopes; moderate depth; subject
to occasional overland flow; may require drainage; moderate risk of damage when cultivated; use crop
rotations, water-control systems or special tillage
practices to control erosion.
III Soils of moderate fertility on moderately steep slopes,
subject to more severe erosion; subject to severe risk of damage but can be used for crops provided adequate plant
cover is maintained; hay or other sod crops should be
grown instead of row crops.
IV Good soils on steep slopes, subject to severe erosion;
very severe risk of damage but may be cultivated
occasionally if handled with great care; keep in hay or
pasture but a grain crop may be grown once in five or six
V Land is too wet or stony for cultivation but of nearly
level slope; subject to only slight erosion if properly
managed; should be used for pasture or forestry but
grazing should be regulated to prevent plant cover from
being destroyed.
VI Shallow soils on steep slopes; use of grazing and
forestry; grazing should be regulated to preserve plant
cover; if the plant cover is destroyed, use should be
restricted until cover is re-established.
VII Steep, rough, eroded land with shallow soils; also
includes droughty an 'd swampy land; severe risk of damage
even when used for pasture or forestry; strict grazing or
forest management must be applied.
VIII Very rough land; not suitable even for woodland or
grazing; reserve for wildlife, recreation or watershed
Classes I-IV denote soils suitable for cultivation. Classes V-VIII denote soils unsuitable for cultivation.
LCC Klingebiel & Montgomery 1966

Since all slopes greater than 70 fall outside lands suitable
for cultivation, the LCC is probably too restrictive in countries when
manual labor and not machinery is the major method of farming. In addition modifications in the LCC for tropical areas is warranted
since generally tropical soils have greater concentrations of iron and
aluminum oxides than mid-latitude soils. This gives them a lower
erodibility than soils of similar texture and structure permitting
their rational utilization on steeper slopes. Nevertheless, the
methodology for classification as presented in the LCC is a useful
planning device as it clearly gives a basis for assessing the relative
erosional risks of tracts of land for agriculture, grazing and
Figure 3.2 illustrates the sequence for assigning units of land
into their appropriate class. The LCC is especially pertinent to the
Haitian situation given the dominant role of agriculture in the
employment and economy of the country. It can, as a first
approximation, rapidly determine current land use that is out of
harmony with the physical environment. As such it is a useful tool for deciding which areas should be targeted into a priority action
designation in order to arrest a region's erosional rates.
3.4.1 Discharge Characteristics
L'Acul River Basin, occupying approximately 30 percent of the
Cayes Plain Basin, is 230 square kilometers in area. L'Acul is
situated along the western margin of the Cayes Plain Basin with its
major river axis oriented from the northwest toward the southeast. In
terms of catchment area, L'Acul Basin is the tenth largest Haitian
river basin (Figure 3.3). The total length of the major stream
segment (Ravine and River Acul) is 34.9 km. With the exception of the
Ravine Caiman all of the headwaters in the basin are intermittent
during a portion of the dry season.
River discharge properties are determined by a complex set of
variables all interacting simultaneously Among the primary variables
affecting discharge are the temperature and rainfall attributes, the
local geology, the topography, the shape of the basin, the vegetation
cover and the soil. Because of the complexity of the interactions
that determine the discharge of a river system, the only accurate
method of obtaining this information is by direct measurements. For
the Cayes Plain Basin as a whole, no continuous stream flow data exist
since 1935 (Harza 1979, p. B-11-4). In L'Acul only one published
figure exists using only one year for measurement, published by
Sheladia Associates (1983). Given this paucity of hydrologic data it
is only possible to crudely estimate that L'Acul's mean annual
discharge at its mouth falls within the range of 2.5 5 m3/sec.
Given the elongated shape of the basin, which should result in an
extreme stream regime having brief periods of large peak discharges

0-1o (1)
50 38 25 50 38 25 100 50 38 25 (3)
75 125 150 125 150 25 125 125 75 125 125 (4)
1-2.5o (1)
A N%%%2
50- 38 25 50 38 25 50 38 25 (3)
7 75 125 150 75 125 (4)
2.5 4.50 4.5 70 (1)
A A\
50-25 38 25 25 (3)
(l)= Slope of land (2)= Topsoil Texture (3)= Effective Soil Depth (CM)
(4)= Permeability of Upper Subsoil (B Horizon) (5)= Class
Figure 3.2

I Igneous Highlands ;. ""'
II Plateau ......... "..
I ~~~ ign'eo"s HigladsLeCye III Slopes IV Flat Lowlands FIGURE 3.3

after periods of rainfall, the steep slopes in the upper and middle
portions of the basin, the removal of the tree cover, and the thin to bare rock soil cover in large areas of the basin, L'Acul River and all of its tributaries must have periods when they flow as raging torrents
with discharge far in excess of their mean flow. The presence of
bedload material over one-half meter in diameter in most of the basins
streams attests to the large velocities experienced during the peak
flows. The general absence of finer materials in the river beds
indicates the highly erosive nature of the river flows in the River
Basin. In order to maximize the water resources of L'Acul Basin,
strategies to dampen the extreme nature of today's river discharges
need to be undertaken. By lowering the peak discharges larger
quantities of water will be available for utilization, especially in the flat lowlands. Today, during the peak flows, a large percentage
of the basins' waters pass through the system and empty into the
Caribbean without being utilized either for irrigation or potable
water supply.
3.4.2 Topography
Using lithologic, slope inclination, and elevation criteria,
L'Acul River Basin can be divided into four major topographic units.
These units differ both in their erosional and agricultural
potential. Because the environmental interaction between these units
is relative to the dynamics of L'Acul river system, the proper
management of these topographic units will both improve their in situ
environmental characteristics as well as improve the lands beyond each
of the units. It is this latter characteristic that necessitates the
need to co-ordinate land use activities within all of the basin's
topographic units if the basin's soil and water resources are to be
used in the most rational manner.
On Figure 3.3 the four major topographic units are delimited.
The greatest relief and highest elevations within the basin are found in the Igneous Highland Unit. This unit comprizes a little less than one percent of the basin or approximately 3 square kilometers (Table 3.2). Elevations in this unit range from 1200 to 2170 meters above
Table 3.2
L'Acul River Basin Topographic Units Name Percentage Area (km 2)
Igneous Highlands 1 3
Platons 6 16
Slopes 68 156
Flat Lowlands 25 55
100 230

sea level. Local relief is 800 meters. It is an area of steep slopes with no flat lands. The northern limit of this unit forms the river divide between those streams flowing southward toward the Caribbean and those flowing northward toward the Atlantic Ocean near Jeremie. The Platons is an area characterized overall as a gently rolling plateau with periodic abrupt limestone hills rising from this rolling terrain. The plateau's surface has a regional dip from the northwest to the southeast. This dip is manifested in the units' elevation which ranges from 1200 meters in the northwest and systematically decreases to 600 meters in the southeast. The underlying lithology of the whole unit is limestone. Because of this rock's permeability and solution weathering properties, the presence of surface water runoff is limited. Most of the area's drainage is underground. The surface runoff flows to the numerous depressions that pock the terrain in short blind valleys. These depressions, known as sinkholes, serve as local collectors of the water runoff. These sinkholes are the major groundwater recharge areas. It appears that the groundwater flow in Les Platons is from the northwest toward the southeast with the greatest outflow emerging at the Tate L'Acul in the lower slope unit. Because most of the water from precipitation in this area flows underground and not on the surface in an integrated river network, a common phenomena in limestone areas, the surface drainage divide on Les Platons is relatively meaningless. The crucial hydrologic divide in this unit is determined by the regional dip of the limestone. For this reason, the Sousbois area in the northwestern portion of Les Platons Unit is included in L'Acul River Basin even though it lies just outside of the topographic divide. All indications are that the groundwater in the vicinity of Sousbois flows toward the southeast and enters L'Acul River System. In its transition to the slope topographic unit, the Platons margins are bounded with steep rocky slopes, in many places greater than 500. These margins are dissected in many places by short but deep ravines. These margins are either occupied by rapidly eroding farmlands, or abandoned fields.
The slope topographic unit makes up 68 percent of L'Acul River Basin. This is the archetypical landscape of Haiti. It is a hodge-podge of fluvially eroded valleys resulting in a complex dissected surface. It is comprised of numerous hills having moderate to steep slopes and narrow-valley bottoms. It is an area where very little flat land is to be seen, and where it exists, it is normally found on the valley bottoms which are susceptible to frequent flash flooding. Within L'Acul River Basin the dominant underlying rocks are either of igneous origin or within the family of rocks comprised of calcium carbonates (OEA, 1972). Slopes range anywhere from 10 to 60 degrees and local relief is between 100 to 200 meters.
The flat lowlands, which are part of the Cayes Plain are an area of low relief and low slopes. The regional dip is from the north to the south. Along the slope unit-flat lowland boundary the elevation of the plain is about 100 meters. From these margins it slopes seaward at relatively uniform rates. The underlying material of the lowlands is primarily alluvium. Being extremely permeable, much of

the runoff flowing from the slope unit infiltrates beneath the
lowlands surface. As such a large reservoir of untapped groundwater likely exists throughout these lowlands. The numerous streams that emerge at springs on the lowlands gives clear evidence of the large
groundwater reservoir existing under this unit.
3.4.3 Climate
Only one meteorological station exists within L'Acul River
Basin. This is at Dubreuil, and this station was only established in
August, 1979. Therefore, no long term rainfall records exist in in
the area. Within the whole of the Cayes Plain Basin, stations are
located only at Les Cayes, Gerard (Torbeck), Levy Farm, Chantal, and
Camp Perrin. Therefore no actual climatological data exist for the
slope, Platons, and Igneous Highlands within L'Acul River Basin. That is, for the whole of the Cayes Plain Basin actual data only exist for the flat lowlands. As a result all statements concerning the upland
areas are based on inference using the roles of elevation, topographic
orientation, and distance from the ocean in determining the
temperature and precipitation values of these areas. It is from these
inferred values that the crucial climatological properties for
agricultural production in the upland areas such as
evapo-transpiration rates are derived.
The minimum rainfall in L'Acul and the Cayes Plain Basin is
along the coastline. In this area the annual rainfall is
approximately 2000 mm (Les Cayes 2035 mm). Moving inland, rainfall
gradually increases on the lowlands (Chantal 2464 mm; Camp Perrin 2272
mm). By extrapolating annual precipitation from the National
Meteorological Service's average annual precipitation map, rainfall in
the upper reaches of the slope unit and Platons must range between
2400 and 2800 mm per year. Along the upper margins of the highlands
precipitation likely over 3000 mm. A humid climate is found
throughout the whole basin. The lowlands have a definite wet-dry
season regime. As you move inland, the elevation rises and the
climate becomes ever more humid until you reach the Igneous Highlands
that experience only a very short dry season. This overall humid
climate, along with the dry seasons, makes the area have very
favorable moisture conditions for agriculture, forestry and livestock
With the average annual temperature at Les Cayes 25.80 C
(elev. 5 m), Chantal 26.10 (elev. 40 m), and Camp Perrin 24.90
(elev. 230 m), we can assume the mean annual temperature at Le Pretre
(180 m) is approximately 25.40 C, at the Citadelle des Platons (680
m) 22.00 C, and at Sousbois (920 m) 20.70 C (Figure 3.3). These
figures are derived assuming the average environmental lapse rate in
the area is 0.60 C per 100 meters, the average rate for the
troposphere. The highest monthly temperature for the lowlands is
about 27.50 C in July and 22.40 C at Sousbois. The lowest mean
annual temperature on the lowlands is about 24.00 C (Jan) and
18.10 C on the Platons.

