DESTRUCTIVE OBEDIENCE
DESTRUCTIVE OBEDIENCE
THE MARTHA’S VINEYARD
INCIDENT AND STUDY
Several years ago, a
small commuter aircraft crashed on final approach to the resort community of
Martha’s Vineyard in visual conditions. In the investigation, it was discovered
that the captain who has been flying the approach, had died at the controls
from a massive heart attack. He was known as an authoritarian captain who
welcomed very little input from his first officers. This fact when combined
with a First Officer (FO) who was on probationary status and striving very hard
to please the captain, created an atmosphere not conducive to delegated authority.
The FO who was not aware of captain’s recent demise (he merely slumped over the
controls), simply sat there on his hands, and allowed the aircraft to fly into
the ground. Although this seemed to be an extreme and isolated incident, this
“hands off at all costs” phenomenon so intrigued researchers that they
conducted a study to see how many, if any, first officers would think and act
in the same manner as the Martha’s Vineyard FO. The results were nothing short
of astounding. Dozens of pilots reacted in exactly in the same way, refusing to
take the controls as the aircraft plunged towards the terrain. (From the
book Culture, Environment and CRM by Tony Kern).
AUTHORITY BIAS (AB)
Authority Bias (AB) is a cognitive bias that
makes people predisposed to believe, support, and obey those that they perceive
as authority figures. Obedience may be rooted into our evolutionary process
where survival depended on being part of a group and respecting the hierarchy.
Society also conditions us, from an early age, to respect authority figures (parents,
elders, teachers, coaches, seniors, appointments etc.). There are awards for
the obedient and punishment for disobedience. History is testament to the fact
that humans have a strong tendency to obey those in positions of power. Freud
recognised this stating that we should “never underestimate the power of the
need to obey.” Submitting to the authority is considered a desirable and
virtuous quality, more so in the military. AB has been clearly visible in many aviation
accidents where the junior member did not caution the senior despite, he making
fatal mistakes. Many of such accidents resulted out of dangerous unauthorised
flying like low level flying, aerobatics, simulating combat in training aircrafts
(HPT-32, Kiran etc.) and taking short cuts and violating SOPs. The other pilot
(mostly trainee), over taken by authority bias, refused to interfere or
caution. The Milgram obedience experiment is the
first and the best-known study about authority bias.
OBEDIENCE TO AUTHORITY EXPERIMENT
Stanley Milgram, professor of psychology at Yale
University, got interested in the topic ‘obedience’ during the trial of German
Nazi war criminal Adolf Eichmann who was one of the architects of the Nazis
“Final Solution” which oversaw the systematic execution of 6 million innocent
jews. Like other high -profile war criminals, he too defended his actions as, ‘merely
following orders of superiors.’ Milgram also,
was intrigued by the acquiesce of normal German population to these mass
killings. The agenda could have been set at top but the actual killings were
carried out by officials down below and local population would have been aware
and acceptable of such mass killings. Were they all demons or psychopaths? What
conditions led people to become partner in such mass scale killings? These were
the questions which Milgram wanted to test.
The experiment had
three participants. The ‘experimenter’ overseeing the research and two
volunteers who were told that the experiment focused on the effect of punishment
on the ability to memorize and that they will be randomly assigned the role of
either a teacher or learner through draw of lots. This was a lie as the
experiment was a test of obedience and one of the two volunteers was a
confederate of the experimenter. The draw was fixed so that the confederate was
always the learner and the true participant always landed up as ‘teacher.’ The learner was strapped on an electric chair
with an electrode attached to his wrist which was further connected to a shock
generator placed with the teacher in the other room. The generator, specially
designed for the experiment, had 30 level of shocks from 15 volts and
increasing in 15-volt increments to 450 volts. In addition to numerical voltage
all switches were labeled with terms describing the intensity of shock like
‘slight shock’, ‘moderate shock’, ‘danger: severe shock’. The final three switches
were labeled with an ominous ‘XXX.’ The learner was given a list of word pairs
to memorize. The test would consist of the teacher calling out the first word
which was to be matched by the learner. If the answer was incorrect the teacher
would administer a shock to the learner via shock generator which would
increase incrementally to maximum 450 volts. Any silence was to be taken as wrong
reply resulting in increment of the shock. The generator, though, was fake. It caused
no shocks but the teacher had no means of discovering this and performed the
experiment believing that the shocks were for real. The generator was integrated
to a tape recorder, kept next to the learner, with prerecorded sounds. With increasing shock level, the recorder had
audible protests ranging from mild cry to severe sounds of agony with loud
banging of doors and cries for stopping the experiment. At 300 volts the
response turned into cries of extreme agony and distress. From 330 volts onwards
there were no sounds indicating that the learner had become unconscious.
