JFK Witness Deaths: Responding to Warren Commission Apologists
Dec. 28, 2013
Updated: Jan.5, 2014
Click Reclaiming Science:The JFK Conspiracy to look inside the book.
Ever since I first posted on the probabilities of JFK-related deaths, Warren Commission (WC) apologists, trolls and various disinformationists have attempted to refute the logic behind the calculation. After all, if it could be proved that the probability of all these deaths was essentially zero, that would mean there was a “cleanup” operation. It would prove a conspiracy to assassinate JFK and a conspiracy by the mainstream media to cover it up.
This is a summary of apologist arguments and my responses.
The basis for WC apologist talking points is a 1967 CIA memo and the 1977-78 House Select Committee on Assassinations (HSCA). http://www.cassiopaea.org/cass/jfkdeaths.htm
These talking points have been promoted for years by John McAdams, the most prolific “Lone nutter”. He has been totally debunked by Michael T. Griffith: http://www.mtgriffith.com/web_documents/vsmcadams.htm
A 1967 memo from CIA headquarters to station chiefs advised:
Such vague accusations as that “more than 10 people have died mysteriously” can always be explained in some rational way: e.g., the individuals concerned have for the most part died of natural causes; the (Warren) Commission staff questioned 418 witnesses – the FBI interviewed far more people, conducting 25,000 interviews and reinterviews – and in such a large group, a certain number of deaths are to be expected.
London Sunday Times and the HSCA
An actuary engaged by the London Sunday times calculated that the odds against 18 material witnesses dying in the three years after the assassination was 100,000 trillion to one.
The Times Legal manager responded to a letter from HSCA:
The Editor has passed me your letter of 25th April. Our piece about the odds against the deaths of the Kennedy witnesses was, I regret to say, based on a careless journalistic mistake and should not have been published. This was realized by The Sunday Times’ editorial staff after the first edition–the one which goes to the United States and which I believe you have–had gone out, and later editions were amended.
There was no question of our actuary having got his answer wrong: It was simply that we asked him the wrong question. He was asked what were the odds against 15 named people out of the population of the United States dying within a short period of time to which he replied–correctly–that they were very high. However, if one asks what are the odds against 15 of those included in the Warren Commission index dying within a given period, the answer is, of course, that they are much lower. Our mistake was to treat the reply to the former question as if it dealt with the latter–hence the fundamental error in our first edition report, for which we apologize.
None of the editorial staff involved in this story can remember the name of the actuary we consulted, but in view of what happened you will, I imagine, agree that his identity is hardly material.
Antony Whitaker, Legal Manager.
The HSCA Statistician testified and noted 21 deaths had been analyzed: http://jfkassassination.net/russ/jfkinfo2/jfk4/hess.htm
Even though the London Sunday Times had not structured its actuarial inquiry properly and, therefore, the 100,000 trillion to 1 odds were invalid, the committee staff looked into the possibility of conducting a valid study, contracting with our own actuarial firms here in the District of Columbia: Edward H. Friend & Co., Towers Perrin, Forster & Co., and the Wyatt Co.
One, to compute valid actuarial statistics, one must be able to determine to a reasonable degree of specificity, the universe of individuals to which the specific group is being compared. In other words, we would have to determine the total number of individuals who exist in each of the categories into which those individuals who have mysteriously died, fall. This means that we would need to establish the number of individuals who in any manner could be considered witnesses to the assassination of President Kennedy and Lee Harvey Oswald, the number of individuals who had any contact with Oswald or Ruby or with Ruby’s nightclubs, the number of individuals who professed to have material knowledge of the case or of the major figures in the case, all news reporters who had expressed interest, taken interviews or investigated the case, and all Members of Congress who sought to introduce legislation concerning the investigation of the case. This, as you can imagine, would have been an impossible task.
As a result of the above, probability analysis confirming the actuary’s odds has been attacked by WC apologists who also cite books by Vincent Bugliosi and Gerald Posner. The following are talking points used by critics of the JFK witness probability analysis – and my response to them:
– The London Sunday Times actuary’s result was invalid (see Legal Manager and HSCA)
I confirmed the actuary’s odds based on the 0.000207 weighted unnatural mortality rate applied to 459 witnesses. The Times did not show the actuary’s calculation, and did not mention that 13 of the 18 deaths were unnatural. No one at the Times remembered the actuary’s name and there was no record of it?
– The universe of witnesses is unknowable, therefore the probabilities cannot be calculated.
That is untrue. We know how many testified at the WC. We also know the approximate number of witnesses sought by New Orleans D.A. Jim Garrison in the Clay Shaw trial, Church Senate Intelligence and the HSCA investigations.
– Bugliosi noted that that the Warren Commission Index has 2479 names and claimed a Metropolitan Life actuary calculated 1 in 1.2 odds.
But how come George Washington, Abraham Lincoln and other non-witnesses are included in the Index? And how come, just like the HSCA, Bugliosi’s actuary did not consider unnatural death mortality rates in the calculation? Even assuming 2479 witnesses, the probability of 78 unnatural deaths from 1964-78 is a very conservative 2.7E-13 (less than 1 in a trillion) assuming the average national unweighted mortality rate (.000818). The correct solution is E-44 (less than 1 in a trillion trillion trillion) assuming the JFK-weighted unnatural rate (0.000253).
