A helicopter crew aboard the destroyer USS Kidd in the Indian Ocean, involved in the search for Malaysia Airlines flight 370.
By Emil Pulsifer, Guest Rogue
There will be much Monday-morning quarterbacking on the Mystery of Malaysia Airlines Flight 370. Not here. Over two days, beginning March 17, I predicted where the missing plane would be found and by whom, providing reasoning to back it up. Thereafter I gave frequent updates of analysis and criticism as new developments occurred. Sift through the full record of date-stamped comments, here.
The errors of investigators and of the media reporting on them can be summed up as three logical fallacies: confirmation bias, argument from authority, and argument from ignorance.
Searchers began their efforts in a part of the ocean known to accumulate vast collections of garbage (it's even called a "garbage gyre"), yet the media treated every stray object floating in the water as if it had a good chance to be plane debris instead of almost certainly being garbage, despite repeated disappointments.
Soon the satellite photographs showed many hundreds of objects, an embarrassment of riches. Suddenly, the search shifted hundreds of miles to the north, to an area which, coincidentally, offered searchers far more congenial weather. The mass of objects in the old search area was summarily dismissed, even though most of these remained unexamined. When new objects were spotted in the new search area, the media response remained the same. This time for sure!
Searchers next heard sonar events, separated by distances ranging from 17 miles (the Australians) to hundreds of miles (Australians-Chinese), all having similar frequency and period characteristics. In addition to the distance between detections, which far exceeded the transmission range of the black box pingers being hunted, both parties initially detected them at depths too shallow to be credible; from the surface (Chinese) and from just 300 meters down (Australians), in an ocean nearly three miles deep at the points of detection.
Pundits and the sources they cited invoked vague, handwaving explanations claiming that oceanic "refraction" increased the range of the black box pingers; but the distance from the ocean bottom to the refractive thermal layer at the surface was itself beyond the range of the pingers; travel through the thermal layer itself added miles more; the journey down from the thermal layer to Ocean Shield sonar detection equipment added still more miles; and the oblique angles of this sonar path added still more miles, since by definition refraction is indirect, not direct from source to listener. So, this was conclusively against identification of these sonar events as black box pings.
Black boxes are designed to emit a regular, continuous signal; by contrast, the sonar events detected were stop-and-start, occurring at irregular intervals and lasting for periods that varied widely each time they occurred. They're also designed to emit a fixed frequency once per second; whereas both the frequency and the period of the detected sonar events was off.
Explanations for these additional inconsistencies varied in quality. Some were nonsensical, such as black boxes confined to undersea canyons that cut off line of sight transmissions except from limited areas of the surface, yet involve signals that cannot be replicated from these same surface positions, and are also separated by tens or hundreds of surface miles.
Some were insufficiently documented, such as the claim that oceanic pressures or temperatures changed the frequency: nobody seems to have "done the math" to determine that frequency deviations in excess of ten percent are predicted by science on the basis of the specific conditions known to exist at those locations in the Indian Ocean.
The sonar events heard by both the Chinese and Australians were so regular as to be deemed artificial in origin. But as anyone who has seen an electrocardiogram knows, the heart is an "electronic device" with regular beats and signals, yet is of natural origin.
The Indian Ocean is remote, inaccessible, and inhospitable, being subject to high swells and nasty weather. As a result, many species native to it remain to be discovered, and the behavior of many others has yet to be studied, much less with any degree of thoroughness. We do know that the area hosts exotic species.
It is an argument from ignorance to suggest that merely because we're not familiar with animals capable of producing such sonar patterns, in an area of the Earth's oceans where we've never looked for emissions of this sort, that these sounds are therefore artificial in origin.
My personal speculation runs to schools of ponyfish, which flash on and off in synchrony as a group, quickly and for extended but irregular periods of time, for reasons poorly understood. These are also known to inhabit nearby waters off Indonesia. Perhaps even schools of rare jellyfish. Deep-sea squid do flash each other in the dark, but individually rather than in synchronous mass groups; so that, while their nervous systems might indirectly produce sonar signals of the right frequency range, these signals wouldn't have a definite period, much less the correct period of about once per second.
Soon, the searchers found an oil slick. Imagine: oil floating in the ocean! What could it possibly be, except from the missing plane? Likely, ships or organic residues or almost anything but the plane, which would be out of jet fuel at the end of its assumed flight to nowhere; and of actual oil or other lubricants, the plane would contain a few tens of liters at most: sure to disperse in a vast ocean and not be visible to visual inspection by the crews of search ships and planes; and at the ocean bottom, subjected to pressures so intense as to reduce lighter than water fluids to sludgy solids unable to rise three miles to the surface. (Water is unusual in that, even at deep ocean pressures, it is nearly incompressible, and does not change state.) Yet, they kept talking about "slow leaks" from a submerged plane, and took samples of this substance for extensive analysis.
