Numb3rs
Season 5
Episode 09: Conspiracy Theory

In this episode, the main theme involves conspiracy theories.

What are conspiracy theories?

At approximately 12:30PM (CST) on 22 November 1963, United States President John F. Kennedy was shot and killed while being driven through a crowded street in a convertible in Dallas, Texas.  The investigation of the assassination was not carried out with the level of rigor that is expected in the current era (especially considering the president was involved), both because standards were lower at the time and because many of the forensic technologies commonly used in criminal investigations had not been developed.  A number of bullet fragments were collected from the scene and another bullet was collected later in the hospital where JFK was pronounced dead on arrival.  A gun was discovered in a nearby building along with three shell casings.  Only a single assassin, Lee Harvey Oswald, was investigated, and he was taken in by police at 2PM on the day of the assassination after allegedly killing a police officer.  Oswald was, curiously enough, assassinated the following morning and was never put to trial.  Additionally, as the investigation was a federal matter, many documents were closed to the public, and a small number of documents remain closed to present (2009).

To many, the assassination seemed too complex for Oswald to have acted alone, and many people believe he must have been involved in a conspiracy to kill the president, meaning he had a number of individuals working with him (either as additional shooters or in the planning process).  The investigation was also too sloppy for many to believe it, especially when compared to modern investigations.  The trajectories of the bullets, the number of bullets obtained from the scene of the crime, and even the number of assassins remains questionable.  Many believe that the documents which are still closed contain information which would debunk the investigation's results.

However, it is seemingly possible that a single individual can produce results like this.  On 19 April 1995, Timothy McVeigh, a veteran of the first Iraq War, blew up the Alfred P. Murrah Federal Building in Oklahoma City, Oklahoma.  The bombing killed 168 people, injured more than 680, and caused hundreds of millions of dollars in damage to the surrounding area.  Until the World Trade Center attacks in 2001, the bombing was the most violent terrorist attack to ever take place in the United States.  There is some evidence which suggests that McVeigh may have had accomplices, but in an interview McVeigh responded to such claims with, "You can't handle the truth. Because the truth is, I blew up the Murrah Building, and isn't it kind of scary that one man could wreak this kind of hell?"

Causes and problems of conspiracy theories.

Conspiracy theories are typically the result of interpretation.  The amount of information collected from an assassination or a terrorist attack is astronomical.  Hundreds of witnesses are interviewed.  Footage from dozens of security cameras is obtained and analyzed.  Autopsies must be undertaken.  Thousands of pieces of evidence must be carefully cataloged.  Typically there are procedural laws which must be obeyed in how all of this information is collected.  There are also laws determining if and when information becomes publicly available.  Since human beings make mistakes, mistakes are made, evidence is mishandled or lost or not collected at all, and witness testimony changes (for a variety of reasons).  Constructing a time-line is difficult because timestamps on the video footage may not line up exactly, testimony from different witnesses may conflict, and huge amounts of time may be completely unaccounted for.

So, some key evidence (the number of bullets, for example) may be questionable.  Or, it may be that all the evidence seems sound.  However the sheer magnitude of evidence allows for drastically different interpretations.  Numb3rs references the Graham Allison Rationality Theorem, which essentially states that, given any action by a government, one can, after the fact, come up with criterion so that this action seems to be the best possible choice.  One can see such behavior surrounding the United States' second war with Iraq.  Prior to the war, the US position was that Iraq was amassing powerful weapons.  However, this turned out not to be the case.  So the government changed its position.  The goal, then, was to free Iraq from a tyrannical regime and bring democracy to an oppressed nation.  Thus, even though the original argument for going to war turned out to be false, the government remains in the position of having made the correct decision.  That is to say that it is not hard to paint any action by a government to seem like it was the rational, necessary choice.  While not an exact analogy, there is a sort of relation to conspiracy theories: by weighting different evidence differently, one can make a particular sequence of events leading to an assassination or terrorist attack seem much more plausible than another, and vice-versa, depending on how important one decides some evidence is.

So, in any investigation, there is a huge amount of subjectivity in how the evidence is weighted.  During an investigation of a serial killer, it may seem logical that it is important to deeply understand the psychology of the killer and study very carefully the way in which he or she kills.  However, it is often the case that seemingly trivial errors on the part of the killer rather than their deranged thinking which wind up being the killer's undoing: David Berkowitz, known as Son of Sam, was caught because he got a ticket for parking too close to a fire hydrant around the time and in the area of a murder.

An additional problem with conspiracy theories is that data can lead to correct but seemingly paradoxical conclusions, causing some people to completely disregard evidence because it doesn't make sense to them.  This is referenced in Numb3rs by Simpson's Paradox.  The paradox can be exemplified by the following example (also referenced in Numb3rs): in 1995, 1996, and 1997, Major League Baseball player Derek Jeter (of the New York Yankees) had batting averages of .250, .314, and  .291, respectively.  David Justice, of the Atlanta Braves (in '95 and '96) and the Cleveland Indians ('97), had averages of .253, .321, and .329, respectively, for the same three-year period.  So in every individual year, Justice had a higher batting average.  It thus seems "obvious" that Justice was better at hitting over those three years than Jeter.  Yet over the whole three-year period,  Jeter had a higher average.  Justice had 104+45+163=312 hits out of 411+140+495=1,046 at bats.  So, his three-year average was .298 (a very, very good batting average).  However, Jeter's average over all three seasons was (12+183+190)/(48+582+654)=0.300 (slightly better).  This is because the three-year average is not just the average of the single-year averages.  Jeter's poor performance in 1995 was the result of far fewer at bats as his good seasons in '96 and '97, and Justice's bad year in 1995 had far more at bats than his good performance in '96.  In 1997, Jeter had more than 4 times as many at bats as Justice.  So as far as overall average over the three seasons is concerned, Justice's 1995 was worse than Jeter's 1995 and Justice's 1996 was not as good as Jeter's 1996.