– There were 25,000 FBI interviews. Many would have died.
Where is the list? How many were relevant? The book Who’s Who in the JFK Assassination includes 1400+ names. But, again, let’s assume 25,000 for the sake of argument. The probability of 80 homicides in 15 years is less than 1 in a trillion.
– The mortality rates were not age-adjusted.
Natural causes of death (heart attack, cancer, other) were age-adjusted. But 78 of the 120 deaths were ruled unnatural and were obviously not age-adjusted. A bullet does not know the age of it’s target. http://www.infoplease.com/ipa/A0005124.html
– It is incorrect to assume that the witnesses were connected to the assassination.
It is not an assumption. Sixty-four (64) of the 120 JFK-related individuals JFK Calc either testified or were sought to testify in four investigations. That makes them connected by definition. The other 56 were were not called by design or because they died suspiciously.
– The witness list is self-selected (not random).
Of course, the list of 120 suspicious deaths is selected from an estimated universe of 1400 material witnesses. In a group of 1400 selected at random, just 2 homicides would normally have been expected in 15 years. But there were 77 official unnatural JFK-related deaths (34 homicides, 24 accidents and 16 suicides, 3 unknown). The number of deaths ruled as accidents, suicides, heart attacks and sudden cancers was far beyond the statistical expectation based on corresponding mortality rates. A reasonable estimate is that there were at least 80 homicides.
– The only reason the witnesses are on the list is because they died.
That is just ridiculous for a number of reasons. As mentioned above, 64 of 120 were called to testify. They are not on the list “because they died”. They are on it because their deaths were unnatural and suspicious. They were part of a much larger group of 1400+ who were connected to the assassination. Eventually, all will die. But only 18 would have been expected to die unnaturally from 1964-78. But there were at least 78 official unnatural deaths. The true number is closer to 100.
– Witness occupations were hazardous.
Is that why 7 top FBI officials died within a six month period in 1977 just before they were due to testify at HSCA? Assuming that 20 FBI were called to testify, the probability of at least 7 deaths (5 heart attacks, 2 accidents) in a 6 month period is 1 in 200 trillion!
– The deaths occurred in high crime rate locations. Yes, that is true. FIFTY-ONE of the 120 in the JFK Calc spreadsheet were in Dallas. HOW DO THE LONE NUTTERS EXPLAIN THAT? THEEY WON’T AND THEY CAN’T – BECAUSE THAT STAT TELLS IT ALL.
In 1967, the Dallas population was 700,000. There were 130 murders (a 0.00019 homicide rate). Therefore a reasonable is that there were 400 Dallas murders in the period 1964-66. The Dallas homicide rate was triple the national rate (0.000059). At least 15 Ruby associates (7.5%) out of an estimated 200 contacts were murdered in 1964-66. If 7.5% of Dallas residents were murdered, that would extrapolate to 50,000 murders. Assuming 200 Ruby contacts, they were 400 times (0.075/0.00019) more likely to be murdered than the general Dallas population. Assuming 1000 contacts, they were 80 times more likely.
Using the Dallas rate, these are the probabilities of 15 Ruby contact deaths:
4.9E-27 for 200 contacts (1 in 200 trillion trillion);
9.4E-17 for 1000 contacts (1 in 10,000 trillion). View Ruby contacts in column E. https://docs.google.com/spreadsheet/ccc?key=0AjAk1JUWDMyRdDFSU3NVd29xWWNyekd2X1ZJYllKTnc#gid=55
Assume the universe of 1400 JFK-related witnesses in “Who’s Who in the JFK Assassination”. Officially 35 were murdered from 1964-78 (actually there were about 80). Given the National average homicide rate (0.000084) for the 15 year period, the probability of 35 homicides is 8E-33. To satisfy the WC apologists who claimed the deaths took place in high crime areas, let’s triple the national rate. The probability rises to 1.1E-17 (1 in 80,000 trillion). But that still will not satisfy the Lone Nutters, so lets double the rate – to six times the National average. The probability goes up to 1 in 500 million.
– The analysis has not been peer-reviewed.
This is not a theoretical exercise. it is based on the official and estimated expected number of witness unnatural and natural deaths and corresponding mortality statistics. We know the actual number of unnatural deaths. We only need to calculate the expected number based on the witness universe, time period and average mortality rates. The actual and expected numbers are input to the Poisson probability spreadsheet function. That’s all there is to it. There is nothing to peer-review. It’s just simple math. In any case, the data and the calculations are available for anyone to look at online. So far, no one has even tried to refute the analysis.
– It has been said that I’m an amateur conspiracy “buff”, not an actuary or statistician.
That ad-hominem will not stand. I have a lifetime of experience as a professional quantitative software developer of computer models for aerospace, defense, Wall Street investment banking, foreign banks and major U.S. consumer goods manufacturers.
– Even if the odds are 1 in a trillion, it’s still possible that the deaths could have occurred by chance and there is no connection to the assassination.
This was a serious response from an individual who was obviously math-phobic. I got a big laugh out of that one.