From an agricultural point of view, these raw temperature and rainfall figures need to be used to derive the evapo-transpiration rates and the moisture balance for the Cayes Plain Basin. Given that no areas of below freezing temperatures exists in the whole basin, it is the moisture balance that determines the growing season for all rainfed vegetation. It is the deficit between potential evapo-transpiration and precipitation that determines the amount of water needed in areas undergoing irrigation. Using the methodology developed by Hargreaves and Samani (April, 1983), the moisture availability index (MAI) is approximated for Le Pr~tre, the Citadelle, and Sousbois. Their published MAI values are used for Les Cayes, Camp Perrin, and Chantal (Hargreaves and Samani, 1983, pp. A-19, A-21, A-43).
The MAI is a particularly useful agroclimatological index for planning purposes in that it is an index of relative rainfall adequacy. It is defined as:
where PD is the 75 percent probability of precipitation (monthly) and ETP is the monthly potential evapo-transpiration. An MAI value of
0.33 or less for one month is considered a danger signal in rainfed agriculture. At this level, production most likely will drop below an economical level. This is especially true if it occurs before the plant has reached its critical growth stage. If the MAI value is less than 0.50 for the growing season, the use of fertilizers is not considered economical since not enough moisture exists to allow the fertilizer to become soluble and then be utilized in plant growth. The MAI value, since it detects moisture deficiencies is useful as a planning guide for irrigated areas too. Conversely, a high monthly MAI index, over 1.20, generally indicates a need for drainage as water logging will occur unless the soils are extremely permeable. Given the poor moisture retention of many Haitian soils because of their degraded state, these high MAI values probably only indicate drainage problems along valley bottom lands and other flat lowlands. In fact, it is probably only during periods of high MAI values ( 0.75) that most soils have enough moisture storage as to prevent moisture stress from occuring.
Table 3.3 lists the published MAI values for Les Cayes, Camp Perrin, and Chantal. The estimated values for Le P~tre, the Citadelle, and Sousbois are also given. These values were derived by estimating the monthly temperature and rainfall of these locations using available data. The probability distribution of this rainfall was assumed to be similar to the closest published station in the Hargreaves and Samani report. Using their computation method the MAI values for these three sites were then derived. A brief inspection of Table 3.3 indicates that all of the Cayes Plain Basin, from a moisture balance perspective, is a good agricultural region.

Ca..-iriin 0.30 0.28 0.39 0.74 1.17 0.66 0.45 0.64 1.11 1.65 0.86 0.18 1.15 2272
Chantal 0.25 0.45 0.46 0.95 1.56 0.90 0.57 0.76 1.59 1.55 0.25 0.25 1.14 2464
Les Cayes 0.36 0.45 0.33 0.82 1.36 0.76 0.63 0.90 1.08 1.57 0.74 0.18 1.20 2035
Citadelle 0.37 0.51 0.57 1.03 1.71 0.92 0.71 0.94 1.72 1.66 0.87 0.36 1.41 2670
Le Pritre 0.34 0.44 0.51 1.00 1.44 0.94. 0.61 0.83 1.49 1.57 0.37 0.35 1.23 2510
Sousbois 0.44 0.52 0.58 1.03 1.80 0.94 0.76 0.99 1.83 1.64 0.86 0.35 1.44 2675
* = Estimated Values
Derived Using Methodology Explained in

At Camp Perrin December through March are relatively dry, while at the opposite extreme, Sousbois only appears to have a moisture problem in December. The flat lowlands and the lower elevations in the slope unit fall within a climatic zone having a rainy season between seven to nine months with either two or three months of excessive rainfall. The upper slopes, Platons, and Igneous Highland Units appear to have a ten month rainy period, with only December and January having really dry periods. From the consideration of rainfall inputs, the whole Cayes Plain Basin appears to be one of the most favorable locales in all of Haiti.
The extreme apparent moisture conditions inferred from Table 3.2 need to be tempered due to local soil conditions and microclimatic properties that are widespread throughout the basin, excepting the flat lowlands. First, visual impressions of the vegetation throughout the Platons seem to indicate that its soil moisture availability is not as high as in the slope valley bottom areas such as Le Pretre. Most likely this reflects the higher permeability rates on the Platons due to its limestone geology (Wooding, etal, 1924). Therefore Table
3.3 most likely overestimates the available moisture in the Platons (Citadelle and Sousbois). Second, the thin nature of the soil throughout most of the slope unit means that despite the favorable macro-moisture conditons, the residual soil moisture storage capacity of the soils is low. Thus, despite no moisture stress being indicated for most of the year, in actuality pseudo-drought conditions could exist whenever rain events do not occur over a two-week period. Evidence for this inference is manifested by the extreme difference in the quality of vegetation between the east and west facing slopes throughout the whole slope unit. In all cases the west facing slopes had a lower quality of vegetation than the east facing slopes. This probably reflects the microclimatic differences resulting from the temporal pattern of exposure to sunlight and the slope's low soil moisture retention capacity.
The east facing slopes have a more favorable micro-climatic environment than the west facing slopes, all other factors held constant. This is due to the presence of ground moisture (dew) being present when the sun shines on the east facing slopes during the morning period. The evaporation of the ground moisture and dew from the plant leaves prevents the temperatures, both ground and plant, from becoming excessive. Thus the plants for a given quantity of solar insolation do not transpire as much water. However, by the time the sunlight strikes the west facing slopes (noon and afternoon) this ground moisture has been evaporated. Therefore the incoming solar insolation raises the ground and plant temperatures to a greater degree requiring the plant to place heavier demands on the soil moisture. Simply put, the east facing slopes are more humid than the west facing slopes. This allows a greater plant density to exist on the west slopes, with the result that the soil has been better protected through the years. In fact, almost all areas having bedrock exposed on the surface in L'Acul River Basin are on west facing slopes. Throughout L'Acul River Basin, the east slopes have been

severely degraded. The existing thin soils cannot store enough
moisture and plants undergo moisture stress despite MAI values far in excess of 0.33. Most of the fields on these slopes have been already
3.4.4 Soils
According to the DATPE classification of soil potential (1982)
the Cayes Plain Basin has some of the highest potential soils in Haiti
(Figure 3.4). Conversely, it also has areas with no agricultural potential. In general the soils having the highest potential are
found solely on the flat low lands; the lowest potential soils are
found in the Igneous Highlands and Slope Units; and good to average
soils are found on the Platons. On the basis of the DATPE
classification, the majority of the slopes unit lands should be solely in forest and where agriculture is practiced, major soil conservation
practices should be in place. In reality, most of the lands in the
slope unit are in cultivation and little soil conservation practices
are in place. Already, as mentioned in the climate section, the west
facing slopes in many areas are devoid of soil. This has been the
product of a large number of years of misuse. In the upper portion of
the Ravine Des Mornes, the Ravine Caiman and on the margins of the
Platons, lands in excess of 450 are being used for sorghum and sweet
potatoes. Clear physical evidence on these fields, such as large
rocks, point to rapid erosion.
The detailed soil mapping done by the USDA-SCS in the upper Acul
Basin indicates that no Ciass I soils exist within the upper Acul
(USDA-SCS, June 1983 p. 13). Only 9.4 percent of the upper Acul Basin
according to the LCC classification should be placed in cultivation
(Table 3.4). Yet, over 45 percent is in cropland and another 18
percent is periodically cultivated. This overutilization of marginal
soils is widespread throughout the Slope and Igneous Mountain Units.
Soil Classes of Upper Acul
LCC Percent of Total Area
A. II 0.13
III 1.7
IV 7.6
B. V 1.9
VI 16.1
VII 68.3
VIII (Wasteland) 4.2
Modified from USDA-SCS June, 1983 P. 14

Distribution of Soils.8 in Cayes Plain Basin
Good to Excellent Les Cayes
Poorly Drained, Limited to Rice Culture
Limited Potential Modified from Datpe-Potentialitee Des Soils, 1982 FIGURE 3.4

It implies that within L'Acul River Basin active land degradation is occurring which is lowering the lands' potential both in the eroded
areas and on the flat lands where excessive soil is deposited.
The Cayes Plain Basin, specifically L'Acul River Basin,
possesses many physical properties that are common throughout most of
Haiti. The utilization of the lands within the area likewise presents a potpourri of conditions typical of the rural countryside. A limited
resource base is and has been stressed over a long period by an ever
growing population pressure resulting in a degrading environmental
situation. This ongoing process must be arrested quickly, otherwise a
large portion of the productive lands will cease to able to be
The highest potential lands are the relatively limited flat
lands concentrated within the Cayes Plain and the Platons. Neither of
these areas are producing their potential agricultural yields. In
both cases, if more water were available for irrigation, farming
practices improved, and selected crop substitution occurred, yields
and income could be increased. To increase the hydrologic potential of the Cayes Plain Basin, a set'of actions need to be undertaken that will alter existing land use and farming practices throughout all of
the interconnected topographic units comprising the area. The
Highlands, Platons, Slopes, and Lowlands all interact with each other
in terms of their physical properties. The co-ordinated strategy as
well as possible interventions required to improve the physical
setting of the area for rural regional basin development is found in
Chapters seven and eight.
Given the information presented in this chapter, three crucial
changes that this strategy must accomplish are:
1. Arrest the pervasive soil erosion occurring in the slope and
highland units;
2. Improve the quantity and availability of water in the Platons
and Lowland units; and
3. Alter the existing farming practices to introduce different land
use and conservation practices to increase the profitability of
the small farmers.

Marketing is a process which coordinates the exchange of
resources, goods, and services over time and space. This process
represents the direct interaction of supply and demand and is the sine
qua non of any but the most rudimentary economic system. The unit
value-of the resource, good or service exchanged should reflect its marginal cost of supply to the seller and its marginal value to the
buyer. If this mechanism is inefficient or costly, scarce resources
will tend to be misallocated and the marketing process becomes a
constraint to economic growth. Conversely, the efficient operation of this mechanism will ensure a satisfactory (if not optimal) allocation of resources among society and will create the potential for improved
economic performance.
The purpose of this chapter is to describe and to analyze the
market structure of L'Acul River Basin 1: to identify the location
of the markets in time and space; to delineate both the functional
hierarchy of this market system and the nature of the commodity flows
between markets;.to examine spatial and temporal price variations between markets; and, finally, to examine the constraints to, and opportunities for, development, associated with this market system.
Before undertaking this detailed analysis of marketing in L'Acul River
Basin, section 4.2 considers some general issues and approaches
relevant to the internal marketing system of Haiti. Note that the
focus of this chapter is the internal (i.e. intraregional or
interregional) market for food stuffs and household articles. Export
commodities such as coffee, sisal, and vetiver are marketed through
independent systems which are quite different from the former in both form and function; these systems are beyond the scope of the present
Over the past two decades a fairly sizeable literature on Haiti's
internal marketing system has developed. This literature covers a
spectrum of issues and approaches. The purpose of this section is to explore the nature and range of topics examined before moving on to a
substantive analysis of L'Acul River Basin.
lAlthough this study focuses upon L'Acul River Basin, the
marketing system of the basin is broadly defined to include
functionally related markets within the broader Cayes Plain Basin.