All of these actions and expectations were conditioned by a model based on the sketchiest of satellite handshake data sent by the plane to an aged Inmarsat satellite that is no longer in true geosynchronous orbit because it lacks the fuel to maintain both latitude and longitude; hence it moves along a north-south path over a 24 hour period, as the sun and moon act upon it.
This could potentially affect the doppler analysis which is at the core of the model, since doppler shifts are determined by relative motion, in this case between the plane and the satellite. The massive inadequacy of the simple handshake data available was filled in by numerous auxiliary assumptions about the plane's (unknown) headings, altitudes, and airspeeds after it disappeared from radar for the last time.
Looking where this jiggered model pointed, searchers found things, which in turn were interpreted as confirmation of the model, regardless of the fact that this required ignoring or twisting their findings to force new data to fit preferred conclusions, instead of tailoring their conclusions to fit the data.
The fact that they kept moving the search area (twenty or thirty times at least), arbitrarily resetting parameters in the hope of getting lucky, might have engendered deep skepticism. Instead, these random resets were called "refinements," with a fawning media and the pundits selected for their panels interpreting this as evidence that those brilliant boffins at Inmarsat really know their business and just kept getting better and better.
Even though the British satellite company never released its basic data, much less the model incorporating them, and even though it was clear from the start that the model was based overwhelmingly on arbitrary assumptions rather than deduction, its authority and expertise was seldom questioned. After all, the international search group acted on the model, so those two authorities validated one another.
All along, it was like a pair of drunks looking for keys in a vast, dark parking lot, with one holding the flashlight and the other crawling on hands and knees, looking only where the tiny beam happened to fall. Every stray glint from assorted pull-tabs and gum wrappers, and every tinkle from someone else's keys in the dark nearby, was seized upon as evidence, no matter how illogically. Because they had never paid attention to the parking lot before they lost something, they never noticed how ubiquitous all of these objects and phenomena are.
If authorities had simply accepted the radar data, which early on showed the plane taking crazy, extremely dangerous maneuvers (climbing beyond the safe maximum altitude, then diving fast for tens of thousands of feet) — maneuvers which no professional pilot or anyone else capable of flying a 777 would deliberately attempt — they might have concluded that the plane itself was in trouble, perhaps being flown by a malfunctioning autopilot (or an autopilot unable to compensate for damage the plane had already sustained), and no longer supervised; hence the crew was already incapacitated or dead, from something which acted either quickly or insidiously enough to give them little or no chance of survival.
Depressurization and hypoxia, or else toxic fumes, are the two most likely scenarios; and fire is the likeliest cause of both. Electrical systems were clearly damaged (i.e., loss of the transponder and ACARS, and erratic flight). The crew could not call for help either because they were unconscious or dead, or because the electrical problems which disabled other systems also disabled the plane's radio. Death from hypoxia or toxic fumes also explains why neither crew nor passengers used their cellphones, even when the plane later entered cellphone range.
The radar data very early on also showed that the plane crossed back over the Malaysian mainland after diverting from its scheduled flight path. This is not something that would be done either by parties attempting to sneak away with the aircraft undetected, or by pilots planning an emergency landing at the nearest airport. Radar then showed the plane turning north (not west or south) in the Strait of Malacca toward Thailand before being permanently lost to radar.
In a part of the world bristling with active radar systems manned by nervous, sometimes paranoid governments jealously guarding their airspace against unidentified aircraft (which is what the airliner was once its transponder failed early on), the fact that the plane was never again detected by anyone, anywhere in the world, strongly suggested that it crashed shortly after its final radar sighting.
Since it was last picked up heading north toward Thailand, not far from its coast, the most natural conclusion was a crash in remote (jungle covered) areas of Thailand, where it would not be observed by non-existent locals, and where the jungle canopy might confuse visual searches by planes or satellites.
The lack of subsequent radar data by the Thai government could be explained by a combination of factors: low altitude as the plane descended to crash and by the geometry of radar stations vis a vis intervening terrain, since active radar operates along strict line-of-sight. This is also consistent with the one, fairly reliable datum released by Inmarsat, the final handshake ping sent by the plane at 8:11 AM, which defines circular arc of possible locations and which passes through Thailand.
Trekkers or perhaps hunters would be the most likely parties to come across the wreckage. One astute reader, Mike G, made the following suggestion in the comments:
I'm still thinking Burma if it crashed in the jungle somewhere. A poor and politically isolated (until very recently) country, most likely with minimal and primitive air defenses, with a dodgy and secretive military government that isn't going to admit that an airliner entered their airspace without challenge. Lots and lots of remote jungle, and big parts of the country are not under control of the central government.
