Electoral Observations in the Americas Series, No.

28

Electoral Observation
in
Haiti

Legislative, Municipal, and Local Elections

February to July 2000
Secretary General
César Gaviria

Assistant Secretary General

Luigi R. Einaudi

Executive Coordinator, Unit for the Promotion of Democracy

Elizabeth M. Spehar

Electoral Observation in Haiti: legislative, municipal, and local elections, February

to July, 2000 / Unit for the Promotion of Democracy.
p. ; cm. - (Electoral Observations in the Americas series, no. 28)
ISBN 0-8270-4331-7
1. Elections--Haiti. 2. Election monitoring--Haiti. I. Organization of American
States. Unit for the Promotion of Democracy. II. Series.
OEA/Ser.D/XX SG/UPD/II.28 (E)

This publication is part of a series of UPD publications of the General

Secretariat of the Organization of American States. The ideas, thoughts, and
opinions expressed are not necessarily those of the OAS or its member states. The
opinions expressed are the responsibility of the authors.
 OEA/Ser.D/XX
SG/UPD/II.28
13 December 2000
Original: French

Electoral Observation
in
Haiti

Legislative, Municipal, and Local Elections

February to July 2000

General Secretariat
Organization of American States
Washington, D.C. 20006
2001
 This report was produced under the technical supervision of Orlando
Marville, Chief of the Mission, Special Advisor of the Unit for the
Promotion of Democracy, and with the assistance of Steve Griner, Deputy
Chief of the Mission and, Senior Specialist of the Unit for the Promotion of
Democracy (UPD).

Design and composition of this publication was done by the

Information and Dialogue Section of the UPD, headed by Caroline Murfitt-
Eller. Kathryn Stafford helped with the editorial review of this report and
Esther Rodriguez with its production.

Copyright @ 2001 by OAS. All rights reserved. This publication

may be reproduced provided credit is given to the source.
 Foreword

Democratic ideals and principles have always been present in the inter-
American system. The Charter of Bogotá, establishing the Organization of
American States (OAS), formally proclaimed in 1948 that "the solidarity of the
American States and the high aims which are sought through it require the political
organization of those States on the basis of the effective exercise of representative
democracy." Forty years later, the Protocol of Cartagena de Indias emphatically
reaffirmed this principle, including among the essential purposes of the
Organization the promotion and strengthening of representative democracy. The
signature of the Protocol gave rise to an unprecedented renewal of the commitment
of the member states to defending and building democracy.

Within this context, the Unit for the Promotion of Democracy (UPD) is one
of the instruments the OAS can draw upon to support member states in their efforts
to strengthen and consolidate democratic institutions. Established by João
Clemente Baena Soares, who was then Secretary General, as agreed at the 1990
General Assembly session, the UPD offers a broad program of support to member
states which, in full exercise of their sovereignty, request advisory services or
assistance in their efforts to preserve or strengthen their political institutions and
democratic processes.

With regard specifically to elections, the UPD provides assistance and

technical advisory services to national electoral agencies and organizes and
dispatches electoral observation missions to member states that so request by
applying to the Secretary General of the Organization. OAS activities in this field
are based on the conviction that the electoral process is always an essential piece of
the transition to or building of democracy.

The electoral observation missions of the Organization are intended to: (a)
observe and report to the Secretary General on the electoral process, using as a
reference point the constitution and electoral law of the host country; (b) express
the international community's support for the electoral process; (c) work with
government, electoral, and party officials and with the population in general to
ensure the integrity, impartiality, and reliability of the electoral process; (d) foster
an atmosphere of public confidence and encourage citizen participation; (e)
discourage attempts to manipulate elections; (f) serve as an informal channel for
vi Foreword

reaching a consensus if disputes arise among the various participants in the

electoral process; and (g) make recommendations for improving the electoral
system.

In order to meet these objectives, the OAS electoral observation missions

deploy observers throughout a country to monitor the different stages of the
electoral process and compile their comments in a final report.

In general, the work of the OAS electoral observation missions focuses on

those aspects and mechanisms of the political and electoral process where
differences or disputes among participants are more likely to arise or that could
jeopardize the integrity and transparency of the results. The missions therefore
closely follow both organizational and political aspects of the electoral process.
On the organizational side, the missions pay special attention to the enforcement of
election law, the actions of the electoral tribunal, logistical arrangements, civic
education campaigns, and the observance of rules governing the registry of
candidates and voters. On the political side, the missions look at promotional
campaigns, the behavior of the mass media, activities linked to elections in
nongovernmental organizations, and government actions that might have an impact
on the course of the electoral process.

The observations and analyses by the electoral observation team, combined

with the Organization's experience with elections, enable the mission to identify
the weaknesses of a system with considerable accuracy and formulate possible
solutions. As a result, mission reports generally include a number of
recommendations for strengthening the electoral process.

This volume is part of a series designed to provide relevant information on

some of the UPD's electoral observation missions, both to the general public and to
readers with special interests. We trust that the study and analysis of these
experiences will help to increase knowledge of circumstances in the countries in
the region and promote democratic values and practices as the 21st century
approaches.

Elizabeth Spehar
Executive Coordinator
Unit for the Promotion of Democracy
 CONTENTS
Page

Foreword.......................................................................................................................v
Executive Summary ....................................................................................................xi

CHAPTER I. STRUCTURE OF THE MISSION ............................................1

CHAPTER II. ELECTORAL STRUCTURE AND ACTORS ..........................5

A. The Provisional Electoral Council (CEP) .................................7
1. The Electoral Structure........................................................7
2. Observations ........................................................................8
3. Relations between the CEP and the Government..............10
4. Relations between the Electoral Apparatus and
the Mission ........................................................................11
B. Haitian Political Parties...........................................................11
C. National Council of Observers (CNO)....................................16
D. The Haitian National Police....................................................18

CHAPTER III. DELAYS IN THE ELECTORAL PROCESS..........................21

CHAPTER IV. NATIONAL REGISTRATION OF VOTERS.........................27

CHAPTER V. OTHER ELECTORAL PREPARATIONS .............................33

A. The Provisional Electoral Council ..........................................35
B. Candidate Registration and the Printing of Ballots.................36
C. Payment of Registration Workers and
Registration Sites....................................................................36
D. The Lists of Bureaux de Vote and Bureaux
de Vote Workers.....................................................................37
E. Training of Bureaux de Vote Workers ....................................37
F. Security and Election Materials...............................................38

CHAPTER VI. THE SECURITY SITUATION .................................................39

A. Overview .................................................................................41
B. Political Parties........................................................................42
C. Incidents ..................................................................................43
CHAPTER VII. THE MEDIA AND THE ELECTORAL CAMPAIGN ..........45
A. Civic Education and the Media...............................................47
B. Campaign Coverage ................................................................47
C. Coverage of the Campaign by Privately Owned
Media ......................................................................................48
D. Freedom of Expression in the Electoral Campaign ................49

CHAPTER VIII. SUMMARY OF OBSERVATION ............................................53

A. Recompilation of Data Collected on Election Day.................55
B. Regional Incident Reports from May 21, 2000.......................58
1. West ...................................................................................58
2. Artibonite...........................................................................60
3. Center.................................................................................62
4. North ..................................................................................65
5. Northeast............................................................................66
6. Northwest...........................................................................68
7. South..................................................................................68
8. Southeast............................................................................72
9. Grande’Anse (June 11, 2000)............................................74
10. Area of Nippes (June 11, 2000)........................................77
11. Maniche, South Department (June 11, 2000) ...................78
C. Table of Elections Affected by Irregularities..........................79

CHAPTER IX. COMPLEMENTARY ELECTIONS........................................81

A. Les Irois, Grand’Anse .............................................................83
B. Dame Marie, Grand’Anse .......................................................84
C. Bahon, North Department .......................................................85

CHAPTER X. ANALYSIS OF THE RESULTS ...............................................89

A. Introduction .............................................................................91
B. The Transmission and Processing of Results..........................92
C. Calculation of Percentages for the Senatorial
Elections .................................................................................93
1. Chronology ........................................................................93
2. The CEP Method of Calculating Absolute
Majority .............................................................................95
3. The Effect on the Results ..................................................96
4. Alternative Methods to Calculate Absolute Majority in
Accordance with the Electoral Law ..................................96
 5. The Position of the OAS Mission......................................98
6. The Alleged Use of this Method in Previous
Elections ............................................................................99
D. Comparison of Results Provided by BEDs with those
Provided by the CEP ..............................................................99
1. Introduction .......................................................................99
2. Elections for Senate .........................................................100
3. Elections for the Chamber of Deputies ...........................109
E. The System of Addressing Electoral Challenges ..................113
1. Legal Framework.............................................................113
2. Considerations of the Mission .........................................114

CHAPTER XI. CONCLUSIONS AND RECOMMENDATIONS .................117

A. Conclusions ...........................................................................119
B. Recommendations .................................................................121
1. Recommendations Regarding the Institutions.................121
2. Recommendations Regarding the Organization of the
Elections ..........................................................................122
3. Miscellaneous ..................................................................123

CHAPTER XII. FINANCIAL REPORT 1999-2000..........................................125

APPENDICES

1. Bilateral Delegations that Provided Observers to the

OAS Electoral Observation Mission for Legislative,
Municipal and Local Elections of May 21, 2000...........................................149

2. Governments Providing Financial Assistance to the OAS

Electoral Mission for Legislative, Municipal and Local
Elections of May 21, 2000 .............................................................................153

3. Press Releases from the OAS Electoral Observation Mission

to the Legislative, Municipal and Local Elections May 21, 2000 .................157
 Executive Summary

After several delays, the Republic of Haiti held legislative, municipal,

and local elections on May 21, 2000. These elections were viewed as vital
for the democratic consolidation of the country. Haiti had not had a working
parliament since January 1999, and hundreds of millions of dollars of much
needed development assistance had been held up as a consequence. The
importance of these elections corresponded in direct proportion to the
magnitude of the logistics. More than twenty-nine thousand candidates vied
for some seventy-five hundred elected posts throughout the country.

The following is the Final Report of the OAS Electoral Observation

Mission in Haiti. The Chief of Mission and his core group of advisors
arrived on February 23, 2000. By early March, some twenty-two observers
had established an office in Port-au-Prince and five regional offices in Cap-
Haïtien, Cayes, Gonaïves, Hinche, and Jérémie. The observers met on a
continuous basis with electoral authorities, political party representatives,
and civil society leaders to assess the developments of the electoral process
in its entirety. On election day, the Mission was supplemented with short-
term observers provided by various governments, and it also coordinated
with other international observer groups, such as that of the Caribbean
Community (CARICOM) and the Secretariat of the Francophone Countries.
In all, the OAS trained, deployed, and received information from more than
two-hundred observers at more than one thousand polling stations on May
21. The information gathered by those observers contributed to this report.

One of the biggest challenges to the conduct of these elections was the
national registration of voters. The registration of voters began on January
24, 2000, and ended on March 19, 2000. According to the statistics
provided by the Provisional Electoral Council (CEP), 3,959,571 voters
registered from a potential voting population of 4,245,384, or 93.27 percent.
In a separate report issued on May 5, the OAS Mission concluded that
registration had been adequately completed and that those who wished to
register were able to do so. The following report recapitulates the findings
of the OAS Mission during this process.
xii Executive Summary

On May 19, 2000, the OAS Mission presented an interim report in

Haiti that detailed the preelectoral situation. The interim report addressed
specific themes, such as the consequences of the delays on the electoral
process, administrative preparations, the security situation, and freedom of
expression and the state of the media. With respect to security, the report
indicated that violence, which appeared to be electorally related, had been
noted during the preelectoral period. Specifically, the OAS recorded
seventy acts of violence that led to seven deaths of political party candidates
and activists.

The following report continues with the observations presented in the

interim report and details the many administrative preparations and the
challenges that were faced in the preelectoral period principally due to a lack
of sufficient communication, transportation, and materials. It notes, at the
same time, that candidate registration was successfully completed, with
some twenty-nine thousand and five hundred candidates registering for an
estimated seventy-five hundred positions. The CEP also completed the
identification of the 11,238 bureaux de vote (BVs), or polling stations, and
recruited poll workers to manage the BVs on election day.

Election day was postponed on at least three separate occasions, and

the Mission estimates that these delays had a deleterious effect on the
electoral process. Political parties marshalled their resources, anticipating
another delay, and the electoral campaign never began in earnest. Despite
the delays and their effects, however, the CEP eventually accomplished the
major tasks necessary to conduct the May 21 elections successfully. The
day was a great success for the Haitian population, which turned out in large
and orderly numbers to choose both their local and national governments,
and for the Haitian National Police, whose capacity had been questioned by
the political parties, by the government, and by the press, but who had been
able to keep order quietly and effectively.

As stated above, election day proceedings on May 21 represented the

high point of the electoral process. An estimated 60 percent of registered
voters went to the polls. Very few incidents of violence were reported. The
Haitian National Police responded efficiently and professionally to situations
that could have deteriorated into violence. Party poll watchers and national
observers were present at almost every polling station observed by the OAS
and performed their jobs, for the most part, in an objective manner. While
voters had to wait in long lines, especially at the beginning of the day, they
 Executive Summary xiii

were eventually able to cast their ballots free of pressure and intimidation.
Most voters were able to find their polling stations with relative ease. The
OAS report provides a breakdown of the election day findings from its
observers deployed throughout the country.

Shortly after the vote count, however, the election process

deteriorated in some localities. Armed groups of men broke into election
offices in the Departments of the Center and the North and burned ballot
boxes. The receipt of the tally sheets and other electoral materials was
extremely disorganized, particularly in Port-au-Prince, Delmas, and Cap-
Haïtien. Exhausted polling officials arrived in overcrowded electoral offices
and threw their materials on the floor. The newspapers the following day
showed ballots and official tally sheets strewn on the street. Even though
most of the tally sheets were recovered and this situation did not appear to
be generalized, confidence in the electoral process ebbed slightly as a result.

In the immediate postelectoral period, the OAS Mission witnessed the

arrest and detention of various opposition candidates. The Mission visited
several of the detainees in prison to assess the reasons for their arrests.
While these opposition figures were eventually released, the Mission
believes that these summary arrests further chipped away at the confidence
achieved on May 21. With respect to incidents of violence, after election
day, the OAS Mission confirmed at least three election-related deaths.

The Mission also observed a state of disorganization and a lack of

transparency in the compilation of results and delays in posting these results
in many of the communes. Days and even weeks passed and many of the
communal offices still had not completed collecting the results. On many
occasions, results were never posted at the communal level. Some high-
ranking electoral officials left their posts and never returned to their duties.

The OAS Mission visited every communal electoral office (BEC) and
every departmental electoral office (BED) and managed to collect most of
the results at these levels. The Mission compared the results from the
different electoral levels and discovered that there were some discrepancies
that altered the outcome in some races, both in the Senate and the Chamber
of Deputies. The Mission identified at least one other Senate and perhaps as
many as three deputy candidates who should have participated in a second-
round election but were declared winners in the first round on the basis of
these discrepancies. To note one important example, in the Department of
xiv Executive Summary

the Northeast, the first-place independent candidate, Luc Fleurinord, lost one
thousand votes, and the second-place candidate gained sixteen thousand
votes from one electoral level to another, changing the order in which the
senatorial candidates won. In these elections, the first-place candidate will
serve a six-year term, while the second-place candidate will serve four years.
Mr. Fleurinord lodged a formal complaint to the CEP, but the Mission is
unaware of any action taken. Moreover, the Mission identified candidates
who had been simply excluded from a second-round election. The Mission
report devotes a chapter to the analysis of the results, focusing on the
calculation of the Senate races, the transmission of results, and the treatment
of challenges by political parties.

Another concern of the Mission is the perceived lack of transparency

and impartiality in the treatment of challenges put forth by the political
parties. For example, in the South Department, it appeared that the BED
treated cases in a manner that favored one political party. In general, the
OAS Mission was unable to obtain the information related to the number of
challenges, the nature of their investigations, or the final decisions and their
impact on the results. Moreover, there did not appear to be a systematic
treatment of challenges at the departmental level.

The CEP called for elections to be reheld in three separate communes

on July 2, 2000. In two of those communes, Bahon and Dame-Marie, there
did not appear to be compelling reasons to redo the elections. Tally sheets,
once believed to be lost, were recovered. The compilation of results was
conducted in public and validated by party poll watchers. The original
results in these two communes showed non-Lavalas parties winning. During
the partial election in Bahon, OAS observers witnessed a serious disregard
for electoral norms, including polling stations without voter registers, voters
without proper identification, and results that appeared to have been
falsified.

With respect to the municipal and local elections, the Mission’s

overall conclusion is that a series of irregularities appear to have affected an
unspecified number of local elections in the country. However, since one
political party won most of the elections by a substantial margin, it is
unlikely that the majority of the final outcomes in local elections has been
affected.
 Executive Summary xv

In the case of the legislative elections, the Mission considers that a

number of irregularities did compromise the credibility of these elections,
particularly with respect to the senatorial race. As noted in this report, the
posting of results at the communal and departmental levels was sporadic and
lacked transparency. OAS observers who were able to obtain results at these
levels noted discrepancies affecting the results in both the Senate and the
Chamber of Deputies.

The gravest irregularity observed by the OAS Mission was the

calculation of results for the Senate candidates. The Constitution and the
Electoral Law of Haiti stipulate that a senatorial candidate must receive an
absolute majority of the valid votes cast. If not, the candidate must
participate in a second-round election.

In late May, the director of operations of the CEP issued preliminary

results in which the absolute majority was based on a limited number of
candidates (usually the first four candidates with the most votes), instead of
the total number of valid votes. By these calculations, nineteen senatorial
races were decided in the first round, eighteen of which went to the Lavalas
Party. If the calculations for an absolute majority had been based on the
total number of valid votes cast, eight senatorial races would have had to go
to a second round. On June 2, the Mission informed the members of the
CEP of this error. The Chief of Mission and his core group also met with
the members of the CEP to explain the error and offer their assistance in
rectifying it. Various sectors of Haitian civil society, the national observers,
and almost all of the opposition parties also supported the strict application
of the Electoral Law in the calculation of an absolute majority in the Senate
races and called on the CEP to change its methodology accordingly.

The controversy concerning the calculation of results had

ramifications within the CEP itself. Two of the three opposition-appointed
members were asked to resign by their party and did so. The CEP president
left his post and the country instead of validating the Senate calculations,
according to his public statement.

Since June 2, the Mission had treated the CEP calculation as a simple
human error. The refusal of the CEP to modify the calculations, however,
eventually led the Mission to conclude that the highest electoral authority of
the country had knowingly violated its own Constitution and Electoral Law.
By excluding some 1.1 million votes for approximately one hundred
xvi Executive Summary

senatorial candidates who were not in the top four, the CEP precluded the
possibility of an election that allowed all participants the same consideration.

Given the failure of the CEP to conduct the second round elections for
the Senate with the proper list of candidates based on Haitian Electoral Law,
the OAS Mission was compelled to consider the very basis of the July 9
elections as fundamentally flawed. Numerous efforts to resolve this
electoral crisis were attempted by various sectors within Haitian society as
well as by important regional entities such as the (CARICOM). However,
since several days before the election date no acceptable solution to the
crisis had been found, the OAS Mission regretfully reported on July 7 of the
suspension of all its observation activity for the second round.

It was the profound wish of the OAS Electoral Observation Mission in

Haiti that a solution to the calculation of senatorial results in the first round
of elections be achieved, in order that the will of the Haitian people as
reflected during the May 21 elections be fully expressed, and so that the
significant achievements of that day not be lost. In a letter dated June 20,
the OAS Mission had offered its good offices to the acting president of the
CEP to review the calculation of the Senate and deputy races and address
any other issues that were of interest to the CEP.

The observations made by the OAS Mission in Haiti are meant to be

constructive and contribute to building an electoral process in which all
Haitians can have confidence. This is a critical step for the strengthening of
democracy and the rule of law in Haiti. Through the coordinated efforts of
the CEP, the government, the political parties, and civil society, the Mission
is convinced that Haiti can yet achieve these goals.
 CHAPTER I
STRUCTURE OF THE MISSION
 In June 1999, the Government of Haiti, through its Ministry of
Foreign Affairs, requested the deployment of an OAS Mission to observe the
legislative, municipal, and local elections in Haiti. In a letter of response by
the Assistant Secretary General in July of the same year, the OAS accepted
the invitation and the Unit for the Promotion of Democracy of the OAS
General Secretariat was instructed to begin preparations in that regard.

The legislative, municipal, and local elections were originally

scheduled to take place on November 28, 1999. The OAS Mission delayed
its deployment when the date was rescheduled for March 19, 2000. With a
core group of seven observers, the OAS Mission arrived in the country on
February 23, 2000 in order to observe the final stages of the registration
process and other important electoral preparations. By the first week of
March, the Mission grew to nineteen observers deployed in six regional
offices in Port-au-Prince, Hinche, Jérémie, Cap-Haïtien, Gonaïves, and Les
Cayes.

In the framework of separate formal agreements signed between the

OAS Electoral Observation Mission (EOM) and the Provisional Electoral
Council (CEP), on the one hand, and the Government of Haiti and the
Secretary General on the other hand, both of which guarantee the unhindered
access to people, places, and information relevant to the electoral process,
the OAS Mission maintained contact with electoral authorities, political
party representatives and civil society leaders. During the preelectoral
period, the EOM observed the nationwide registration of voters, other
electoral preparations, the security situation in the country, the campaign,
and the organization of the domestic observation effort. The OAS Mission
also followed closely the political conflict in the commune of Anse-
d´Hainault in the Department of the Grand’Anse and offered its good offices
to facilitate a peaceful resolution to this problem.
4 Structure of the Mission

Shortly before election day, contributing governments and

international nongovernmental organizations provided the Mission with
short-term observers who remained in the country for between five days and
one month. Those governments and organizations included the
Governments of the United Kingdom, the United States, Canada, Norway,
Spain, France, and the Concertation for Haiti and the World Lutheran
Federation (See Appendix I, List of Bilateral Delegations). While those
governments and organizations covered the expenses of the observers, they
were trained and deployed by the OAS Mission and reported directly to the
OAS Chief of Mission.