LaGra, Fanfan, and Charleston (1975) have produced a fairly
comprehensive enumeration of 519 public markets throughout Haiti. For each market they identify its location, principal day(s) of functioning, type of construction, approximate size, principal modes of transportation, origins of motorized transport, and the state of the roads leading to the market. In addition, they develop and apply a three-level typology of markets: semi-rural markets, which serve local producers and consumers, as well as some Madame Saras (traveling intermediaries that assemble and break bulk and transport goods between regions); regional markets, which are assembly points characterized by a variety and abundance of products, and which serve as principal points of exchange between producers and Madame Sarans; and urban markets, large commercial centers in which exchange occurs through several levels of intermediaries. Note that a regional market may function as a semi-rural market (i.e., with a lower range and quantity of goods, with a smaller number of buyers and sellers, and exclusively for a local market) on certain days of the week. LaGra et al argue that the prices in both semi-rural and regional markets r-flect prices paid at the producer-level; urban markets are characterized by higher prices than the other two types, due to transportation costs and the number of intermediaries involved.
A number of other authors have developed similar typologies. These are summarized in Table 4.1. The most utilized of these typologies is that of Mintz (1960b) who identifies the following four market types: 1) strategic marketplaces which play a central role with respect to the entire market system, serving as bulking and break-bulk points for interregional trades; 2) "captured" town marketplaces in which shops dominate the market and which provision local middle-class merchants; 3) mixed marketplaces, local markets with associated subsidiary trading activities such that its relationship to the outside is not as unilateral in character as a local market; and 4) local marketplaces which are outlets for consumer goods via local exchange (and which may also be a strategic marketplace on one of its non-principal days). The other typologies presented are self-explanatory; perhaps the only one requiring comment is that used by the Institut Francais d'Haiti. The type 1, 2, and 3 markets are based upon the role of Madame Sarahs from Port-au-Prince, indicating respectively: 1) direct interaction between Port-au-Prince Madame Sarahs and producers; 2) interaction between local revendeuses and producers; and 3) interaction between Port-au-Prince Madame Sarahs and revendeuses. A type 4 market is also implied but not observed in

the study area; this corresponds to the "urban market" of LaGra et al
(1975). It should also be noted that Le Bureau des Contributions
employs the simple distinction of the urban market, which takes place
in a center in which a communal magistrate resides, and the rural
market, which is located in a section rural.
Typologies of Haitian Markets
Author l(highest) 2 3 4Clowest)
LaGra etal (1975) Urban Regional Semi-rural
Mintz (1960b) Strategic Captured Town Mixed Local
Moral (1959) Sea Cost Regional Town Local
Wood (1963) Port Interregional Intraregional
Johnson (1970) Port Major inland Local
Assembly and
Institut Francais Type 4 Type 3 Type 2 Type 1
d'Haiti (1980)
After Lundahl (1980)

A second major issue evident in the marketing literature on
Haiti involves the extent to which the market system is characterized by perfect competition and efficiency, as opposed to imperfect competition, inefficiency, and exploitation of the peasant consumers. Moral (1959) was one of the first authors to argue that Haiti's internal market system operates under imperfect competition with the end result that the peasants are victimized. Jaffe (1983) describes the market system as monopolistic-monopsonistic in its process of price determination. The lack of competition is responsible for a dendritic market structure which abets economic stagnation. Johnson (1970) takes a similar view with regard to both the dendritic structure and its consequences. Opposing views are taken by a number of authors. Johnson and LaGra (1975) found price variation characteristic of imperfect competition between geographic areas but not between markets within the same area; transport costs between areas seem to account for what variations do exist. Mintz (1964), citing a high level of competition between market intermediaries, concludes that the system is free from exploitation. The level of market services may be low but society pays a low cost for them. A large number of participants is substituted for a high level of productivity. Lundahl (1980) characterizes the market system as dendritic ("vertical" in his terminology) but argues (1983) that it is relatively efficient and that there is relatively little exploitation of the peasants by middlemen; no monopsony can be detected and prices are set by the interaction of supply and demand.
Two further related issues involve levels of price variation between markets and the distribution of profits among market participants. As noted above, Johnson and LaGra'(1975), using average monthly prices for 6 commodities over 37 markets for the period 1971-1974 found significant levels of price variation between geographic areas but did not detect significant variation by market type within a geographic area. Lundahl (1983), using 1969-1974 average monthly price data collected for regional markets by the Inter-American Institute for Agricultural Sciences, found some degree of price variation beyond the expected differences caused by transportation and processing. With regard to the distribution of profits among market participants, Murray and Alvarez (1973) found a 25 percent mark-up between the price paid to the producer and that paid by the final consumer; Madame Sarahs gained, on average, a 10 percent profit on their role as intermediary. Lundahl (1983) estimates that both revendeuses and Madame Sarahs make a profit of 10
- 20 percent on their transactions, but notes that the margin is, in fact, somewhat less since the cost of their labor is not taken into account. L'Institut Francais d'Haiti (1980) arrives at the following ranges in relation to the allocation of final selling price: producers, 45-78 percent (mean = 61 percent); Madame Sarah, 18-48 percent (mean = 31 percent); transportation, 1-24 percent (mean = 7 percent); depot operators, 0.1 4.4 percent (mean = 1 percent).

A detailed analysis of the roles of various actors in the
marketing process has been made by a number of authors. The
Inter-American Institute for Agricultural Sciences (1974) presents an overview of the various types of producers, consumers, and middlemen.
Murray and Alvarez (1973) focus upon the role of the Madame Sarah, and
Mintz (1960a; 1960b; 1964) presents a detailed account of the
functioning of both Madame Sarahs and revendeuses.
Other approaches have examined the marketing of individual
commodities. Murray and Alvarez (1973) have focused upon beans;
Sorenson and Chung (1973) upon corn, rice, and millet; and Roe (1978)
upon a variety of grains, vegetables, and fruits.
Finally, Rouzier (1981) has used the structure of the Haitian
marketing system (both internal and international) as an example in
the context of his theoretical micro-econometric discussion of
international and interregional trade flows. Rouzier's principal
purpose is the development of an abstract conceptual framework and
his discussion contains little substantive information about marketing
in Haiti.
In the following sections an examination of the marketing system
of the study area is undertaken in light of some of the major issues
and approaches that have been raised in the literature. By gaining an understanding of the manner in which this system functions it becomes
possible to assess its potentialities for enhancing the development
process in L'Acul River Basin. Unless otherwise noted, the following
observations are based upon this Study Team's field work, undertaken
in L'Acul River Basin between November 15 and 29, 1983.
Thirteen markets in L'Acul River Basin, as broadly defined
above, were examined. The location of these markets and the routes connecting them are shown in Figure 4.1; their location in time is
presented in Table 4.2. Tuesday is the most significant marKet day in L'Acul River Basin; the Study Team was told repeatedly and in various
locations that Tuesday is the market day of the "professionals",
meaning the Madame Sarahs and the revendeuses. This statement also reflects the central role of the Ducis market in provisioning other
markets in the area. This will be discussed below.
Figure 4.2 combines the spatial and temporal dimensions of the
market system. A definite regionalization of principal market days
may be observed. Taken together, there is a temporal complementarity
among the larger markets, i.e. Ducis meets on Tuesday, Cayes on
Wednesday; Chantal on Friday. This permits a maximum flow of goods and intermediaries between these major markets. Table 4.3 presents
additional information about each market: Type of infrastructure,

Market Days, L'Acul River Basin
Market Principal Day(s) Secondary Day(s)
Arniquet Monday Thursday
Carrefour Valbre Monday Thursday
Ducis Tuesday Friday
Canon Tuesday Friday
Le Pritre Tuesday Friday*
Caiman Tuesday Friday*
Platons Tuesday All Days
Except Sunday
Cayes Wednesday Saturday
Chantal Friday
Camp Perrin Friday Monday
Sousbois Saturday Thursday
Carrefour Joute Saturday Wednesday
Torbeck Every Day
*Market also functions in a reduced capacity every other day of the
week except Sunday.

Camp Perrin Platons
Le Pretre
a Torbec

Camp Perrin ( Platons
Le Pretre B Canon OT T*Caiman DucisOT % \
* MARKET Chant
F BE (E Everyday)
Ar n iq u e V a r e o L'ACUL RIVER BASIN ArniquetM Carrefour
Va \lee

Table 4.3
Market Characteristics
Market Infrastructure No. of Modes of Index of
vendors transportation road conditions*
Arniquet 120 permanent 150 Foot, animal truck 25
tin & straw
Carrefour 38 permanent 50 Foot, animal truck 25
Valbre tin & straw
Ducis 200 permanent tin 300 Foot, animal truck 25
& straw tonnelles
Canon 15 Foot, animal 8.5
Le Pritre 8 permanent 30 Foot, animal 7.5
Caiman 4 permanent 20 Foot, animal 3.5
Platons 15 Foot, animal 3.5
Cayes Permanent 300 Foot, car, truck 40
semi covered
Chantal 200 permanent tin 300 Foot, animal, truck 25
& straw tonnelles
Camp Perrin 100 permanent tin 120 Foot, animal, truck 30
& straw tonnelles
Sousbois 35 permanent straw 50 Foot, animal 3.0**
Carrefour 8 permanent tin 100 Foot, animal, truck 25
Joute & straw tonnelles
Torbeck 13 permanent tin 20 Foot, animal, truck 30
& straw tonnelles
*Defined as a "safe" speed in a four wheel drive vehicle, miles per hour.
**Represents walking speed; inaccessible to vehicles.

approximate number of vendors, principal modes of transportation, and
an index of road condition for access roads.
Marketing systems consist of interrelated individual elements
which may be distinguished from one another on the basis of the function that each provides in the context of the entire system.
These functions are expressed in terms of level in the hierarchical system of markets. Such systems are conventionally analyzed within
the framework of central place theory. In the present study the
central place approach is employed as a frame of reference; as will be
seen subsequently, the divergence of L'Acul River Basin system from central place theory is more striking than any similarity. Central
place theory is based upon six propositions:
1. There exists a hierarchical set of markets which are distinguished
from one another on the basis of the range of goods and services that
each provides. Higher level centers (markets) exhibit all of the goods and services available at lower level centers, plus a set of
goods and services that the lower level centers do not provide.
2. Higher order goods and services require a higher population
concentration to support them than do lower order goods and services.
This is referred to as the threshold of a good.
3. In order to achieve the higher threshold required,' higher order
centers are characterized by a larger range than are lower order
centers. That is to say, higher order centers have more extensive
market areas associated with them than do lower order centers.
4. Consumer movements to lower order centers tend to be frequent and
to cover shorter distances (a smaller range), while movements to
higher order centers tend to be less frequent and cover larger
5. In most instances goods tend to move vertically, up and down the
hierarchy from one level to an adjacent level rather than horizontally.
6. Market areas are arranged in a nested fashion such that the market
area for a higher order center will contain several market areas for each of one or more centers tend to be spaced farther apart than are
lower order centers.
In contrast to the market hierarchies summarized in Table 4.1
which are, in general, based upon the nature of the settlement in
which the market is located (e.g., rural, town, urban), and to a lesser extent, upon the area that the market serves (e.g., local, interregional), the present study defines the hierarchical market
system on the basis of the types and quantities of goods available at
individual markets. Having examined the range of goods provided at
each of the 13 markets in the study area, a hierarchy was established
on the following basis:

Level 5. Basic items: foodstuffs, soap, thread,
(lowest) matches, notebooks, pens, small amount of cloth, small amount of meat.
Level 4. Basic items plus sundries: charcoal, batteries, plastic ware, small amount and limited variety of clothing.
Level 3. Basic items and sundries plus houseware/clothing: metal
cookware, cups, glasses, hardware, shoes, large amount and variety of clothing.
Level 2. Basic items, sundries, and houseware/clothing plus
specialty items: briefcases, handbags, straw products and hats, furniture, livestock in large quantities (alive and butchered) tools, rubber boots.
Level 1. Basic items, sundries, houseware, clothing, and
(highest) specialty items in greater quantities and variety than in Level 4.
Table 4.4 and Figure 4.3 illustrate the nature and location of
the hierarchy observed in L'Acul River Basin marketing system. The large urban market at Cayes is distinguished from the next highest
level (Ducis and Chantal) principally in terms of the amounts and variety of goods available; the Cayes market does not represent a
discrete step up in terms of the order of goods provided. Referring
to Table 4.3, one can also note that the size of a market, in terms of
the number of vendors, does not necessarily accurately reflect its
level in the hierarchy. For example, Arniquet with 150 vendors and
Caiman with 20 are placed at Level 3; similarly, Platons and Canon
with 15 vendors each and Carrefour Joute with 100 are placed at Level 5. In both instances the seemingly disparate markets are
functionally analogous to one another. The wide spreads in the number of vendors may be explained by the population densities supporting the
markets. Arniquet and Carrefour Joute have much higher population densities than do Caiman, Platons, or Canon, and can thus support a
greater number of vendors.
It is interesting to note that no clear central place spacing is
evident from Figure 4.3, although Level 3 and 4 markets do tend to be
rather evenly distributed. The two Level 2 markets, Ducis and
Chantal, are much closer together than would be expected on the basis
of central place theory. However, one must consider the time
dimension; the two markets are not in direct competition, but rather,
are complementary.
The flow of commodities between markets is shown in Table 4.5.
This summarizes both origin and destination of goods. To facilitate
the interpretation of this table, Figures 4.4A 4.4M depict the
origins of goods flowing to each market. Several observations may be

