This page includes my comments on risks and links to other sites.
My writings are in black. Plagiarized text is in maroon, sometimes highlighted by me in red.
Go Home.
. . . Articles about risk theory
. . . Does a red light mean "Stop" or does it mean "Go"? Mixed signals.
. . . Escape from a burning building
. . . Head-on car crashes
. . . Last words of pilots
. . . Was EgyptAir 990 a suicide flight?
. . . Safe use of language in flying. English-to-English translation problem: 582 dead.
. . . Thetis: a 1939 sub disaster. A toothpick could have saved lives.
Directory of links about risk, and more articles
. . . Activities that increase chance of premature death by 1 in a million
. . . Behavioral research on drivers and driving ___ Bicycles ___ Causation: law vs science
. . . Challenger tragedy ___ Chances of dying ___ Crash: why vehicles crash page
. . . Richard Feynman ___ How safe are safety devices? ___ Links about risk
. . . Literature Review ___ The Myth of Man the Killer ___ Normal Accidents
. . . Normal Accident at Three Mile Island ___ Nuclear power
. . . number of days lost per lifetime (on average), due to various causes
. . . Other theories ___ Perceived vs actual risk ___ racing drivers -- have higher accident rates
. . . Risk Homeostasis Theory by Wilde ___ Risk resources ___ Risk-theory links
. . . Risk-theory page ___ Safety devices ___ skydiving ___ Smoking does not cause lung cancer
. . . target level of accident risk ___ Target Risk book
. . . Traffic-calming page ___ Various links ___ Warnings can be counterproductive if they exaggerate the risk
. . . Gerald J. S. Wilde [See also Professor Wilde (Queen's U.)]
What would you be saying during the last few seconds when you knew that you were about to be killed? Here's what some pilots said. [ Caution; some of it is not too polite! I wouldn't blame them. ] Go to Plane Crash Info.com .
Go up to Page contents.
Was EgyptAir 990 a suicide flight?
To:
Christopher Dornan
Director, School of Journalism and Communications,Carleton
University, chris_dornan@carleton.ca
Copy to: Editor, The
Globe and Mail, letters@GlobeAndMail.ca
Globe Technology Team editor@globetechnology.com
Re:
"Technical Difficulties" column "When the machines
we build betray us"
by Christopher Dornan, Globe and Mail,
Nov. 11, p. T1. C:\Up\Egypt990 12-Nov-99, revised 17-Nov-99
This message has the following purposes: (a) to thank the Globe for having such skilled people as Chris Dornan on your Technology Team; (b) to give some insight into my reaction to the subject article, as a kind of "lesson learned" for future journalists; and, (c) to indicate why it is difficult to write a letter of this sort which will be short enough to be published, yet long enough to include the supporting logic for the thesis. (No, I can't claim that I did not have time enough to write it shorter!)
An additional reason is to touch on the question of how far a journalist should push the boundary of speculation, especially in the matter of a possible suicide by a respectable professional (pilot), without being accused of sensationalism.
What was intended to be a Letter to the Editor is enclosed in the delimiters "=======" below, unchanged from its date of writing, 12 Nov 99, except for the addition of footnote pointers, "*", "**", etc., added today, 17 Nov.
=============
Was EgyptAir 990 a suicide flight?
(About 951 words to follow, from ~~~ to ~~~.)
~~~~
Thank you Christopher Dornan for your insight into the fallibility of machines and systems. (Column "Technical Difficulties: When the machines we build betray us" by Christopher Dornan, Nov. 11, p. T1.)
You indicated that there were ". only four reasons an airliner suddenly isappears from radar: sabotage, bad weather, pilot error, or catastrophic electro-mechanical failure . ". The word "sabotage" usually conjures up an image of a (usually foreign) terrorist. A possibility that is not usually envisioned is suicide and murder by a "friendly" party.