Additionally, the OAS Mission provided training and deployment for

a number of observers of other international organizations that arrived
shortly before election day, including the Caribbean Community
(CARICOM) and the Secretariat of the Francophone Countries. Those
groups utilized the election day forms designed by the OAS and provided
information to the Mission about election day proceedings.

The financing for the basic infrastructure of the Mission was made
possible through generous contributions from the Governments of the United
States, Canada, France, Argentina, Chile, and Japan. (See Appendix II,
Donor Countries.)

After election day, the OAS observers remained in the regional offices
to monitor the posting of results and their transmission to the CEP and the
manner in which challenges to those results were investigated. The Mission
did not observe the proceedings of the second round on July 9.

Since its arrival, the EOM issued numerous press releases about
different aspects of the electoral process, a report on the state of voter
registration, and an interim report detailing the proceedings of the
preelectoral process. (See Appendix III, OAS Press Releases.)

The OAS Mission departed Haiti on July 14, 2000.

 CHAPTER II
ELECTORAL STRUCTURE AND ACTORS
A. The Provisional Electoral Council (CEP)

1. The Electoral Structure

The Provisional Electoral Council (CEP) is an independent body

responsible for the organization and control of the elections up until the
proclamation of the results (Article 1 of the Electoral Law). It has
administrative autonomy. Its independence vis-à-vis the executive power is
guaranteed by the law although the CEP depends on the executive power for
its material and financial resources and for the collaboration of ministries
and state agencies at all levels (i.e., police, use of public places for
registration, and voting sites, etc.). The CEP is composed of nine members,
who for these elections were named by presidential decree after consultation
with political parties and party coalitions. The choice of the members was a
result of consensus among the parties concerned.

The CEP establishes a central litigation office for electoral matters

(BCEC) to hear disputes already heard by departmental electoral offices.
This body is composed of the CEP members who are assisted by two
lawyers (Articles 11-12).

The CEP has the power to appoint (Article 9) the members of the
departmental electoral offices (BEDs). It designated eleven BEDs for the
legislative, municipal, and local elections of May 21, 2000; that is, one per
department except for Grand’Anse and the West where, due to geographical
and logistical considerations, there are two. Each BED is made up of three
members: a president, vice president, and a secretary, all named by the CEP
according to the criteria delineated in the Electoral Law. No political party
is represented, but the political sensibilities or party allegiance of a member
is not, by law, a factor to impede his or her eligibility. The members of the
8 Electoral Structure and Actors

BED hear all disagreements of an administrative nature. Aided by a lawyer,

the BED convenes then as the departmental litigation office for electoral
matters (BCED).

The CEP also appoints the three members of the communal electoral
offices (BECs). The BECs have responsibilities and prerogatives similar to
those of the BEDs, to which they refer matters that cannot be resolved at a
local level.

The CEP designates all of the registration offices (BIs), placing at

least one in each neighborhood, communal section, commune and town. In
zones of high population density more than one office is opened, if possible
within public buildings. Each BI has four members. The president of the BI
is chosen and named by the CEP, while the three other members (one
secretary and two clerks) are named by the CEP from a list of suggestions
presented by political parties, coalitions, organizations, bureaux de vote and
civil society. Two members of the same political party may not be in the
same BI.

The Bureaux de Vote (BV) are installed at the same location as the BI
or, otherwise, in close proximity of the BI where the register was filled. The
BV is made up of three persons named by the CEP, one president, one vice
president, and a secretary. The president is chosen and named by the CEP.
He or she is responsible for the voting operations, the counting of the ballots,
and all the electoral documents of the BV over which he or she presides.

2. Observations

Resulting from an agreement between the government and the

political party group Espace de Concertation, the CEP was created on March
6, 1999, and was charged with the organization and the control of the
elections throughout the country. According to the Electoral Law, it had
administrative authority and was independent from governmental
institutions. Nine members comprised the CEP. For the first-round
elections, the nine members included Léon Manus, president; Debussy
Damier, vicepresident; Irma Rateau, secretary general; Micheline Figaro,
treasurer; Yva Youance; Ernst Mirville; Macajou Médard; Emmanuel
Charles; and Carlo Dupiton. This CEP remained in place until June 16,
when Léon Manus left the country and two members, Debussy Damier and
Emmanuel Charles, resigned. On July 3, 2000, the President of Haiti
 Electoral Structure and Actors 9

replaced the three departed members with Richard Chéry, Jean Alexandre,
and Yves Bertrand. They were sworn in on July 6 by the Supreme Court.
The political parties from the opposition requested that the Court not
confirm those appointments.

For most of the electoral process, the CEP enjoyed the support of the
political parties, civil society, and the international community. The CEP’s
decisions were based on consensus, so as to include the opinions of a
majority of the political parties in Haiti. Although this choice of procedure
by the CEP slowed the process, it also made its decisions acceptable to most
candidates and political parties.

Although the CEP suffered from several logistical, technical, and

financial difficulties, it eventually managed to accomplish the tasks required
to hold elections. Despite criticisms, the CEP was able to register almost all
eligible voters and provide them with an electoral card with their
photograph. The electoral card was well received by the majority of Haitian
society and understood as an acquisition, in particular by those Haitians in
rural areas who found themselves for the first time with a picture
identification card. This card ensured an increased transparency of the
process and contributed largely to reducing the risk of fraud.

Nonetheless, the deficiencies of the CEP gradually worsened and

eventually culminated with the departure and/or resignation of three of its
members. From almost its inception, the nine members of the CEP failed to
meet in weekly plenary sessions, as required by internal rules. Moreover,
the members of the CEP and of its management often gave out contradictory
statements to the press, creating confusion among electoral personnel and
the public.

A lack of transparency further damaged the credibility of the CEP.

For instance, the selection of polling officials at all levels required the
participation of all the political parties. Many of the parties should be
faulted for not providing timely or accurate information. However, part of
the responsibility must be placed on the CEP, as it failed to communicate
clearly the criteria to be utilized. Likewise, political parties complained to
the OAS Mission that the treatment of challenges by the CEP appeared
arbitrary and lacked transparency.
10 Electoral Structure and Actors

In addition, the relations among electoral offices seemed overly

hierarchical, the higher levels rarely consulting lower levels. The CEP often
issued orders disregarding the concerns of the BEDs; the BEDs duplicated
this behavior with regard to BECs, and those offices did the same with poll
workers. Every level suffered from lack of access and consideration from
the superior level. Observers often noted among the members of BEDs,
BECs, and registration and polling stations a sentiment of professional
frustration that affected the daily work of organizing the elections.

Given the problem of communication, the CEP decided to appoint

each of its members responsible for one department. This division of
responsibilities was often theoretical and, in fact, other constraints prevailed.

Between the first and second rounds many electoral officials resigned
or abandoned their posts. The presidents of the BEDs of the Departments of
Grand'Anse, the North and the Center all resigned. (The BED president of
the Center explained his departure by his refusal to condone the grave
irregularities committed in his department.) Likewise, at least ten BEC
presidents resigned on the eve of the second round due to a variety of
reasons, including security concerns and alleged fraud. After the
complementary elections in Bahon on July 2, members at all levels of the
electoral apparatus in the Department of the North resigned, citing their
nonconformity with the conduct of those and other elections in the
department. Most of these officials had not been replaced when the Mission
departed the country.

3. Relations between the CEP and the Government

The channels of communication between the President of the republic

and the president of the CEP sometimes appeared insufficient. For instance,
when it became clear that the March 19 date could not be respected, the CEP
formulated another electoral calendar and announced elections for April 9,
citing the need to seat Parliament by the second Monday in June, as
stipulated by the Constitution. The CEP, however, had not consulted the
President, who had to issue an official decree.(See Chapter III, Delays in the
Electoral Process.)
 Electoral Structure and Actors 11

4. Relations between the Electoral Apparatus and the Mission

The relation between the Mission and the Haitian electoral apparatus,
including the CEP, BEDs, and BECs, was generally positive, and observers
were given full access to information without delay.

A debate ensued, however, following the publication of the letter sent

by the OAS Chief of Mission to the CEP concerning the method of
calculating the absolute majority for the Senate elections. The OAS Mission
intended for this letter to be confidential, but upon arrival at the CEP office
to discuss the issue, its contents were leaked to the media. Soon after the
leak, the CEP member in charge of public relations accused the OAS Chief
of Mission of making the letter public. The Mission persevered in its work
and continued to contact the CEP confidentially each time it observed
irregularities.

For the most part, the CEP continued to provide the OAS Mission
with access to information, although there were some important exceptions.
In June, the BED secretary of the Artibonite refused to provide information
to OAS observers and requested that other members of the BED deny access
as well. Similarly, in Jacmel, before the BED of the South refused access to
OAS observers, its president had provided incorrect information on
challenges to the results for deputy.

B. Haitian Political Parties

Since 1987, the democratic process of Haiti has been characterized by

a multiparty system. Political parties are numerous; their size, geographical
roots, and tendencies vary. They have scarce resources and other than
receiving free airtime on national television and radio, receive no state
support. Many of these parties merely represent individual candidates and
have little popular support. With the exception of Fanmi Lavalas, most
parties did not conduct many campaign activities during the electoral
process. Political leaders opposing Lavalas usually transmitted their parties’
messages through unpaid media such as news programs. With few
exceptions, these parties did not present platforms, and their rhetoric focused
primarily on discrediting the government and the Fanmi Lavalas Party.
12 Electoral Structure and Actors

According to a study conducted by a private-sector group, Économie,

Finances Gestion & Société (ECOSOF), in November 1999, there were 19
active political parties and approximately thirty organizations considered
political parties. Some regional and national political parties formed
coalitions for the May elections while, at the municipal level, candidates to
the mayoral position were grouped in cartels of three individuals, disputing
the seats for mayor, first adjunct, and second adjunct.

Communication between candidates and the parties’ national

leadership was practically nonexistent. Frequently, regional representatives
did not know about decisions taken by their leadership. Moreover,
candidates received little or no financial or logistical support from their
national leaders, and they were therefore hesitant to follow decisions that
were unfavorable to them. For instance, most political parties opposing
Lavalas announced their decision not to participate in elections in the
Grand’Anse and in the second round. Nonetheless, they explained the
participation of their candidates as personal stands adopted independent
from the parties. Most of the candidates, who had made personal
investments in the campaign and had an electoral base of their own, did not
request to be excluded and participated in the elections.

Political parties opposed to Fanmi Lavalas alleged various

irregularities in the electoral process, many of which could not be
substantiated. The supposed expulsion of poll watchers from BVs is
illustrative. In the BVs observed by the Mission, party poll watchers left at
night due to fatigue and were not denied entry or expelled. The hurried and
negligent position of these parties harmed their image. The view that they
desired to delegitimize the process because they had lost the election gained
credence. After the first round, the parties in opposition to Lavalas decided
not to participate in the second round due to their belief that the increasing
irregularities precluded a fair chance of winning. Their lack of effort to
substantiate their allegations further discredited their position, especially in
the view of the international community.

Organizasyon Fanmi Lavalas (OFL). Fanmi Lavalas was created in

November 1996 by former President Jean-Bertrand Aristide, after the
breakup of the coalition Organisation Populaire Lavalas. Fanmi Lavalas
was the only party to have a presence throughout the country. There was
heavy publicity at the national level aimed at familiarizing the electorate
 Electoral Structure and Actors 13

with its number on the ballot, “Number 11,” and with the party symbol, “ the
Table.”

Fanmi Lavalas presented the greatest number of candidates and won

the vast majority of the elections contested. According to the results
proclaimed by the CEP, Lavalas won eighteen of nineteen Senate seats, all
elected in the first round. According to the CEP results, the party likewise
won twenty-seven deputies in the first round, and forty-two candidates to
deputy participated in the second round (excluding the results from the
Grand’Anse, which had not yet been published by the time the Mission left
the country). During the electoral process, the Mission met with its leader
and several candidates to Senate and deputy.

Parti Louvri Barye (PLB). The Mission met its director, Renaud
Bernardin; its secretary general, Willy Louis; members of its management;
and a number of candidates, several times. The PLB presented three
candidates for the Senate and thirty-nine for the Chamber of Deputies, of
whom four candidates participated in the second round (excluding the results
from the Grand’Anse). The PLB proposed the creation by the President of a
national commission to evaluate the electoral process and to propose
measures to resolve the electoral impasse.

Espas Solidarite Kan Popile (ESKANP) and Kowodisyon

Resistans Grandans (KOREGA). ESKANP-KOREGA is an electoral
alliance, which gathers members of the former anti-neoliberal parliamentary
bloc and the KOREGA. The Mission met with its spokesperson, Kelly
Bastien, Senator Maxim Roumier, as well as other candidates. The Mission
also met with Father Joachim Samedy, leader of the KOREGA. The
ESKANP-KOREGA, which denounced fraud during the elections in
Grand’Anse on June 11, announced it would not participate in the second
round.

Mouvement Organisation du Pays (MOP). The MOP is considered

part of the Lavalas movement and was a member of the Plateforme
Politique Lavalas during the 1995 elections. Gesner Comeau and Jean
Molière are its principal leaders. Three MOP candidates for deputy
participated in the second round. The Mission met several times with the
MOP regional leaders and candidates.
14 Electoral Structure and Actors

Tet Ansamn Pou Sovtaj Ayisyen. This party, created in early 1999
by Gérard Blot, former minister of the Aristide administration, had two
candidates running for deputy in the second round. The Mission met several
times with the regional leaders of Tet Ansamn.

Espace De Concentration. The Espace de Concertation was

officially founded in July 1999 as a political coalition. In March 1999, its
leaders reached an agreement with the President of the republic in order to
form a new government and to create the Provisional Electoral Council.

Espace de Concertation brings together Ayiti Kapab, a party founded

by Ernst Verdieu and led by Robert Auguste since 1998; Génération 2004, a
party directed by Claude Roumain and Daniel Supplice, who was candidate
to the senate in the Department of the West; KID (Konfederasyon Inite
Demokratik), a group created by Evans Paul, one of the main
spokespersons of the Espace and former mayor of Port-au-Prince;
KONAKOM (Congrès National des Mouvements Démocratiques), led by
Victor Benoit, who was candidate to the Senate in the Artibonite, and Micha
Gaillard, candidate for mayor in Port-au-Prince; and PANPRA (Part
National Progressiste Révolutionnaire d’Haïti), led by Serge Gilles,
former senator during 1991-1994 and candidate to the Senate in the Center.

The Mission met with the leaders of the Espace several times, either
jointly in representation of the coalition or separately in representation of the
separate parties. Observers visited candidates at all levels. The Espace
presented ninety-two candidates to parliamentary elections, sixteen to the
Senate and seventy-six to the Chamber of Deputies. None was elected in the
first round; thirteen candidates to deputy were to participate in the second
round.

To protest the irregularities of the electoral process and prior to the

second round, the Espace requested that its ministers in the Préval
government resign, which they did. After the first round, the Espace also
demanded the resignation of its three CEP members. Two of the three left
the CEP. The third, the CEP secretary general, was later excluded from the
movement. The Espace called first for a commission of evaluation and then
for the invalidation of the elections.

Mouvement Chrétien pour une Nouvelle Haiti (MOCHRENA)

was created recently in order to mobilize Protestant churches. The Mission
 Electoral Structure and Actors 15

met with its leader, Pastor Luc Mésadieu; its spokesperson, Ernst Colon;
several of its managers; and many candidates.

The movement presented eighty-one parliamentary candidates,

fourteen to the Senate, and sixty-seven to the Chamber of Deputies, of
whom Five candidates for deputy were to participate in the second round.
The MOCHRENA officially contested the counting methodology and called
for the repeal of the elections, the resignation of the President of the
republic, and the organization of general elections.

Organisation du Peuple en Lutte (OPL), formerly Organisation

Politique Lavalas. The Mission met its leader Gérard Pierre-Charles; its
spokesperson Sauveur Pierre Etienne; former Senator Paul Denis; and many
candidates. Seventy parliamentary candidates participated in the first round
under its banner, sixteen to the Senate, and sixty-three to the Chamber of
Deputies, of whom thirteen candidates to deputy were to stand for the
second round. The OPL demanded the invalidation of the elections.

Mouvement Patriotique pour le Sauvetage National (MPSN), a

union of conservative politicians bringing together the Mobilisation pour le
Développement National (MDN), Mouvement National Patriotique 28
Noviembre (MNP-28), and the Alliance pour la Libération et
l’Avancement d’Haïti (ALAH), Parti Démocrate Chrétien Haitien
(PDCH) and a wing of the Mouvement d’Organisation du Pays (MOP).
The Mission met with its spokesperson, Hubert de Ronceray, and several
activists and candidates. The movement fielded twenty-three parliamentary
candidates, seven to the Senate and sixteen to the Chamber of Deputies,
none of whom was elected. They also called for the invalidation of the
elections.

Rassemblement des Démocrates Nationaux Progressistes (RDNP).

Party created by former President Leslie Manigat, a well-known public
figure who had an important presence in political debates in the country.
The Mission met with him and with several candidates. This party presented
fifty-five parliamentary candidates, ten to the Senate, and forty-five to the
Chamber of Deputies, of whom only one was to stand in the second round.
It called for the invalidation of the elections.
16 Electoral Structure and Actors

C. National Council of Observers (CNO)

Created by the Electoral Law, the National Council of observers

(CNO) was established in January 2000 and served as a CEP-sanctioned
umbrella organization for the forty-five Haitian organizations that observed
the legislative, municipal, and local elections. Of its members, twelve
groups provided most of the observers and resources and comprised the
CNO coordinating council.1/ While collectively the CNO was nonpartisan,
its members individually represented the many political tendencies in the
country.

The coordinator of the CNO, Léopold Bélanger, headed the RCN,

which brings together members of the business sector and other private
institutions. (RENO) grouped several trade unions that had a presence in all
of the communes in the country and trained departmental coordinators and
instructors, who were then in charge of the training of all observers in the
other departments. Another group, Democratic Initiatives, brought together
likeminded businessmen. KOZEPEP, an organization linked to agrarian
reform and funded by the Office of the President of the republic, also
provided a number of observers throughout the country.

Human rights organizations were also active in the CNO. The

network of the National Commission on Justice and Peace covered seven
departments. The Catholic church already had commissions in each diocese,
making up a network of human rights observation spread through practically
every commune. Throughout the country, they held, jointly with the
POHDT, training sessions for coordinators and instructors, then for
observers.

The relations between the CNO and the CEP were generally
characterized by a lack of dialogue. On February 25, the coordinator of the
CNO and the president of the CEP, Léon Manus, signed an agreement in
order to define the procedures of accreditation, but the CEP did not ratify
1
The coordinating council includes Réseau Civique National (RCN), Réseau d’Observation du Bas-
Artibonite (ROBA), Organisation Nationale pour l’Alphabétisation et la Formation des Adultes
(ONAFPA), Coordination Nationale du Mouvement des Femmes Soleil Levé (CONAMOFS), Justice et
Paix (JILAP), Comité d’Observation Electorale du Département de l’Ouest (COEDO), Fondation pour le
Développement et le Progrès du Centre (FDPC), KOZEPEP, Réseau National d’Observation (RENO),
Plateforme des Organisations Haitiennes des Droits Humains (POHDH), Réseau d’Observation Electorale
du Nord’Est (ROEDNE) and Organisation des Jeunes Patriotes du Sud (OJPS).
 Electoral Structure and Actors 17

this agreement, and doubts about accreditation and other procedures

continued until the eve of election day. Eventually, the CEP provided
accreditation, but only after much prompting by the CNO, including a public
letter demanding immediate accreditation of its observers.

The CNO complained that it lacked financial support and relied

primarily on the resources of its member organizations. The National
Democratic Institute for International Affairs, a Washington-based
nongovernmental organization, provided training to many of the groups
participating in the CNO.

On election day, the CNO fielded approximately seven thousand

observers throughout the country. OAS observers noted their presence in
many BVs and indicated that they accomplished their tasks according to
their mandate. The OAS Mission met with the principal groups of the CNO
several times after election day. For the most part, the CNO and the OAS
Mission concurred that, despite its shortcomings, the electoral process
immediately after May 21 had proceeded positively.

During the elections in the Department of the Grand’Anse, the OAS

observers noticed that many CNO observers did not appear to belong to the
organizations that comprised its membership. Indeed, a select number of
observers had communications equipment and transportation that appeared
to be provided by the government.

After election day, differences among the CNO began to appear. In

particular, the membership split on its assessment of the calculation of the
senate results. On June 8, the CNO issued a press release in which it
denounced the counting methodology as “illegal and unacceptable." Twelve
days later, on June 20, eight of twelve members of the CNO organizations
signed a resolution in which they announced that the CNO would not
observe the second-round election. The letter cited that the method of
calculation of the absolute majority for the Senate was illegal and expressed
concern about the sudden departure of the president of the CEP. Four of the
twelve members (JILAP, KOZEPEP, POHDH, and OJPS), however, did not
sign the letter.
18 Electoral Structure and Actors

D. The Haitian National Police

On May 21 in eight departments and on June 11 in the department of

Grand’Anse, OAS observers received few reports of violent incidents,
thanks in large part to the work of the Haitian National Police (HNP). In the
days immediately before election day, the actions of the HNP prevented or
controlled potential incidents that could have degenerated into violence.

Prior, during, and after the vote, observers took notice of the hard
work and the diligence shown by policemen who, for the most part,
accomplished their tasks professionally and within their mandate. After the
vote, the police helped to transport voting materials from polling sites to the
BECs. OAS observers often witnessed that this operation, while perceived
differently by opposition parties, protected electoral materials and personnel
and ultimately had a positive impact on the process.

Likewise, polling officials performed many tasks in the police stations

normally done elsewhere. It is important to emphasize that electoral
officials themselves took the decision to utilize the police stations, as they
believed their security to be in jeopardy in the polling stations. OAS
observers did not note any irregularities or efforts to alter the results,
although drafting official records and counting votes in the police stations
caused a problem of perception in the eyes of many political actors.

Despite its professional and efficient work during election day and
immediately after, the HNP sometimes did not provide adequate security or
respond efficiently during the preelectoral process. Some of the incidents
observed by the Mission in which there was minimal or inadequate police
response include the demonstrations in Port-au-Prince on March 28-29, the
fire at the headquarters of the Espace de Concertation, and the violent
reactions that followed the funeral of Jean Léopold Dominique. (See
chapter VI, “The Security Situation.”)