L'Acul River Basin Marketing System Hierarchy
Level Characteristics* Market
1 Basic items, sundries, houseware/clothing Cayes
specialty items in larger variety and quantity
2 Basic items, sundries, houseware/clothing Ducis,
specialty items Chantal
3 Basic items, sundries, houseware/clothing Arniquet
4 Basic items, sundries Camp Perrin
Carrefour Valere Torbeck, Le Pretre 5 Basic items Platons, Canon,
Carrefour Joute
*Defined in text

Sousb Ois*3
3 (Lowhest) 5
L'AUL IVE BAINArniquet Carrefour

So**.*o FIGURE 4.4 A-M
0 Camp Pri
LES CAYES-c ----n*
Major Flow
Minor Flow sCaaalO CAye
Carrefour Vallere o t. Satu L'ACUL DIVIDE SUBSURFACE
areusoue Frt-u-Pinc


- L'AAC~ 1wf 1111UBASIN

* C......,V.U.

made concerning Table 4.5. First, it is clear that the classical
central place-type flow of goods is not found in L'Acul River Basin marketing system. That is to say, although goods move vertically up
and down the hierarchy they tend not to pass through markets at
adjacent levels of the hierarchy. Contrary to the suggestions made by Johnson (1970) and Jaffe (1983), the marketing system is not dendritic but, rather, may be described as primate. This term is generally used
to describe the non-hierarchical direct connections between the
largest (primate) city and all other cities in developing countries.
Its use here indicates that direct connections exist between the
higher order markets (e.g. Cayes, Ducis, Chantal, Sousbois) and the
majority of other markets, irrespective of their levels in the
hierarchy. Note that there are several instances of direct links to Port-au-Prince. These findings help to reinforce the delineation of
the market hierarchy specified above.
Second, with few exceptions, markets at the same or similar
level of the hierarchy tend to have similar linkage patterns, for
example Cayes, Ducis, Chantal; Arniquet, Caiman; Platons, Canon. The
major exceptions to this generalization are: Sousbois, which has a
more extensive linkage pattern than other level 3 markets due to its strategic role as an interface between L'Acul Basin marketing system
and markets in the mountains and on the northern coast of the
peninsula; and Carrefour Valere and Carrefour Joute which serve
principally as markets for local produce.
Finally, one should note the role of the population in and
around Platons in maintaining linkages between, on one hand, Sousbois and the markets toward the north; and, on the other hand, those in the
study area. It is common for the residents of Platons to make the 4
hour round trip to Sousbois on a Thursday or Saturday and the 5-6 hour
round trip to Ducis or Chantal on Tuesday or Friday. In this manner
the flow of goods between the uplands and lowlands is facilitated.
This is not to imply, however that other linkages between the uplands
and lowlands do not exist; it is not uncommon for a 10-12 hour (or
longer) round trip to be made between the uplands and lowlands in one day. It is possible for the market women from the uplands to catch a truck to Ducis at Le Pretre, if the river is sufficiently low (cost 6
G per person round trip), or, if not, at Dubreuil (cost: 4 G per
person round trip). From Ducis transport can be obtained to Cayes at
cost of 2 G per person and an additional 2 G per sack each way.
Direct connections by truck may be made between Ducis and Port au
Prince every Tuesday for a cost of 15 G per person each way.
Over the two-week period in which the marketing system was
studied, (the major portion of markets being visited in the first
week) no significant difference in the individual prices of a common

Origin-Destination Matrix:
Commodity Flows
Destination/ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Origin CY D CH A CA S CP CV LP T P CN CJ PaP 0
1 Cayes / + + + + + + + + + + + + +
2 Ducis + / + + + + + + +
3 Chantal + + / + + + + + +
4 Arniquet + + /
5 Caiman /
6 Sousbois + / + +
7 Camp Perrin- + /
8 Carrefour
Valere /
9 Le Pritre / +
10 TorDeck /
11 Platons /
12 Canon /
13 Carrefour
Joute /
14 Port au
Prince + + + + +
15 Other + + /
+ Major Flows
- Minor Flows

list of products was found between markets. 1- This finding is somewhat surprising given the functioning of all other marketing systems with which the authors are familiar. The lack of variation can, however, be explained by two factors. First, time has no intrinsic monetary value in the "accounting system" of the Haitian individual or household. Therefore consumers are willing to spend several extra hours travelling to a market where they can obtain the lowest price on a specific item. This high degree of consumer mobility creates strong competition between markets and enforces a high degree of price uniformity. Second, as noted above, L'Acul River Basin marketing system is primate rather than dendritic in nature. The direct vertical connections imply a reduced level of intermediary activity (i.e. intermediate linkages between adjacent levels in the hierarchy) and thus a minimal level of mark-ups as the commodities reach their final destination. Both Johnson (1970) and Jaffe (1983) argue that a dendritic system, characterized by lack of horizontal competition, leads to price discrimination and exploitation of the consumers by intermediaries. This may be so but the present evidence indicates that a primate structure can compensate for lack of horizontal competition. In summary, then, there is every indication that L'Acul River Basin marketing system possesses a price structure that responds directly to the stimulus of supply and demand and tnat the system is both relatively efficient and non-discriminatory.
Table 4.6 summarizes typical system-wide prices for a list of directly comparable and commonly purchased commodities. It must be stressed that these prices and the general lack of price variation between markets are associated with a specific time period; without further study the results cannot be generalized for longer time periods or for other markets. Figures 4.5 (A through E) illustrates the temporal variations (in terms of average monthly prices) for some specific commodities at three markets in the study area. These data suggest varying degrees of price variation, at least between Cayes Camp Perrin, and Ducis/Chantal. The latter pair are treated as one since there is no variation in the price of commodities between them.
Figure 4.6 shows temporal price variations for four commodities at the Cayes market. Both sets of data are presented for illustrative purposes and no attempt is made to interpret them due to their unknown level of reliability. Variations in the supply of certain commodities are, however, evident in the data. For example, the relative supplies of sorghum and corn, as reflected in the market prices, are exactly out of phase; the ieak in the corn supply corresponds to the lowest availability of sorghum and vice-versa.
1 Market prices for commodities were collected by two
Caucasians, one fluent in Creole. It was made clear to the vendors that the information was required for a study; there was no intention of purchase. In this way the true market price rather than the "1prix blanc" was obtained. For verification, 3-5 vendors were asked the price of an individual commodity.

Market Prices:
L'Acul River Basin Marketing System November 15 29, 1983
Item Price per Unit
Congo Beans 20 Kob/verre*
Black Beans 1 Gourd/verre
Red Beans 90 Kob/verre
Country Rice 1 Gourd/verre
Milled Corn 50 Kob/verre
Unmilled Corn 35 Kob/verre
Wheat 50 KoD/verre
Brown Sugar 20 Kob/small roll
Milled Sorghum** 60 Kob/verre
Salt 15 Kob/verre
Okra 20 Kob/low
Matches 2.5 Gourd/package
Farine France 60 Kob/verre
Garlic 20 Kob/head
Tobacco 1 Gourd/4 leaves
Soap 1.10 Gourd/bar
Peanuts 25 Kob/petit verre
Cigarettes 3.25 Gourd/package
* There are 11.5 verre in a marmite.
** In short supply; not available everywhere.

.95G Gourds/Pound
Figure 4.5 A thru E s A
Nov, Dec. Jan. Feb. Mar. April May June July Aug. Sept.
1983 1983
2.90G C
BLACK BEANS 2.70G Gourds/Pound
2.3G 2.50G
1. 81 2. 1OG
1. 7G
1. 6G \. 90G
1 5 G
I.4G '- 1. 70G
1. 3G
Oct. Nov. Dec. Jan. Feb. Mar.April May June July Aug.Sept. Oct. Nov. Dec. Jan. Feb. Mar, April May June July Aug. Sept.
1982 1983 1983 1982 1983 1983
1.251 D 2.00G E
1.15G Gourds/Pound CAMP P 1.80G Gourds/Pound
1.05G 1.60G
0.95G 1.40G
0.85G 1.20G CAES
0.75G 1.000
0.65G 0.80G CAMPERIN
0.55G0 0.60G
Oct. Nov. Dec. Jan. Feb. Mar.April May June July Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar.April May June July Au.Sept.
1982 1983 1983 1982 1983 1983

19 AVERAGES/MONTH, 4/80 9/82
SOURCE: IDAI; compiled by M. Yeats 18 17 16 15
14 13 RICE
Apr. M J J A S J A M J J A S N f
1980 1981 1982

A final consideration relating to market prices is the amount of
profit accruing to the various participants in the marketing process.
Table 4.7 illustrates typical profit margins for vendors selling
directly to consumers (i.e., retailers rather than intermediaries) for a variety of commodities. Due to time and manpower constraints it was
not possible to measure the profit margin realized by other
participants. In general, there tends to be a lower mark up on
commonly utilized foodstuffs and a higher one on tools, h 'ousewares,
and imported foodstuffs, e.g. garlic. Where appropriate, the profit
margin has been adjusted to include the cost of motorized
transportation. Although there is a relatively wide range in level of
profit, values clustering around 10% are the most commonly
encountered. It should be noted that one way in which the vendor may
increase her profit margin is to process raw or semi-finished
products, thus enhancing the final selling price. This increase in
value added frequently occurs in the case of milling corn en grain and
coffee en paille, drying green rice, and baking bread.
L'Acul River Basin marketing system has been described and
analyzed. The general impression left by this examination is that the system is a fairly uniform and relatively closed functional unit with few lateral external linkages. The system is a competitive one which
is non-exploitative and which is relatively efficient (from a
viewpoint which does not place a monetary value on the expenditure of
time). It remains to discuss the implications for development that
are raised by this analysis of the system. These implications are
relevant not only to the marketing system per se but also to the
broader economic system to which it is intimately related. It should be stressed that at this stage of the study (i.e., before passing the
results through our two social-institutional filters) the following
observations are only that; they are not yet specific recommendations
for development
Export Development. Certain agricultural products that are
grown or that could be grown in the Cayes Plain Basin have the
potential for serving export markets at both the interregional (i.e.,
Haitian) and international levels. Among the existing crops the
expansion of coffee production is the most fruitful possibility, given
the introduction of better management practices. In addition, coffee has the added benefit of enhancing soil conservation. It is difficult
to identify other crops with the potential for international export
given the large, complex, organized, and independent marketing
infrastructures required. Existing crops with the potential for
interregional export include bananas (which have had in the past an international significance) and the various citrus tree crops. Both
bananas and citrus crops are conducive to soil conservation.

