#4425: Elections & the CEP : Kozyn comments

[Robert Corbett <bcorbett@netcom.com>]

From: John Kozyn <jckozyn@hotmail.com>

I read with some interest most of the posts made to this virtual community 
regarding 1) the urgent need to hold legislative and regional elections in 
Haiti as soon as possible - as the State Department noted repeatedly - there 
was "election fever"; 2) reports from both journalists and accredited 
observers all saying that the elections took place peacefully and 
successfully; and 3) charges that the CEP did not tabulate the vote 
properly, placing its declarations (faits accomplis) at odds with 
interpretations made by some mebers of the international community.

I don't mind reminding some people of the last paragraph in Article 1 of the 
1999 Electoral Law <http://www.haiti.org/lwaelek9901.htm> which states as 
follows:

"Il est le Contentieux de toutes les contestations soulevées à l'occasion 
soit des élections, soit de l'application ou de la violation de la Loi 
électorale sous réserve de toute poursuite légale à entreprendre contre le 
ou les coupables par-devant les tribunaux compétents."

Others here are more competent than me translating French, but my rough 
reading would indicate that this means that the CEP is the final arbiter in 
any and all disputes regarding the electoral process even in the event of 
_any legal arbitration_. Period. Not the OAS, not the UN, not Canada, not 
France, not the USA, not any NGO, nor any individual. (The pontificating I 
have read here recently is, well... just that).

Some of you may seize on my categorization above and retort, "But Mr. Kozyn, 
you cite the 1999 Electoral Law for justification of your argument, yet the 
CEP rejected its own Articles when it came to tabulating the vote!"

There was no rejection of the spirit of the law, but unfortunately -  it 
seems to me anyway - the law itself had weaknesses when it came to 
tabulating the vote in the senatorial contests. I imagine that the law's 
articles in question weren't addressed fully in their drafting because a 
certain precedent had already been set IIRC in 1990 and 1995. This was a 
practice the OAS and other international observers were certainly aware of.

What follows are remarks (translated - no, not by me) which are attributed 
to the CEP Director of Operations. It is a bit long, so if you're not 
interested anymore feel free to delete this post now :)

=======

The methodology employed to count the votes in the first round of Senatorial 
elections was the same methodology employed by the CEP in previous elections 
  when more than one Senate position was open in each departmental race in 
an election.  The methodology was used in 1990 and 1995 elections without  
objection either within Haiti or by the international community.

The Constitutional Mandate under Article 197

The Electoral law of July 1999 provides under Article 64 that to avoid a 
run-off in an election a candidate for the Senate must obtain an absolute 
majority of votes which is defined as 50%+1 of the valid votes.  Article 197 
of the Constitution of the Republic of Haiti vests the authority to resolve 
all disputes "arising either in the elections or in the enforcement or the 
violation of the Electoral law" with the CEP.  Unfortunately, neither the 
Election Law nor the Constitution describes how to compute 50%+1 of the 
valid vote when determining the winner in a contested election. The question 
that arose in this election is what methodology should be used to determine 
whether candidates had won the 50% +1 of the valid vote for the purposes of 
avoiding a run-off election.

The Methodology for Single Member Elections

Obviously, where there is one position for each separate election contest, 
the methodology is easy to understand and to apply.  The CEP simply tallies 
all the votes for all the persons running for that office and determines 
whether or not any candidate obtained 50% +1. In this situation the 
denominator is the total number of votes cast and the numerator is the votes 
cast for the individual candidate.  Thus for the deputies, where there was a 
separate election for each post, this principle was applied without any 
difficulty.

Under Article 95-3 of the Constitution one-third of the Senate is supposed 
to be elected every two years.  Thus only one of the three Senate seats from 
each department would be contested.  This means that all the Senate 
candidates in each department would be vying for a single seat and their 
constituents would only vote for one Senator for that position.  Here again, 
the principle of "absolute majority" is clearly understood because we would 
simply tally all the votes and determine if any candidate had 50% + 1 as we 
have done in the election for deputies.  Due to a substantial number of 
vacancies in the Senate, however, the May 21, 2000 election required that 
the Haitian people elect more than one Senator from each department.  As a 
result the methodology that would be appropriate where only one Senate seat 
per Department was contested was not appropriate in this election.

The methodology used for one senatorial election per Department does not 
apply where more than one Senator is being elected in each department. It is 
obvious that the method used to calculate a majority for a single seat 
election cannot apply where multiple candidates are running at large for two 
or more Senate seats in one department.  In a two-candidate race no 
individual Senate candidate could obtain a majority if all the votes were 
simply tallied and divided among the candidates.  For example, let us say 
that there are 100 people voting in an election for two Senators in a given 
department.  This means that if all 100 people voted for the two positions 
there would be 200 total votes.  If each person voted for two different 
candidates, than no candidate could obtain a majority when all the votes are 
tallied and divided.  If there are six candidates running in an election in 
the election and one is extremely popular and everyone votes once for that 
candidate he or she will obtain 100 votes.  However, the same people will 
divide their votes for the second seat among the reaming 5 candidates. Under 
this scenario the person who obtained a vote from every single person in the 
department would still not have a majority because absolute majority would 
be 101 and he would have only 100 of the 200 votes cast.  Thus, the process 
of simply tallying the votes of the candidate as the numerator would never 
yield a winner on the first round unless people did not vote for  two 
candidates in a Departmental election.