During the preelectoral process, the HNP and the CEP were charged
with drafting a comprehensive security plan. Unfortunately, middle- and
lower-level electoral authorities were unaware of the plan. They often did
not receive a response when they transmitted their concerns about the lack of
security measures to departmental police directors. Despite the repeated
demands of local authorities in the Grand’Anse, for instance, the police did
not intervene during the violent incidents in this department.
 Electoral Structure and Actors 19

The Mission also noted that high-level transfers within the

administration of the HNP at departmental levels (i.e., in the South and
Nippes) hindered the implementation of electoral security plans. Political
parties in the South complained to the Mission about those transfers. In
Port-à-Piment in the South, for example, the commissaire was transferred
forty-eight hours before the May 21 election.
 CHAPTER III
DELAYS IN THE ELECTORAL PROCESS
 In January 1999, the Parliament of Haiti was dissolved, although
elections to install a new parliament had not yet been contemplated. An
important step towards having the elections was achieved in March 1999
when President René Préval reached an agreement with one of the
opposition political party group and created the new Provisional Electoral
Council (CEP). Shortly thereafter, the President, after consultations with the
CEP, decreed that elections would take place on November 28, 1999.

Because of administrative delays related specifically to the

registration of candidates and voters, it soon became apparent that elections
would not be held on the November date. Again, the CEP formulated a new
electoral calendar, and the President decreed a new date of March 19, 2000.
Most believed that the additional four months would provide ample time to
complete the necessary preparations for election day. In this context, the
OAS Mission was finally deployed on February 23, 2000.

Shortly after the arrival of the OAS Mission, it became apparent that
actions did not accompany the declared commitment to the March 19 date.
Deadlines on the electoral calendar lapsed, and efforts to rectify the delays
were slow in coming or nonexistent. For instance, the departmental electoral
offices (BEDs) complained that new registration materials arrived late,
sometimes on the last scheduled day of registration, and they did not have
the means to transport the materials to the communal electoral offices
(BECs). The officials of the BEDs and BECs rarely had contact with the
CEP and learned through the press, for instance, that the period for voter
registration had been extended.

Oftentimes, the Government of Haiti did not provide in a timely

manner the resources necessary for the transport of materials or payment of
the salaries of the electoral authorities. (Additionally, the CEP did not seem
24 Delays in the Electoral Process

to know precisely whom from its offices it should pay.) Through its highest
authorities, the government informed the public that it did not have the
means to guarantee adequate security to electoral authorities. Few attempts
were made to reach a political compromise or provide security for voter
registration in Anse- d’Hainault, and even those attempts were well after the
close of national registration and close to election day.

Of particular concern to the OAS Mission was the lack of fluid

communication between the Office of the President and the CEP. When it
became apparent that the March 19 date was impossible to maintain, the
CEP yet again drafted a new electoral calendar and announced an April 9
election date. The CEP’s president and secretary made this date public in a
press release in March, stating that the elections must take place in order to
allow for the seating of parliament by the constitutionally mandated second
Monday of June.

According to Article 92.2 of the Constitution, the second

parliamentary session lasts from the second Monday of June until the second
Monday of September. The June deadline was introduced by the CEP in its
press release announcing the new election date and was supported by the
opposition political parties and the international community. The CEP,
however, failed to consult with the President of the republic, who had to
issue a decree in order to make that date official. The President of the
republic questioned the CEP’s readiness to conduct the elections and stated
that it was more important to have well-run elections than respect the June
date. The electoral process in Haiti entered a precarious stage in which a
postponement of the March 19 date was inevitable, but a new date was not
agreed upon.

After a period of uncertainty, the President and the CEP finally agreed
on a new date. On April 10, President Préval issued a decree stating that the
first round would take place on May 21 and the second round on June 25,
the dates agreed to in his consultations with the CEP.

The Mission considers that the continuous delays in the holding of the
elections had a deleterious effect on the electoral process. The inability of
the CEP to respect its own electoral calendars had a negative impact on its
image of professionalism. The government’s vacillation also led some to
question its commitment to the timely and transparent conduct of these
elections.
 Delays in the Electoral Process 25

The delays had economic and security-related consequences as well.

The OAS Mission witnessed precious few campaign events, originally due
to the doubts of the political parties that the elections would take place on
the dates previously announced and later due to a lack of funds. The
deteriorating security situation in the face of this uncertainty also impeded
candidates from campaigning.
.
 CHAPTER IV
NATIONAL REGISTRATION OF VOTERS
 The registration of voters began on January 24, 2000, and ended on
March 19, 2000. According to the statistics provided by the Provisional
Electoral Council (CEP), 3,959,571 voters registered from a potential voting
population of 4,245,384, or 93.27 percent.2/

In a report issued on May 5, the OAS Mission concluded that

registration had been adequately completed based on the following
observations: 1) the high percentage of registered voters relative to the
estimated voting population; 2) many new registers that had been distributed
were not completely filled, indicating that those who wished to register were
able to do so; and 3) the overall satisfaction expressed by the political party
representatives and other actors in the electoral process. One political party
did allege that there was a numerically important number of Haitians who
had not been able to register. However, it was unable to cite numbers or
provide documentation supporting this contention.

As noted in the OAS registration report, the process nevertheless had

many flaws. Originally, registration officers could not meet the demand for
cards due to a shortage of materials. When new materials arrived, the BEDs
often could not distribute them quickly to the BECs or the registration
offices (BIs) due to a lack of transportation. The BEDs complained that they
rarely received official communications from the CEP. For instance, when
the CEP decided to extend voter registration from March 15 to March 19, no

2
The Mission understands that the estimated voting population was calculated on the last census taken in
1982, with a projected estimated increase in population. The Mission was unable to obtain the exact
methodology for this calculation.
30 National Registration of Voters

official notice was given to the BEDs or the BECs. The electoral authorities
in the field all heard about the extension through the media.

For the first time, many BIs were grouped together in electoral
registration centers, obliging the CEP to create corresponding bureaux de
vote (BVs) in nearby geographical locations. In many cases, particularly in
urban concentrations, this resulted in logistical difficulties in that it became
difficult for the CEP to find suitable sites for the BVs, which require more
physical space than the corresponding BIs.

Another difficulty in the process was the CEP’s late payment of the
salaries of the BI workers. In many locations, discouraged and skeptical BI
workers observed erratic office hours, lengthening in some locations the
long lines of voters waiting to register. Towards the end of registration,
some BI workers retained the registers as a means of exerting pressure on
the CEP to pay them, contributing to the CEP’s delay in pronouncing a final
registration figure. In the most extreme case, frustrated unpaid BI workers
attempted to set fire to the Port-au-Prince BEC at the end of registration.

It was difficult to quantify the exact percentage of registered voters

relative to the voting population. In an April 13 opinion poll carried out by a
private polling company, SEMA in Port-au-Prince, 84.1 percent of
respondents said they had registered. In almost every department observed,
the voting population estimates appeared to be low. At the CEP’s own
admission, some voters registered more than once, which would inflate the
percentage of those registered to the estimated voting population. It did not
appear, however, that duplicate registration was significant on a nationwide
scale.
 National Registration of Voters 31

The Republic of Haiti – Registration January 24 to March 19, 2000 3/

Department Registered Estimate BI % BI BV BV

Est. Total Est. Total
Artibonite 580,975 576,400 100.79% 504 514 1,441 1,565
Center 282,899 301,600 93.80% 242 244 674 754
Grande’Anse 196,865 206,800 95.20% 251 246 517 547
Nippes 138,431 144,800 95.60% 172 173 378 385
North 420,075 439,200 95.65% 482 483 1,098 1,219
Northeast 156,402 148,400 105.39% 98 116 370 422
Northwest 208,880 205,660 191 81 514 267
South 340,053 338,400 100.49% 391 393 870 962
Southeast 253,578 240,600 105.39% 226 245 601 672
Port-au- 1,158,169 1,176,000 98.48% 470 833 412 3,110
Prince 1
Port-au- 309,591 319,000 97.05% 280 290 570 787
Prince 2
TOTAL 4,045,918 4,096,860 101.08% 3,307 3,618 7,445 10,690

3
The numbers cited above were collected at the BED level by the OAS electoral observers. The OAS
Electoral Observation Mission’s total number of registered voters (3,933,165) varies slightly with the final
number published by the CEP (3,959,571). Moreover, the OAS Mission calculated a total of 10,690 BVs
on the BED level, although the CEP noted that there were a total of 11,238 BVs.
 CHAPTER V
OTHER ELECTORAL PREPARATIONS
A. The Provisional Electoral Council

The Provisional Electoral Council (CEP) is responsible for the

organization and control of the elections throughout the territory of Haiti and
has administrative autonomy from the other branches of government. The
CEP, which organized the legislative, municipal, and local elections,
consisted of nine members and for the most part enjoyed the respect of the
political parties and civil society. Its inability to respect deadlines and the
postponement of elections on three separate occasions, however, sorely
tested this credibility.

Probably one of the biggest reasons that the CEP had difficulties in
accomplishing its tasks was the absence of middle management to
administer the daily work of elections and maintain contacts with the
departmental and communal offices. The consequences of this absence
included too few registration officers and offices, poorly trained registration
clerks, lack of a public awareness campaign, nonpayment of registration
officers, nonpayment of rent for electoral offices, and the lack of a system
for distributing registration materials.

OAS observers noticed that on many occasions, both in the CEP in

Port-au-Prince and in the field, the responsibility of supervision reverted by
default to the international consultants of the International Foundation for
Electoral Systems (IFES) and the United Nations Development Program
(UNDP).

Cooperation and open channels of communication between the

executive branch and the CEP are essential for a smooth electoral process.
In its observations, the Mission noted periodic difficulties in this
36 Other Electoral Preparations

relationship, which at times inhibited the work of the CEP and added an
element of uncertainty to the electoral process.

B. Candidate Registration and the Printing of Ballots

Probably the two most substantial challenges that the CEP faced in the
preparation for elections were candidate and voter registration. The OAS
Mission was not present for candidate registration, but the CEP informed the
Mission that 29,490 candidates had registered for 7,526 positions for the
Senate, Chamber of Deputies, mayors, and the members of the Assembly of
Communal Section (ASECs) and the Administrative Council of Communal
Section (CASECs). Candidate registration lasted from December 10, 1999,
to January 9, 2000, and was extended for three additional days subsequent to
that period. In a likely effort not to exclude candidates, few challenges to
candidate selection were upheld, and few were disqualified.

A consortium of three Haitian printers was charged with printing the

ballots: Henri Deschamps, Donatal, and Desanti. Henri Deschamps has
printed ballots in every election in the country since 1987 except in 1995,
when a California-based company was contracted. Originally, the printers
produced approximately 4.4 million ballots and later produced another 2
million to cover the supplemental BVs added during the extension of
registration.

The printers acknowledged that producing the ballots took much

longer than expected due to the many errors in the CEP data base. Errors
included names and/or faces appearing twice on the same ballot and
incorrect or omitted logos. According to the printers, approximately 80
percent of the ballots on the original data base provided by the CEP
contained one type of error or another. Eventually, the CEP set up an office
on the printers’ premises to review the ballots against the original
documentation and make the necessary corrections. The ballots did not have
consecutive numbering, and thus it was difficult to control their exact
distribution.

C. Payment of Registration Workers and Registration Sites

A recurring problem during the registration process was the CEP’s

failure to pay the workers on time. As noted in the registration section of
this report, many disgruntled workers seized registers, demanding payment.
 Other Electoral Preparations 37

Eventually, the CEP received the names of the workers from the BEDs and
passed on the information to the Ministry of Finance. Moreover, the CEP
was slow in paying rent for private homes used both as BIs and BVs.
Private homes comprised an estimated 20 percent of all BVs.

D. The Lists of Bureaux de Vote and Bureaux de Vote Workers

Upon the completion of voter registration, the BECs and BEDs

submitted the addresses of the polling sites to the CEP. According to the
CEP, some 11,238 polling sites were entered on a computer data base.
Additionally, the names of the poll workers (three members and one
electoral security agent per polling station) were also computerized. The
BECs posted the lists of the bureaux de vote and their corresponding
workers in order for voters to locate exactly where they should vote.

For the first time in Haiti, the Electoral Law required the political
parties to submit names of candidates to run the bureaux de vote. Some
political parties were slow in submitting the names (and some never
submitted them), which caused a delay in inputting this information on the
computerized data base. Some political parties alleged that their poll worker
candidates had been excluded from the BVs.

E. Training of Bureaux de Vote Workers

IFES technicians had recommended the design of the forms that were
to be used to open and close the polling tables and to tally the votes. The
CEP approved these forms and incorporated them into the manual of
procedures to be used in the training of polling officials. IFES proceeded to
print the forms and the manuals at an estimated cost of US$30,000. On
April 17, the CEP, with the assistance of IFES, began its training of BEDs,
BECs, and supervisors (departmental trainers). After training had been
completed in three departments, the CEP operations manager recommended
changes in the forms.

The lack of consultation between the CEP and IFES about the utility
of the different forms resulted in confusion and caused the suspension of the
training program for one week. Eventually, the CEP opted to use modified
forms and committed to retraining the officials in the departments where the
original forms had been utilized. IFES continued to provide infrastructure
38 Other Electoral Preparations

support to the training session, but it did not provide assistance in the
interpretation or use of the forms themselves.

The OAS Mission did not review the forms and did not judge their
utility. It appeared, however, that the timing of the changes was unfortunate,
creating delays and wasting precious resources.

F. Security and Election Materials

With regard to the election materials, ample security measures

appeared to have been taken to ensure that unauthorized persons did not
obtain the materials. Where it was able, the Haitian National Police
provided security for electoral materials in Port-au-Prince and in the
departments. They also provided adequate security in the transport of
materials.

The Mission maintained a continuous presence at the printers’ and the

CEP’s warehouses where the ballots were stored. It can attest that adequate
security was present at both of these sites.
 CHAPTER VI
THE SECURITY SITUATION
A. Overview

While violence escalated in the days leading up to the elections, the

overall environment in the departments was that of calm, accompanied by a
fear that the situation could deteriorate at any moment. In several
communes, protests ignited violence that led to loss of life and property
damage. The violent acts underscored the fragility of the security situation
and demonstrated the real possibility that those acts could repeat themselves
on a nationwide scale.

The political conflict in Anse-d’Hainault manifested why the political

environment in the Department of Grand’Anse was particularly precarious.
In addition to the crisis in Anse-d’Hainault, that department suffered attacks
on half of the BECs, a fire in the house of the BED president, and the forced
closure of the BED and the BEC in Jérémie by groups of armed men. In the
Nippes area of the Grand’Anse, the BECs of Miragoane and Petit Trou de
Nippes were set on fire by unidentified assailants two days before election
day, and all registers were destroyed.

In the Department of the South, many political parties questioned the

objectivity of the BEC officials in Roche à Bateau, Côteaux, and Port-á-
Piment. In the latter commune, the protests of the parties eventually led to
demonstrations, which ended in fighting among Fanmi Lavalas,
MOCHRENA and Organisation du Peuple en Lutte (OPL) supporters.
While the authorities eventually reported a normalization of the situation in
these communes, the climate of uncertainty led to fear of a possible repeat of
these events.

In the Department of Artibonite, the communes of Marchand

Dessalines and Saint Marc required special attention. The OAS Mission
42 The Security Situation

received reports of several violent incidents that appeared to be related to the

electoral process and could have had repercussions during election day.

In the department of the North (in the communes of Borgne, Grande

Rivière du Nord, Trou de Nord, and Pignon) and in the Department of the
Northeast (in the communes of Monbin-Crochu, Caracol, and Trou du
Nord), the OAS Mission observed with concern various incidents of
intimidation and aggression by political bosses who appeared to act with
impunity.

In the commune of Maïssade in the Central Department, political

party representatives demanded the removal of the president of the BEC for
alleged irregularities in the registration. After being closed temporarily, the
BEC reopened with the vicepresident and the secretary assuming the daily
responsibilities of the office. Several political parties, however, continued to
threaten to impede elections from taking place until the president was
formally removed from office.

B. Political Parties

On several occasions, through its press releases and in individual

interviews, the Mission called for the use of moderate language so as not to
incite violent confrontation. Nevertheless, some party leaders utilized
aggressive language that contributed to the deterioration of the electoral
environment. The Mission was especially concerned by the declarations
made by one of the political leaders in Port-au-Prince on March 29, in which
the people were incited to defend themselves by running aggressors over and
creating “brigades de vigilance.”

While not to minimize the deteriorating security situation, the Mission

also noted that some political parties and the media exaggerated events and
even distorted facts surrounding certain incidents to dramatize the security
problems. Moreover, some political parties and the media attempted to
portray incidents of common crime as politically motivated and related to
the electoral process in the country.
 The Security Situation 43

C. Incidents

Most of the election-related violence occurred in Port-au-Prince and

its surroundings. Groups known as “popular organizations” appeared to be
the perpetrators of much of the unrest, especially during the violent street
protests in Port-au-Prince.

Other violent incidents included attacks perpetrated against the

candidates to the Senate for the West Department, Mirlande Manigat
(RDNP) and Marie Laurence Lassegue (PLB), both of whom were
interviewed by the OAS Mission. In the first case, a truck with men armed
with machetes blocked the way of the Manigat vehicle when she was
traveling in the section of Palmes in Petit-Goâve. In the second case, about
seven armed men surrounded and pounded on the vehicle of Ms. Lassegue
when she was leaving her party headquarters.

On April 9, Jean Dominique, a renowned Haitian journalist and close

friend of the President, was murdered just in front of his office. Following
the funeral of Dominique, violent protesters from the popular organizations
burned the headquarters of Espace de Concertation.

Other journalists, too, were subject to threats and violence. Through

its press releases of March 28 and 30, and April 3, 4, 9, and 24, the OAS
Mission expressed its concern about those incidences of violence and called
upon the civil and judicial authorities to reestablish order and to bring those
responsible to justice.

Since its arrival, the OAS Mission catalogued different reports of

violent incidents during the electoral process. The incidents were reported
to the Mission through its observers, political party representatives,
concerned citizens, and press releases. In most cases, the Mission
interviewed the victims or their families and inquired about measures taken
by the appropriate authorities. In the majority of the cases, no significant
progress had been made and there were few arrests. By election day, the
Mission recorded seventy incidents of violence that had led to seven deaths
of candidates or political activists.

The most direct consequence of the violence or threat of violence was

the curtailment and the eventual suspension of the campaigns of several
44 The Security Situation

candidates. On several occasions, the preelectoral violence also interrupted

preparations undertaken by the electoral authorities.

On May 2, the Prime Minister assured the public that the Haitian
National Police had completed the preparations and training necessary to
guarantee security on election day. He further called on dialogue among the
government, civil society, and the CEP to coordinate efforts to prevent
violent interruptions in the voting process and the subsequent vote count.
 CHAPTER VII
THE MEDIA AND THE ELECTORAL CAMPAIGN
A. Civic Education and the Media

In the absence of an official civic education campaign carried out at

the local level, the population largely relied on the media to be informed
about the elections and candidates’ campaigns. Haitian observers also
pointed out that the days of mass political rallies in which candidates sought
to develop personal contact with the electorate had gone, and candidates
now relied on the media to diffuse their message. Additionally, security
concerns and a lack of sufficient financing for a campaign period prolonged
by repeated postponements of elections contributed to candidates’ reliance
on the media. This reliance is a heavy burden for the Haitian media, which
operate under a series of infrastructural and operational constraints. Those
include the centralized political and economic structures of the country, with
a news focus on the capital of Port-au-Prince, as well as the standards
created by poorly paid, trained, and equipped provincial correspondents and
the country’s poor communications infrastructure.

Journalists nationwide complained of the lack of access to sources of

objective information. This problem was particularly in evidence during the
electoral campaign. Although the CEP had one member who was officially
responsible for providing information to the press, Haitian journalists
complained that, in practice, it was often impossible to obtain clarification of
news.

B. Campaign Coverage

While most political parties had some complaints about the media,
especially the state-run media, they concurred that there has been an overall
improvement in freedom of the press since the election of President René
Préval and that the media were able to cover opposition parties’ campaigns.
48 The Media and the Electoral Campaign

Under the July 1999 Electoral Law (Article 112), state-run media
were obliged to provide equitable coverage of the electoral campaign.
Throughout the electoral campaign, the state-owned Télévision Nationale
d’Haïti (TNH) and Radio Nationale d’Haïti both offered one hour of air time
to each of the thirty-three parties participating in the elections. Parties,
however, complained that this time was insufficient, especially given the
protracted length of the electoral campaign.

Parties also objected that the state media offered viewing time in the
middle of the day to some of the opposition parties, rather than during peak
hours. Some opposition parties refused to appear on TNH panel discussions
or debates because they considered the hosts biased.

Throughout the campaign, the OAS Mission observed that the Fanmi
Lavalas Party received the most coverage on state television News programs
generally gave more coverage to items involving Fanmi Lavalas, and
advertising on state television and radio was dominated by the
advertisements for this party and those of popular organizations with known
affiliations to Lavalas, such as KOZEPEP.

The length of the electoral campaign negatively affected candidates’

advertising, on both state and private media. Parties complained they could
not continue to finance expensive national advertising, with smaller parties
being the most seriously affected. The U.S. Association for Rural
Development provided a total of US$500,000 in advertising and assistance
to political parties for campaign advertising expenses, but the parties deemed
the amount insufficient. The National Democratic Institute for International
Affairs, a Washington, D.C.-based nongovernmental organization, also
provided “electoral information centers” throughout the country as a
resource for political parties.

C. Coverage of the Campaign by Privately Owned Media

Private media contributed positively with civic education campaigns

during the pre-electoral period, a contribution that became all the more
important when the CEP was unable to conduct its own education campaign.
Almost all radio and television stations organized special programs to cover
the electoral campaign, including debates, panel discussions, and increased
news coverage. This programming was offered free of charge to the
 The Media and the Electoral Campaign 49

political parties. Of note were efforts by the business sector’s Groupe

Croissance, which organized a weekly prime-time seventy-five-minute
program, "L’Heure des Choix," which invited candidates and parties to be
questioned about their electoral platform. "L’Heure des Choix" was shown
on the private television station Télémax and broadcast simultaneously on
fifteen Port-au-Prince private radio stations, with relays to the provinces,
reaching an estimated one million people.