Retailer Profits
November 15 -29, 1983
Commodity Profit M%
Unmilled Corn 5
Soap, Cloth 6
Sorghum, Coffee, Straw Hats
Country Rice 7
Peanuts, Wheat 8
Milled Corn, Black Beans 9
Bread, Brown Sugar, Matches 10
Fish 15
Flour, Rice Jean-Claude 15
Rubber Boots 18
Cigarettes, Plastic Bowls, Machetes Coconut pieces 20
Garlic, Plantain 25
Carbonated Beverages 29
Batteries (flashlight type) 38
Charcoal 40
Salt 43
Glassware 50

Onions, radishes, and possibly garlic and potatoes are among the potential crops with significance for interregional export. With time and good management, wood may become a source of export income. The difficulty here is that because of the time lag between the initial planting and the first "harvest", the small farmers must find alternative short-term sources of income. The practice of continuously planting and selectively harvesting trees is, of course, one of the better soil conservation practices. Finally,,there is the possibility that some of the lowlands within the Cayes Plain Basin may be suitable for the introduction of cacao, a crop with international export potential.
Import Substitution. A related consideration is the production for local consumption of commodities imported into the region from interregional or international origin points. Again, onions and potatoes (presently from Kenscoff) and garlic (presently from Spain) are obvious possibilities. I~n addition, many imported small manufactured items that require neither large capital investments nor state-of-the-art technologies could be subsituted for by locally produced goods. Among these are various tools and implements, houseware and hardware items, and clothing.
Creation of Value Added. A further consideration that is
directly linked to both export development and import substitution involves the processing of raw materials, especially of agricultural products; i.e., turning perishable produce into-more durable commodities with a higher market value., It should be noted that one of the few existing processing operations in the Cayes Plain Basin, a tomato paste plant, has rectwitly been shut down because of competition from lower priced imports from the Dominican Republic. Although the situation is highly complex, it is rumored that the imported product has been able to attain a lower price because the Haitian government has released one large importer from the obligation of paying import duties. It is evident that, if this is in fact true, this type of practice should be curtailed and indigenous industries encouraged rather than discouraged. Sugar refining is another possibility for the creation of value added. Therefore, it is essential that the linkage between the cane and its processing site be improved through more efficient collection and transportation. With each hour that elapses between the cutting of the cane and its processing the cane's sucrose content diminishes. In order to compete on the world market, however, the existing underutilized Centrale Dessalines will need to achieve an economies-of-scale threshold. Related to the production of cane is the distillation of rum. It appears, however, that a major world-wide marketing campaign would be necessary to increase the demand for Haitian rum. Further, in the long-term the possibility exists for the processing of wood products and lumber.
Transport and Storage Infrastructure. In order to enhance the economic viability of the region certain improvements in its transportation infrastructure would be desirable. The roads above

Dubreuil linking it with Canon, Caiman, Le Pr~tre, and Platons need to be upgraded for improved vehicular access. The three river crossings between Dubreuil and Le Pretre, in particular, present serious problems during the rainy season. The road from Le Pr~tre to Platons, which is barely passable in a four wheel drive vehicle (which can attain a speed equivalent only to foot traffic) will need to be upgraded if the Platons area is to realize its full potential. The peasants in the Platons area have aspirations of growing certain crops, such as tomatoes, which would likely suffer considerable damage in transit over the present road. Also, other farming changes such as the introduction of livestock would likely be facilitated, as would access to the area for agronomes and other technical workers. In addition, foot paths in the slopes area that link the Ravine des Mornes and the Ravine Duclerc hinterlands to the principal L'Acul road network are presently dangerous during the rainy season even to tne Haitian peasant who is accustomed to navigating them. In general, however, "desenclavement" is not a problem as it is in the Central Plateau. Critical issues relating to domestic and international transportation linkages with Cayes also need to be addressed. These will be considered in the following section. Clearly, improvements are required in both areas in order to promote the development of the Cayes Plain Basin. Facilities for the storage of agricultural produce is a second required infrastructural component. As shown in Figures 4.5 and 4.6, there are seasonal fluctuations in the market prices of agricultural commodities due to the seasonal variations in the supply of these commodities. In many instances the peasant producers sell their produce for low prices when the supply is abundant and buy it back at higher prices later in the year when it is scarce. Improved rural storage facilities would assist in the interseasonal stabilization of market supply and permit the farmer to retain the surplus value that his efforts have created. Beans, sorghum, corn and, to a somewhat lesser extent, rice are among the commodities for which the need for storage facilitis are most acute.
The Role of Cayes as a Central Place Growth Center. The Cayes Plain Basin is linked to the rest of Haiti and to the rest of the world through Cayes, the nation's third largest city, and thence through Port-au-Prince. As shown above in section 4.3.3, there are direct connections between Ducis, and also some of the lower order markets, and Port-au-Prince, but Cayes generally functions as the region's interface with the exterior. Thus Cayes has the potential for serving a very crucial function in a regional development context. There are several aspects to this strategic function. First, Cayes may serve as a regional growth center from which investments, technology, and the impetus to economic growth will "spread" or "trickle down" to the areas with which it has contact.
At present Cayes, as all other Haitian cities outside of
Port-au-Prince, is not being utilized to its full economic potential. The surrounding rural areas would benefit not only from the development of an urban-industrial base in Cayes but also from the innovations, information and social services that a vital urban center

generates. Second, the development of a more developed industrial structure would provide employment opportunities which might help to reduce some of the pressure of the farm population upon agricultural land (particularly on the slopes), the productive capability of which is diminishing visibly with each passing year. Third, Cayes has the potential for the re-establishment of its historical function as an active port with linkages to the rest of Haiti and to the rest of the world. Not only would this facilitate interregional and international exports, but it would also raise the possiblity of developing a quasi-freeport assembly function similar to that of Port-au-Prince. The lower land prices in Cayes might make this a very economically attractive alternative. Note, however, that all of the above alternatives relating to Cayes' potential as a growth center would require a major reorientation on the part of the Government of Haiti, whose explicit and implicit economic development policies tend to centralize all activities and direct all economic flows to Port-au-Prince. For example, the recently completed paved road between Port-au-Prince and Cayes has actually worked to the detriment of local merchants in Cayes; many of their former customers now do at least a portion of their shopping in Port-au-Prince.
Other Issues.
A number of other issues that may have implications for regionaL development are worthy of exploration. First, the markets are an effective medium of communication with the Haitian population. With a high degree of frequency some member of every household visits one of the lower level markets and, with somewhat lower frequency, one of the higher level ones. The implication of this is that information, seeds, and various social services can be distributed using the market system. Second, levels of agricultural production may be enhanced by crop type substitution and by selective crop variety substitution.
The former refers to the substitution of soil-conserving perennials, such as coffee, for annuals, as well as to the introduction of new types such as onion and garlic. The latter refers to the substitution of crop varieties that have a different date of maturation than those presently utilized. The effect of this is to even out the peaks and valleys in. the supply of a crop. The substitution must be accomplished selectively, for if the new variety is adopted by all farmers the peaks and valleys in the supply of the crop will simply be shifted in time. A final economic consideration is the availability of credit to the farmers. Credit is important for the purchase of land, fertilizer, and seeds. It was generally observed in the course of this study that there is a wide variation in the access to, and utilization of, credit; the lowland farmers tend to avail themselves of available credit to a much greater extent than do farmers on the slopes or uplands. The problem of credit is discussed in more detail in Chapter Six.

From the perspective of developing a rational co-ordinated rural
regional development strategy for the Cayes Plain Basin in general and
L'Acul River Basin in particular, it is necessary to consider the plain first from a regional perspective, and only then examine its
site specific attributes. The optimal approach to initiate the
crucial changes in rural Haiti for improving the environmental
situation is simultaneously combining and developing all physical and
human resouces while never taxing the resource capabilities above
their ecologic capabilities. Improvements in the physical environment
will increase the long-term economic situtation by developing a resource base able to support improvements in per capita income.
Equally, existing limitations of the human resource need to be
countered since without concurrent improvements in the socio-economic
situation the physical improvements would likely be only exceedingly
Only in a regional river basin approach can all resources be
controlled in a rational manner from the basin divides to the lowest lands without external forces preventing the re-establishment of the
ecologic balance. This balance is essential for improved agriculture,
grazing, and silvacultural production and social progress in a
primarily land resource oriented economy.
Table 5.1 lists the major agricultural crops, forestry, and
animal production activities potentially viable for the climatic
moisture conditions found within the Cayes Plain Basin. Because the
soils throughout the uplands generally are either thin or drain
rapidly, in spite of the area being classified as climatically humid (0.80), in terms of soil moisture available for plant growth most of the lands within the upper Acul act as subhumid (0.65 0.80) zones.
The west facing slopes, due to the already discussed microclimatic
attributes of these areas, act as a semi-humid to sub-humid moisture
zone. In addition to moisture, temperature affects the range of
possible crops. In L'Acul River Basin the variation in temperature from one portion to another is largely a result of elevation. The lowlands are generally below 100 meters, the slopes between 100 and
800 meters, the Platons between 600 and 1200 meters, and the Igneous
Highlands above 1200 meters. Given that mean annual temperature
decreases at an average value of 0.60 C/100 meters, a temperature
range of over 70 C exists between a lowland location such as Chantal

Potential Rainfed Crops for Cayes Plain Basin
R/EO Moisture Range Suitable for Animal rest Species
UNIT W% Various Crops without Irrigation Production
Platoflj, Iji~aeuus >. 80 A
Highlands, i
Slopes I
ciiJ CU I
Cal Wc -4 0 1 0 Cl)
Lowlands .65-.80 0 a) 0 4- in 0 ci)1
1 oo 0 : 0-O aU)W4 0u4-W OCU->4-U 0)-4,W
P O 4 i 0 :J4~ l 4 C ~ J C a~ U U r-, > -4C 4 4
WC 0010 W*cu0CU O I064~WCC~C 00 NUI. 0C 4 r
:JJ 10 005 CO caUU I o )u)CO r i)f 1Cd00
____________________k Major potential growing zones (optimal)
-- - ------ Sub-optimal moisture conditions
R =Average Annual Rainfall
EO =Annual Potential Evapotranspiration
EO(mrn) 2422-O.358h(meters) h = elevation
Modified for Haiti based on H.M.H. Braun Agroclimatic Zone Map of Kenya, Nairobi, 1982
Table 5.1

and a Platons site such as Formond (Sousbois). With the increasing
elevation and cooler temperatures you would expect to find a
differentiation in crop types between the four major units in L'Mcul River Basin. Table 5.2 presents the crops according to land unit and
tenure system as they were observed by the Study Team during November
and December, 1983 throughout the whole Acul River Basin. Very little
differentiation is evident in terms of land units. Many crops are
found in all of the units. Among these widespread crops are: sweet
potatoes, cocoyam, beans, sorghum, mango, corn, and banana. Only a
few crops such as sugarcane, citrus, rice, and Irish potatoes are
grown in only a portion of the river basin. It appears that little
crop specialization has evolved within L'Acul due to ecological differences. This can be partially explained by the relatively
similar moisture available for plant growth in the uplands and
lowlands given the soil moisture retention properties of the upland
soils. However, this lack of ecological crop specialization might reflect two other critical factors determining crop choice. First, the absence of a wide selection of seed or nursery stock limits the
selection process. Many farmers, especially in the Platons indicated
this was an important factor that limits their crop selection.
Second, there is not a large demand in L'Acul River Basin marketing
system for other crops. However, this second possiblity is likely
invalid since foodstuffs from as far away as Kenscoff-Furcy reach ail
markets within the system.
A greater differentiation in crop selection appears to result
from land tenure. On table 5.2, in the sharecropped and rented
categories, *no tree crops are evident. All of the crops grown are
annuals. Clearly investment in perennial crops on these types of
lands is not considered by the individuals working the lands.
Finally, from an environmental perspective it is important to
note that on the lands within the slope unit, very few of the crops
grown minimize soil erosion. In spite of active field abandonment
resulting from severe land degradation, farmers continue to crop their
land until it is impossible. Fields over 550 are continuously being
planted in sorghum, a particular crop that exacerbates erosion.
5.3.1 Introduction
Many of the environmental problems existing within the Cayes
Plain Basin including low land productivity, water shortages, and
flooding result from the improper use of the land. Because both soil
and water are transported within the river system, all of the land
units are interconnected regarding these two resources. Within the Cayes Plain Basin the best lands from an agriculture perspective are the rolling to flat uplands on Les Platons and the low flat lands on