I propose the possibility that the crash of Flight 990 was intentionally caused by either the pilot* or, less probably, a terrorist. There is no positive evidence for or against either hypothesis. However, during the first eight seconds of its initial dive, air-traffic control does not receive any message, and the recovered data recorder does not sense any failure. Odd? In a hostage takeover, one might expect that one of the crew might be able to transmit a brief distress message, or at least leave a radio transmitter activated, especially considering that it is standard practice for the door behind the pilots to be locked, implying that there could be several seconds of warning that a terrorist was trying to breach that doorway. There is no such indication. An argument favoring terrorism is that there were several Egyptian military officers aboard returning from training courses in the USA. Perhaps Egypt has some enemies? In the case of suicide by the pilot, the pilot could have surreptitiously deactivated the radio transmitter. Therefore, it might appear that suicide by the pilot* is the more probable of these possibilities. As for the probability of a highly controlled professional committing suicide, note that the incidence of suicide among physicians is about the same as for other professionals. The equivalent probability for women physicians is four times as high. We can assume that the FBI is investigating the backgrounds and recent activities of all of the flight crew.
Here is the proposed suicide scenario. The plane is proceeding routinely on course on autopilot, in level flight, at a prescribed safe altitude of about 6 miles. The pilot* decides to commit suicide. He disengages the autopilot. The co-pilot* notices this, thinks that it is slightly unexpected and only somewhat unusual in the early stage of a long flight. It might not seem unusual enough for him to ask the captain why, nor to report it by radio. (The cockpit voice recorder has not yet (as of Nov 12) been recovered, although the flight data recorder has been.) According to the data recorder, eight seconds pass during which nothing very unusual was recorded.
The captain* then pushes the plane into a steep dive. Whether the co-pilot* is in his seat or possibly in the washroom***, for example, he senses what is happening immediately and with great alarm, so much so that he would have voiced his alarm. Why did air-traffic control not receive some such message? Two possible reasons: (a) the co-pilot was out of reach of a radio (in the washroom?***); or, (b) the captain had already disabled the radio. In either case, the co-pilot confronts the captain while the plane continues to dive steeply. It is accelerating in its dive at a rate explainable only by there being positive** power applied, by the captain in this scenario. The two struggle. The co-pilot gains temporary control, bringing the jet out of its dive and returning toward normal cruising altitude. (A crucial piece of evidence would be to know whether or not the radio system had been purposely disabled. Is such data available on the data recorder?) Another explanation for such an ascent might be some automatic altitude recovery mechanism in the absence of pilot action.
The captain then regains control, probably by injuring or killing the co-pilot, and purposely forces the plane into a vertical dive, accelerating** straight down at as high a speed as possible -- a speed high enough that parts of the plane begin to be torn off, as the plane is now operating, under pilot control, far beyond its design specifications. The passengers are now suffering from extreme panic, and from pain because of rapid air-pressure changes. Some of them realize that they have only about a minute left to live. (Descending at an average vertical speed of 300 mph from a height of 5 miles would take only one minute.)
Why would the captain want to accelerate** downward (past 800 mph and the speed of sound) and not just fall at terminal speed (of about only 125 to 175 vertical mph)? For fear of being overpowered and thwarted by the rest of the crew or passengers. Why endure pain, when sudden death offers a painless way out? Another lingering question: why, instead, couldn't the pilot just buy a pistol and do the job alone in his hotel room? Not enough of the "right" kind of stuff? Too hard to buy guns in Chicago or New York? (Maybe gun control can be too strict for some purposes!) Maybe he didn't want his family or an insurance company to realize that the "accident" was a suicide? Why wouldn't the captain kill the co-pilot before starting the initial dive? Perhaps he thought it too repugnant to kill someone directly or was afraid of seeing blood? (A psychologist could probably explain why this is a possibility, unlikely as it might seem.)
In the days of Minuteman missiles, we had systems in place, including dual operators, to prevent such suicides. Why not, somehow, in planes? After all, the toll could have been much higher than 217 deaths had the plane been aimed at, for example, a biological warfare lab or Times Square.
~~~
Signed: Wayne R. Paulson, Ottawa
=============
Footnotes added on 17 Nov.:
* It since appears that it may have been the co-pilot rather than the main pilot (captain) who forced the plane into the dive. (Globe and Mail, "FBI probe of EgyptAir crash put off", 17 Nov., p. A11. The diagram accompanying that article is an excellent adjunct to the text.)
** It appears that once the co-pilot forced the plane to dive, he did not force acceleration, but that the engines were turned off within a few seconds.
*** The cited article* indicates that the pilot was away from his seat when the initial dive was initiated (maybe in the washroom?). This would be the logical time for the co-pilot to initiate his action, giving him a crucial few seconds of delay with the captain away from radios and out of range of physical retaliation, which would have been instantaneous.