As the lengthy campaign continued, the privately owned media

complained that audience interest was declining in election-related
broadcasts and that it was becoming more and more costly for them to
continue this “patriotic effort.” Télémax administrator Pierre Richard
Desmornes told the OAS Mission on April 12 that by covering the
campaigns of all the candidates and all the major electoral issues, the private
station was losing potential advertising revenues. Advertisers, Desmornes
said, were less inclined to place commercial spots around those broadcasts,
preferring instead entertainment programs.

D. Freedom of Expression in the Electoral Campaign

Since its arrival in late February, the OAS Mission monitored the state
of freedom of the press through interviews with Haitian journalists, human
rights groups, and political parties. The political instability and uncertainty
surrounding the electoral process negatively affected the freedom of the
press and illustrated the precariousness of freedom of expression in Haiti.
Some Haitian journalists commented that the increasing polarization and
intolerance observed over the past year, and especially during the electoral
period, resulted in decisions to suspend critical analysis and commentary.

The most publicized violation of press freedom in Haiti during this

period was the murder of the director of Radio Haïti-Inter, Jean Dominique,
well known for his progovernment positions. Dominique’s style of news
analysis was unique to Haitian journalism, and he had expressed strong
opinions on the electoral process on his daily programs. The journalist was
also well known for his passionate advocacy of the government’s land-
reform program and campaign against impunity in relation to human rights
crimes. Dominique and the radio station’s janitor, Jean-Claude Louissaint,
were shot dead on April 3 in the radio station’s courtyard by an unidentified
gunman who was waiting for Dominique to arrive to broadcast his 7 A.M.
daily news program.
50 The Media and the Electoral Campaign

The daylight murder of the renowned journalist had a dramatic effect

on public confidence in the government’s capacity to guarantee security
during the electoral period. The killing was also largely interpreted as a
chilling message to journalists covering political issues, including the
electoral process. As a result of Dominique’s assassination, the station
closed for a month. When Radio Haïti-Inter resumed broadcasting in early
May, the station dropped its daily editorial and decided to forego in-depth
reporting on a variety of election-related issues.

The OAS Mission issued a press release on April 3 condemning the

violence and urging the government to bring the perpetrators of this crime to
justice. At least one person was arrested for questioning in the murder. The
suspect died while being treated in a private hospital after being taken into
police custody.

A spate of threats against the radio station and against newsroom

director Daly Valet and Leontès Dorzilmé, host of the phone-in show
“Vision 2000 à l’Écoute” resulted in the two journalists going into hiding.
The radio station stopped airing news for five days and suspended the
program "Vision 2000 à l’Écoute" indefinitely. After the journalists of
“Vision 2000” wrote an open letter to the Minister of Justice denouncing the
threats, police sent a contingent of the CIMO (riot squad) police to stand
guard outside the radio station, on April 7. On April 8, the radio was visited
by a mob who had just burnt down the headquarters of the Espace de
Concertation coalition. The mob threw stones and threatened to burn down
the building. They eventually left without doing any serious damage. As
the result of continuing threats, newsroom director Dalet Valy left Haiti
during the month of May. Reporter Leontès Dorzilmé resumed working for
"Vision 2000" in June, after the first-round elections.

During this period, Espace de Concertation leader Evans Paul called

for violence against those holding up vehicles at barricades (chimères). He
recommended that drivers held up by the chimères should run their vehicles
straight over them. This interview was aired by "Vision 2000" as well as
several other main radio stations. It is worth noting that this call to violence
was made in a prerecorded taped interview, and could have been cut out by
editors without altering the message of the interview. In the context of
mounting violence, this might have been an appropriate measure to take.
 The Media and the Electoral Campaign 51

News presenter Gary Pierre-Paul Charles also received threats after

his coverage of the March 26-28 demonstrations. According to staff at the
private Television channel, Charles’s coverage of the barricades was
analytical and critical. The Television presenter had called for respect of
lives and private property. On April 3, after the murder of Haïti-Inter
director Jean Dominique was announced, Charles received two telephone
calls in which an anonymous caller warned him that he would be next on the
list. Charles decided to stop presenting the news for a period. He returned
to work on April 15.

Petit-Goâve Radio Echo 2000 stopped newscasts after an incident on

March 22 in which OPL candidate for deputy Sergo Erly Philistin was
threatened as he spoke on a live electoral coverage program. A vehicle
drove by the radio station several times as Philistin spoke, calling out threats
that they were waiting for the OPL candidate to come out so they could kill
him. Philistin returned to his home in Grand-Goâve under police escort.
After the incident, staff at the radio station reported receiving a number of
telephoned threats. Newsroom director Brignolle Lindor and journalist
Jean-Pierre Louis were specifically threatened. The radio station decided to
suspend newscasts.

During the night of April 3, unidentified individuals broke into the

community's Radio Unité and stole all of the station's equipment. Many
believe this incident was an attempt by some members of the opposition
parties to silence a radio station whose director was the Fanmi Lavalas
mayoral candidate. Director Etienne Oles told the OAS Mission that he had
received threats from other candidates and their supporters, who had said
that the pro-Fanmi Lavalas radio station must close. Throughout the
campaign, Oles said, the radio station had only aired advertisements from
the Fanmi Lavalas candidates and those of the Parti Louvri Baryè.

In Gonaïves, an anonymous leaflet, in Creole, was posted during the

week beginning April 24 under the doors of several radio stations, referring
to the fate of Jean Dominique and the threats against "Vision 2000"
presenters Daly Valet and Leontès Dorzilmé.
 CHAPTER VIII
SUMMARY OF OBSERVATION
A. Recompilation of Data Collected on Election Day

On election day, observers were provided with information forms to

fill out, which were intended to cover many organizational aspects of the
conduct of elections. They obtained this information by observing and
interviewing poll workers at specific BVs. Usually, the observers
interviewed poll workers from only one BVs in each polling site. Thus, the
proceedings at many other BVs were also observed, but not recorded.

The following global summary includes data from the nine

departments, from observations carried out during both the May 21 elections
in eight departments and the June 11 elections in the Department of the
Grande’Anse.

OAS observers conducted interviews at 1,109 BVs, which represent

approximately 10 percent of all the BVs in the country. Of these, 96 percent
were functioning at their officially designated locations. Only 34 percent
opened on time, although most opened less than an hour late. Delays were
generally due to the time taken to prepare the BV and count ballots,
sometimes due to problems in receiving materials - 8 percent of BVs
observed were missing materials. In 97 percent of BVs observed, all three
authorized BV members were present, and in 83 percent of BVs no
unauthorized people were present.

Political party representatives (mandataires) were present in almost

all BVs observed, with FL participating more than any other political party.
FL representatives were present in 638 BVs observed, Espace de
Concertation representatives in 384, OPL in 372, MOCHRENA in 359,
ESKANP in 152, PLB in 150, RDNP in 125, Tet Ansanm in 124. Other
parties were represented in fewer than 100 BVs. A mixture of independents
56 Summary of Observation

were represented in 390 BVs. National observers were present in 85 percent

of the BVs observed.

In terms of security, national police were present in slightly more than

50 percent of BVs observed. Those BVs tended to be in the towns, and
police were rarely seen in rural BVs. Electoral security agents were present
in 96 percent of BVs. Security problems affecting the polling were observed
in 30 BVs (4.5 percent) and reported but not observed in forty-five BVs.
Armed individuals were seen in twenty-nine BVs (3.5 percent). Lines of
voters were generally well controlled (84 percent), although in 11 percent of
BVs, observers stated that voters had problems accessing the BV.

In 91 percent of BVs observed, voting procedures were followed,

although in 38 percent of BVs, ballot boxes were not correctly sealed. The
failure to seal the ballot boxes appeared to be due to a lack of training and
not deliberate. The BV president gave impartial instructions to the voters in
79 percent of the polling stations observed. In 23 percent, party
representatives also gave instructions to the voters.

Irregularities were neither observed nor reported in 37 percent of BVs.

Normal operations but minor irregularities, which were not considered to
affect the integrity of the process, were observed or reported in 39 percent of
BVs. Normal operations but many irregularities, still not considered to
affect the integrity of the process, were observed or reported in 15 percent of
BVs. Major irregularities which may have affected the integrity of the
process were reported in thirty-three BVs (3.4 percent) and observed in fifty-
three BVs (5.4 percent).

Other irregularities included acts of intimidation (observed in 3.8

percent of the BVs), illegal political activities (observed in 5.1 percent of the
BVs), and the presence of people trying to disturb the polling (observed in
2.3 percent of the BVs).
 Summary of Observation 57

Compilation of Election Day Results

(Sample of 1,109 interviews)

OBSERVATION OF BUREAUX DE VOTE (BV) % 4/ NUMBER

OF BVs
At their original location 96% 1,052
Opened on time 34% 352
Received all voting materials 92% 808
All members were present 97% 1,078
No unauthorized people were present 83% 895
National observers were present 85% 820
Haitian National Police were present 50% 481
An electoral security agent was present 96% 920
Security problems affecting the vote were observed 4.5% 30
Armed individuals were seen 3.5% 29
Lines of voters were well controlled 84% 873
Voters had access problems 11% 108
Voting procedures were followed 91% 955
Correctly sealed ballot boxes 62% 655
BV president gave impartial instructions to the voters 79% 739
Party representatives gave instructions to the voters 23% 203
Acts of intimidation were observed 3.8% 23
Acts of intimidation were reported but not observed 5.6% 34
Illegal political activities were observed 5.1% 35
Illegal political activities were reported but not observed 18% 123
Presence of people trying to disrupt the polling was observed 2.3% 14
Presence of people trying to disrupt the polling was reported 3% 18
Irregularities were neither observed not reported 37% 358
Operations were normal, but there were minor irregularities 39% 374
Operations were normal, but there were many irregularities not affecting 15% 148
the integrity of the process
Major irregularities that could affect the integrity of the process were 3.4% 33
reported
Major irregularities that could affect the integrity of the process were 5.4% 53
observed

4
Percentages given relate to the number of responses provided to each question rather than to the number
of BVs observed. In some interviews, observers did not obtain answers to all of the questions on the form.
58 Summary of Observation

B. Regional Incident Reports from May 21, 2000

1. West

BVs visited: 369 (10 communes out of 18)

Although the voting took place in an atmosphere of calm and

enthusiasm marked by high voter turnout, overall electoral operations were
poorly managed. The majority of polling stations in the metropolitan area of
Port-au-Prince opened late while, in provincial towns they fared somewhat
better, with approximately 50 percent opening on time. In addition, the
tardiness or absence of poll workers added to delays. According to BED
officials, this was due to last-minute changes in the list of poll workers on or
just before May 21. Thus, the BED estimated that 30 percent of poll
workers present had not received any training at all. Another common
irregularity noted was the violation of the secrecy of the vote, mostly due to
attempts by various party poll watchers to influence voters’ choices. In most
of the polling stations in the capital, officials permitted voting after 5:00
P.M., even if they were not informed of the CEP’s decision to extend voting
hours until 6:00 P.M.

The Haitian National Police (HNP) played a fundamental role during

election day in this department. The police presence insured that people
could vote without fear of violence. Despite the limited number of police
available for duty (estimated at three thousand five hundred nationwide),
observation teams saw police officers in 50 percent of the BVs visited. In
Port-au-Prince, the HNP also provided most of the urban BVs with transport
to their respective BECs or, in the case of Pétionville, the police station.

The police transported election officials and materials so quickly that

the BED officials could not handle the sudden influx. Long lines of polling
station officers waiting to turn in their results quickly grew. Unfortunately,
those lines gave way to frustration and the eventual abandonment of
materials in the streets of the capital. According to the BED, only 30 BVs,
out of a total of 1,220, were not accounted for, but the OAS observers
questioned the accuracy of this number. Furthermore, the provision of
transport by the HNP compromised the transparency of the process, as party
representatives were not informed of the pickup plan, and poll watchers
were not allowed to accompany BV workers on their ride to the BEC. In
many cases, BV officers had not completed the count and made the final
 Summary of Observation 59

tallying of the results upon arrival at their destination in the absence of poll
watchers.

One violent incident marred a generally peaceful election day. At

approximately 1:00 p.m., a group of armed men attacked five polling
stations in the Bois Neuf neighborhood of Cité Soleil, commune of Delmas.
They first entered into three BVs located in Zouk Disco and demanded to
vote without electoral cards. According to poll workers, they were refused
until they brandished guns. In the ensuing chaos, the assailants seized a
register and marked and stuffed ballots in one or more ballot boxes. They
then moved to a nearby private home, which housed two BVs, and repeated
the operation. The five BVs affected remained closed from 2:00 P.M. to 5:00
P.M., when poll workers moved to the police station to carry out the count.

Following the election, a candidate to town delegate for the

Rassemblement Citoyen Patriotique was killed in suspect circumstances,
while several other candidates from the opposition were arrested. In Petit-
Goâve, Jean Limogène, candidate to deputy for Espace de Concertation, was
jailed following a day of violent protests during which he was accused of
disturbing the peace. According to the police, he had led a group of Espace
supporters in barricading the streets, throwing rocks at the police, and
shooting at the police from a radio station he owns. He was arrested with a
warrant, incarcerated first at the Carrefour police station, then detained on a
judicial order in the National Penitentiary. Espace claimed the arrest was
unprovoked.

In Thomazeau, Moise Jean, a PLB mayoral candidate, was arrested

after two rooms of the BEC were set on fire. The police accused him of
having threatened to burn the BEC on May 22 during protests organized by
PLB supporters.

The PADEMH candidate for mayor of Pétionville visited the OAS

Electoral Mission to denounce grave irregularities in that race. The
candidate stated that the number of votes she obtained were altered in order
to favor the Lavalas candidate. She presented as proof, copies of a number
of procès-verbaux (PV) (election records) and tally sheets for compilation.
The Mission verified that according to those documents, there were
inconsistencies between the numbers recorded in the PV and the data
registered in the tally sheets for compilation. The Mission also confirmed
that the differences in such documents reduced the number of votes obtained
60 Summary of Observation

by this candidate. The Mission visited the competent electoral authorities to

verify the information received, but the observers were denied access to the
necessary documents.

2. Artibonite

BVs visited: 106 (7 out of 15 communes)

The OAS regional office confirmed a number of violent incidents

reported after election day, some of which may have affected local results.
In addition, technical problems, such as insufficient ballots, could have also
affected races in two communes.

In Marchand Dessalines, elections for town delegate took place even

though the emblem of one of the cartels was missing on the ballots, while
elections for CASEC in this commune’s third section were not held due to
the omission of the OPL logo on ballots.5/ In addition, ten polling stations
located in the commune of Saint-Michel de l’Attalaye, out of a total of 197
for the electoral jurisdiction (or approximately 5 percent of the polling
stations), did not receive ballots for the deputy race.

In Verrettes, BEC officials confirmed that unidentified armed men

stopped the vote at approximately 9:00 A.M. and stole and burned the
electoral materials for nine out of the twelve polling stations in the sixth
section. (Local elections in this section were to be reheld during the second
round.)

In the first section of Grande Saline, the BEC reported that Fanmi
Lavalas supporters marked ballots and stuffed ballot boxes in two polling
stations. However, the results of only one of those polling stations was
annulled. Likewise, the Mission confirmed that in Petite Rivière
unidentified individuals removed the ballot boxes from one of the twenty
polling stations in the second section and returned with them already filled.
Polling station workers submitted a report to the BED, requesting the
cancellation of the results. The Mission was not informed of the BED’s
decision.

5
CASEC elections were scheduled to be reheld in this section during the second round.
 Summary of Observation 61

Observers also confirmed that one violent incident took place during
the vote count. In two polling stations located in the first section of
L’Estère, armed Fanmi Lavalas supporters interrupted the count and forced
poll watchers to leave. Opposition parties further alleged that the assailants
marked unused ballots for Fanmi Lavalas candidates and stuffed ballot
boxes in the presence of the BEC president. The police, dispatched to the
site following complaints made by poll watchers, seized all the electoral
material. The next day the BEC president was beaten, supposedly due to his
intention to annul the results for both polling stations, by an individual he
identified as affiliated with KOZEPEP.

Reception of voting materials was disorderly and without a thorough

control in six of the seven BECs visited between election night and May 22.
BEC officials in Saint-Marc initially reported to observers that they had
misplaced procès-verbaux for 11 polling stations. However, at a later date,
the BEC president told observers that the procès-verbaux for all of those
polling stations had been found. The Mission’s regional team was not able
to verify their recovery. Observers also witnessed polling station workers
filling out additional copies of procès-verbaux at the BECs, allegedly from
the original procès-verbaux or the tally sheets, often at the request of BEC
members.

On May 24, for security reasons, the BED instructed all BECs to
move compilation proceedings, which had already began in most communes,
to the main police station in Gonaïves. At least seven BECs (L’Estère,
Ennery, Grande Saline, Petite Rivière de l’Artibonite, Saint-Michel de
l’Attalaye, Marchand Dessalines, and La Chapelle) finished their work there
in the presence of few party representatives. During the compilation,
observers noticed a variety of irregularities in the procès-verbaux:
inconsistencies between the totals of ballots received and the totals of used
and unused ballots, omission of results for some posts, and no signatures or
several signatures in the same handwriting.

On May 26, the BED completed the compilation for senate and deputy
races. Since BED officials did not announce in advance the date and the
venue for the final compilation of those races, party representatives were not
present.

The regional office staff was unable to obtain official information on

missing procès-verbaux for any polling stations, electoral incidents, nor the
62 Summary of Observation

electoral process in general, following a decision by the BED secretary to no

longer collaborate with the Mission.

3. Center

BVs visited: 50 (9 communes out of 12)

Observers reported violent incidents in the rural sections of the

communes of Hinche, Savanette, and Thomonde, which took place during
the vote and after the closing of the polls.

As for technical errors, elections for CASEC in the first section of

Cerca-la-Source were not held due to the omission of the OPL cartel on the
ballots, while ASEC elections did not take place in Lascahobas due to
errors on the ballots.

At least twelve polling stations in the commune of Hinche, including

six in the second section of Marmont, three in the first section of Juanaria,
and three in the fourth section of Aguahedionde, were annulled due to the
interruption of the vote by armed groups, who forced BV workers to
abandon their stations and stuffed ballot boxes in favor of Fanmi Lavalas
candidates. The cancellation of six out of the fifteen polling stations (or 40
percent of the total) in Marmont called into question the validity of local
elections in this section. According to eyewitnesses interviewed by
observers in the field, the perpetrators included several Fanmi Lavalas
candidates and police officers. BEC officials later told observers that the
results for all those polling stations were annulled.

BEC officials in Thomonde reported to departmental electoral

authorities that armed individuals entered two polling stations in the first
section, forced party poll watchers to leave, and overtly influenced voters. It
is not clear if ballot-stuffing occurred during those attacks, if the assailants
tried to coerce people to vote for a specific party, or if they simply attempted
to discourage voting. At least one of the polling stations in that section was
not able to complete the count on site. The BEC in this commune later
moved compilation operations to the main police station.

Two polling stations in the second section of the commune of

Savanette were annulled after armed individuals attacked the stations during
the vote, forcing poll workers to flee and abandon voting materials.
 Summary of Observation 63

In Cerca-la-Source, the president, secretary, and security agent of

one polling station alleged that the vicepresident and a party poll watcher put
ballots marked in favor of MOCHRENA into ballot boxes, leading them to
request that the results of the polling station be annulled.

On the evening of Tuesday, May 23, an armed group attacked the

Boucan Carré BEC, expelled the president and vicepresident from the
premises, and destroyed an unknown but significant number of procès-
verbaux. According to the official report filed by the first justice of the
peace who arrived on the scene on the morning of May 24, empty ballot
boxes, scattered and ripped-up ballots, and procès-verbaux were found
strewn throughout the BEC offices. This report also contained a statement
by the BEC president that the armed attackers had first pressured him to alter
vote results and then began ripping up procès-verbaux. However, the report
filed by a second justice of the peace, who visited the BEC in the afternoon
of the same day accompanied by police officers, indicated that plastic bags
containing procès-verbaux for an unspecified number of polling stations
were found in the BEC offices. This report also contains a statement by the
Mirebalais police commissioner attesting to the recovery of the procès-
verbaux for all fifty-five polling stations in this commune. Although the
president and vice-president of the BEC doubted the authenticity of the
procès-verbaux, a CEP member visiting the BED in the department of the
Center decided that the procès-verbaux should be considered legitimate.

In the commune of Maïssade, tensions ran high in the main township

from the start of the vote. Well before election day, political parties accused
Espace de Concertation of fraud involving electoral officials linked to that
party. Those allegations led to the arrest of the Maïssade BEC president on
charges of falsifying voter registration cards.

The security situation at the Maïssade BEC quickly deteriorated after

the closing of the polls. In the case of one polling station, workers seeking
increased safety moved vote-count operations to the BEC but were unable to
proceed because of harassment by representatives of several political parties.
On the morning of May 22, a group of armed individuals attacked poll
workers transporting voting materials from a station located in the third
section to the BEC, stealing all the ballot boxes. The same morning, while
workers for some polling stations continued the vote count at the BEC,
demonstrators threw stones and threatened to burn down the BEC. As a
64 Summary of Observation

result, BEC members fled the premises and poll workers, fearing for their
security, abandoned voting materials in the courtyard. Members of an
international observer delegation from Canada submitted a report to the
BED in which they stated that during the afternoon of May 22 they found
several hundred ballots in the streets of Maïssade and witnessed the burning
of ballot boxes behind a private house, as well as the arrests of Espace,
Fanmi Lavalas, and OPL candidates. Demonstrations and threats of
violence continued on May 23, leading to the decision to transfer under
police escort all voting materials remaining at the BEC, whether counted or
not, to the BED in Hinche.

According to information provided by BEC officials to the Mission,

by May 22 the Maïssade BEC had only recorded delivery of materials for
thirty polling stations, out of a total of sixty-six, where votes had been fully
counted for all races and procès-verbaux completed. Materials for most or
all of the other thirty-six polling stations had apparently been delivered to
the BEC by this time but had not yet been completely counted nor officially
recorded. It is not known how many of those materials, if any, may have
been destroyed or lost in the confusion of the events of May 22 and 23.
After the transfer of remaining materials to the BED, the Mission could not
obtain access to any procès-verbaux for this commune, despite claims by the
BED vice president that materials for all sixty-six polling stations had been
recovered and subsequently counted. Results posted later at the Maïssade
BEC were allegedly torn down by residents and could not be verified either.