Crops Observed: November December 1983, L'Acul River Basin
Platons Inherited Land Sweet Potato Irish Potato
Manioc Peas
Sorghum Beans
Cocoyam Taro
Coffee Banana
Citrus Mango
Purchased Land Coffee Yam
Mahaut Peas
Sweet Potato Beans Sorghum Corn (maize)
Citrus Mango
Sharecropped Land Peas Beans
Sweet Potato Manioc
Rented Land Manioc Irish Potato
Sweet Potato Sorghum
Slopes Owned Land (Include inherited and purchased)
Congo Peas Sweet Potato
Cocoyam Avocado
Sorghum Coffee
Orange (sour and sweet) Manioc Banana
Mango Peanut
Karayib Sugar Cane
Rented Land Yam Manioc
Karayib (?) Banana Sweet Potato Sorghum
Sharecropped Yam Sweet Potato
Karayib Beans
Sorghum Corn
Lowlands Non Irrigated Sorghum Karayib
Corn Cocoyam
Sweet Potato Peanuts Banana Tobacco
Beans Mango
Sugar Cane
Irrigated Rice Beans
(?) English Name could not be determined but like a cocoyam

the Cayes Plain. It is the uplands--The Igneous Mountains, Les
Platons, and Slope Units-- which are the major watersheds of the
rivers flowing across the Cayes Plain. In addition many of the
characteristics of these uplands, their infiltration rates, slope
steepness, and interception characteristics--determine the ratio of the precipitation that flows as direct runoff contrasted to ground water flows to the river channels. As this ratio increases, river
discharges for low frequency flows become larger, while at high
frequency flows they become smaller. For example if because of a
decrease in the vegetation cover, a smaller proportion of
precipitation infiltrates into the groundwater, the river will flood
more often (low frequency event) and have lower baseflow (high
frequency). Since the occupation of the Cayes Plain Basin by the
French and continuing unabated to the present, all land use practices
in operation have resulted in increases in this ratio. This signifies that land use practices have been and are diametrically opposed to the rational use of the water resource. In the following sections of this chapter, the term, agriculture is used in its broadest sense. That is
it includes not only foodcrops but also livestock production and
5.3.2 L'Acul River Basin Agriculture and Environmental Impacts
During the field portions of this S.I.P. it became obvious to
the Study Team that within the river basin the complex mixed forest
that was indigenous to the whole area has .been completely cleared.
Only in a small portion of the Igneous Highlands does there exist a
semblance of forest cover, and here it is in a degraded state with few
tree species and large areas having no protecting underbrush and few
saplings, the latter being continuously cleared for firewood.
Therefore the natural regeneration of the forest, even in its one
small remaining area, is threatened by existing practices.
"A brief inspection of Table 5.2 clearly indicates that the
overwhelming proportion of land in cultivation is in annual crops.
Only sugar cane, scattered fruit trees and other scattered types of
trees, "wild" coffee bushes, banana and vetiver are perennial. From
an environmental perspective perennial crops are soil conserving
compared to annuals. Thus, the current mix of crops is not favorable
for a conservation strategy. This latter statement attains even
greater credibility when the spatial distribution of these crops is
examined. No systematic pattern of the perennial crops exists
relative to the environmental setting with the exception of rice which
is only grown on the plain. Sugar cane is grown on both steep and
gentle slopes, trees are found scattered throughout the lands without
any development of complete areal coverage. Thus little continuous tree cover exists throughout the basin. Individual trees scattered
throughout the landscape provide only minimal erosional control since
runoff generated in other areas flows freely past the individual
trees, often exposing some of the tree roots. Vetiver actually
encourages soil erosion throughout L'Acul River Basin since, when it
is harvested, it is pulled out of the ground resulting in a loose base

soil. As the harvest often occurs just prior to the advent of the
rainy season, the potential for soil erosion is extremely high.
Coffee, as it is only grown in the immediate area of the kay, which is
often situated on a level parcel of land, likewise is not being
utilized as a soil conserving crop. Similarly, annual crops are grown
under widely disparate conditions independent of any apparent
environmental concern. Sorghum, congo peas, beans, corn, and manioc
are planted over the whole spectrum of slope conditions existing in
the river basin. While congo peas are considered a soil 'enriching
crop due to their ability to fix nitrogen, because of their poor leaf
cover the soil is continuously washed down slope when they are grown
on moderate to steep slopes. Thus the benefits of increasing soil
fertility are more than countered by the low degree of erosional
Similar to cropping practices, livestock grazing as it occurs in
L'Acul River Basin is not conducive to environmental maintenance.
Generally, animals are tethered and left to graze in a limited area
each day. This results in a process of discontinuous overgrazing. In
addition, as livestock are moved on a daily basis along footpaths in
their search for pasture, there develop zones of overgrazed land.
These overgrazed zones act as catalysts for the initiation of
excessive rill and, at times, gully erosion in many locales. In
summary, given the distribution of agricultural activities throughout
the whole basin, it is evident that either little concern exists for the soil resources or the constraints placed upon the farmers do not
allow the peasants to consider the options necessary for soil
5.3.3 Existing Farming Practices and Environmental Impacts
Throughout L'Acul River Basin, little obvious conservation
practices are evident. From the preparation of the fields for
planting through the harvest, little evidence exists that indicates
soil conservation enters into the minds of the small farmers. For
example, in the process of preparing the soil for seed planting, the
farmers almost universally 'hoe' the soils downslope, thereby
contributing to the downslope movement of the soil. Even in the
relatively well managed slope areas of Furcy-Kenscoff (Port au Prince)
this practice is utilized. Clearly the ease of using this hoeing
technique results in its ubiquitous use.
The piling-up of soil in mounds might appear to be a soil
conserving device. Yet the evidence indicates that it is the need to
increase the depth of the soil to a sufficient depth to permit plant
growth that is the major motivation for this practice. That is, in many places the thinness of the soil profile requires this practice.
The lack of a specific orientation bias in the plane view of the
mounds results in these features having a minimal impact on decreasing
the movement of the overland flow of the water. In fact, the
placement of mounds on the steep slopes might even increase the
likelihood of erosion.

The clearing of rocks from the fields gives further evidence of
the minimal consideration of soil erosion control in the daily
activities of the farmer. Throughout the Slope and Platons units many fields are 'pocked' with cairn-like mounds. Nowhere were these cairns
oriented parallel to the slope contour to act as barriers to the
downslope movement of materials even though this practice would not have resulted in more work for the individuals. Minimal contouring
and mulching are practiced throughout L'Acul. When contouring is
evident, it is often on steep slopes ( 100) that minimize its impact.
Two bits of contradictory evidence exist to this apparent lack
of erosion control consciousness on the part of the small farmer.
First on Les Platons, relics of stonewalls are evident in the vicinity of the Citadelle. These were built for soil erosion control. However
today they are gradually going into disrepair. This reversal in the
erosion control process began when the agronome who mobilized the
farmers was transferred and no government replacement was sent to the area. Likewise, terrace building near Canoe initiated under the PDAI
program is evident. But these, as the stonewalls, appear to be relics
on the landscape.
In summary, current farming practices clearly place a low
priority on soil conservation. In fact some of the practices appear
to accelerate the erosional processes. Evidence exists that the small
farmers are willing to become involved in soil erosion control. But
for this activity to be successful, methods must be found to make it a
part of their daily routine; otherwise, when the outside force that
initiated the erosional activities leaves the area, the practices will
5.3.4 Existing Irrigation Practices and Environment
Irrigation in L'Acul River Basin for all practical purposes is
synonymous with the Dubreuil Project. The evaluation of this project is reported in great detail (Stutler, et al. 1983). F or thle purposes
of this report only two aspects of the water management in this
project need to be mentioned. Neither of these aspects are stressed in the Stutler report. First, it appears that minimal monitoring of
the quantities of water delivered to individual fields occurs. Visual
observations by the Study Team appear to indicate that most of the
fields are being over irrigated. Worldwide, many irrigation projects have failed due to problems associated with water logging. Care needs
to be practiced to insure that the lands being irrigated remain
viable. Second, this wasteful use of the water minimizes the areal
coverage of the Cayes Plain that is potentially available for
irrigation. One other aspect of the irrigation project that was noted
appears to be the lack of ongoing maintenance in the canals. Some
damage was observed but no repair crews were ever observed. If this
low maintenace policy continues, when the system is under stress major
damage could occur.

Another environmental component that could limit the Dubreuil
Project occurs in the vicinity of the diversion dam. The reservoir
behind the barrage is already completely silted-up. If a major flood
occurs on L'Acul, in spite of the engineering precautions, the
diversion channel could become blocked by the massive quantities of
bedload materials available for transport upstream of the dam.
Whether or not this event occurs will be decided by the agricultural
activities occurring not on the Cayes Plain lowlands, but by the
farming practices on the Slopes, Platons, and Highland units.
5.4 Land Tenure
It is widely recognized that the specific characteristics of a
region's (or a nation's) land tenure system exert a strong influence
upon agricultural practices and productivity. A number of authors
have focused upon this issue in the Haitian context (Murray 1977,
1978a, 1978b, 1978c; Smucker 1982; Zuvekas 1978). The purpose of the present section is to examine patterns of land tenure in L'Acul River
Basin and to explore their implications for agricultural activity.
5.4.1 Land Tenure Patterns
In his discussion of land tenure, Murray (1977: 177) suggests
the existence of "genuine subcultural differences between lowland
Haiti and highland Haiti, differences that appear to be related more
to ecology than to simple regional variation." Most previous research
in the Cayes Plain Basin has been in the lowlands. This is in
contrast to Murray's opinion that the literature on Haiti has "leaned
somewhat heavily toward the study of highland life", (Murray 1977:
77). As a result, the Study Team felt it was necessary to obtain land
tenure information in both the uplands and lowlands in spite of the
limited field period.
The present examination utilizes three of the four sub-regional
units defined in Chapter 3: Les Platons, the Slopes, and the
Lowlands. A fourth unit, the Igneous Highlands, has been omitted due
to time and accessibility constraints. As is shown, the tenure
situation in Les Platons and slope areas of L'Acul River Basin doesn't
differ significantly from what has been written about other
mountainous regions of Haiti (Zuvekas 1978).
In order to compare Les Platons, Slope and Lowlands units, a
formal land tenure questionnaire was applied to 12 household heads on
a random sample basis in each of the three subzones. For the
Lowlands, a distinction was made between irrigated and non-irrigated
land: 6 farm units were sampled in each of the two lowland areas.
Besides questions dealing with tenure, the questionnaire examined the
areas of credit, wage labor, and informal forms of cooperative labor
such as eskwad.

Most farmers were interviewed in their homes, and the majority were men. However, when a woman considered herself to be a household head, she was interviewed. The interviews were conducted in Creole. After informal conversation of 10 to 15 minutes, the formal questionnaire was introduced. The survey was done within a two week time frame (Nov. 18 Dec. 1, 1983), and covered a total of 129 plots worked by 36 farmers for a total of approximately 53 karo.
While the results from Les Platons and the Slopes likely are
representative of the land situation in these areas, the sample of 12 farmers in the plain certainly does not represent the lowland land tenure situation in totality.' In this study the large landowners, which are important in the lowlands, were not interviewed. Therefore it is better to consider this sample a "1sondage"l which can be compared with caution to the Slopes and Les Platons.
To facilitate the interpretation of this survey, the
introduction of Serge Larose's (undated) diagram of possible tenure status in Haiti is introduced (Table 5.3):
Table 5.3
Potek refers to land which has been rented and paid for, for two or' more years. Fem is land rented on an annual basis.
Because of important developments over the last two or three
decades, in particular the phenomenon of buying inheritable land, the Larose breakdown of land tenure needs to be modified. Many respondents, explaining about plots they purchased, bought lands that were part of the family lands. Thus, the term "inherited" becomes ambivalent because up to now "inherited land" referred by definition to non-purchased land. It thus seems better to divide the section "titled land" into three categories: (1) purchased non-inherited, (2) purchased inherited, and (3) inherited non-purchased land. An explanation of this recent development is presented below. The term "inherited", as used in our tabulations, refers to non-purchased inherited land, usually divided and rarely titled. In Larose's scrieme it would fit the "divided t6 min~l" category. The inherited land is worked by the farmer for His personal benefit.