My background:
A bit about my background, as it undoubtedly influences my view of the possible cause of the crash. I am perhaps not a "typical" reader. I have a Ph.D. in Nuclear Physics, but have spent my career designing and programming computer systems, and later contracting for their acquisition. I am now retired and happily writing a computer program that exploits fractal image-generation techniques.
I am not a conspiracy buff. I approach life with an air of skepticism, although, I hope, not cynicism. One of my hobbies is to look around me and wonder why we manage to do complex things so very well, but often fail when it comes to simple things. There are some unwanted activities that could be prevented through the use of technology rather than more police and jails. For example, many people complain that our (already very tough) penalties for drunk driving are not tough enough. We have the technology now to make it impossible for a drunk person to drive a car for more than about two minutes. (The "smart" steering wheel that senses skin-perspiration pH values, invented at least four years ago, about which I can supply a letter upon request.) We would not need any laws against drunk driving for the same reason that we do not have laws against operating bicycles that weigh more than ten tonnes. It would simply be impossible. What could be simpler, and cheaper for that matter, than building more expensive courts, prisons, and hospitals?
Anyone is free to use or to publish any or all of the above.
Signed: Wayne R. Paulson, Ottawa
Go up to Page contents.
Safe use of language in flying. English-to-English translation problem: 582 dead.
C:\Up\OCATCL.rtf 30-Mar-00
To: Letters Editor, Ottawa Citizen
Re: Article: "Bilingual flight risky, aviation expert warns" 28 Mar 00, p. A1
Safe use of language --- or --- I'd rather fly skies that are unilingual English!
(About 328 words to follow, from ~~~~ to ~~~~.)
~~~~~~~~~~~~~~
I gain no comfort when Kathy Fox, director of air traffic services for Nav Canada, states that bilingual air-traffic control is not unsafe. ("Bilingual flight risky, aviation expert warns", Mar 28.)
Even when communicating in one language, ambiguity is possible, especially when pilots are operating under stress, monitoring four things at once, over a noisy radio, in clipped language, in which there is often no time to repeat a statement. I offer two examples.
A Brazilian pilot on an approach to a European airport was told: "Set altimeter to 997". He took this as "2997" and set his altimeter to 29.97 (inches of mercury) instead of the 997 (millibars) meant by the controller. The altimeter thus showed the aircraft to be 500 feet higher than it was.
On 27 March 1977, traffic was busy at Los Rodeos Airport on Tenerife Island, with extra flights diverted because of an explosion at another airport. A KLM captain was preparing his Boeing 747 for takeoff, worried that unless he did so soon, he might have to cancel the flight in order to observe strict limits on working hours. At the far end of that runway was a Pan Am 747, taxiing to leave the runway, but not visible in fog. The situation was clear, however, to both pilots and air-traffic control (ATC), all communicating in English. The KLM captain started the plane rolling. His co-pilot told him that they were not cleared to take off. The captain asked him to get the clearance. When asked by ATC for status, the co-pilot replied that they "are now at take-off ". 582 people killed.
What did "at takeoff" mean? To the ATC, it meant that the KLM was ready to take off, but, of course, was not taking off yet, as it had not been cleared by ATC to do so. No problem. To the KLM crew it meant that they were (in the Dutch language) "taking off", i.e., "moving and lifting off".
~~~~~~~~~~~~~~
Signed: Wayne R. Paulson, Ottawa
For a detailed 47-page account,
fasten your seat belts and fly to:
http://aviation-safety.net/reports/770327-2.htm
KLM, B-747, PH-BUF and Pan Am B-747 N736 collision at Tenerife Airport Spain on 27 March 1977
Report dated October 1978 released by the Subsecretaria de Aviacion Civil, Spain, in both Spanish and English
Prepared by Harro Ranter
Source: Aircraft Accident Digest (ICAO Circular 153-AN/56) page 22-68
The KLM pilot was KLM's most experienced, with a 25-year unblemished record (until that morning). Hear the last words of the Pan Am pilot at Last words.
This was the worst accident in aviation history. And language is not a problem?
Thetis: a 1939 sub disaster. A toothpick could have saved lives.