At the BED, the operations of compilation and transmission of results

were performed in the absence of the president, who abandoned his post on
May 23. During an interview in Port-au-Prince with members of the OAS
Mission’s central office, he informed the Mission that he feared for his life
and thus had gone into hiding.

The regional observer team was not able to corroborate the exact
number of missing procès-verbaux at the level of the BED. However, the
authenticity of results for ninety-nine polling stations (fifty-five from
Boucan-Carré and at least thirty-six from Maïssade), out of a total of 754 in
the department, is at the very least questionable.
 Summary of Observation 65

4. North

BVs visited: 58 (7 communes out of 19)

Voting in Limbé took place in an atmosphere of intimidation.

Observers witnessed campaign activity by pro-Fanmi Lavalas groups in this
town and blatant attempts by Fanmi Lavalas representatives to influence the
vote. While observers noted the presence of up to seven accredited Fanmi
Lavalas poll watchers in several polling stations, other parties reported to the
Mission that their poll watchers had been expelled. In one case, observers
themselves were not allowed to enter a polling station.

Observers confirmed that several violent incidents occurred during

and after the vote count in areas throughout the department. In the
commune of Bahon, the ballot boxes for seventeen polling stations out of a
total of thirty-five for the commune were burned by unidentified armed
groups which forcibly entered the BEC on May 22 and, 24. In the case of
four polling stations, procès-verbaux were also destroyed, meaning that all
results were lost for those polling stations. Although the BED had initially
decided to re-hold elections in this commune, its members reported in mid-
June that copies of all procès-verbaux had been gathered from party
representatives and thus, legitimate results could be compiled for the entire
commune. In early June, a recompilation of results, based on recovered
procès-verbaux, was done at the BED in the presence of candidates, party
representatives and the press. Results showed OPL winning by a solid
margin in deputy, municipal and local elections. Nevertheless, the CEP
ordered BED to rehold elections in this commune on July 2. (See Chapter
IX, Complementary Elections.)

In Bas Limbé, six armed men interrupted the count in one polling
station at approximately 9:00 P.M., marked an unknown number of ballots
for OPL and an independent candidate, and put the ballots into ballot boxes.
It was alleged that an OPL CASEC candidate was among the perpetrators.
The results of this polling station were later canceled by the electoral
authorities.

A group of armed individuals entered a polling station in the third

section of Grande Rivière and stuffed the ballot box for the deputy race
with unmarked ballots during the count, while in the town of Dondon
unidentified armed men interrupted the count at one polling station and stole
66 Summary of Observation

the ballot boxes. Also in Dondon, a polling station moved vote-count

operations to the police station after polling station workers heard shots in
the surrounding area.

In Grande Ravine, the fourth section of Acul-du-Nord, the number of

votes recorded in the first, second, and third polling stations was higher than
the total of registered voters. As a result, the BEC decided to annul the
results for all three polling stations. In this section, observers also confirmed
that an armed group stole ballot boxes from one polling station. The
cancellation of results for four out of a total of thirteen polling stations (or
31 percent) for the section may have had an impact on the final results for
ASEC and CASEC races. Also, BED officials informed the Mission that in
the commune of Acul-du-Nord, numerous individuals had stuffed the ballot
boxes of two polling stations in the section of Buisson, which were later
annulled by the BEC.

In the commune of Le Borgne, the BEC cancelled seven out of a total

of eighty-five polling stations (or 8.2 percent) due to a variety of
irregularities, including the fact that ballots were found that had been filled
out with markers rather than the pens provided by the CEP. The suspect
ballots were discovered during a recount at the BEC.

5. Northeast

BVs visited: 53 (4 communes out 13)

In the commune of Mombin Crochu, poll workers refused to open

polling stations on the morning of the vote after learning that the president of
the BEC had resigned and, fearing arrest, had gone into hiding the night
before.6/ After the police persuaded poll workers to begin operations,
stations in this commune began functioning normally at 10:00 A.M.

In the commune of Terrier Rouge, one violent incident took place

during election day. BEC officials confirmed that unidentified armed

6
On May 20, the Commission on Security (composed of the police director, the commissaire du
gouvernement, and the president of the BED) had ordered the arrest of the BEC president of Mombin
Crochu for his alleged involvement in a plot to burn down the BEC. In addition, the commission had
confiscated ten voter registers for Mombin Crochu considered to be fraudulent, an allegation that the
regional EOM team was not able to verify.
 Summary of Observation 67

individuals entered two polling stations located in the main township and
destroyed ballots. Given the contradictory reports of this incident provided
by the justice of the peace, BEC officials, and party representatives,
observers were not able to verify whether the incident occurred before or
after the vote count, and if the procès-verbaux for these two polling stations
were destroyed during these attacks. However, during the compilation of
results on May 23, BEC officials were able to produce one of the procès-
verbaux for one of those polling stations. Fanmi Lavalas representatives
challenged its authenticity as well as that of one of the procès-verbaux for a
third polling station in the township, which was incomplete and unsigned by
poll workers. Fanmi Lavalas demanded that results for all three of those
polling stations out of a total of fifteen located in the town of Terrier Rouge
(or 20 percent) be annulled. The Mission could not confirm the decision
taken by the electoral authorities. Cancellation of those three polling
stations could have altered results for municipal elections in the town of
Terrier Rouge.

On May 23, the BEC of Mont Organisé, along with the Tribunal de
Paix and half of the market, was burned down. Procès-verbaux for only six
out of the twenty-six polling stations in this commune were salvaged by
BEC officials. Fearing for their safety, BEC officials then fled to Fort
Liberté, where they were arrested on the orders of the departmental director
of the police. The departmental director told observers that he found it
suspicious that BEC members were unable to save all the procès-verbaux.
Although the BED initially asked the CEP to decide whether elections here
should be annulled, the OAS Mission’s regional office was informed in mid-
June that copies for sixteen out of the twenty destroyed procès-verbaux had
been recovered from party poll watchers and thus, new elections were not
necessary. Observers were able to consult the recovered procès-verbaux,
but could not confirm their authenticity. The BED president objected to the
legality of using the recovered procès-verbaux to compile results for this
commune. The Mission believes that the results obtained for at least sixteen
out of the twenty-six polling stations in this commune (or 61.5 percent of the
total) may be questionable.

In Ouanaminthe, OPL challenged the validity of results for thirteen

out of the eighty-one polling stations (or 16 percent of the total) in the
commune. At the BEC, observers verified that in the case of at least one of
the contested polling stations, the number of votes cast exceeded the number
of ballots received. BED officials later told observers that many of the
68 Summary of Observation

ballot boxes for the thirteen contested polling stations contained between
450 and 600 ballots.

6. Northwest

BVs observed: 88 (10 communes out of 10 total)

Observers reported virtually no major incidents or irregularities,

except in one polling station in the commune of Baie de Henne during the
vote and in one polling station in Port-de-Paix during the vote count.
Relatively minor delays in the opening of polling stations were common
throughout the department, as all of the polling stations visited reported to
have opened approximately one hour late.

Three polling stations functioned only partially in Jean Rabel,

because the owner of the house where they operated forced them to close
due to nonpayment of the rent by the BEC. One polling station did not open
at all in Port-de-Paix as a result of the loss of the voter register by the BEC.
In Jean Rabel, the BEC president stopped the count at 6:00 p.m., when it
started getting dark, and moved operations to the police station, which had a
generator.

There were challenges filed in the communes of Saint Louis du

Nord, Môle Saint Nicholas, and Jean Rabel, after most polling station
workers failed to fill in the procès-verbaux immediately after the count,
leaving this for the next day. Observers reported that on May 25 a recount
for the commune of Saint. Louis du Nord took place in the Port-de-Paix
BED in the presence of candidates, poll watchers and national observers.
The recount revealed exactly the same numbers that had already been
compiled at the Saint. Louis du Nord BEC. Compilation for the communes
of Jean Rabel and Môle Saint Nicholas also took place at the BED.

7. South

BVs visited: 115 (14 communes out of 18)

Receipt of materials in most BECs was orderly and calm. However,

on May 22 BEC officials in Port-à-Piment did not permit observers to enter
working areas. There, observers also witnessed a representative for an
independent candidate being denied access. The BEC completed
 Summary of Observation 69

compilation of results in the sole presence of Fanmi Lavalas representatives

and observers from KOZEPEP.

Following election day, the security situation in the South

deteriorated. Five BECs (Chantal, Ile-à-Vache, Les Anglais, Côteaux, and
Roche-à-Bateau) had to close their doors for several days for security
reasons, while two others (Saint Louis du Sud and Aquin) initiated
compilation proceedings in a tense atmosphere and under heavy police
protection. As of May 30, the BEC of Ile-à-Vache had still not begun
compiling results for local elections. A few BECs compiled results for
Senate and deputy seats, while others were ordered by the CEP liaison to
postpone compilation, which took place later at a central location set up by
the BED in Les Cayes.

OAS observers witnessed various irregularities during compilation at

the BED: missing procès-verbaux, blank or partially filled-in procès-
verbaux, extensive use of correction fluid, incorrect additions, procès-
verbaux filled for some posts and not others, large discrepancies in the
number of voters voting in different races in the same polling station,
differences between spelled-out figures and the recorded figures, erasures of
figures, scotch tape on several procès-verbaux covering candidates’ names
and figures, no signatures or several signatures in the same handwriting.
BED officials also confirmed that results showed one polling station in
Aquin with a 100 percent turnout in favor of Fanmi Lavalas candidates.

Observers also confirmed a number of serious incidents that took

place during or after the vote count. Several of those incidents raised doubts
about the legitimacy of vote results for different races in certain communes.

The arrest on May 23 of Paul Denis, OPL Senate candidate and a

prominent member of OPL’s national coordination, further destabilized the
security situation in the department. Denis was arrested at his home in Les
Cayes by police from the National Palace Security Unit based in Port-au-
Prince, taken before the commissaire du gouvernement in Les Cayes who
issued a warrant for his arrest and then transferred to the Pétionville prison
in the capital. According to the police, a number of heavy-caliber weapons
intended for use in violent protests were found during the search of Denis’s
home, a fact denied by eyewitnesses interviewed by the OAS election
observers. Following Denis’ detention, the OAS Mission’s regional office
received dozens of complaints by OPL supporters and candidates who feared
70 Summary of Observation

for their security, particularly those residing in Saint-Louis du Nord, where

a warrant had been issued against the OPL mayoral candidate, allegedly for
assaulting a poll worker on election day.

In the commune of Côteaux, registers and four ballot boxes for two
polling stations located in the sixth section were stolen and burned after the
count. Although observers were originally told that both polling stations
would be annulled, on May 29 the president of the BED reported that the
results of those polling stations would be included in the compilation
because someone had managed to save the procès-verbaux posted on the
polling stations’ doors. Although the two polling stations were located in
close proximity to the BEC, both procès-verbaux were delivered to the BEC
late on the morning of May 22. The opposition contested the authenticity of
the procès-verbaux.

Electoral authorities also confirmed that, in one polling station located

in Sinai in the fourth section of Côteaux, the security agent collected twenty-
nine electoral cards, ran to the BEC, and accused the BV president of
confiscating the cards. Although all poll watchers and voters present at the
time denied it, the BEC president and vicepresident believed the security
agent’s version and ordered the BV to close around 3:00 P.M. The problems
in this polling station also resulted in the early closing of a nearby polling
station, where it is alleged that the polling station president had threatened
voters with a gun. Because the count was not done for either one of those
polling stations after closing, it was agreed that the count would take place at
the BEC on May 22. However, following several demonstrations that
required the intervention of the antiriot police, the president and
vicepresident of the BEC decided to leave for Les Cayes with all the voting
materials. At the request of the president of the BED, they returned to
Côteaux on May 24. That night they took all voting materials to their
homes, including ballots. Both officials were accused of being pro-Fanmi
Lavalas. In January, opposition parties had filed a formal complaint with the
CEP concerning their partisan affiliation.

The president of the BED subsequently informed the OAS Mission’s

regional office that only the results for the two Sinai polling stations would
be annulled in the commune of Côteaux. However, the fact that all voting
materials for the commune remained overnight in the private homes of two
BEC officials casts doubts on the transparency of the process in this
commune.
 Summary of Observation 71

In the commune of Les Anglais, the chain of custody of electoral

material was broken on May 22 when an independent candidate for deputy
and his supporters took over the BEC, expelled BEC members, and
remained alone with the voting materials until noon the following day. The
incident took place after BEC members publicly announced that a
verification of procès-verbaux and a possible recount of ballots would take
place. After antiriot police retook control of the BEC, Fanmi Lavalas
representatives formally requested that BEC officials annul the votes for the
independent candidate in the six polling stations originally reported as
having been lost prior to takeover of the BEC. The president of the BED
ordered BEC members to annul all results for all races in those six stations.

In addition, observers confirmed that a BEC supervisor closed two

polling stations in Les Anglais during the vote. In one of them, he claimed
the polling station president was marking ballots for MOCHRENA. The
president of the BED told observers that as many as eight polling stations in
the commune may have been closed in an irregular fashion during the vote.

In Ile-à-Vache, BEC members reported to observers that, on the

morning of May 22, the Fanmi Lavalas mayoral candidate requested that
ballots for one polling station in Balais Raise be recounted. Notification of
the recount was sent to the other two cartels participating in the mayoral
election, both independent candidates - one identified by a boat emblem, the
other by a fish emblem. Members of the boat cartel then demanded that all
fourteen polling stations in the commune be recounted. As the recount
proceeded, it was clear that the boat cartel had won the race for mayor.
When Fanmi Lavalas supporters started gathering around the BEC, BEC
members requested police protection to leave for Les Cayes with all the
procès-verbaux. The president of the BED confirmed that the security of
BEC members was at risk. He also confirmed that several BEC workers,
who had gone into hiding, had been harassed by the police because of
accusations that they had committed fraud on behalf of the boat cartel.

The Fanmi Lavalas mayoral candidate continued to demand the

annulment of several polling stations, particularly two in Balais Raise,
alleging that voters had been allowed to vote several times in these polling
stations. Fanmi Lavalas also alleged that a number of polling stations in the
commune had allowed voting to continue until 8:00 or 9:00 P.M. and that the
BEC vicepresident had plotted with the boat cartel to steal the vote from
72 Summary of Observation

Fanmi Lavalas. The president of the BED sent the case to the CEP, which
decided in late June to annul the results of the two polling stations of Balais
Raise.

In the commune of Tiburon, observers confirmed that, while

materials for one polling station were delivered to the BEC on the evening of
May 22, one poll worker fled with two ballot boxes, one for the CASEC race
and one for the ASEC race. The next morning, after he returned the boxes to
the BEC, BEC officials, with the consent of party representatives, decided to
annul the ballots for CASEC and ASEC.

In Camp Perrin, four masked persons, identifying themselves in the

dark as BEC members, intercepted members of one polling station located in
Haut Levy, stole the five ballot boxes and burned them. Three individuals
were arrested on charges related to this incident. One of them, a PDI
mayoral candidate, was later released. In the commune of Roche-à-Bateau,
the antiriot police reestablished order after armed groups attempted to steal
voting materials from poll workers on election night. In addition, policemen
and BEC officials in the commune of Chantal confirmed that a supervisor
found 156 marked ballots, belonging to a polling station located in Rivière
des Mornes, under a bed in the private house where the polling station had
operated.

8. Southeast

BVs visited: 75 (6 communes out of 10)

With the exception of one commune, most of the polling stations

observed reported to have opened on time or with some minor delays. In the
first section of Belle-Anse, however, polling stations opened with delays of
up to six hours, due to the late arrival of materials. Because poll workers
were not informed of the CEP’s decision to extend voting hours, the eight
BVs located there closed at 5:00 P.M.

Elections for town delegate in Bainet did not take place due to the
omission of one candidate on the ballots. Also, one polling station in the
third section of Cayes-Jacmel received ASEC ballots intended for the
Northwest Department.
 Summary of Observation 73

The reception of materials in the commune of Bainet was disorderly.

On May 23 OAS electoral observers saw unused ballots, unsealed
envelopes, procès-verbaux, ballot boxes, and registers tossed in a heap in the
cramped and unsecured BEC office. Meanwhile, a crowd of approximately
eight-hundred persons had gathered in front of the BEC, requiring police to
cordon off the BEC, while candidates met with the Commission on
Security.7/ Opposition parties requested the annulment of the election in this
commune alleging, among a number of irregularities, the falsification of a
number of procès-verbaux and BEC officials’ complicity in fraud. All the
opposition parties and independent candidates signed a formal request to
cancel the election results in this commune. Given the magnitude of the
allegations and the lack of confidence in BEC officials, the commission
transferred compilation proceedings to the BED.

Compilation of results for the commune of Bainet, partially observed

by the Mission, finally began on May 27 under the supervision of BED
officials and in the absence of almost all opposition candidates and
representatives. Recounts were first done for the fourteen polling stations
located in the first section of the commune and the twenty-three polling
stations in the ninth section. As the compilation advanced, it became clear
that the independent cartel “pigeon” had won the mayoral race. OAS
electoral observers witnessed the Fanmi Lavalas mayoral candidate declare
that proceedings should be halted, and then knock over the table and the
results. The recount for the mayoral race was begun again and the
independent candidates ultimately offered their votes to the Fanmi Lavalas
cartel, stating that “pigeon and table are one”.

In the commune of Belle-Anse, candidates and party poll watchers

reported numerous incidents of armed intimidation, threats and vote buying
in several rural sections during the vote and at the closing of the polls,
leading them to call for the cancellation of all election results in this
commune. Following the presentation of these complaints, some of which
were confirmed by electoral authorities, the Commission on Security also
moved compilation of results for the commune of Belle Anse to the Jacmel
BED. The BED president reported to OAS electoral observers that the
results for a minimum of five polling stations out a total of sixty for the

7
The commission was composed of the departmental delegate, the departmental director of the police, the
commissaire du gouvernement, and the BED president, as in other departments.
74 Summary of Observation

commune (or 8.3 percent) were questionable and, therefore, not included in
the compilation of results.

Opposition parties also requested the cancellation of the results for the
communes of Cayes-Jacmel, Marigot, and Jacmel. In Cayes-Jacmel,
approximately thirty out of fifty-one procès-verbaux contained major
inconsistencies that required a recount, which was partially observed by the
OAS regional electoral team. A recount of results, ordered by the BED
president, for certain contested polling stations began in the Cayes-Jacmel
police station on May 23, in the presence of international and national
observers as well as one representative for each political party concerned.
The recount continued for several days.

9. Grand’ Anse (June 11, 2000)

BVs visited: 90 (12 out of 12 communes)

On December 22, 1999, the BEC president of the commune of Anse-

d’Hainault in the Department of the Grand’Anse removed from the list of
candidates Georges Simon, the then mayor and a ESKANP-KOREGA
candidate for reelection. Simon’s removal was based on an allegation by the
Espace de Concertation candidate that the mayor had made threats against
his party during an interview on Radio Pipiritte. Following the BEC
president’s decision, persons claiming to be supporters of Simon ransacked
the BEC office and burnt a vehicle belonging to the International Foundation
for Electoral Systems (IFES).

In early January, the CEP reinstated Simon’s candidacy, but Simon

refused to recognize the legitimacy of the BEC in Anse-d’Hainault or the
BED in Jérémie. On April 3 a group of unidentified men burned down the
house of the BED president. The next day, a group of armed men closed the
BED and BEC in Jérémie. By this time, electoral authorities had completed
registration in all the communes except Anse-d’Hainault.

Both electoral offices remain closed until May 12, when the CEP
installed a new BED. On May 15, a new BEC in the commune of Anse-
d’Hainault was also installed. Elections in the Department of Grand’Anse
were postponed until June 11 to allow for the registration of voters in Anse-
d’Hainault.
 Summary of Observation 75

Because of previous political problems and violence, the government

deployed a significant number of police to maintain order on June 11, 2000.
The presence of police from other departments helped to lessen the
incidence of violent outbreaks, especially in the commune of Anse-
d’Hainault. However, the large number of police officers, including the
CIMO riot squad, created an atmosphere of fear among some parts of the
local population.

The Prime Minister and the chief of police traveled to the department
the day before the elections and remained until after the elections were
completed. While the OAS electoral observers noted some crowd-control
problems (according to reports, two BVs in Galete, Potonier, third section of
Les Irois, were closed by the area supervisor), the voting proceeded
relatively normally.

In order to avoid the confusion of May 21 during the vote count and
the compilation of results, the Prime Minister ordered the police to transfer
BV staff to a common compilation center in Jérémie. However, because of
the relocation, many party representatives were not present at the
compilation of results. Additionally, since polling officials had relocated,
they did not post results on the main entrance of the BV, as stipulated by
law. The OAS Mission did not witness any attempt to alter the results or the
count in this exercise, although it noted that the decision to transfer the BVs
had been made by the Prime Minister rather than the electoral authorities.

The most serious incident occurred in the commune of Dame-Marie,

where various BVs were attacked on election night, resulting in the
destruction of the procès-verbaux and boxes of ballots. In one BV at
Plonket, Barriadelle, first section, a Lavalas candidate stole the ballot boxes
for ASEC and CASEC races. Additionally, a Fanmi Lavalas candidate and
ten of his supporters stole all of the ballot boxes and procès-verbaux from
BV staff in Tet Plance, first section. The men were armed with knives and
tear gas. The procès-verbaux posted at the BV appeared to have been
salvaged. In La Haie, fourth section, two BVs were ransacked by armed
assailants. Although no material had arrived at the BEC, staff believed that
the procès-verbaux may have been saved. In Farouge, third section an
ESKANP supporter allegedly shot his rifle into the air while he held the BV
president by the neck. The ESKANP supporter and his companions took the
boxes of ballots, procès-verbaux and some personal belongings from the BV
president.
76 Summary of Observation

Around midday on June 11, in Les Irois, a fight broke out in a city
hall BV and a Fanmi Lavalas supporter allegedly stole the list of voters from
the BV security guard. The list was recovered shortly thereafter, and the BV
was reopened some hours later. In Galete, Potonier, third section, the
election supervisor closed two BVs in the mid-morning for security reasons
when the population became uncontrollable. Polling booths were reportedly
kicked over by voters who pushed and shoved their way into the BV,
interrupting operations. In the third section of Jorgue, the election
supervisor for the area closed a BV where there had been problems among
the personnel. The secretary, a Fanmi Lavalas member, had previously
walked out together with a Fanmi Lavalas poll watcher and had left the
ESKANP president and OPL vice president to run the BV.