At the time I discounted Myanmar (Burma) since the final ping-arc does not pass through it. But on further reflection, if the position of the Inmarsat satellite was not accurately assessed because of its movement, the ping-arc shown on maps issued to the media would also be inaccurate; and the Isthmus of Kra, just north of where the plane was last spotted on radar, is divided lengthwise into long strips belonging to both Thailand and Myanmar.
Instead, search authorities assumed that the plane kept flying for nearly seven hours, simply because the engine sensors kept sending a simple handshake ping ("I'm still here") once an hour to the Inmarsat satellite. Inmarsat appears to have reversed-engineered all ping arcs except for the final one.
Note the difference between keeping a record of the receipt of ping signals, and having the actual physical signals saved and available; only the latter could be subjected to the doppler analysis which formed the heart of the model. Alternatively, the north-south movement of the satellite, rather than the plane, could have accounted for the illusion of the plane's continued movement.
The idea that, after a crash, the pinger could continue to be powered by either an idling engine or by a backup battery system doesn't seem to have been considered, much less seriously. Passenger jets have been known to crash on rocky mountains without destroying the plane as a whole: one infamous incident in the Andes in 1972 had many passengers survive, only to resort to cannibalism of the dead to feed themselves.
A jungle canopy might help cushion the fall of a low-flying plane in a stall, and woody materials would yield somewhat when crushed by a heavy, metallic object like a 777, which is engineered to withstand the severe structural stresses of maintaining a heavy plane in flight at high speeds through air resistance.
The news media gallingly kept referring to "two" arcs, a northern and a southern, pointedly ignoring the fact that the arcs meet in the middle where the plane was last seen. Elementary search theory: look near the last known location, extrapolating from available data, before searching far afield.
The absence of signals from the plane's emergency locator transmitter (ELT), designed to activate when a land-based crash occurs and capable of sending signals to satellites, can be explained by the unreliability of older versions of the device: their activation success rate is only 25 percent; so in three out of four cases, no signal can be expected.
The statistic applies to accidents rather than individual devices: while the plane is equipped with four ELTs, only two are designed to function in land-based crashes; and the activation of each depends on the same crash conditions involving the same plane at the same time.
Tellingly, when Boeing was asked whether the missing jet was equipped with older versions or the new, far more reliable version, it refused to say. In my view this is tantamount to admitting that the plane was equipped with the older version. Why admit to something that might be unfairly used against you in a firestorm of public controversy, at a point in the investigation when the relevance of this issue had yet to be proved?
Yet, if journalists had demonstrated appropriate professional skepticism, independent critical reasoning, and the kind of persistent digging which is the hallmark of good investigative journalism, much of the confusion could have been avoided.
Instead, they treated the Australian military like tin gods, despite the fact that the Australian Maritime Safety Authority showed habitual technical incompetence by miking international search-update press conferences so that reporters' questions were inaudible.
Even something as obvious as failure to obtain and deploy adequate search resources was interpreted by the media as evidence that the international teams must have "secret knowledge" and as a consequence didn't need to exert themselves.
This was repeated by the media at nearly every step, especially if the steps were baffling. If Angus Houston had suffered an epileptic seizure, I'm convinced CNN would have called it ballet dancing and admired his grace.
Australia is best known for tolerably good wines, a sandwich paste called Vegemite, atavistic mammals, Rupert Murdoch, didgeridoos, and apocalyptic biker movies where toughs inexplicably sport lavender scarves. You locate the cosmic balance point, then explain to me the genesis of mystical wisdom which makes their claims unassailable.
Unreasoning deference to authority by the media is no mystery: it's easier, uses fewer costly resources (since investigative journalism is time consuming and often leads to dead ends on the road to truth), and doesn't step on powerful toes, thus avoiding annoying those whose continued access, favor and cooperation is professionally necessary.
Making a habit of asking sharp questions which might reveal foolishness on the part of the powerful and influential is professionally undiplomatic. Mavericks may be resented even when they're right, whereas supporters of the status quo are appreciated by it for their service even when wrong; and errors are more easily forgotten when they are also made by "everybody else," so being on the wrong side of the fence with mainstream wisdom is scarcely a career ender, whereas bucking the trend and slipping on a banana peel can damage a professional reputation.
Going beyond press conferences and the summaries distributed to the media by authorities who have their own agendas, requires difficult, time consuming investigation and research. In effect, good investigative reporters need to gain expertise on every subject they report.
One lesson to be learned from all this trumps all others: the other lessons will not be learned. History shows they are routinely, repeatedly forgotten or never properly grasped to begin with. If they weren't, we wouldn't be having this discussion yet again. Until next time.
Want to be a guest contributor to Rogue Columnist? Email me at firstname.lastname@example.org