To: Letters Editor, Globe and Mail , 21 Aug 00
(About 196 words to follow, from ~~~~ to ~~~~.)
~~~~
The sinking of the British submarine Thetis in 1939 illustrates how a simple error can lead to tragedy.
On its initial test run on the surface it was found that trying to steer the sub to the left caused it to veer to the right. Odd that this was not discovered in the acceptance test!
On its first test dive the sub proved to be too buoyant to sink properly, so torpedo chambers were checked to see if one could flood more of them, a routine procedure. Of course, one does not want to open the inner door of a torpedo chamber unless one is sure that the outer door is closed. To that end, each chamber had a small hole drilled through the inner wall to let in a small stream of water to signify that the chamber was filled with water. No such water was observed, so the inner door was opened to inspect the chamber. Surprise! Water rushed in, futile attempts were made to close it, and 99 people were drowned. Four survived.
The problem was that, in painting the inner walls of the sub, the small test hole had been painted over.
~~~~
Signed: Wayne R. Paulson, Ottawa
You are free to publish any or all of the above, or as edited, including my name and city.
The irony of this is that a toothpick inserted into the litttle hole by the painters could have kept the hole open. But, did the painters even know what the hole meant? Where were the acceptance-test inspectors? Asleep?
Information source:
The peace time tragic loss of HMS Submarine "Thetis" and 99 men: http://indigo.ie/~mattc/thetis.htm
See also longer version, at: Thetis: a sub disaster in 1939 below.
Thetis: a sub disaster in 1939
To: As it Happen, CBC Radio 1, 22 Aug 00
(About 293 words to follow, from ~~~~ to ~~~~.)
~~~~
The sinking of the British submarine HMS Thetis in 1939 illustrates how a simple error can lead to tragedy.
On its initial test run on the surface it was found that trying to steer the sub to the right caused it to veer to the left. How odd that this was not discovered in previous acceptance inspections! Was this an omen of things to come?
On its first test dive the sub was too buoyant to sink properly, so torpedo chambers were checked to see if one could flood more of them, a routine procedure. Of course, one does not want to open the inner door of a torpedo chamber unless one is sure that the outer door is closed. For confirming that, each chamber had a small hole drilled through the inner wall which would let in a small stream of water if the chamber was flooded. No such water was observed, so the inner door was opened to inspect the chamber. Surprise! Water rushed in, futile attempts were made to close it, and 99 people were to lose their lives. Four survived by means of an escape hatch.
The problem was that, in painting the inner walls of the sub, the small test hole had been painted over, thereby not letting any water through.
There were some additional cruel twists to this story, in that: for some time the stern was projecting from the water; there were many hours of delay in locating the sub, even though it had an escort ship monitoring it; men were still alive inside; acetylene torches were sent for to cut through the hull; and an attempt was made to open a manhole; but the 270-foot sub shifted position and sank 160 feet to the bottom..
~~~~
Signed: Wayne R. Paulson, Ottawa
You are free to broadcast or publish any or all of the above, or as edited, including my name and city.
=======
Information sources:
The peace time tragic loss of HMS Submarine "Thetis" and 99 men: http://indigo.ie/~mattc/thetis.htm
Bike Lanes vs Wide Outside Lanes: http://bicyclinglife.com/EffectiveAdvocacy/blvswol.htm
CRAG Links Page: http://www.pcug.org.au/~psvansch/crag/links.htm . . . links to sites which are in some way connected to bicycle helmet laws . . .
Accidental Death -- Do You Know the Risks: http://www.anesi.com/accdeath.htm Detailed statistics and comments on preceived vs actual risk.
If you're like most people, you are more afraid of an airplane crash or gun accident than an auto accident, or accidental fall.
In fact, auto accidents and falls kill 25 times more people in the U.S. each year than airplane crashes and firearms accidents. Indeed, accidental drowning alone kills far more people than airplane crashes and firearms accidents combined.
So, why do people people grossly overestimate the danger of trivial risks, while ignoring the danger of greater ones?
Psychology gives an answer: the "availability error". Airplane crashes are far more dramatic than auto accidents, and remain more accessible in a person's memory. . . .
For more information on irrational thinking and how to avoid it, consult the works of John Paulos (mathematician), Garrett Hardin (physical scientist), and Stuart Sutherland (psychologist).