On June 12, a crowd of OPL and ESKANP supporters entered the Les
Irois BEC and destroyed the radio, removed the battery and solar panels, and
destroyed procès-verbaux as well as boxes of ballots and other materials in
the building. This action was a consequence of the BEC members’
reluctance to accept the ballot boxes from the Jorgue BV, which had closed
at 4:00 P.M. on June 11. The remaining two BV members reportedly arrived
at the BEC between 8:30 and 9:30 P.M. on election night. The BEC
members refused to accept the boxes because, according to them, they had
not received any official report on the closing of the BV and could not
proceed to count the votes in the absence of the Fanmi Lavalas secretary.
The antiriot police who were called in to restore order, arrested three men
who, according to members of the three political parties contesting seats in
that commune, had nothing to do with the ransacking of the BEC.

In Les Roseaux, the Mission was informed of the alleged beating and
kidnapping of a Tet Ansam party poll watcher on the morning of June 12 in
Leon (second section of Jérémie), allegedly by an individual wearing a CNO
observer T-shirt who was in a vehicle marked “Press.” On June 11, the poll
watcher had asked a BV president in Bois Sec, second section of Roseaux, to
expel a Fanmi Lavalas poll watcher who was reportedly influencing the
voters. According to a Tet Ansamn chamber of deputies candidate, the
observer had threatened to kill the poll watcher after the vote.
 Summary of Observation 77

10. Area of Nippes (June 11, 2000)

BVs visited: 88 (6 out of 6 communes)

For logistical reasons, the Nippes area of the Grand’Anse constitutes

an independent electoral jurisdiction made up of six communes. During the
electoral process, this region was not affected by the political and electoral
problems observed in the rest of the Grand’Anse. Therefore, electoral
authorities were unaware until late on Saturday, May 20, that the
cancellation of the election in Grand’Anse covered the region of Nippes.
Electoral authorities confirmed that, due to lack of communication and
infrastructure, they were unable to inform all poll workers. Thus, 50 percent
of BVs operated normally on May 21; of those, 75 percent reported results.
The Mission also verified that an unconfirmed number of unused ballot
boxes from the May 21 election were never returned to the BECs.

Several incidents of violence took place on the days before the June
11 election. Firstly, the BECs of Petit Trou de Nippes and Miragoane
were set on fire two days before election day. Nonetheless, the personnel of
the Nippes BED managed to supply electoral authorities in those communes
with duplicates of registers, thus ensuring the normal conduct of the election.
In addition, on June 10, armed individuals stole ballot boxes for eight BVs in
Anse-à-Veau. An independent candidate for deputy was later arrested for
this crime.

A few minor incidents marred a peaceful day. In the commune of

Barradères, at approximately 4:00 A.M., unidentified individuals attacked
poll workers from one BV and stole the register and ballot boxes. Also in
Barradères, two individuals stoned poll workers and stole voting materials
from two BVs in Fond Palmiste, causing their early closing at 11:00 A.M. In
the rural area of Miragoane, armed individuals entered one BV and forced
members of one BV to flee, leaving voting materials abandoned. BED
officials confirmed the results for that BV were annulled. In Anse-à-Veau,
none of the BVs of the fifth communal section operated, as three out of six
BVs did not receive materials. Additionally, in the area of Brossard, an
armed observer from Fanmi Lavalas attacked the president of a BV.
78 Summary of Observation

11. Maniche, South Department (June 11, 2000)

BVs visited: 20

The Mission visited twenty out of the forty BVs that operated in the
commune of Maniche and concluded that elections were fairly well
managed. One incident took place in the communal section of Melon, where
two candidates for deputy were arrested after being accused of disorderly
conduct. The candidates were released after the closing of the poll.
 Summary of Observation 79

C. Table of Elections Affected by Irregularities

Department Commune Communal section Race Date Situation

- Artibonite - Verretes Sixth, CASEC May 21 9 out of 12 BVs had election materials stolen and burned. Election
(Terre Natte) ASEC was to be reheld; the Mission did not confirm repetition of the election
- Center - Hinche Second CASEC May 21 Operation of 6 out of 15 BVs canceled due to the interruption of the
(Marmont) ASEC vote by armed groups
- Center - Boucan All CASEC May 21 Doubtful legitimacy of procès-verbaux for all BVs
Carré ASEC
Mayor
Deputy
- Center - Maïssade All CASEC May 21 Doubtful legitimacy of procès-verbaux for a significant portion (at
ASEC least 50%) of BVs
Mayor
Deputy
- North - Acul-du- Fourth CASEC May 21 Results from 4 out of 13 BVs annulled (more ballots than registered
Nord (Grande Ravine) ASEC voters in three BVs; ballots for one BV stolen)
- North - Limbé All CASEC May 21 Attempts to influence the vote by OFL representatives
ASEC
Mayor
Deputy
- North - Bas Limbé All CASEC May 21 Attempts to influence the vote by OFL representatives
ASEC
Mayor
Deputy
- Northeast - Mont All CASEC May 21 Doubtful legitimacy of at least 60% of procès-verbaux
Organisé ASEC
Mayor
Deputy
- Northeast - Terrier All CASEC May 21 Results of 20% of BVs may have been annulled. Armed attack on two
Rouge ASEC BVs
80 Summary of Observation

- Northeast - All CASEC May 21 13 BVs out of 81 BVs found to have between 450 and 600 ballots
Ouanaminthe ASEC
- South - Les Anglais All CASEC May 21 6 to 8 BVs out of 41 may have been irregularly closed on election day.
ASEC Independent candidate to deputy and supporters occupied BEC for
Mayor over a day
Deputy
- South - Ile-à-Vache All Mayor May 21 Irregular cancellation of two BVs resulted in alteration of results
- Grand’Anse - Dame-Marie All CASEC May 21 Irregular cancellation of results following OPL’s victory on May 21.
ASEC July 2 Elections on July 2 were plagued with irregularities. (See Chapter IX,
Mayor “Complementary Elections.”)
Deputy
- North - Bahon All CASEC May 21 Initial loss of results for 17 out of 25 BVs on May 21; recovery of
ASEC July 2 procès-verbaux and cancellation of results following OPL’s victory.
Mayor July 2 elections plagued with irregularities (See Chapter IX,
“Complementary Elections.”)
 CHAPTER IX
COMPLEMENTARY ELECTIONS
A. Les Irois, Grand’Anse

On Sunday, July 2, complementary elections were held in Les Irois.

These elections became necessary following the invasion of the BEC on
June 12 by a crowd of angry OPL and ESKANP supporters, during which all
of the ballot boxes and procès-verbaux from the twenty-six polling stations
were allegedly destroyed.

All of the polling stations visited started late. This was due to the late
arrival of voting materials in Les Irois at around 7:00 A.M. on July 2.
However, the late start did not cause undue problems throughout the rest of
the day, and by 4:00 P.M. most of the polling stations visited were virtually
empty. Members of the HNP were present outside most of the BVs visited,
and they helped to maintain law and order in an area where the population is
politically polarized.

The electorate voted in a peaceful manner. There were no major

incidents or crowd control problems. However, due to the very polarized
nature of local politics, heated arguments between polling station members
over how the voting process should be conducted, especially the manner in
which instructions should be given to the elderly and illiterate voters, were
observed. In some cases, voting had to be stopped while the BV members
sorted out their differences.

Generally, polling officials correctly followed the procedures for the

vote count, and political party representatives were present at most of the
BVs observed. OAS observers noticed a high percentage of spoiled votes,
as many as 20 percent in some cases. Most of the procès-verbaux were
properly filled out and duly signed.
84 Complementary Elections

The presence of the president and other members of the BED and two
members of the CEP contributed significantly to a very well organized
postelectoral operation. By 8:00 P.M. on July 3, the compilation of the
results at the BEC level had begun, and by 11:30 P.M. the whole process was
completed.

Half way through the count, the police interrupted the process to
transport poll workers and materials to a school, where the count continued.
While the poll workers were inconvenienced, the transfer appeared to be
professional and facilitated the vote count. Shortly after the BV workers and
party poll watchers arrived at the school they were provided with tables,
chairs, and adequate lighting to facilitate their task. Due to the transfer,
though, copies of the procès-verbaux were not posted on the doors of the
BVs.

B. Dame-Marie, Grand’Anse

On Thursday, July 6, complementary elections were held in the

commune of Dame Marie. These elections were organized following a
decision of the Grand’Anse BED to annul fifteen of the forty-one polling
stations after the June 11 elections.

The OAS Electoral Mission did not have strong evidence that it was
necessary to hold complementary elections in Dame-Marie. On June 11, the
OAS observers in this commune reported that armed individuals had
attacked five polling stations. A few days later, however, the BEC president
of Dame-Marie claimed that fifteen had been destroyed. At the same time,
OPL, which had won in almost all the BVs, presented thirty-four procès-
verbaux, which had the special code of the BED president, but the BED did
not review this evidence. Some of those procès-verbaux had been handed
over to the OPL representatives in the presence of two OAS observers
assigned to Dame-Marie.

During the June 11 elections, the OPL candidates for the chamber of
deputies, Riche Andris and Arthur Papillon, obtained 5,074 and 3,266 votes,
respectively, while the two Fanmi Lavalas candidates, Gerard Gilles and
Clones Lans, obtained 786 and 581, respectively. Based on those results
only Andris would have been elected in the first round. Only ESKANP-
KOREGA and Fanmi Lavalas fielded poll watchers. OPL, which had won
on June 11, was not present in any of the twenty polling stations visited. No
 Complementary Elections 85

ballots for the post of senator were sent to the polling stations because,
according to the BEC, the two senators had already been elected.

An extremely high number of votes were cast in many polling stations

after only three or four hours of voting. For instance, OAS observers visited
a polling station in Docan, fourth section, at 9:15 A.M. and noted that 145
persons had already voted. Similar high numbers were observed in
Desormeaux, Lesson and in the town of Dame-Marie. The OAS electoral
observers also noted voters not registered at the BVs, but claiming to be
national observers, forced the BV workers to allow them to vote. Poll
workers noted their names and card numbers on the back of the register. As
early as 11:00 A.M., BV members were seen carrying electoral materials to
the BEC because supposedly everyone registered had already voted.

At 10:15 A.M., OAS observers visited a polling station in a remote

area called Montagnac, in the third section where there were no ballots for
deputies left on the table. The polling station workers told the observers that
two groups of armed men had invaded the BV and filled the ballot boxes.
After inspecting the ballot box, the OAS observers discovered that all the
ballots at the top had been cast in favor of Fanmi Lavalas. Those ballots
were not folded. In another polling station, all four ballot boxes had been
filled, and there were no remaining ballots on the table.

The deployment of the HNP did not take into consideration the areas
that had been affected during the June 11 elections. Consequently, four
polling stations (two in Farouge, one in La Haie, and one in Tete Planche)
that had been attacked on June 11 were once again the objects of attack
during the July 6 complementary elections.

Based on information obtained from the BED, there were cases of

fraud, serious irregularities or armed attack in at least eighteen of the forty-
one polling stations. It is important to note that complementary elections
were held because of problems in fifteen polling stations during the June 11
elections.

C. Bahon, North Department

After the burning of seventeen ballot boxes and one of the procès-
verbaux in Bahon on May 21, the North BED succeeded in obtaining
procès-verbaux of thirty-four of the thirty-five total BVs. Under intense
86 Complementary Elections

pressure from all sides, compilation of results from those procès-verbaux

was organized in the North BED on June 12. Political party representatives,
electoral officials, and journalists were present for the compilation of results.
The three BEC members signed the compilation sheets. Nevertheless.
Nawoon Marcellus, the Lavalas candidate for deputy, alleged fraud and
demanded a new election. On May 25, one of Marcellus’s supporters
entered the BED with an Uzi, while his candidate was in the BED and other
supporters carried hand weapons. On May 26, in the presence of OAS
electoral observers, Marcellus threatened to kill the BED president and the
three members of the Bahon BEC.

The complementary elections in Bahon were held on July 2 and were

tainted by numerous irregularities. Voters did not cast ballots for senator
since, according to the CEP, the number of votes could not alter the results
for senator from May 21, although the results for senator had not yet been
posted. Additionally, the BED failed to send the procès-verbaux for the
thirty-five polling stations; therefore none of the polling stations was able to
record the opening of the voting. Of the fifteen polling stations observed,
only one received a complete register (as all of the other registers were
allegedly destroyed). Ten polling stations used copies provided by the BED,
of which three were not complete. Two polling stations did not receive any
register at all. A polling station in Saint Maurice did not have a BV number,
and it had a register that was assigned for another polling station in Lea.

In about half of the polling stations observed, people voted despite not
being registered at that site. In Haute Bourge College BV, for example,
twenty-seven people not appearing in the register voted. According to the
secretary of the BV, many of those who voted were actually registered in
Grande Rivière and not in Bahon.

Throughout the three communal sections observed, aggressive

supporters (chimères) strategically mobilized groups of people to vote
multiple times. A voter told observers that he had voted in Glacis Bourri,
and he was found outside a polling station in Nicolas, another communal
section several miles away from Glacis Bourri. The voter’s thumb was
clean, indicating that the voting procedures were not properly carried out.
The rate of voting was mysteriously rapid. Observers spent a minimum of
half an hour in each polling station. During the thirty minutes spent in the
polling stations, nobody voted. However, after returning only an hour later,
OAS electoral observers noticed that approximately two-hundred people
 Complementary Elections 87

voted. Only two of the fifteen polling stations observed had a line of about
twenty people waiting to vote.

Two polling stations observed ran out of ballots for the deputy
election. By 11:15 A.M. more than four-hundred ballots had been cast for
deputy despite only eighty-eight people being registered at that particular
station. At the other polling station, some ninety-five people had voted, but
all of the four-hundred ballots for deputy had been used.

In all of the fifteen polling stations observed, the polling station

members had been replaced from those who had worked on May 21. The
new polling station members received training on Saturday, July 1, for three
hours.

The majority of the polling stations in the communal section of Bahon

arrived at the BEC between 3:30 P.M. and 4:00 P.M., as all registered voters
had already voted. All of the polling stations were instructed to conduct the
vote count in the school next to the BEC in Bahon. The OAS Electoral
Mission observed the count in two polling stations and did not observer any
irregularities.

The compilation of results was relatively well organized. As each BV

finished counting, the BV members were instructed to transport the ballot
boxes next door to the BEC. Once the BV members arrived at the BEC, the
electoral material was taken from them and the BV number and address
were recorded. During compilation, OAS observers noticed an anomaly in
at least one BV, in which all 357 voters cast their votes for deputy for Fanmi
Lavalas. At the end of the compilation, the results were not posted on the
door of the BEC.

According to the results of the July 2 complementary elections,

Marcellus, Lavalas candidate for deputy, won the election in the first round
with 7,919 votes. On May 21, he had received 1,297 votes. According to
the official results from July 2, some 89 percent of registered voters cast
ballots (an increase of about a third from May 21).
 CHAPTER X
ANALYSIS OF THE RESULTS
A. Introduction

According to the Electoral Law, the process of transmitting results

entails the following: once the vote is over, poll workers count the votes,
election by election, then take note of the results for each candidate on the
tally sheets. At that time, complete results are obtained for each polling
station with the number of votes for each candidate. The poll workers also
write in the procès-verbaux or election records all the relevant information,
including incidents that took place throughout the voting procedures. Poll
workers take the results to their respective BEC. For each polling station,
the BEC obtains a copy of the procès-verbaux, containing the results of each
election.

With these procès-verbaux, the BEC compiles the results for all of the
elections in the commune. After this task is accomplished, it is possible to
determine the winners of the local elections (mayor, ASEC, CASEC, and
town delegate) and have the number of votes received by senator and deputy
candidates for that commune. BEC members are supposed to post all the
results on the BEC’s door (Article 161). BEC members must also forward
to the BED compilation sheets for the elections for deputy and senator as
well as all the results for their communes. BEC members also annex a
report containing all pertinent information on incidents that may have had an
impact on the results.

For the deputy race, it is at the BED level that the compilation of
results for the different communes forming a district (circonscription) takes
place. Similarly, the BED compiles the results of all communes for the
senator race. Once the compilations are completed, results for deputy and
senator are obtained and posted by the BED.
92 Analysis of the Results

The Electoral Law stipulates that within forty-eight hours, the results
must be forwarded to the CEP. The results transmitted by the BED must be
accompanied by a report mentioning “relevant facts stated in the procès-
verbaux, as well as the challenges of candidates, political parties, and
coalitions” (Article 162). The results are not final and can still be legally
changed by the CEP. In order to change the results, however, the CEP must
respect certain conditions as specified in the report submitted by the BED.

B. The Transmission and Processing of Results

On the evening of May 21, the count proceeded with little incident.
However, the transporting of the materials to the BECs, and in particular, the
storage of the materials at the BECs, were extremely disorganized. The lack
of organization was evident in the way in which the procès-verbaux and the
rest of the materials were stored in several BECs. The OAS electoral
observers reported that the compilation of the results for the senate and
deputy races in the BEDs was generally carried out with a certain degree of
transparency. Additionally, the procedures for the transmission and
processing of the results were more or less respected at the BV, BEC, and
BED levels, and the problems observed by the OAS electoral observers did
not appear to be intentional. The discrepancies and irregularities pertaining
to the transmission of results appeared more often at the level of the CEP.

According to Article 162 of the Electoral Law, each BED must

publish the results of the elections in its respective department after having
first examined the records and reports from the BECs, taking into
consideration complaints and challenges presented by candidates or party
representatives. It is only at this stage, according to the same article, that the
BED has forty-eight hours to transmit the results to the CEP.

Those procedures, however, were not always observed. On May 23,

while the BEC and BED members were still preparing the compilation of
results, the Prime Minister announced that the CEP should immediately
proclaim the final results of the first round of elections. A few days later the
director of operations announced unofficial results for the senatorial race.
On June 5, the CEP reminded “all candidates, parties, and political groups
that the partial results published by the CEP were based on information
provided by the BEDs.” Yet, only one BED (West 2) had announced
results. On June 6, the president of the CEP announced that no candidate
had been elected up to that time, contradicting the earlier statements of the
 Analysis of the Results 93

director of operations. The Mission was not aware of the source the CEP
used to publish its results.

The premature announcement of the results caused confusion and

these results did not always correspond with those published at the
departmental level. For instance, in the Department of the Northwest, the
CEP announced that the Fanmi Lavalas senatorial candidate, Fedner Pierre
Louis obtained more votes than the independent candidate, Luc Fleurinor,
while the BED had posted results that gave the victory to the latter. Shortly
after the announcement by the CEP, the BED members in this department
removed the results from where they were posted.

On June 19, the director of operations of the CEP published a list of

the results of the deputy race and, on June 21, the list of results for the
municipal and local elections. This announcement once again was made
before the BEDs themselves had completed the process of tabulating,
reviewing challenges, and posting the results. The directorate of operations
published only the list of the winners and in another document, the number
of votes obtained by the two candidates going to the second round. The CEP
never published a complete list of the results of the deputy race indicating
the number of votes obtained by each candidate.

C. Calculation of Percentages for the Senatorial Elections

1. Chronology

In late May, the CEP director of electoral operations announced

partial results of the senatorial elections, reflecting percentages based on the
total valid votes obtained by a limited number of candidates. On May 30,
the Chief of Mission met with the president of the CEP, Leon Manus, to
bring the error to his attention. Manus requested that the OAS Chief of
Mission detail his concerns in a letter, which was sent on the following day.
The letter stated:

A thorough review of the results announced by the CEP and, in

particular, of the percentages attributed to the leading candidates to
the Senate, reveals a grave miscalculation that has an impact on the
number of seats filled for the Senate in the first round. […] The
Electoral Law clearly states in Articles 53 and 64 that, in order to win
a seat for the Parliament in the first round, a candidate must have an
94 Analysis of the Results

absolute majority of the total of valid votes. […] The figures

distributed by the Directorate of Operations of the CEP indicate that it
did not follow the procedure established in the aforementioned
articles, but chose instead to add up only the votes gained by a small
number of leading candidates. Therefore, it arrived at incorrect
percentages.

The next day, the Chief of Mission met with the President of the
republic, upon the President’s request, to discuss the matter. Although the
Mission had hoped to maintain the strictest confidentiality, the letter was
leaked to the press by unknown sources.

A few days later, Manus responded to the OAS concerns in a letter

explaining that the results were to be considered partial and noting the issue
of publication of the OAS letter. In this letter, Manus agreed that “Articles
53 and 64 [of the Electoral Law] state that in order to win a seat for the
Parliament in the first round, a candidate must have the absolute majority of
the total of valid votes. I accept and fully agree with these dispositions that
we have to respect and follow.”

The Mission met in a private meeting with all CEP members to

reiterate its position and to explain to them that the identification and
communication of this problem were covered by the terms specified in the
Mission’s mandate. A few days later, the president of the CEP left the
country. On June 21, from abroad, he addressed the Haitian people to
explain the reasons for his departure. He stated:

While we continue to maintain our concern about the way the OAS
statement was made public, I must admit that some of the issues it
raised helped us delve more deeply into some of the technical issues
and to recalculate the percentages of votes in strict adherence to the
stipulations of the Electoral Law. Thus, the definitive results for the
senatorial candidates are that only five were elected outright in the
first round. The majority of those who had the largest vote in the
initial count should participate in a second round. This is the result of
the final tabulations according to the terms of the Electoral Law.

Confronted by an ultimatum to immediately proclaim results that I

considered illegitimate and incorrect, I found myself incapable of
 Analysis of the Results 95

committing such an act of treason to my country at such a decisive

moment in its history.

Shortly thereafter, the CEP published its results using its own method
of calculations. On June 26, the President of the republic decreed the second
round elections to be held on July 9, 2000.

