29139: Dailey re 29133 Holmstead (fwd)

[Bob Corbett <corbetre@webster.edu>]

From: Peter Dailey <phdailey@msn.com>

Last week the results of an elaborate statistical survey of violence and human 
rights violations in Port-au-Prince since Aristide's departure in February, 
2004 was published in The Lancet. The study revealed that out of 8,000 murders 
and 35,000 sexual assaults that it estimated had occurred during this period, 
not a single one had been committed by either the U.N. forces or Lavalas 
partisans.

It then transpired that the principal author "Athena Kolbe" was in fact Lyn 
Duff, a well-known pro-Lavalas hack, and Kevin Pina's partner at Pacifica 
Radio, Flashpoints, the SF Bay View etc, a fact well-known in the Aristide 
camp.

Holmstead's three page attempt at damage control states that concerns had been 
raised by the fact that the author of the study "once wrote articles under the 
name Lyn Duff." This hardly begins to do her justice. During the period- 
December, 2005- when Duff/Kolbe was conducting her survey in Port-au-Prince, 
articles were appearing under her byline such as one in the San Francisco Bay 
view on December 7, 2005 headed "Bloody U.N. siege on Cite Soleil" which begins 
"At least 15 residents were killed and dozens wounded by United Nations troops 
during incursions in the zone of Cite Soleil this past week." Persons curious 
about the scope of her activities can consult the three page entry in 
Wikipedia- presumably submitted by Athena Kolbe- which states that since 
February 29, 2004, Lyn Duff "has regularly covered the situation in Haiti for 
the San Francisco Bay View, Pacifica Radio's Flashpoints and Pacific News 
Service... Subjects have included politically motivated mass rape, the United 
Nations mission in Haiti, killings by American marines in Port-au-Prince, 
civilians taking over the neighborhood of Bel Air, murders of street children 
by police and ex-soldiers, presidential/legislative elections, and the general 
human rights situation."

Within hours, the whitewash of Lavalas partisans and finding that U.N. troops 
had threatened civilians with death and sexual violence- and how seriously 
should such threats be taken if in fact no deaths or sexual assaults ever 
occurred?- were being trumpeted by Brian Concannon, Ira Kurzban, Amy Goodman 
etc. It is probably not surprising that Aristide's attorneys, who have been 
operating a "Lavalas Defense Fund" under the guise of an independent 
non-partisan human rights organization, would see no conflict in the Duff/Kolbe 
situation. (However, for an exercise in hypocritical sanctimoniousness, see the 
same parties comments on the news that a reporter for NPR was also broadcasting 
reports- under her own name- for VOA.)

Which brings us back to the question of what Lyn Duff would have us believe: 
her December, 2005 survey that absolves the UN from any deaths during the 22 
month period or her December 7, 2005 article in the SF Bayview in which states 
that during the past week UN troops had been responsible for at least 15 deaths 
and dozens of wounded? There is no way to reconcile the two.

All of the reports of reputable human rights organizations like Amnesty and HRW 
confirm what everyone even remotely familiar with the situation in 
Port-au-Prince since Aristide's departure knows already: that UN troops, gangs 
of Aristide partisans, the HNP, former soldiers, government security forces, 
and anti-Lavalas gangs have all been responsible for a substantial number of 
deaths and other civilian casualties. How reliable is a survey that fails to 
register this elemental fact? How does it account for the remarkable drop in 
violent crime, kidnappings etc. in the four month period following Preval's 
election?

There are significant flaws in the methodological underpinnings, design of the 
questionnaire, and way the survey was carried out that make much of the data 
worthless. One is the reliance on the "GPS." The 1250 households were chosen at 
random geographically from the greater Port-au-Prince area through the GPS 
(Global Positioning System), a method ordinarily used in geological surveys 
that had apparently been applied to a study of this nature once before. Under 
this system of random selection, a household in Turgeau would be given equal 
weight with one in La Saline. This flies in the face of the general assumption 
that violent crime is greater in poor, densely-populated areas than in upper 
middle-class neighborhoods, and suggests that the number of deaths, sexual 
attacks etc. was much greater than the survey asserts. However, Duff/Kolbe 
states that a comparison of the results from Cite Soleil and those of 
Petionville did not reveal a statistically significant difference. Again, this 
contradicts what everyone in Port-au-Prince knows: that there are "zones of 
danger" official or otherwise.

Another instance was the survey's estimate that 9.6% of all restaveks had been 
victims of sexual assault during the past 22 months. The survey was made on the 
basis of randomly chosen households. Just as the reliance on the "GPS" failed 
to give sufficient weight to the results of households in densely-populated 
neighborhoods, so it failed to give sufficient weight to middle-class enclaves 
in estimating assaults on restaveks. However, an even more fundamental defect 
lies in the way the survey was conducted. In each of the randomly selected 
households, the head of the household or other adult was questioned about the 
experience of all other members of the household. So in other words, the sexual 
assaults on restaveks documented by the survey were only those reported by the 
head of household or family member. If you believe that family members are 
responsible for a significant portion of such assaults, then it follows that 
the figure of 9.6% substantially underreports the problem.

Another problem with the survey is that although it purports to measure levels 
of crime since February 28, 2004, there are, not surprisingly, no similar 
statistics for the period prior to 2004. How are we to interpret these figures? 
Does the number of assaults represent an increase, and if so by how much, etc.

It is hard to see what if any value this survey contains.

I had assumed that the Wayne State Dept. of Sociology, like the Lancet, were 
victims of the Lyn Duff/Kolbe hoax but this was apparently not the case. Duff's 
supervisor and co-author Prof. Hutson is quoted as saying "The charges of bias 
are baseless. We were aware that Athena had written (note: and was writing) 
under another name and found no conflict. Our concern is the way UN soldiers 
are interacting with Haitians." Well, at least the Professor is up front about 
it. Some of the rest of us had been mislead into thinking that the purpose was 
to determine the number of human rights abuses and other criminal violations in 
Port-au-Prince through a rigorous, and impartial, scientific survey.

Peter Dailey