Location and Tenure Type. Figures 5.1 to 5.3 indicate the
relationship between sub-region and type of ownership. With a same number of respondents in each level of the basin (Platons, Slope, Lowlands), there are a significantly smaller number of plots owned in the Lowlands (Figure 5.3). Out of a total of 129 plots surveyed, 83 were owned (i.e inherited or purchased). All those who bought plots have a clear legal title to their land. For those who own divided inherited plots, the legal title grann pyes) is in the hands of one of the family members. The average number of parcels owned (purchased or inherited) for the 36 farmers is 2.5 plots. As illustrated in Figure 5.4, there are no significant differences between Platons, Slope and Lowland units in total average number of plots per farmers: 3.16 in the Platon, 3.58 in the Slope, and 3.91 in the Lowlands.
The literature reports a great variety in the average number of plots held per farmer. Smucker (1982) found 5.2 plots per "economic unit" in the northern Haitian hills. Zuvekas (1978) cites Kulakow et al.'s findings in the Cayes Plain as 2.6 plots per farmer; White'sreport in the same area found 1.7 The 1971 census showed an average of 1.8 plots. Murray's (1977) average in the Cul-de-Sac Plain was much higher: 5.8 plots per "holding".
In our survey, three farmers out of 36 did not own any land; one in each of the three land units. The Platons farmer is a future land owner. Today he works with other family members on a 21.3 karo family holding (anpil 6ritiy6). The slope farmer sharecropped 5 karo from his cousin, a coffee speculator who stole the land from his-uncle by bribing the notary. Otherwise this land would have become the inheritance of the actual sharecropper. The non-owner on the lowland rents one plot and sharecrops two non-irrigated plots. His mother is still alive and owns 2.5 s'z'm (or 2.5 sixteenth of a karo), often called ventsenk (25) in Creole. The farmer works his mother's land but not for his own benefit. In the future he will inherit at least part of this plot.
The reason these three exceptions are emphasized relates to the issue of landlessness. The only person who will not own any land through future inheritance is the speculator's cousin. Zuvekas' percentages of people who own at least part of the land they work range from 65 to 100 percent. L'Acul Basin findings of 33 out of 36 farmers owning land is 91.6 percent. It increases to 97.'2 percent if we consider the future ownership of two of the three other farmers. Smucker found 17 percent of landless farmers (Smucker, 1982: 265); Plotkin's survey in the North revealed only 2 percent of landless household heads (Plotkin, undated:2). From the evidence gathered in this study and reported in most other studies, landlessness appears to be a non-issue in most areas when only the current farmers are considered. However, from this, one should not infer that enough land exists for the rural population. Perhaps one reason why landlessness is not an issue is that most rural people without land migrate out of the countryside.

TOTAL: 35 plots
12 34.2%
10 (N-12) I25.7%
8 (N-9)
4 2
TOTAL: 48 plots 41.6%
1 8 ( 2 0 ) 3 %
14 25%
o 12
lO (N-12)
6 4 2
TOTAL: 83 plots
32 36.1%
28 I(N-32)
24 25.3%
o 20 S(N-21)
11 12

5 per farmer
0 4 3.58 plots 3.91 plots
. 3.16 plots
TOTAL: 28 plots
18 60.7%
16 (N-17)
' 14
12 10
8 21.4
6 17.8%
4 (N-5) (N-6)
TOTAL: 18 plots
10io 50%
38% ] )N-9)
6 (N-7)
2 11.1%

Figures 5.5 and 5.6 refer to rented and sharecropped plots. They show that in the Lowland many more plots are sharecropped and rented than in the Slope and Platons areas. In the Platons, two respondents indicated that they would like to sharecrop, but that no land was avalaible. The same two farmers were also owners (of less than 1 karo); one of them also rents a plot. Thus, it appears that farmers' land demands are not met by the current availability of land.
Of the total 129 plots surveyed, 18 are sharecropped. This is 13.9 percent, which is close to the 14.4 percent found in the 1970 national survey. (Zuvekas, 1978: 16).
Figures 5.7 5.10 illustrate the meaning of the word "owner". They indicate that "ownership" doesn't mean that the farmer necessarily can make a sufficient living from his land. In the same way that inherited and purchased land is governed by a regime of microproperty (81.8 percent own less than 1/2 a karo), so are rented and sharecropped plots extremely small in size. Of the rented plots, 92.8 percent are less than 1/2 karo; 94.3 percent of the sharecropped plots are also less than 1/2 a karo. In our survey in the Cayes Plain Basin, 62 percent of the plots measure less than 1/4 of a karo, and 23.2 percent are between 1/4 and 1/2 a karo.
Signs of increasing pressure on the land are revealed in many of the respondents' reactions to the question on what they do to prevent erosion. They can no longer afford to fallow their plots for one or more seasons. In Smucker.'s research area, less than a decade ago,. over 75 percent of the household units still maintained some land in fallow (Smucker 1982: 259). In L'Acul Basin no such freedom remains. In Le Pretre (Slope), family members have conflicts even over which part of the non-divided family land will be worked by whom. This "first-come, first-take" method of acquiring family plots was observed by Murray (1978) in the same area. The Study Team was told that this was not a general practice, but that only mechan yo (mean people) had this bad habit. Both phenomena, the disappearance of the fallowing custom and the "grabbing" of family lands, illustrate increased pressure on the land. The farmer's most obvious reason for maintaining land in fallow is r~f6 tb-a (rebuild the land fertility). They all considered it a good, and traditionally probably the only, erosional control method.
Farm Size and Price. Figure 5.11 illustrates average farm sizes in each sub-region. The 36 farmers work a total of 52.1 karo, an average of 1.44 karo. The plots in the Lowlands are only about half as large as the Slope and Platon's plots. One explanation may be the higher land prices in the irrigated plain and the lack of cash availability for buying larger plots. Zuvekas' (1978) statement that Haitian farmers "often deliberately seek to have fragmented landholdings, particularly in different ecological zones, because this

TOTAL: 28 plots
30 25
2 71.4%
'04 15
o 10 21.4%
5 7.1%
0.25 & less 0.26-0.50 0.51 1
TOTAL: 83 plots 61.4%
50 45
o 25
z 20 20.4%
10 I80.8%
5 42.4%
7 7 77 7/ 14 ..,,....,,,,
less 0.25 0.25-0.50 0.51 1 1.01 2 3
karo karo karo karo karo

TOTAL: 18 plots
o 10 a0 38.8%
o 6
z 4
2 L % r5.5% -1
0.25 &less 0.26-0.5 (0.51.4) 5 karo
80, 70
0 40
40 23.2%
2 0 8 .5 %Z
0.25 & 0.26-0.5 0.51-1 1-2.01 2 4 5
2.00 1.845 karo
1.00 0.888 karo
0.50 0.25

provides some security against crop failure in any area" seems suspect. This might be a post facto explanation for not finding land close to their (farmers') houses. In the study area, each farmer's lands were in the same ecological zone even when fragmented.
Compared to other published estimates of Haitian holdings, the average 1.44 karo is high. Murray (1977) found 1.3 karo the average in the Cul-de-Sac. In Comhaire-Sylvain, the average size is 5 acres (1.8 karo), and in Simpson, the size is 3 to 6 acres (1.08 to 2.17 karo). It should be noted that in Murray's report (1977 pp. 208-209) he incorrectly converts his land holding at the rate of 1 karo equals
1.43 ha instead of 1.29 ha.
Within the Lowlands, rent for irrigated plots is almost twice the price of dry plots. Also, in the Lowlands more plots are rented than elsewhere. Out of 129 plots surveyed, 28 were rented, 17 in the Lowlands and 14 in irrigated areas. Irrigation is the only significant factor differentiating land sale prices. Irrigated land is 7 to 8 times more expensive than dry land, and jumped to an average of US$ 2,120 per karo in the 1980s (Figure 5.14). Palmer (1976: 147) found that in Belladere "renting is increasingly common as more and more farmers inherit impracticably small plots. At the same time, however, as land becomes scarce and more valuable, large landowners are more hesitant to put their properties up for rent." This can neither be confirmed nor denied for L'Acul based on results from our sample. The fact that in L'Acul River Basin the average size of all rented plots (0.24 karo) is smaller than the average size of all inherited plots (0.35-aro) would make the first part of Palmer's statement seem invalid for the area.
Land Security. Out of 129 plots, 35 (or 27.1%) were purchased. Not many authors mention comparable figures. Purchasing land implies to some form of relative wealth at one or another moment in life. Fifteen out of 36 farmers have purchased land: 5 in the Platons, 4 in the slope area, and 6 in the plain. Thus, there is no significant difference between sub-regions. It is interesting to note that land prices are higher in the Lowlands, especially for irrigated lands (Figure 5.12-5.14)
As mentioned earlier, many farmers purchase land from family members (inheritable land). One interpretation is that land is sold intergenerationnally (Murray 1977), and that the rituals, which are often performed for the dead, function as an intergenerational land circulating mechanism. Findings in Les Platons might confirm these findings. Farmers in Les Platons have the custom (at least ideally) of performing at least once in a lifetime an expensive ritual for the ancestors called devwa, literally "obligation". It consists of a Catholic mass by Ifi-ocal priest, a ritual by the local oungan (voodoo-priest), and a big meal, for which animals are killed. The' price of such a ritual might run between $100 and $500. Often, in

$ 1,800 120
1,800 1,600 1,400 1,00 1,000
800 600
400 $407
$ 200 2 375 4 $305
(N-12) (N-1LA)
120 $131.7
100 (N-14)
80 $88.9
60 $65.3 (N-8)
$ 20 (N-5)
Irrig. Non-irr.
Overall average 407.- 37.5 2,120. 305.6
per Karo
Before 1960 100.- 122.- 96.7
1960-1969 408 367.1970-1979 264 338 617.7
980+ 563 2,640.Location Platon Slope Plain

order to pay for this feast, people have to sell land. None of the respondents had yet performed their devwa. People tend to postpone this obligation until late in life. The devwa custom does not exist anymore in the Lowlands. People referred to- t as vye koutim (old custom); vye has also a morally negative connotation ("only backward people do this"). Some people said they were Protestants and would not participate in this custom, but to the question if Protestantism was the cause of the custom's disappearance, it appears that it was not the only reason: "depi Janklod moun yo pi eklere" (Since Jean-Claude, people are more educated). On the slopes, devwa was also considered not to be fashionable but some people still perform it. One must make special mention of the custom of building tombs in gardens, a custom common in the South. It might just be a "folkloric custom," with no rational for it. But all tombs are built on family land, never on purchased or rented land. Thus, one can assume that the dead, symbolised by tombs, are perhaps at the core of the land tenure issue in L'Acul River Basin.
It does not appear that peasants feel very insecure about their land. The main complaints were that it is hard to buy or rent land. The land tenure security theme runs through most of the land tenure literature in Haiti; the threat of losing land seems to be a slowly increasing pressure growing from generation to generation. The presence of tombs (kavo) on inherited land could carry two opposite symbolic meanings in -relation to land security: (1) "This land belongs to my (extended) family, and should never go into hands of non-family buyers"; or (2) "There is a constant danger that this land will be lost for the family, so, this stone (kavo) and my eternal presence here will make it impossible for you, my progeny, to sell this land." Both seem to mean the same but the first way of formulating it, refers to true tenure security, while the second formulation expresses the fragility of ownership, and the lifelong fight of the ancestor for this piece of property.
Tenure and Location.. Figures 5.15 5.18 deal with the size of the total amount of land differentiated by type of tenure and location in the study area. Since there were less owned plots in the Lowlands, the highest total acreage of rented and sharecropped land is found there. The acreage of inherited land is almost the same as the acreage of the purchased plots (Figure 5.15). Much more land is owned, than is either rented and sharecropped (Figure 5.14); 35.75 karo are owned while 16.38 karo are rented or sharecropped.
Common to Haiti and existing in L'Acul Basin, mixed tenure is
the dominant method of land holding; 15 farmers out of 36 did not work any land besides they owned (non-mixed tenure pattern of 44.4 percent), 7 in Les Platons, 4 in the Slopes, and 4 in the Lowlands. For the 19 other farmers, besides working their own inherited or purchased plots, 7 farmers also rent land (Platon 4, Slope 2, Lowlands 1), and 7 farmers both rent and sharecrop some land (Platon 1, Slope 3, Lowlands 3).