2. The CEP Method of Calculating Absolute Majority

As previously noted, the CEP calculated its percentages using the

votes obtained by a limited number of candidates. In the departments where
two senators were being elected (eight of nine departments), the CEP used
the sum of the votes cast from the top four candidates. In the Center
Department, where three senators were being elected, the CEP used the sum
of the votes cast from the top six candidates. After taking the sum of these
votes cast, the CEP divided by two (or in the case of the Center, by three) to
arrive at its baseline to calculate absolute majority. In some departments,
specifically the West, the (BED) calculated its absolute majority based on
the total number of valid votes, divided by two. The CEP, however,
recalculated these results using its own methodology.

For example, if in a Senate election, with two seats in dispute, there

were one-thousand valid votes obtained by eight candidates.

Candidate Votes Obtained Percentage Official

Percentage

A 200 20 % 66. 66 %
B 150 15 % 50. 00%
C 130 13 %
D 120 12 %
E 115 11.5%
F 100 10 %
G 100 10 %
H 85 8.5 %

8 candidates 1000 votes 100%

This method excludes four-hundred votes that make up 40 percent of

valid votes.
96 Analysis of the Results

3. The Effect on the Results

The use of the CEP method of calculating absolute majority affected

the results in nine Senate races in the departments of the North, Northwest,
Northeast, the Center, and the Artibonite. Furthermore, this method
excluded more than 1.1 million votes obtained by nearly 10one-hundred
candidates.

4. Alternative Methods to Calculate Absolute Majority in

Accordance with the Electoral Law

Since many seats were in dispute, voters could vote for several
candidates at the same time (two in eight departments and three in the
Center) or choose to cast one single choice. Determining the exact number
of valid votes cast is made difficult for two reasons. Firstly, voters could
cast one or multiple votes on the same ballot and the votes were still
considered valid. Secondly, the forms provided by the CEP did not
contemplate the recording of valid ballots cast in the Senate elections. As
demonstrated below, the number of valid ballots would have been the most
accurate reflection of the number of valid votes and could have served as the
baseline for the calculation of the absolute majority.

While the OAS does not advocate any particular methodology to

calculate absolute majority, two potential alternatives that would have
conformed to the Electoral Law are explained in this chapter.

The first method takes into account valid ballots, no matter the
number of choices cast by the voter; it uses this number as denominator and
the number of votes obtained by each candidate as numerator.

For example: If there were one-thousand valid ballots in which

eighteen-hundred votes were cast for eight candidates; two-hundred voters
chose to cast only one choice.
 Analysis of the Results 97

Candidate Valid votes Percentage CEP Percentage

A 310 31 % 59.33%
B 300 30 % 57.47%
C 220 22 % 42.10%
D 215 21.5%
E 210 21.0 %
F 205 20.5 %
G 200 20 %
H 140 14.0 %

8 candidates 1,800 votes 180% (the choices not cast by 200

voters represent the 20% left over).

In the above example, the number of votes obtained by each candidate

is divided by the total of valid ballots. In this example five-hundred
represents 50 percent of valid votes.

The second method takes into account that fact that two separate
elections are taking place in each department (except in the Center). In this
method, the baseline to establish absolute majority is calculated by dividing
the total number of valid votes cast for all candidates by two. Using the
same example, the eighteen-hundred valid votes are divided by two. Thus,
nine-hundred valid votes provide the basis on which to calculate the
percentages.

Candidate Valid Votes Percentage CEP´s Percentage

A 310 34.00 % 59.33%
B 300 33.33 % 57.47%
C 220 24.44 % 42.10%
D 215 23.88 %
E 210 23.33 %
F 205 22.77 %
G 200 22.22 %
H 140 15.55%

8 candidates 1,800 votes 200%*

* Given that there are two seats in dispute and each voter can vote for each
seat, we obtain here 100 percent of votes for each seat, that is 200 percent.
98 Analysis of the Results

5. The Position of the OAS Mission

As was stated in separate letters to the president of the CEP and the
President of the republic, the OAS Mission believed that the method used by
the CEP to calculate the absolute majority violated the Constitution and the
electoral law of Haiti. By not taking into account the votes cast for every
candidate, the methodology additionally violated the principal of equality of
the vote: one person, one vote. Specifically, the methodology of the CEP
excluded approximately 1.1 million votes.

Analysis of Votes and Candidates Excluded from Consideration

in Senate Elections

Depart. Votes Excluded Voters Excluded Candidates %Votes

Excluded Excluded

West 362,263 181,131 21 23.7

South 107,813 53,906 12 32.32
Southeast 72,804 36,402 11 28.65
Center 140,993 46,998 15 37.75
North 119,667 59,833 11 28.65
Northwest 67,951 33,975 7 31.87
Northeast 41,637 20,818 8 31.39
Artibonite 218,923 109,461 14 35.49

Total 1,132,051 542,524 99

Since the calculations for the senate seats did not contemplate a
second round for the departments of the North, Northwest, Northeast, the
Center, and the Artibonite, the OAS Mission did not observe the proceedings
of the second round. In a press release dated July 9, the Mission
“determined that according to the provisions of Haiti’s own electoral
legislation, the final results for the Senate elections as proclaimed by the
Provisional Electoral Council (CEP) are incorrect, and the Mission cannot
consider them either accurate or fair.”
 Analysis of the Results 99

6. The Alleged Use of this Method in Previous Elections

To justify the use of its methodology, the CEP explained that this
procedure was used in the elections of 1990, 1995, and 1997 and that the
procedure was then validated by the international community. The Mission
believes that whether or not such a procedure was used in past elections is of
little relevance to the current electoral process, as it still excludes from
consideration a large number of votes and voters and does not accurately
reflect the requisite “absolute majority of all valid votes.”

Since the CEP claims that the past use of this procedure constitutes
jurisprudence, however, it is important to clarify that the procedure
introduced by the CEP for the legislative, municipal, and local elections on
May 21, 2000, was not utilized in the elections of 1990, 1995, nor 1997.

In 1990, the CEP, in its fourth press release, indicated that the
absolute majority of votes was required to elect the President of the republic,
the senators and the deputies. The CEP added that this majority was
calculated based on the total number of voters, counted as the total number
of ballots cast for each election. As the CEP was unable to obtain the
accurate total number of voters in the Senate races, it used the number of
votes obtained by the presidential candidates in each department.

In 1995, when final results of the elections of June 25 and August 13

were announced, the CEP showed the number of votes obtained by each
candidate to the Senate and the respective percentage. As seen in the results
issued by the CEP, these percentages were calculated on the basis of the
total of votes cast for all the candidates. Likewise, the results of the 1997
elections for the Senate proclaimed by the CEP based its percentages on the
total number of valid votes obtained by all candidates. (See Appendix IV,
CEP Results and Proclamation of Winners in 1995 and 1997 Elections.)

D. Comparison of Results Provided by BEDs with those Provided by

the CEP

1. Introduction

What follows is an analysis of the final results of the May 21 elections

proclaimed by the (CEP) on June 19 and 21. The Mission compared the
100 Analysis of the Results

results from the CEP to those posted by the departmental electoral offices
(BEDs) and the communal electoral offices (BECs). For the Senate race,
results in the West, South, Southeast, and Grand’Anse were identical. In
other departments, the BED and CEP results varied, although the differences
did not affect the final outcome in all of the elections. According to the
Electoral Law, the CEP has the prerogative to modify the results posted by
the BEDs, but only after it has been formally presented with a challenge to
these results. The OAS Mission observed that the process for receiving
challenges was rarely respected.

2. Elections for Senate

Northwest

In the Northwest, according to compilations by the Port-de-Paix BED,

independent candidate for Senate Luc Fleurinor obtained 40,115 votes, and
the Fanmi Lavalas candidate, Fedner Pierre Louis, obtained 32,647.
According to the final results of the CEP, independent candidate Fleurinor
obtained 39,120 votes, and the Lavalas candidate, Louis obtained 48,949
votes. The first senator is the candidate elected with the highest number of
votes under the condition that he or she obtain the absolute majority of
votes. He is elected for six years, while the second is elected for four years
only. Results for the nine other candidates for the Senate race varied also, as
shown in the table below.

The OAS was provided by the BEDs with a final result sheet for the
Senate race with a breakdown per commune from the BED. The CEP
results, however, were not substantiated with a breakdown of results per
commune, which prevents verifying the results from the commune to the
department. According to the CEP lawyer, Jean Louis, Fleurinor made a
complaint to the CEP, and his complaint was rejected because “that
candidate had no proof to substantiate his case” and “the CEP took results
from the Northwest BED.” Until early July, the Northwest BED maintained
its results and stated to an OAS observer that “the CEP made a mistake”.

The table below shows how the change of final results promotes the
Lavalas candidate to be elected “first senator” and drops the independent
candidate to that of “second senator.” The table also shows how the CEP
method of calculating the total number of votes allows both candidates to be
 Analysis of the Results 101

elected in the first round. However, according to the Electoral Law, no

candidate should have been elected in the first round.

CANDIDATES PARTIES CEP BED DIFFERENCE % %CEP

FEDNER PIERRE LOUIS FANMI 48,949 32,647 16,302 45.92 67,40
LUC FLEURINOR IND 39,120 40,115 -995 36.70 53,87
MELIUS HYPPOLITE OPL 29,800 30,646 -846 27.96
EVALLIERE BEAUPLAN FANMI 27,366 28,876 -1,510 25.67
VOLNEY JEAN EC 21,404 23,072 -1,668 20.08
ALEXIS GUERRIER MOP 11,455 11,596 -141 10.75
FASTIN JOSEPH IND 10,867 11,066 -199 10.19
MARC L. THEOPHIN PSR 7,853 8,139 -286 7.37
HENRY VIXAMAR IND 7,070 7,302 -232 6.63
OSNI EUGENE IND 4,146 6,040 -1,894 3.89
WALTERD DUGARD IND 5,156 5,035 121 4.84
TOTALS 213,186 204,534 8,652 200.00

North

The Mission also noted discrepancies between the BED and the CEP
results in the race for senate in the North. Houseline Bell Augustin was
credited with 8,172 more votes in the results proclaimed by the CEP than in
the BED-posted results. There was no recounting of votes for Senate
organized in the North, and the BED transmitted the same sheet of final
results it first posted. The BED of the North did not later transmit other
results for the senate race.

On May 28, the CEP announced that Bell Augustin, a Fanmi Lavalas
candidate, was elected first senator, ahead of Martineau Guerrier, also a
candidate for that party. However, the North BED had not yet sent official
results, and the Mission is unaware of the source of the CEP’s information.
Interviewed several times by the OAS regional coordinator, the North BED
president and secretary defended the results they posted and could not
explain the discrepancies. The CEP later retracted the announcement and
stated that Guerrier was first.

The increase of 8,172 votes in the CEP-proclaimed results allowed

Bell Augustin to obtain the absolute majority regardless of the method used.
Taking the number of votes credited to her by the CEP (97,130) and
calculating the percentage based as the total number of valid votes, Bell
102 Analysis of the Results

received almost exactly 51 percent. Using the results posted by the BED,
Guerrier would have been elected in the first round, but Bell Augustin would
have received 47.36 percent and would have gone to a second round against
the OPL candidate, Milton Chery. It is important to note that the OPL
candidate received 2,462 fewer votes in the CEP-proclaimed results than in
the results posted by the BED. (See table below.)

CANDIDATES PARTIES CEP BED DIFFERENCE % %CEP

MARTINEAU GUERRIER FANMI 105,603 102,197 3,406 55.46 56.19
YOUSELINE BELL AUGUSTIN FANMI 97,130 88,958 8,172 51.01 51.69
MILTON CHERY OPL 30,836 33,298 -2,462 16.19 21.39
ANTOINE RENE SAMSON IND 27,579 27,954 -375 14.48 20.89
MANDENAVE JEAN EC 19,734 20,462 -728 10.36
GABRIEL ARLET GRAND CHAMPS EC 16,367 17,171 -804 8.60
PIERRE CHENT OSIAS MOP 14,478 15,424 -946 7.60
JOSEPH JEAN BERNARDIN IND 14,454 14,735 -281 7.59
JACKSON NOEL MOCHRENA 13,375 13,511 -136 7.02
ANTOINE PIERRE RANFFO 10,226 10,391 -165 5.37
LOUIS PIERROT RNDP 8,000 8,124 -124 4.20
JOSEPH PIERRE LOUIS PADH 7,683 7,765 -82 4.04
JEPHTAT ROSEME IND 6,694 6,865 -171 3.52
KENOLD JOSEPH FRONCIPH 4,875 4,931 -56 2.56
JEAN BAPTISTE JULIEN FRONCIPH 3,781 3,847 -66 1.99
TOTALS 380,815 375,633 200.00

Northeast

In the Northeast, OPL candidate for Senate, Philome Firmin Jean

Louis, obtained 15,966 votes according to results first published by the Fort-
Liberté BED; in results proclaimed by the CEP on June 19, he was credited
with only 11,082 votes.

Independent candidate Argais Joseph was credited with 8,547 votes in

the BED count and with 5,547 in the CEP final results. The number of votes
for both Lavalas candidates did not change. Utilizing the CEP method of
calculation, both Lavalas candidates won in the first round. Basing the
percentages on the total number of valid votes, however, neither candidate
should have won in the first round.
 Analysis of the Results 103

CANDIDATES PARTIES CEP BED DIFFERENCE % %CEP

JEAN RODOLPHE JOAZILE FANMI 32,969 32,969 0 49.72 72.47
LUCIEN DELINX PIERRE LOUIS FANMI 30,736 30,736 0 46.35 67.56
ARMAND JOACHIM OPL 15,966 15,966 0 24.08 35.00
PHILOME JEAN LOUIS OPL 11,082 15,966 -4,884 16.71
JOSE JOSEPH IND 11,305 11,305 0 17.05 24.00
ARGAIS JOSEPH IND 5,547 8,547 -3,000 8.37
GERARD PIERRE EC 6,703 6,703 0 10.11
JUDNEL JEAN EC 6,459 6,459 0 9.74
EMMANUEL TINORD TET-ANSANM 3,762 3,762 0 5.67
ABRAHAM BONHOMME MOCHRENA 3,427 3,427 0 5.17
AMOS ANDRE MOCHRENA 3,201 3,201 0 4.83
LUNICE MATHIEU MPSN 1,456 1,456 0 2.20
TOTALS 132,613 140,497

Artibonite

In the Artibonite, the results of seven candidates out of seventeen

changed, as shown in the table below. These changes, however, did not
affect the final results. The table shows that basing the percentages on the
total number of valid votes, only one candidate would have won in the first
round.

CANDIDATES PARTIES CEP BED DIFFERENCE % % CEP

MEDARD JOSEPH FANMI 158,664 158,664 0 51.45 79.75%
NADIA MONDESIR SANON FANMI 139,074 139,074 0 45.10 69.91%
REVENEL BENOIT MOCHRENA 54,808 49,286 5,522 17.77 27.55%
YVES GUSTINVIL MOCHRENA 45,318 45,318 0 14.69 22.78%
ALEX LARSEN EC 27,371 27,371 0 8.88
VICTOR BENOIT EC 31,681 31,681 0 10.27
BEABRUM LAROSE OPL 23,881 23,881 0 7.74
RONALD DESORME OPL 23,653 23,653 0 7.67
PIERRE CORVIL PLB 23,368 23,368 0 7.58
SERGE JEAN LOUIS CFD 13,663 13,663 0 4.43
FRANCOIS FOUCHARD BERGROME IND 12,237 12,167 70 3.97
GEORGES ANAPHTESE ORMEJUSTE FRONCIPH 11,240 11,220 20 3.64
EDOUARD PETIT HOMME TET-ANSANM 10,679 10,679 0 3.46
SERGE JOSEPH IND 16,280 8,450 7,830 5.28
JOSEPH ROLAND MECKLEMBOURG RNDP 7,571 7,580 -9 2.45
ARNOLD ARELUS MPSN 6,576 6,576 0 2.13
MARC S. BENOIT IND 5,485 5,495 -10 1.78
J.ERIE METELLUS RNDP 5,238 5,169 69 1.70
TOTALS 616,787 603,295 13,492
104 Analysis of the Results

Center

In the Center, the Lavalas candidate for Senate, Jean Marcellin

Hercule Casseus, was credited by the CEP with 5,027 more votes than in the
results released by the BED. With these added votes, he became the first
elected senator instead of the second. Opposition candidates, such as
Gabriel Jolly, candidate for Tet Ansanm, were also credited with more votes
than originally announced by the Hinche BED.

Basing the percentages on the total number of valid votes, no senator

in the Center would have been elected in the first round. According to the
CEP, all Lavalas candidates were elected in the first round. (In the Center,
three senators were elected, which means that the total number of votes
should be divided by three, since voters marked as many as three candidates
on the same ballot.)

CANDIDATES PARTIES CEP BED DIFFERENCE % CEP%

SIMON DIEUSEUL DESRAS FANMI 56,542 56,542 0 45.42 72.99
JEAN MARCELIN HERCULE CASSEUS FANMI 58,463 53,436 5,027 46.96 75.45
PIERRE GERARD MICHEL FANMI 51,965 51,965 0 41.74 67.06
SERGE GILLES EC 24,003 24,003 0 19.28 30.90
SALNAVE DESARMES EC 20,315 20,315 0 16.32 26.21
SALNAVE SILVESTRE EC 21,166 21,166 0 17.00 27.31
GARRY GUITEAU OPL 18,888 17,448 1,440 15.17
MARINO ETIENNE RANFO 14,372 14,372 0 11.55
JOSUE SALOMON OPL 14,244 14,244 0 11.44
JOSEPH HUGUES SYLVAIN IND 10,602 10,612 -10 8.52
JEAN SAINT-LOT IND 10,252 10,252 0 8.24
JOSEPH FIGNOLE JEAN LOUIS CFD 10,128 10,128 0 8.14
PIERRE BOINEUS AUGUSTAVE RANFO 9,406 9,406 0 7.56
EVENS DORCE MOCHRENA 8,621 8,621 0 6.93
PAUL DESTINE MOCHRENA 8,047 8,047 0 6.46
EDZER DESARMES IND 7,656 7,656 0 6.15
ESCARNE BOSQUETTE RNDP 7,455 7,455 0 5.99
REYNOLD GEORGES MPSN 4,772 4,772 0 3.83
GABRIEL JOLLY TASH 11,071 4,771 6,300 8.89
SMITH METELLUS FRONCIPH 2,889 2,849 40 2.32
GREGOIRE GUERRIER FRONCIPH 2,590 2,590 0 2.08
TOTALS 373,447 10,612 300.00
 Analysis of the Results 105

West

There is no discrepancy between the results posted by the BEDs and

those proclaimed by the CEP for the Senate race in the West. Both Lavalas
candidates are elected no matter which method is used.

CANDIDATES PARTIES CEP WEST-I WEST-II TOTAL BED DIFFERENCE % %CEP

DANY' TOUISSAINT FANMI 536,802 447,020 89,782 536,802 0 70.25 92.06
YVON NEPTUNE FANMI 483,782 399,653 84,129 483,782 0 63.31 82.97
MIRLANDE MANIGAT RNDP 81,381 67,258 14,123 81,381 0 10.65 10.64
M.L.J 'LASSEGUE PLB 64,121 41,410 22,711 64,121 0 8.39 8.40
DANIEL SUPPLICE EC 59,757 44,834 14,923 59,757 0 7.82
SUZY CASTOR OPL 43,842 28,379 15,463 43,842 0 5.74
J.R 'LOUIMA MOCHRENA 30,156 25,962 4,821 30,783 -627 3.95
RAYMOND PIOU MOCHRENA 30,034 25,712 4,322 30,034 0 3.93
M.D CLAUDE FRONCIPH 21,889 15,864 6,625 22,489 -600 2.86
E.P CASSEUS MOP 19,926 12,162 7,764 19,926 0 2.61
FRANTZ MONCHAIS MPSN 15,480 8,774 3,020 11,794 3,686 2.03
EVANS NICOLAS URN 15,400 10,474 4,926 15,400 0 2.02
LUCIEN PETIT APPA 14,679 9,589 5,090 14,679 0 1.92
ALIX FILS-AIME IND 13,554 9,975 3,579 13,554 0 1.77
ANDRE CALIXTE FRONCIPH 12,976 8,136 4,840 12,976 0 1.70
JEAN MAX ADAM CDSH 12,891 9,736 3,155 12,891 0 1.69
ELINER CADET RANFO 12,275 8,280 3,984 12,264 11 1.61
M.A MAXIMILIEN PENH 9,462 6,456 3,006 9,462 0 1.24
AMOS GEORGES IND 9,376 6,915 2,461 9,376 0 1.23
A. JOSEPH IND 8,631 6,353 2,278 8,631 0 1.13
J.A BELLERICE PENH 7,970 5,923 2,047 7,970 0 1.04
PHITO CALIXTE MODEJHA 7,403 5,378 2,025 7,403 0 0.97
CLOTAIRE ST NATUS IND 6,258 5,223 2,035 7,258 -1,000 0.82
J.P.R MATHIEU IND 5,169 3,662 1,507 5,169 0 0.68
WILNER PREMPIN IND 5,135 3,713 1,422 5,135 0 0.67
TOTALS 1,528,349 1,216,841 310,038 1,526,879 200.00
106 Analysis of the Results

South

There is no discrepancy in results posted by the BED and proclaimed

by the CEP for Senate race in the South. Both Lavalas candidates are
elected in the first round, regardless of the method to calculate final results.

CANDIDATES PARTIES CEP BED DIFFERENCE % %CEP

JOSEPH YVON FEUILLE FANMI 91,275 91,275 0 54.72 80,79
VICTOIRE MAGLOIRE FANMI 86,866 86,866 0 52.08 79,19
PAUL DENIS OPL 27,748 27,748 0 16.63 21,62
GERALD EXELLENT EC 19,910 19,910 0 11.94 18,38
ANTONY CYRON MOCHRENA 19,444 19,444 0 11.66
PIERRE PAUL MARSAN OPL 19,221 19,221 0 11.52
JEAN GABRIEL FORTUNE IND 18,450 18,450 0 11.06
NELSON FLECOURT EC 14,893 14,893 0 8.93
WILLIAM FRANCOIS CHERESTAL MRN 6,413 6,413 0 3.84
MARC MARIE YVES MAZILE MOCID 5,876 5,876 0 3.52
GARCIA LOURDES RDNP 4,739 4,739 0 2.84
GABRIEL ROLAND CASTOR IND 4,676 4,676 0 2.80
LOUIS RENE MISERE RDNP 4,344 4,344 0 2.60
FRANCK FOUGERE MORPEAU FRONCIPH 3,446 3,446 0 2.07
YVES ROUSSEAU PSR 3,181 3,181 0 1.91
PIERRE FRANCOIS VITAL MPSN 3,130 3,130 0 1.88
TOTAL 333,612 333,612 200.00
 Analysis of the Results 107

Southeast

There is no discrepancy between the results posted by the BED and

those proclaimed by the CEP for the Senate race in the Southeast. Both
Lavalas candidates are elected in the first round, regardless of the method to
calculate final results.

