Philosophy of Search

The progress of the search has exposed many of the failings in human nature that tend to saboutage research, particularly when the outcome is extremely desirable and the time is limited.

As the search evolved, the search area jumped to a new location each time a new piece of evidence became available and drew the focus of attention. The technologically advanced the evidence, the more trust it was given and the more resources were comitted to the new search area. However, it seems that very little effort was put into assessing the fundamental assumptions on which each new search site was based.

For example, the search area moved to the southern Indian Ocean because the time delay from the satellite to engine communications provided an arc where the last position might be and the search area was selected by intersecting that arc with a vector representing the assumed cruise speed of a Boeing 777 aircraft. It subsequently emerged (from radar data) that the speed assumption was wrong and the aircraft was travelling below cruise speed (at least for the period on radar) So the search area moved several thousand kilometers.

Later again it emerged that the assumption of a geostationary satellite was also not correct. Due to a decay in orbit, the satellite moves in an ellipse and the "ping arcs" had to be moved to account for this.

Later still, "further analysis" of the satellite data indicated that the aircraft had been travelling even slower and this moved the calculated crash site further north again.

Each time a new calculation came to light, it seemed more certain than the last and more resources were focussed more tightly on the new search area. When acoustic recorders picked up a number of strong and consistent signals, in the newest search area Confident statements were made that the signals were almost certainly from the lost aircraft and it would bo only a matter of days before an autonomous submarine with sidescan sonar would locate the wreckage.

Alas, after 2 weeks the submarine had scanned the entire bottom area where the acoustic signals had been recorded and found nothing. The search agencies went very quiet after that and information releases all but terminated.

Unfortunately, similar stories are played out in mineral exploration programmes worldwide on a regular basis. Theories are proposed, and assets focussed on targets based on remotely acquired data with a host of built in assumptions that are never tested and alternatives not considered to select the most probable target.

Part of the problem is the classic research trap. If a researcher proposes a theory and it seems to explain the facts at hand, there is a strong human tendency to search for evidence that supports that theory, to ignore facts that contradict the theory and to stop looking for alternative theories that might do a better job of explaining the facts. The more the outcome of the theory is desired, by the researcher, the stronger the motivation to ignore contradictory evidence.

In the case of the acoustic signals, a number of very strong contradictions were ignored and alternate sources for the signals were not considered.

The pings were recorded after the end of the predicted battery life for the black boxes.

The frequency of the signals was well outside the known range transmitted by the acoustic pingers on the black boxes.

The signals were heard for periods longer than would have been possible for a stationary object on the sea floor passed by a towed receiver.

Any one of these contradictions should have initiated investigation of alternate theories, but all 3 combined should have raised a very large red flag.

The search has fallen into a series of research traps because the outcome is desired so strongly and there is no external peer review to assess proppossed theories and propose alternatives. Crowd sourcing could provide this kind of review , but as discussed here, the search agencies have been too scared, proud or uninformed to take advantage of this resource.

From the point of view of the explorer (or any other researcher) the takeaway lessons are:

1. Contradictory evidence should be sought with equal vigour as confirmatory evidence.

2. When contradictory evidence is discovered, investigate alternative theories.

3. Have all theories reviewed by external peers before comitting resources to targets.

4. Dont comitt all resources to a single target.