Bayesian: Criminal Investigation Opened

Probability theory, detective work, and a Sicilian prosecutor who smells a rat.

This morning I was delighted to wake up to an e-mail from my friend and favorite epidemiologist, Dr. Harvey Risch (Professor Emeritus of Epidemiology in the Department of Epidemiology and Public Health at the Yale School of Public Health). He wrote to share his reflections on my late night musings about a Bayesian Interpretation of Two Deaths.

John, the answer to this is not statistics, but detective work.  IMO, ask, "how likely would these individuals die--as opposed to being injured--if these events were to happen." … What about all the people on the ship who didn't die?

Seems to me that the methods of murder are not viable enough to have succeeded with little chance of failure, which means that if these had been intentional hits, they weren't all that professional. Or, they were just accidents.

I am glad (and not surprised) that he noted this, and I completely agree with him. Indeed, the main argument of my most recent book—The Meaning of Malice: On the Trail of the Black Widow of Highland Park—is that the Dallas County Medical Examiner attributed far too much significance to statistical data showing that suicide (by a single gunshot to the head) is far more common than homicide (by a single gunshot to the head). I contend that the Medical Examiner should have told the police to do far more detective work on the two violent deaths in question.

I was motivated to write my post last night mostly because I was charmed by the discovery that Thomas Bayes, who formulated Bayes' Theorem, was an 18th century English Presbyterian Minister. Students of the Presbyterian theology are aware that it derives from the theology of John Calvin, who posited the doctrine of Predestination—that is, that all events have been willed by God, and because God’s consciousness is not constrained by time, everything that happens over time was predestined at the moment God created the universe. Thus, I wondered how Minister Bayes reconciled his fascination with probability with presbyterian theology.

As a true crime writer, I have long found probability assessments to be a useful starting point for an investigation. As I noted in my earlier post, Stephen Chamberlain was struck by a car in Cambridgeshire, England, on Saturday, August 17 at around 10:10 a.m. Less than 48 hours later, Chamberlain’s former business partner, Mike Lynch, apparently drowned when his super yacht named Bayesian sank while at anchor off the coast of Sicily.

What are the odds that these former business partners and co-plaintiffs in a recent litigation over billions of dollars would die of unusual accidental deaths within 48 hours of each other? These circumstances strike me as highly suspicious, but as Professor Risch pointed out, the question of possible foul play can only be answered with detective work.

Regarding the Bayesian disaster—many readers and yachtsmen have proposed that special scrutiny be directed as the vessel’s retractable keel, which—according to news reports—was found by divers to be “partially raised” but still intact.

I wonder about this. Had the retractable keel sheered off, this could explain why the boat quickly sank. However, the fact that the keel was intact but “partially raised” does not strike me as sufficient to explain the catastrophe.

The retractable keel enables a huge sailboat like the Bayesian to enter relatively shallow water near marinas and beaches. In deep water, when the sails are set, the keel is fully lowered to provide lateral resistance to the enormous lift generated by the sails. The keel is essentially a wing under water that generates lift (as it moves through the water) that counteracts the lift generated by the sails. These two forces (or vectors) acting together produce a resultant that drives the boat forward in the direction it is being steered, with some leeward drift known as leeway.

Had the weighted keel been fully lowered (instead of partially raised) it would have lowered the vessel’s center of gravity, producing more counter leverage against the force that caused the boat to heel. Nevertheless, it seems to me that the most salient fact in this disaster is that the vessel was at anchor, with sails furled.

And so, the question remains: If the boat was flooded by catastrophic heeling, what exactly caused the boat to heel so violently? Cyclonic wind blowing on the hull and superstructure would have pushed the boat laterally, causing it to swing violently around the anchor. However, the windage on the bare mast strikes me as grossly insufficient to cause such a big and heavy sailboat to roll over, even with the weighted keel partially retracted.

Thus, it seems to me that investigators should consider the possibility that some act of sabotage was performed on the vessel to coincide with the arrival of forecasted bad weather.

Regarding Professor Risch’s observation: “Seems to me that the methods of murder are not viable enough to have succeeded with little chance of failure…”

I believe investigators should consider the possibility that nefarious actors did not necessarily want to murder Chamberlain and Lynch, but rather to send them a message in the same way that Don Corleone sent an equestrian message to Jack Woltz in The Godfather — that is, “No matter how safe you think you are, we can still get to you.” Sicily, the ancestral home of Don Corleone, strikes me as a suitable place for such an operation.

POSTSCRIPT: Right after I wrote this post this morning in Europe, I did a Google search for any possible updates, and saw that Italian police have opened a criminal investigation based on the suspicious circumstances of this disaster. According to a report just published in Palermo Today:

The Prosecutor's Office of Termini Imerese, led by Ambrogio Cartosio, is investigating the offenses of shipwreck and multiple culpable homicide against unknown persons. … Coastguard officials have visited all of the homes and public places with surveillance cameras pointed out to sea in the area.

The Prosecutor’s Office is also reviewing the actions and decisions made by the Bayesian’s Captain James Cutfield, 51, and Matthew Griffith, 22, who is understood to be the first officer.