CANDIDATES PARTIES CEP BED DIFFERENCE % %CEP

JEAN MARIE FOUREL CELESTIN FANMI 73,136 73,136 0 57.56 80,67%
PIERRE PRINCE FANMI 68,700 68,700 0 54.07 75,78%
RICHARD PIERRE OPL 20,111 20,111 0 15.83 22,18%
ROSEMOND PRADEL EC 19,357 19,357 0 15.24 21,35%
ANTONY SAMDI MOCHRENA 12,038 12,038 0 9.47
SANTELIAN MARCELIN MOCHRENA 12,013 12,013 0 9.46
SERGE PIERRE LOUIS OPL 10,578 10,578 0 8.33
ELIE ARMAND RDNP 7,370 7,370 0 5.80
WILKENS GILLES PLB 6,910 6,910 0 5.44
MILO GOUSSE RDNP 6,481 6,481 0 5.10
SAINT PHORD BALTHAZAR IND 4,600 4,600 0 3.62
BASTIEN METELUS PENH 4,581 4,581 0 3.61
EMIL PENN MPSN 3,841 3,841 0 3.02
RICOT PAUL MPSN 2,885 2,885 0 2.27
OMAR GARAT MRN 1,507 1,507 0 1.19
TOTALS 254,108 254,108 200.00
108 Analysis of the Results

Grand’Anse

There is no discrepancy between the results posted by the BED and

those proclaimed by the CEP for the Senate race in the Department of the
Grand’Anse. Both Lavalas candidates are elected in the first round
regardless of the method to calculate final results.

CANDIDATES PARTIES CEP BED DIFFERENCE % %CEP

LOUIS GERALD GILLES FANMI 86,404 86,404 0 59.25 76.71%
CLONES LANS FANMI 78,661 78,661 0 53.94 69.83%
JEAN MAXIME ROUMER ESCAMP 33,137 33,137 0 22.72 29.41%
MICHEL CLERIE ESPACE 27,067 27,067 0 18.56 24.00%
JEAN SYLVERA SIMON ESPACE 16,105 16,105 0 11.04
RICHE ANDRIS OPL 15,841 15,841 0 10.86
ARTHUR PAPILLON OPL 8,755 8,755 0 6.00
GEORGES KENOL FRONTCIPH 2,514 2,514 0 1.72
JEAN HERVE BEAUPORT TET ANSANM 7,061 7,061 0 4.84
ZABULON VILMONT MOCHRENA 6,082 4,999 1,083 4.17
MARCEL DESIR RDNP 5,010 5,010 0 3.44
J. VILLEVALEIX LAURENT MOCHRENA 4,999 4,999 0 3.43
TOTALS 291,636 254,108 200.00

Overview

If the Senate seats had been calculated based on the number of total
valid votes instead of those of a limited number of candidates, the senatorial
elections in the Northwest, Northeast, Artibonite, and Center should have
had a second round of voting, as no candidate received the requisite absolute
majority. Moreover, utilizing the results from the BED and calculating
percentages for all candidates, only one candidate in the North (Guerrier)
would have won in the first round. The difference in the results between the
BED in the Northwest and the Center and those of the CEP affected the
order in which the candidates finished as well as the need for a second
round.
 Analysis of the Results 109

Departments Elected senators8/ Elected senators (CEP results)

West 2 Lavalas (D. Toussaint, Y. 2 Lavalas (Toussaint, Yvon
Neptune) Neptune)
South 2 Lavalas (Feuille, Magloire) 2 Lavalas (Feuille, Magloire)
Southeast 2 Lavalas (Celestin, Prince) 2 Lavalas (Celestin, Prince
Grand-Anse 2 Lavalas (Gerald, Clones) 2 Lavalas (Gerald, Clones)
Center 0 3 Lavalas (Casseus, Desras,
Renard)
Artibonite 1, Lavalas (Joseph) 2 Lavalas (Joseph, Sanon)
9/
North 2 Lavalas, (Guerrier, Augustin) 2 Lavalas (Bell, Martineau
Guerrier)
Northwest 0 1 Lavalas, 1 Independent (Luis,
Fleurinor)
Northeast 0 2 Lavalas (Joazile, Pierre-Louis)
Total 11 win in first round 19 win in first round

3. Elections for the Chamber of Deputies

In the Department of the Center, there are discrepancies between the

results posted by the BED and those proclaimed by the CEP. According to
results calculated by the BED in Hinche, there should have been a second
round in three races. According to the CEP, however, the Lavalas
candidates were elected in the first round.

Center

Detailed below are three districts (circonscriptions) where major

changes in the number of votes or percentages resulted in three Lavalas
candidates being elected in the first round. In Maïssade, the Lavalas
candidate was credited with two-thousand more votes than in the BED
results. In Mirebalais, the Lavalas candidate was credited with one-thousand
more votes.

In the Las Cahobas district, the number of votes remained unchanged,

but the percentage obtained by candidates varied from BED to CEP. The
CEP only provided the number of votes of the winner, and hence it is

8
Based on BED results and calculating percentages for all candidates.
9
Utilizing the results from the BED and taking into consideration all of the candidates for the calculation of
absolute majority, Youseline Bell Augustin would have obtained 47.36 percent of the vote and would not
have won election in the first round.
110 Analysis of the Results

impossible to verify its percentage calculation. According to the CEP

results, the Lavalas candidate was elected with 51.84 percent, although
basing the percentages on the total number of valid votes, the Lavalas
candidate only obtained 46.9 percent.

Maïssade

Candidates Parties BED BED CEP results CEP % Difference

votes %
Max F.Lavalas 2,899 38% 4,899 51.6% +2,000
Théodort (elected)
Renold Espace 2,043 27% Not Not available
Beaugé available
J.Ravel OPL 1,089 14% Not Not available
Pierre available
J.R. Mochrena 998 13% Not Not available
Belmaire available
Wilson RDNP 456 6% Not Not available
Joseph available
Total 7,485 Not Not available
available

Las Cahobas

Candidates Parties BED BED CEP results CEP % Difference

votes %
Babeau Lavalas 9,606 46.9% 9,606 51,84% Diff. in %
Villers
G.Lyonel OPL 3,031 14,8% Not Not
Jean available available
Markentz PLB 2,603 12,7% Not Not
Sigue available available
Roland CFD 2,342 11,4% Not Not
Désir available available
C.Denaud Ind. 1,895 9,2% Not Not
available available
Es. Ind. 706 5,4% Not Not
Dubuisson available available
J.F Michel RDNP 275 1,3% Not Not
available available
Total 20,458 100% Not Not
available available
 Analysis of the Results 111

Mirebalais (Saut d’Eau-Boucan Carré)

Candidates Parties BED BED CEP CEP % Difference

votes % results
Olide Bruno F.Lavalas 17,284 37% 28,284 50.2% +11,000
(elected)
Urguintz Espace 5,681 12.4% Not Not
Louis available available
Walk. OPL 5,321 11.6% Not Not
Novembre available available
Y.A. Gardère CFD 4,287 9% Not Not
available available
A. MPSN 2,632 5.7% Not Not
Toussainvil available available
Ravix Emile MPSN 2,342 5.1% Not Not
available available
Jorel Lubin PADH 1,899 4% Not Not
available available
J.R. Jasmin RDNP 1,601 3.5% Not Not
available available
Y. Desrose CFD 1,485 3.2% Not Not
available available
Jeantel FRONTC 1,379 3% Not Not
Joseph available available
E. Ambroise Ind. 986 2.2% Not Not
available available
J.H.Médor KNA 935 2% Not Not
available available
Total 45,832 100% Not Not
available available

South

In the third district, Camp Perrin, J. François Augustin, Lavalas

candidate for deputy, was credited with 5,269 more votes in the CEP results
than in the BED results, even though this did not change his being elected.

According to the results first posted by the South BED for the Port-
Salut district, Jean Robert Placide, Lavalas candidate for deputy, had
obtained 10,266 votes, which meant that a runoff would be required. After a
112 Analysis of the Results

complaint by Placide, the BED organized a recounting, which took place

without the presence of observers nor J.N. Dutreil, the independent
candidate against whom Placide would have run for the second round. After
the recount, Placide was credited with 1,003 more votes, all of them from
the fourth section rurale de Barbois, Port-Salut. Placide thus obtained
11,269 votes, which put him just above the 50 percent plus one absolute
majority. According to the CEP final results, Placide obtained 14,285 votes,
enough to be elected in the first round, with 54 percent.

North

According to the BED results, in the district of Le Borgnes and Port-

Margot, MOP candidate for deputy, Gaston Obas, qualified for the second
round, as he obtained the second highest number of votes, 5,893. However,
the CEP first proclaimed on June 26 that the second highest number of votes
was obtained by the ESKANP candidate, Gabriel Ducatel, who only came in
third, according to the BED results. According to the first CEP results,
Ducatel had obtained exactly the same number of votes as Obas did, 5,893.

Obas challenged the results with copies of the compilation sheets from
both Le Borgnes and Port-Margot and explained that it was a simple mistake
with an inversion of names in the CEP results. On July 3, the CEP issued
new results, giving Ducatel a total number of 5,906 votes. The CEP did not
specify how many votes Obas had received. The OAS Mission verified that
the BED had not sent any new information.

Candidates Parties BED First CEP votes Second CEP votes

votes (after 1st challenge)
Innocent F. Lavalas 8939 8939 8939
Joseph
Gaston Obas MOP 5893 Not communicated Not communicated
Gabriel Eskanp 4894 5893 5906
Ducatel
Fritz Odalma Independent 3848 Not communicated Not communicated
Vickson Espace 2702 Not communicated Not communicated
Severe
Dieufaite PENH 1193 Not communicated Not communicated
Valsonis
Telusma PDI 966 Not communicated Not communicated
Joseph
J.L. Innocent OPL 878 Not communicated Not communicated
 Analysis of the Results 113

Summary of Differences for Elected Deputies10/

District BED results CEP results
Mirebalais (Center) Runoff (ballotage) Lavalas candidate elected
Las Cahobas (Center) Runoff (ballotage) Lavalas candidate elected
Maïssade (Center) Runoff (ballotage) Lavalas candidate elected
Hinche (Center) Lavalas candidate elected Lavalas candidate elected
Cerca-la-Source (Center) Lavalas candidate elected Lavalas candidate elected
PAP 1 (West) Lavalas candidate elected Lavalas candidate elected
PAP 2 (West) Lavalas candidate elected Lavalas candidate elected
PAP 3 (West) Lavalas candidate elected Lavalas candidate elected
Pétionville (West) Lavalas candidate elected Lavalas candidate elected
Carrefour (West) Lavalas candidate elected Lavalas candidate elected
Delmas (West) Lavalas candidate elected Lavalas candidate elected
Cx. Des Bouquets (West) Lavalas candidate elected Lavalas candidate elected
Léogane (West) Lavalas candidate elected Lavalas candidate elected
Petit-Goâve (West) Lavalas candidate elected Lavalas candidate elected
Arcahaie (West) Lavalas candidate elected Lavalas candidate elected
La Gonave (West) Lavalas candidate elected Lavalas candidate elected
Les Cayes-Ile-à-Vache (South) Lavalas candidate elected Lavalas candidate elected
CampPerrin (South) Lavalas candidate elected Lavalas candidate elected
Port-Salut (South) Lavalas candidate elected Lavalas candidate elected
P.-à- Piment- Coteaux (South) Lavalas candidate elected Lavalas candidate elected
Milot-Plaine (North) Lavalas candidate elected Lavalas candidate elected
Limbé (North) Lavalas candidate elected Lavalas candidate elected
Bahon-G.Rivière Runoff (ballotage) Lavalas candidate elected
Total 19 elected deputies 23 elected deputies

E. The System of Addressing Electoral Challenges

1. Legal Framework

Articles 10, 11, and 12 of the Electoral Law address the process of
challenges of election results at different levels. Article 11 establishes a
central litigation office for electoral matters (BCEC) to guarantee the rights
of candidates and serve as a mechanism to review decisions made by the
departmental litigation office for electoral matters (BCED). The BCED
serves as the last instance in challenges related to municipal elections and a
court of appeal for the races for deputy and the Senate. Article 12
establishes that the decisions of the communal electoral office (BEC) related
to the CASEC and ASEC elections are not subject to appeal.

10
Based on information obtained from the BEDs by the OAS Electoral Observation Mission regarding
number of votes and/or percentages of vote obtained.
114 Analysis of the Results

The BCEC reviews the decisions of the BCED. Other challenges that
are not of an administrative nature fall under the exclusive jurisdiction of the
BCEC. Challenges of an “administrative nature” are understood to be those
which seek compliance with the formal requirements outlined in the
Electoral Law. The law also establishes that any challenge presented after
three days will not be accepted.

2. Considerations of the Mission

Given the limited information received by the Mission in relation to

challenges presented and the manner in which those challenges were
addressed, the Mission wishes to make the following observations.

Firstly, it was extremely difficult to specify the term of three days as

by the Electoral Law for the presentation of challenges. The CEP did not
stipulate an official date by which results should be posted by the communal
and departmental electoral offices. Hence, those offices posted their results
at different times or, as was often the case, not at all. Both circumstances
made it difficult to determine exactly at what time the three-day period
began and if this three-day period could be considered valid at all.

During the period of challenges, a significant number of BECs and

some BEDs were closed, making it practically impossible to present
challenges to those authorities. For instance, none of the members of the
BED in the Department of the Center were present during the challenge
period, and the Mission attested that no challenges were reviewed. All
challenges presented to the BED in the Center were eventually transferred to
the BCEC in a sealed envelope.

In other departments, such as the Northeast, the large majority of

challenges was likewise transmitted for review by the BCEC. In the South,
the Mission observed that the challenges were reviewed selectively,
appearing to favor one party. At the level of the BCEC, too, it appeared that
challenges selected for consideration and the eventual decisions were not
determined in a neutral manner. Such was the case in the election for mayor
of Ile-à-Vache, described in chapterVIII.2 of this report, “Regional Incident
Reports from May 21, 2000.”

Despite its various requests, the Mission was not able to observe the
proceedings of the BCED. While the Mission was able to obtain some
 Analysis of the Results 115

information regarding the decisions made at this level, it nevertheless was

compelled to raise serious questions about the conduct of the meetings and
the manner in which these challenges were considered and resolved. The
Mission likewise requested that the CEP provide a list of challenges received
by the BCEC, as well as information on the manner in which these
challenges were resolved. As was the case at the departmental level, the
Mission was unable to observe the proceedings of the BCEC and therefore
cannot attest that these challenges were handled professionally and
impartially, as stipulated by the Electoral Law.

Based on the limited information received, the Mission considers that

the system of addressing electoral challenges did not rigorously follow the
prescript of the Electoral Law and Therefore placed in jeopardy its role as a
juridical guarantor of the electoral process.
 CHAPTER XI
CONCLUSIONS AND RECOMMENDATIONS
A. Conclusions

During its stay in Haiti, the OAS Electoral Mission observed the
electoral process in its entirety. From the beginning, various aspects of the
process appeared problematic. In particular, the many delays in actually
holding the elections eroded public confidence, and frustration mounted
among party leaders, candidates, and their supporters.

The first major challenge for the CEP was the national registration of
voters. Despite the many administrative and logistical difficulties it
encountered, the Mission believes registration constituted a significant
achievement. Practically everyone who sought to register was able to do so.
Thus, it is to the credit of electoral authorities that they extended the
registration period and supplied additional materials in order to guarantee
the citizens’ right to vote. In addition, registration allowed thousands of
Haitians to obtain a photo identification card for the first time.

The CEP faced a number of problems in its preparations for election

day. The lack of communication and transportation infrastructure,
compounded with the inexperience of the majority of electoral authorities at
all levels, including most members of the CEP, poor or no communication of
the CEP’s decisions to departmental and communal authorities, and delays
in the payment of electoral workers, constituted the biggest problems.

On election day, polling took place in an atmosphere of calm

characterized by a high voter turnout (estimated at 60 percent) and a massive
participation of party poll watchers and national observers. Few violent
incidents were reported during the day; hence, the professional behavior of
the Haitian National Police, which responded efficiently to threats of
violence, should be singled out.
120 Conclusions and Recommendations

Nonetheless, the process began to deteriorate at the closing of the

polls. The Mission confirmed that armed groups stole and burned ballot
boxes in the departments of the Center, North, and Artibonite. OAS
observers also reported the chaotic delivery of materials in Port-au-Prince
and Cap-Haïtien, during which a substantial number of materials may have
been lost. Once the compilation of results began, observers noted numerous
irregularities: missing, incomplete, and/or unsigned tally sheets, use of
correcting liquid on tally sheets, and loss of ballots, making a recount
impossible for many polling stations. Most importantly, the Mission
confirmed that the manipulation of results during the postelectoral phase,
including the choice of methodology for calculating the percentages in
determining Senate seats, the faulty transmission of results, and the arbitrary
treatment of challenges, had an impact on the results for several races.
Therefore, the Mission concludes the following:

• The choice of the method of calculating the percentage of votes obtained

for Senate seats violates both the Constitution of Haiti and the Electoral
Law, as it is does not take into account the total majority of valid votes,
excluding thus a significant part of the Haitian voting population and
breaching the internationally accepted principle of one person, one vote.
Consequently, the Mission considers the Senate results flawed.

• The faulty transmission of results led to an irregular proclamation of

first-round results for several deputies. This, coupled with the irregular
conduct of complementary elections in several districts, casts doubts on
the legitimacy of some of the candidates elected to the Chamber of
Deputies.

• The loss of materials and the arbitrary treatment of challenges also had an
impact on several local elections.

• The Mission noted with regret the irresponsibility of some party leaders
who, soon after the election, made unfounded accusations of fraud that
exacerbated the political crisis and the precarious security situation.

• The Mission also took note of the partisan behavior of some mass media,
which often reported uncorroborated and strongly biased information
seeking a maximum political effect.
 Conclusions and Recommendations 121

B. Recommendations

To date, the OAS has observed four separate electoral processes in

Haiti since 1990 and notes that many of the same problems repeatedly arise.
The OAS Electoral Mission believes that the principal cause for the
insufficiencies in the administration of elections stems from the lack of a
permanent electoral body. Once a permanent electoral council has been
constituted in Haiti, it can serve as the institutional memory of the electoral
processes in the country and can be the object of more systematic and longer
term institutional strengthening. In the meantime, the current CEP can learn
from past difficulties and build on its acquired knowledge to strengthen the
process in the areas identified below.

A permanent electoral body, however, is not a panacea. Public

confidence in the electoral process of the country can only be built through a
concerted and coordinated effort among the many sectors of Haitian society
including the government, political parties, and civil society. The
international community, too, should be available to provide assistance, but
only after these sectors have demonstrated the political will to provide for a
permanent, professional, and transparent electoral process.

1. Recommendations Regarding the Institutions

a. An autonomous and independent permanent electoral council,

as required by the Constitution, should be established promptly
after the presidential elections. This council should include
representation of the various political tendencies in the country.

b. The electoral structure should be decentralized in order to avoid

a paralysis of the highest electoral authority due to massive
referrals of decisions that could have been taken at communal
and departmental levels. At the same time, BEDs and BECs
should make use of the authority granted to them and refrain
from transmitting all controversial cases to the CEP.

c. In addition, the CEP should ensure that BEDs and BECs are
provided with the financial and human resources to guarantee
their efficient operation.
122 Conclusions and Recommendations

d. The CEP should ensure that channels of communication among

the various levels of the electoral system allow for a continuous
and unhindered flow of information.

e. The creation of a permanent electoral register should be one of

the priorities of a newly formed permanent electoral council.
Once created, the council should set up a system of ongoing
registration.

2. Recommendations Regarding the Organization of the

Elections:

a. The treatment of challenges should be made systematic, public,

and transparent. Communal and departmental authorities
should be available to receive challenges within the period of
time stipulated by law, duly review them. and rule on them. If
the challenge for a departmental post is sent to the CEP for
review, the complainants should be so informed with a detailed
explanation of the process to be followed by the CEP.

b. The process of selection of poll workers should be made more

transparent. For example, a public session for their random
selection could be held. As a minimum, the list of poll workers
should clearly identify their political affiliation so that
interested parties and observation groups may monitor the
respect for the fair representation of all political parties. This
list should be made public well in advance of election day.

c. Ballots should be numbered and printed on special paper to

facilitate distribution and control. Ballots for all posts should
have colored bands in order to avoid duplication.

d. The assistance provided to illiterate voters should be more

clearly regulated in order to avoid pressure by poll workers and
poll watchers.

e. Procés-verbaux (election records) should include carbon or

chemical paper, in order to produce various copies and to
ensure the authenticity of these copies.
 Conclusions and Recommendations 123

f. To better calculate the absolute majority of valid ballots for

each post, the compilation sheets for communal and
departmental posts should include the total number of valid
ballots cast.

g. The use of private houses for the operation of registration and

polling stations should be avoided. By utilizing public places,
the CEP will not have to pay for the rental of localities. Also,
the CEP will not risk losing the use of these localities if the rent
is not promptly paid.

h. The training program for poll workers should be one of the

priorities of the electoral process.

i. In order to preserve the neutrality of the vote count, electoral

authorities should avoid compiling results in police stations. If
no other alternative is available, the decision to use a police
station should be communicated to all the actors in the electoral
process well in advance.

3. Miscellaneous

a. The CEP, the government, and civil society should work

closely to develop a nationwide, permanent civic education
campaign that would explain the rights and responsibilities of
citizens in a democracy and motivate them to participate
actively in the choosing of their leaders.

b. The CEP should be provided with substantial infrastructure

support and training.

c. The Parliament should consider meaningful campaign funding

and spending legislation to permit participating parties and
candidates to campaign.
124 Conclusions and Recommendations

CHAPTER XII
FINANCIAL REPORT 1999-2000