When Amtrak’s Coast Starlight train cuts through the unspoilt coastal countryside of the Vandenberg Air Force Base, most eyes turn to the right, to pick out the missile launch platforms and the late Space Shuttle’s emergency landing strip. A few eyes turn to the left, however, and look for a series of rocky outcroppings jutting into the Pacific, known collectively as Honda Point. It was here, on 8 September 1923, that seven destroyers, travelling at full speed, ran aground. Two other ships grounded and twenty-three sailors died. It was the U.S. Navy’s largest peacetime loss of ships. Though Christian Morel does not cite this particular case in his seminal Décisions absurdes (Gallimard, 2002), it was also a perfect example of an absurd decision – a term, as a concept, invented by Morel.
Fleet Commodore Edward Howe Watson, on the flagship destroyer USS Delphy, was leading a squadron of fourteen destroyers, all less than five years old, south from San Francisco to San Diego. The squadron was performing an exercise simulating wartime conditions. Despite heavy fog, Commodore Watson therefore ordered the ships to travel in close formation, each behind the other. They were navigating by dead reckoning – estimates based on charts and headings and speeds, as measured by screw revolutions per minute. (The coast sported notoriously treacherous rocks and shallows but to have taken soundings would have slowed their speed.)
At 21.00 the ships turned east into what they thought would be the Santa Barbara Channel. Radio navigation equipment was just coming into use and was not yet entirely trusted. The radio direction finder (RDF) on the USS Delphy indicated different bearings but the navigator and the captain assumed it was unreliable and ignored it. In fact, the RDF was right. The USS Delphy had turned east several miles too soon and ran at 20 knots onto the rocks. In a tragic scene worthy of the Keystone Cops, ship after ship ran aground or turned too late and ran onto other offshore rocks. At the subsequent mass courts-martial, due account was taken of a series of mitigating factors – the wind and the waves, the fog and the darkness and, above all, significant and unpredictable changes brought about in the local currents by the 1 September Great Kanto Earthquake in Japan. To the commendation of his peers, Watson accepted full responsibility for the errors and was relieved of his seniority.
According to Morel’s definition, absurd decisions are drastic and persistent mistakes. The actors involved behave consistently with a certain logic (in the example above, the calculation of positions, distrust for unproven technology, the following of orders) but effectively against the desired aim (to get the fleet back to base intact). In his study Morel singled out a dozen such decisions and, based on detailed analyses, attempted a classification and an explanation. All of Morel’s chosen examples demonstrate two common aspects; the drastic nature of the consequences of the decisions, and the persistence shown by the institution or organisation concerned with the original decisions. The cases he examines range from the 1986 Space Shuttle Challenger disaster (the explosion at lift-off being the result of a problem that had been identified five years previously) through to a company that for many years carried out internal opinion surveys drawn from sample populations that were too small for any valid conclusions to be drawn. They include a plane crash due to a correctly functioning engine being turned off (whilst the malfunctioning engine was left on) and a crash between two oil tankers where each had, by taking evasive action, steered itself into the path of the other.
Morel conducts his analysis from three perspectives. The first is the logic, the ‘cognitive interpretation’, underpinning the wrong decision itself. The second is the collective behaviour that locks the concerned organisation into an ‘absurd solution’. The third is the teleological explanation showing ‘senselessness’ at different stages of the action. The study’s parting shot – and perhaps Morel’s most surprising finding – is the great level of social tolerance for such absurd decisions (witness the way in which Commodore Watson was demoted and yet commended).
With regard to the errors in the original decisions that Morel studies, technology is almost always involved and either mistrusted or misunderstood or, just as frequently, passed over in favour of human logic (for example, turning off the good engine because, despite what the instruments say, the crew thinks it knows better). Organisational, or technological, deference is frequently as much to blame as hierarchical arrogance, for the erroneousness of the initial decision is not enough by itself; it has not only to be respected but reinforced. Indeed, in some cases Morel even identifies ‘an interactive process which forces a group of protagonists into an absurd solution, whereas each individual person was against the collective decision initially.’ Morel considers various models of such collective behaviour, each involving managers, experts and what he terms ‘ingénues’. But what all of the models have in common are two factors; a leader (or leaders) and a hierarchical organisation. Bureaucratic organisations, strong in both factors, are particularly prone.
Morel identifies communication, or the lack of it, as another key factor in the collective process. In hierarchical bureaucracies the voicing of dissent, or of doubt, can be regarded as being disloyal or as simply being inappropriate and may put career prospects at risk. In such organisations – particularly when they are contested entities or engaged in evolutionary transformation – group cohesion is at a premium and information should always permeate slowly upwards, through the voies hiérarchiques. Within such organisations silence, Morel points out, does not necessarily signify agreement, nor even acquiescence, and can therefore be misleading.
Morel also points to another common aspect of the communication habits of bureaucratic organisations (a category in which he includes airlines), which is the premium placed on understatement and on sang froid. Who can easily forget Captain Chesley B ‘Sully’ Sullenberger’s calm, clipped language as he landed his stricken passenger plane on the Hudson River in January 2009? In one of Morel’s examples, the pilot of a plane repeatedly reminded the Control Tower in similar calm language that his aircraft was running out of fuel. The absence of any sense of urgency in his communication led to the Control Tower underestimating the risk until, fatally, the plane did run out of fuel and crashed.
At the teleological level, Morel shows how easy it is for decisions and the processes that follow them to lose their meaning. Objectives are ill-defined. Evaluation exercises are directed to quantifiable aspects of an activity even though those aspects may be tangential. Organisations find it impossible to accept that situations do not necessarily have solutions, and therefore create ‘solutions’ which may be irrelevant, useless or counterproductive.
The main operational recommendations that can be drawn from Morel’s analysis – that leaders/managers should recognise that they are capable of taking absurd decisions whilst sincerely believing them to be absolutely right, and that leaders/managers should encourage reasoned dissent and consider such dissent carefully before proceeding with any decision – are not easy to learn and might even, if they encourage self-doubt or undermine authority, be dangerous. The absurd decision is not, therefore, about to go away, but it is as well to know that it exists and to have some idea what it might look like. The ‘good news’ for senior managers is that, as Commodore Watson’s example shows, organisations are remarkably tolerant about such decisions and their consequences.
Christian Morel, Décisions absurdes, Gallimard, 2002
Les amis de l’Ecole de Paris published a report in English of an interview with Christian Morel that summarised the main themes of his book. It is available at: http://christian.morel5.perso.sfr.fr/English%20report.pdf.