Accident statistics: http://planecrashinfo.com/cause.htm Airplane accident causes by category.
Fielding's DangerFinder
- What Danger Awaits: http://www.comebackalive.com/df/dgrawait.htm What
Danger Awaits
the Weary Traveler? Includes tables: International Vehicular Deaths
(per 100,000 population), Causes of Vehicular Accident, . . .
Fielding's DangerFinder - What Is Dangerous: http://www.comebackalive.com/df/whatis.htm
SADVI Example of Quantitative Risk Assessment: http://www.imsa.edu/team/spi/SADVI/sadvi97/riskexample.html Illinois Mathematics and Science Academy. Let's say that tomorrow you are traveling from New York City to Chicago and you want to know what the risk will be that you will die from both natural and unnatural causes. Here is a simple way to evaluate how risky the travel day will be for you.
The Myth of Man the Killer: http://www.tuxedo.org/~esr/writings/killer-myth.html One of the most dangerous errors of our time is the belief that human beings are uniquely violent animals, barely restrained from committing atrocities on each other by the constraints of ethics, religion, and the state. . . . Desmond Morris, in his fascinating book ``Manwatching'', for example, shows that the instinctive fighting style of human beings seems to be rather carefully optimized to keep us from injuring one another. . . .
Individual human beings, outside of a tiny minority of sociopaths and psychopaths, are simply not natural killers. Why, then, is the belief in innate human viciousness so pervasive in our culture? . . .
The historical roots of this belief are not hard to trace. The Judeo-Christian creation story claims that human beings exist in a fallen, sinful state; and Genesis narrates two great acts of revolt against God, the second of which is the first murder. Cain kills Abel, and we inherit the ``mark of Cain'', and the myth of Cain --- the belief that we are all somehow murderers at bottom.
Until the twentieth century, Judeo-Christianity tended to focus on the first one; the Serpent's apple, popularly if not theologically equated with the discovery of sexuality. But as sexual taboos have lost their old forbidding force, the ``mark of Cain'' has become relatively more important in the Judeo-Christian idea of ``original sin''. The same churches and synagogues that blessed ``just wars'' in former centuries have become strongholds of ideological pacifism.
But there is a second, probably more important source of the man-as-killer --- the reactionary naturism of Rousseau and the post-Enlightenment Romantics. . . .
And, in fact, less than one half of one percent of the present human population ever kills in peacetime; murders are more than an order of magnitude less common than fatal household accidents. Furthermore, all but a vanishingly small number of murders are performed by males between the ages of 15 and 25, and the overwhelming majority of those by unmarried males. One's odds of being killed by a human outside that demographic bracket are comparable to one's chances of being killed by a lightning strike.
War is the great exception, the great legitimizer of murder, the one arena in which ordinary humans routinely become killers. . . .
there is a crucial commonality between war and genocide, one not shared with the comparatively negligible lethalities of criminals and the individually insane. Both war and genocide depend, critically, on the habit of killing on orders. Pierson observes, tellingly, that atrocities "are generally initiated by overcontrolled personality types in second-in-command positions, not by undercontrolled personality types."
This is part of what Hannah Arendt was describing when, after the Nuremberg trials, she penned her unforgettable phrase ``the banality of evil''. The instinct that facilitated the atrocities at Belsen-Bergen and Treblinka and Dachau was not a red-handed delight in murder, but rather uncritical submission to the orders of alpha males --- even when those orders were for horror and death. . . .
And this brings us to the final reason for the prevalence of the myth of man the killer; that it encourages obedience and legitimizes social control of the individual. The man who sees every one of his neighbors as a potential murderer will surrender nearly anything to be protected from them. He will call for a strong hand from above; he will become a willing instrument in the oppression of his fellows. He may even allow himself to be turned into a killer in fact. . . .
There is no near-term hope that we can edit either aggression or docility out of the human genome. And the individual small-scale violence of criminals and the insane is a mere distraction from the horrific and vast reality that is government-sanctioned murder and the government-sanctioned threat of murder.
To address the real problem in an effective way, we must therefore change our cultures so that either alpha males calling themselves `government' cease giving orders to perform aggression, or young men cease obeying those orders.
NSC Statistics - What are the Odds of Dying: http://www.nsc.org/lrs/statinfo/odds.htm National Safety Council. Very detailed table re injuries.