TOTAL: 35.75 karos 100% 18 17.951 Karo
14 (N-12) 12.808 Karo
8 6
4.919 Kato
4 2.81K
2 8.39K 7.28K (44
N- 5 0N-6 .1KA9i (H-I
50.2% 35.8% 13.7%
3.403 karo
3. 2.687karo (N17)
2.5 (N-6)
1.5 0.686karo
0.5 (N-5) .
10.1% 39.6% 50.2%
TOTAL: 4.612 karo 100% (N.B. one case of 5 karo was left out)on slope
2.5 2.342 karo
2. 1.545 karo (N=9)
1. 0.625 karo
0.5 (N-2)
13.5% 39.9% 50.7%

0.9 0.8 0.7 0.6
0 0.5
~ 0.4
TOTAL: 36 farmers
20 rent + sh.own
sh. 7 16
15 rent
own own 15
10 sh.
own 8
mixed non-mixed
tenure tenure
55.5% 44.4%

Of the 36 farmers, only 3 farmers did not own any land; two
farmers both rented and sharecropped some plots, the other
sharecropped only his cousin's 5 karo (Figure 5.19).
Thus, 20 farmers out of 36 (55.5 percent) work their land under
mixed tenure conditions. In Smucker's northern research village "the modes of tenure ... is thoroughly mixed" (Smucker, 1982: 278). "The 1950 census does not provide any information on mixed tenure patterns"
(Zuvekas 1978: 9). Only one farmer out of 36 had a tenant working
on his land; he was ebenis, so besides his farm activity, he had also
other cash income. Thus, one does not observe Murray's Cul-de-Sac pattern of sharecroppers, who give land out to others (38 percent),
and of landlords, who have tenants working on their land, working as
tenants on the land of others (60 percent); this illustrates the
ambivalence of terms like "sharecropper, "owner" or "renter" (Murray
1977: 393) in other areas of Haiti. For L'Acul area, there is no
such confusion in the surveyed sample.
In concluding the land tenure analysis of L'Acul River Basin, it
is possible to make the following generalizations: (1) the regime of
microproperty is the dominant pattern, and the majority of farmers are
not capable of making a living on their own lands; (2) most farmers
work under mixed tenure conditions; and (3) with an average farm size
of 1.44 Karo, along with the risks of drought and crop failure, and
the low productivity per plot, fundamental changes must be introduced
in this area. These three generalizations apply only for the fields
in Les Platons and Slope areas.- Further studies are needed for the
Lowlands area. Also the impact of large land holdings and ownership
in the Lowlands ought to be studied.
5.4.2 Constraints On Agriculture
The following three aspects of land tenure need to be addressed
in terms of their impact on the potential options for rural
development. First, how do the size of land holdings affect the
potential of co-ordinating rural development while improving the river
basin's soil and hydrologic resource bases? Second, to what degree
are the farm units fragmented? And third, what strategies are viable
given the range of ownership/tenant categories?
Size of Land Holdings and Fragmentation of Farm Fields. Figure
5.11 illustrates the average farm size for Les Platons, Slope, and
Lowland units. In a general manner the relative size of the farming units reflect the-quality of the land. The lowlands, possessing the
highest quality of land have the smallest ownership units. However
this figure (0.89 karo) is biased toward underestimating the farm
size, as no large landowners were part of our sample. Nevertheless,
it is fair to state, that on the lowlands farm size is, on the

average smaller than in any of the other units. One explanation for the smaller plot size on the lowlands is that the higher productive lands command higher land prices. The lack of capital liquidity among the small farmers has resulted in the plot size being relatively small on the plain. On the Slopes and Platons, the lower productivity of these lands clearly requires a greater area to meet the demands of the farming unit. Yet given the relative quality of the lands, the size of farms on Les Platons indicate that the average Platons farmer has a greater agricultural potential than the Slope farmers.
.From the perspective of farm size affecting rural development, in all cases the small nature of the land holdings in all of these units compounds the problems in developing an agricultural strategy. On the lowlands where irrigation is involved, the allocation of waters during periods of water shortage presents major policy decisions if equity is a consideration. Given the extremely small size of holdings, if water is withheld from any farmer, major financial losses will entail. Co-ordination and maintenance of the system is further complicated by the fragmentation of the land holdings. On the average, each lowland farm unit is comprized of 3 3/8 fields (average field size approximately 0.27 karo), none of which are necessarily contiguous to each other.
These same general properties exist on the Slope and Platons farms. On the slopes, the average size of a plot is 0.37 karo; and each farming unit on the Slopes is made-up of 5 fields. Les Platons unit appears to have the largest blocks of land. Here the average field size is 0.43 karo, with each farming Unit on the average made of 3 7/10 fields. As on the lowlands, the fields of an individual farmer are not usually contiguous to each other.
Both the small size of the farming units as well as the
dispersal of the fields compounds the problem of developing a strategy that will increase the agricultural returns from the fields while improving the quality of the resource. Unfortunately nature does not recognize the sovereignty of field boundaries. Thus to minimize erosional and runoff problems, it is necessary to co-ordinate all farming practices existing on the hillslope system. The actions of a farmer practicing good farming techniques could be completely negated by a farmer on the same hillslope system that ignores conservation. It is further complicated in Haiti due to the fragmentation of fields which results in such small parcels and in a multitude of owners farming parcels of lands juxtaposed to each other operating under numerous categories of tenure. This interrelation of farming units located in the same erosional depositional system (stream basin) has been legally recognized in the State of Iowa. A farmer in that State has the right to sue for damages occurring on his farmlands because of improper land use on another farmer's land.

Ownership/Tenancy Arrangements. As in other regions of the
world, in the study area a strong relationship between the nature of land tenure and the degree of investment on the land may be observed
(Table 5.2). As is expected intuitively, the ownership of land
implies a relatively long-term time horizon; the land is a resource
which, if maintained properly, will provide both food and income over
a lifetime. The rental of land implies a medium-term time horizon.
In the Cayes Plain Basin where land is generally rented in five year periods, often renewed, there is an incentive to preserve the land's
productive capacity. Finally, sharecropping (deux-moiti6) is
associated with a relatively unstable form of land tenure. There is a
definite disincentive to the expenditure of time and capital on the
preservation of land that maybe utilized for a short-term.
Table 5.4 presents a summary of the ownership/tenancy
arrangements for three of the four topographic units found in L'Acul
River Basin. From the table it is clear that Les Platons unit has the
most favorable land tenancy conditions in terms of incentives for the
preservation of the land resource. Here 92 percent of the farmers
cultivate lands of which they own at least some of their cropped lands. Fifty-four percent of Les Platons' farmers only cultivate lands that they own. Given this property along with the slightly
larger land units found in this area, previously discussed, it appears
the residents of Les Platons would be amenable to interventions that
would improve the land and water resources of the area.
Ownership/Tenancy Arrangement by Topographic Unit (November December 1983)
Land Farmed Platons Slopes Lowlands
Owned Land Only 54% 36% 36%
Owned and Rented Land 30% 18% 36%
Owned and Sharecropped 8% 18% 8%
Rented and/or Sharecropped Land 8% 28% 28%

Conversely, the land unit that requires immediate interventions
to arrest its rapidly deteriorating environmental situation, The Slope
Unit, has apparently the least favorable land tenancy conditions.
While ownership percentages of the land are identical for the Lowlands
and Slopes (72%), the Slopes have the largest percentage of lands under sharecropping arrangements. Nevertheless, for L'Acul River
Basin as a whole, it presents generally favorable land tenancy
conditions for interventions. The overwhelming percentage of land in
the basin is owner occupied, the group that should be most open to
long term investment.
The preceeding portions of this chapter examined the
interrelations between the physical and socio-economic domains in the Cayes Plain Basin. Within the Haitian context, it is the agricultural
activities that most clearly manifest these interrelations.
Throughout the life of a rural Haitian the direct links of the
physical environment have life and death implications almost on a
daily basis. As little slack exists in both the physical and
socio-economic areas, whenever any component becomes stressed, the
small farmers' low level of well being is placed in jeopardy. In the following two sections the likely outcomes of non-intervention in the
existing rural framework are explored.
5.5.1 Land Resources
In each of the four topographic units within L'Acul River Basin,
if the current situation is allowed to continue, the environmental
situation will deteriorate in each unit. This will place ever
increasing demands on the land resources. Because the topographic
units are linked together via the river system network, the negative impacts occurring upstream will have a multiplier-like effect as they
move downstream. Therefore the best lands from an agricultural
perspective, the lowland Unit being located at the end of the system
will experience most of the negative impacts of the deleterious events
occurring upstream.
By extrapolating the existing trends into the future, unless
interventions are introduced and become successful the following
conditions in terms of the natural resources will likely occur:
1. The Igneous Highlands will become completely deforested. Runoff
will increase; infiltration decrease. Soil erosion will remain
a problem, and will likely become exacerbated.
2. Les Platons will experience local erosional deterioration. A
gradual decline in land productivity and resulting lower yields
will occur.

3. The Slopes will become non-productive except for the valley
bottoms. The valley bottoms will become more susceptible to
flooding making the area both less productive and more hazardous.
4. The non-irrigated lowlands w ill remain non-irrigated as less
water will be available for irrigation.. The currently irrigated
lands will both decrease in-areal extent as well as experience
higher costs because:
a. water supply will become more extreme. L'Aculls flow regime
will fluctuate between torrents and trickles. Moderate flows
will decrease in frequency and less waters will therefore be
available for irrigation;
b. Siltation will increase due to deterioration of upstream
areas. This will result in high maintenance costs; and
c. Actual destruction of irrigation works could occur as
extreme hydrologic events (i.e. a discharge that today represents the 100 year flow) will become more common.
5. In all topographic units, the carrying capacity of the land for
livestock will decrease as a need will exist to bring even more marginal land under cultivation. Thus only the extreme marginal
lands will be available for grazing.
As it is easier to-improve a degrading environment than to
com lately rehabilitate.a degraded area, from the physical
perspective, there is every reason to immediately begin the actions required to arrest the ongoing processes throughout the Cayes Plain
Basin. The need for such immediate action is further strengthened
when the socio-economic impacts of the current degrading system are
5.5.2 Social Economic Implications
The physical degradation of land resources described in the
previous section will have significant impacts upon the social and
economic structure of the study area. These include: .
1. A deteriorating level of food supply, both crops and livestock
will occur. This will induce a decline in nutrition and
health. Ultimately this will lower the productivity of the human resources in the area. In addition, a decreased food
supply will create the necessity for the import of food, thereby
creating "leakages" from the regional economy and diminishing
its ability to retain the surplus value that has been generated.
2. The level of cash income from the marketing of products will be
reduced. This will lead to a reduction in the range and
quantity of goods and services that can be purchased, resulting in a lower quality of life. The impact of this income reduction
will be amplified due to continuing inflationary trends in the
world economy.

Regional rural development programs necessarily depend, in
practice, on the organizations and people through which such programs
can be implemented and the institutional context within which the
program activities operate.
This chapter will inventory and generally describe principal
institutions that d 'irectly or indirectly affect rural development in
the Cayes Plain Basin. For the purposes of this report, these
institutions are classified as popular, governmental, or
non-governmental. The term "institution" is used broadly to encompass
agencies, organizations, officials, and a variety of permanent,
semi-permanent or ad hoc groups. Limited time for field work did not permit a full inventory of the organizations and their activities, or an evaluation of their effectiveness. Discussion here focuses on the
range of activities the groups or institutions are involved in, and
their strengths and weaknesses insofar as regional agricultural
development activities are concerned. As many institutions work on
similar or common activities, discussion of different institutions will
occasionally overlap. This chapter concludes with a. brief discussion
of the effects of land tenure on development options.
For purposes of discussion in this chapter, institutions will be
referenced as follows:
6.2 Popular Institutions
6.2.1 Lakou
6.2.2 Coumbites and Escouads
6.2.3 Groupments
6.2.4 Irrigation Groups
6.2.5 Cooperatives
6.2.6 Community Action Councils
6.3 Government
6.3.1 Government Agencies Agriculture Plan Public Works

92 National Education Public Health and Population Finance Justice Interior IDAI BCA ONAAC SNEM SNEP CORCOPLAN
6.3.2 Subnational government entities Prefect Mayor's Office
6.3.3 Other government or quasi-government VSN CONAJEC
6.4 Private organizations
6.4.1 Religious Organizations Catholic organizations Protestant organizations
6.4.2 Non-religious organizations Pan American Development Foundation AteliersAgricoles et Ecoles de Camp Perrin Interntl. Maize and Wheat Imp-rovement Center ASDEC nursery
6.4.3 Private sector Comme I1 Faut FACOLEV Other agroindustries Credit institutions Other citizens groups in Les Cayes
6.5 Land Tenure
6.2.1 Lakous
The lakou, or extended family compound characterized by several
houses and a yard, has long been considered the fundamental institution
of rural Haitian society. In addition to being the central focus of
family life, it serves a variety of other functions related directly or
indirectly to agricultural development. It enables family members to
recruit unpaid labor to assist them in assuring their common or
separate subsistence. It provides an orderly system for access to
land and its transfer between generations. Typically, it is the first