The OAS Proposed Solution

To avoid the problem of never having a majority where two Senate positions 
were contested in a department, the OAS Observer Mission, through a letter 
sent by the Hon. Orlando Marville, recommended that the number of votes be 
divided by two in order to see if anyone obtained more than 50% of that 
number. Thus, Mr. Marville's  methodology, when applied to the above 
example, would result in dividing the total votes cast in half and using 
that number as the denominator.  This would treat the departmental election 
as if it were two separate elections between which votes were evenly 
divided.  Under this methodology we would be required to treat the total 
number of voters as 100 for each election in the above example.  Any Senate 
candidate who obtained 51 votes or more would then have won a majority and 
would not be required to have a runoff.

The problem with the OAS methodology is twofold.  First, it assumes that 
each  department is holding two separate Senate elections, when, in fact, it 
was holding an at large election in which all the candidates in the 
department were running for all the Senate seats.  Second, it assumes that 
everyone voted two times in the election in the department.  That is why 
they are suggesting that the total number of votes be divided in half and 
that that number be used as the denominator.  We know that this was not the 
case.

In all Senate elections in each department many people did not vote for two 
candidates even though they were allowed to vote for two different Senate 
candidates.  Under the OAS formula, the CEP would be required to assume that 
everyone voted for two candidates when clearly this was not the case. Using 
OAS methodology would introduce phantom votes in the calculation of the 
denominator of one Senate seat or another when they did not exist.  It would 
both dilute the vote of a person who voted for only one candidate, and at 
the same time, allow for an improper distribution of votes.  The CEP 
believes that it could not utilize this methodology under the Haitian 
Constitution
because it would determine the outcome of an election using an inaccurate 
percentage of persons as if they had voted in each of two elections when in 
fact they made only one selection.

The CEP Methodology

The methodology used by the CEP was simply to take the top four candidates 
where there were two positions, add their votes, and determine whether any 
person had a majority in relation to the total of the four candidates' 
votes. This parallels the procedure used without objection in 1990, when 
three Senators were to be selected in each  Department and consequently the 
CEP added the votes of the top six candidates, as the denominator in the 
fraction used to determine the percentages won by each and to see who had 
won an absolute majority.  Admittedly this methodology also affects the 
percentages of the total votes by discounting the marginal candidates who 
obtained a small number of votes in the election.  But it comes closest to 
following the intent of the law's reference to an absolute majority of 
"valid votes" and does not rely on a distribution of votes that may not 
exist, as does the OAS methodology.  The criteria the CEP used to determine 
a 50%+1 requirement for a first round victory is consistent with the 
electoral law, because a person must still obtain a majority of the votes 
cast among all but the most marginal candidates in the elections to be 
declared a winner.  This was the methodology used in all previous elections 
where there was more than one Senate seat up for election.

=======

I know that this was somewhat turgid to digest, but I'm given to understand 
this accurately represents the position of the CEP.

Now, I certainly can't speak to the defection of the once-president of the 
CEP, Mr. Léon Manus, but I was hoping to be able to provide a link to the 
website of the Center for International Policy for a facsimile of his angry 
letter to the OAS in response to their questioning of the CEP's tabulation 
in the first place. Alas, no such letter was to be found, which is 
interesting given that Manus' language in that letter seems opposite to what 
he seems to be saying these days. (If someone has an English translation of 
that letter handy, I think a lot of people would be interested to read it to 
see how much Mr. Manus' tone had changed and how dramatically opposite his 
remarks were.) Apropos, I'm looking forward to reports from Cortbett list 
participants when the CIP trots out Manus at their invitation-only meeting 
of the "Haiti Study Group" tomorrow, June 28, 2000.

Anyway, no matter. These elections were not anyone's but the Haitian 
people's. As I indicated earlier and as was reiterated elsewhere above, two 
legal implements cite the CEP as the final authority in any electoral 
dispute. Not any other country - or its emissaries - nor any other foreign 
organization.

I might only wish to add that Fanmi Lavlas won, what? about 30 seats in the 
Chamber of Deputies contests according to preliminary figures I saw. At 
first glance, this sure doesn't sound like any "obliteration" of the 
opposition to me! In fact, given that there are 83 seats up for grabs, I 
wonder why the "opposition" doesn't crow about this - I mean, were it me, I 
might.

John Kozyn

________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com