Internet Resources: http://faculty.washington.edu/pmay/pols398/398resdocs.html Internet
Resources -- Risk Research
Risk Resources -- Including Sites with Internet Links
. . . GENERAL SITES -- Multiple categories
of risk with Internet links
. . . SPECIFIC RISKS -- Sites about particular
risks, some with Internet links
. . . RISK STATISTICS AND RESOURCES
Governmental Resources -- also see links above
Polls and Surveys
Media and Interest Groups
Library Research -- Information and Links
Search Engines
Relative Risks: Before clicking on this link, see note ** following next paragraph below.
http://muller.lbl.gov/teaching/Physics10/physics%2010%20notes/Risk.html COMPARATIVE RISK.
Bernard Cohen, now a retired professor of physics from Carnegie Mellon University, created two tables to illustrate the relative risk of various activities. For the first list, he used the best known risk estimates, along with the linear hypothesis, to estimate how much of various activities it would take to increase your risk of premature death by 1 part in a million. . . . The list also shows that if you are worried about radiation, it is much safer to live near a nuclear power plant than to live far from such a plant but be in Denver Colorado. (Denver has high radiation because of the presence of large amounts of granite and other similar rock, which has high uranium content.) . . .
Activities that increase chance of premature death by 1 in a million . . .
See note ** a few lines below.
Cohen made a second list, in which he calculated the
number of days lost per lifetime (on average), due to various causes. . . .
** This link is no longer active (as of 7 Mar 03). I shall try to find a suitable alternative. For now, see the extensive list of links at Risk resources.
SkyFry - SKYDIVE!: http://skyfry.com/skydive.php Skydiving! An important reminder for those with a fear of flying: In America, your chances of dying in an airplane crash are 1 in 4.6 million. Your chances of dying in a car crash are 1 in 125. (Source: Institute for Pyschology of Air Travel)
Cellular Phones: Are They Safe to Use?: http://www.the-scientist.com/yr2000/nov/opin_001127.html. The Scientist 14[23]:47, Nov. 27, 2000. This area of research in the United States did not evolve as biological research normally does. It basically had its origin in the physics and engineering community's concern about the hazards of their high-power radio equipment in the late 1930s. . . What must be done to provide the decision makers with a biological input? A sampling of documented events will indicate the answer. The key fact is that the mind-set of those who control the funding determines what is looked at and thus what is found. And this must change . .
Lead candles a health hazard: http://www.straightgoods.com/item420.asp U.S. bans leaded candles while Canada drags its feet in favour of voluntary controls. Straight Goods, Tuesday, February 20, 2001.
PEMBROKE, ONTARIO: Despite action to protect American consumers from health hazards of leaded candles, Canada continues to resist calls to restrict their import and sale. On February 13, US federal regulators voted to ban all candles imported into the United States that contain lead in their wicks. Some candles, especially those imported from China or Taiwan, contain dangerously high amounts of lead. The lead is found in the core wick, a metal piece inserted into the candles to support the wick as it is being burned. Votives, pillars and tea lights, candles that produce puddles of wax, are more likely to contain a lead core than other types. . . .
See also my letter to the editor: Lead poisoning via candle smoke.
pickover.com Clifford Pickover home page. The Risks Digest Volume 16: Issue 18
Plane Crash Info.com: http://www.planecrashinfo.com/
Database of about 2000 plane crashes with statistics, unusual accidents, famous people who died in plane crashes, photos, last words of pilots . . . all kinds of fun, morbid stuff. Click on "Last Words..." for text and sounds, and "Accident Pictures" buttons near top left, and "Warning Sounds" near bottom left.
An unusual one: Photos: A U.S. Army Air Force B-25 crashed into the 79th floor of the Empire State Building in New York in 1945. The crash killed both crew members and eleven people in the building. See also . . .
Cockpit Voice Recorder Accounts of In-Flight Accidents: http://www.ralphmag.org/newR.html. Text only.
The Risks Digest: http://catless.ncl.ac.uk/Risks/16.18.html. Forum on Risks to the Public in Computers and Related Systems. ACM Committee on Computers and Public Policy, Peter G. Neumann, moderator.
Go Home. Go up to Page contents.
You can e-mail me at waynerp@sympatico.ca