All the
participants gave the initial mild level shocks. Few refused after around 150
volts, but many continued. Every participant paused the experiment at least
once worrying and expressing concern about the learner but most continued after
being prompted by the experimenter to continue. They continued despite being
visibly uncomfortable about their actions. They were seen sweating, trembling,
stuttering, biting their lips, groaning, digging their finger nails into their
skin and some were displaying nervous laughing fits or seizures. In the end
almost two third of the participants (26 out of 40) had shocked another person
to the max 450 volts (marked clearly as XXX: DANGER: SEVERE SHOCK), even when
they felt that it was wrong and wanted to stop. They all felt pressured to
continue by the perceived authority of the person leading the experiment
despite getting no response from the student indicating that he is either dead
or severely incapacitated by the shocks. The point to note here is that there
was no penalty for refusing. The participants were told that they could
withdraw from the experiment anytime. Also, it was a benign lab setting and the
authority was a lab coat draped scientist. The results shook the world. Normal,
well intentioned and educated people with no hostility could be compelled to
unimaginable hostilities merely on command of a perceived authority. The
results also showed that holocaust could happen anywhere.
TRANSPORTING MILGRAM STUDIES INTO COCKPIT
In 1994 the NTSB reviewed all serious
accidents between 1978 and 1990 where flight crew’s action was causal or
contributing. It was found that in more than 80% of the 37 accidents reviewed,
the captain was the flying pilot and the First Officer was the nonflying pilot.
Monitoring and challenging failures were identified in 31 of the 37 accidents. In
19 of the 37 accidents a monitoring /challenging error followed a causal error.
The data reveals that 25% of all accidents could have been prevented by better
monitoring and challenging. Also, factoring that 80% of all accidents the
captain was the flying pilot, about 20% of all accidents could have been prevented
if the first officer had monitored and challenged the captain. By focussing on one issue i.e., why are FO not able to
challenge or monitor the captain, it is possible to reduce accidents by 20%.
To understand the hesitation of first officers,
Dr. Eugen Tarnow Ph D in his article “Towards the zero-accident goal: Assisting
the first officer monitor and challenge captain errors,” compares authority
system in aircraft cockpit to Milgram’s obedience to authority experiment. The
experimenter represents the erring captain. The teacher is the first officer/Copilot
and the student (getting electric shocks) symbolises a crashing aircraft. The following two accidents highlights the
authority dynamics at play in the cockpit.
The accident happened about 27 years ago. A
low-level aircraft detecting radar was to be tested for its capabilities. The task
of undertaking the Low-level sortie was passed to the Sqn which was busy in
planning a battalion level drop. The CO decided to take on the task with two
youngsters who had reported only a few weeks back, from training stage. The
aircraft took off and landed after about one hour with tell-tale signs of going
through high tension wires. The aircraft had flown below the authorised height
of 500 feet AOL. As expected, there was silence all around.
The second case is of an accident involving diamond
DA-42 aircraft VT-NFM operated by National flying training institute on
26/04/2017 at Gondia. The aircraft hit a
ropeway cable while flying low and crashed killing both the instructor and
trainee. The captain had wilfully disobeyed the ATC instructions and went to a
different sector than the one allotted. There, he indulged in dangerous low-level
flying over the river and in the process hit the ropeway cable which sliced the
canopy and the vertical stabilizer. Random analysis found that he had routinely
indulged in such dangerous manoeuvres with trainees.
The authority gradient in the cockpit was
extreme in both accidents. The forces in play were excessive obedience,
hesitant challenging (if at all there was any) which would have been easily
overruled by the captain and acceptance of the version of the captain that the
task (flying) must be done his way. The same forces were in display in the
experiment.
The Milgram experiment had shown that the presence
of strong authority creates a ‘situation’ of obedience. This can be further
explained by the psychological theory called ‘Situationism’ which argues
that changes in human behaviour are influenced by external, situational factors
rather than personality traits or motivations of individuals. ‘Situational
strength’ is defined as implicit or explicit cues provided by external
entities regarding the desirability of potential behaviours. Strong situations can pressure individuals to
act in similar ways by providing very clear indicators as to what behaviour is
most appropriate. It bypasses the personality traits and individuals act contrary
to their convictions. In the above two accidents most of the copilots/ trainees
would have acted in the same manner due to the strong situation created by the presence
of an authority figure (CO and Instructor).
RESISTING AUTHORITY BIAS IN
COCKPIT
Milgram experiment
highlighted the already existing difficulty of monitoring and challenging
authority in our society. Any hierarchical system where powers are concentrated
in top individual would be vulnerable to compliant subordinates who are not
likely to challenge the superior even when they are in error. Aircraft cockpit
is one such place which is excessively captain-centric. One way to encourage first officers to
question and monitor the actions and decisions of captain is to flatten out the
authority gradient between captain and first officers and give respectability
to the role assigned to the first officers. This can be done by revisiting captain’s
authority which is absolute, final, and backed by law, “The pilot in command of an aircraft is directly
responsible for and is the final authority on the operation of the aircraft.” The
extraordinary authority gives too much power to the captain and places him at a
disproportionately higher pedestal to other crew members which makes it
extremely difficult for crew members to challenge a captain even when he is
wrong and dangerous. The concept of command authority and its inviolable nature
inhibits CRM. Flying is a team effort where every member is a vital and
integral part of flying operations. There is a need to balance out the authority
level of the captain by introducing the concept of collective responsibility,
authority with participation and encouraging crew members to exercise
leadership in their own sphere of influence.
This may sound like a drastic proposal
but an honest understanding of human limitations will encourage effective
and pro-active monitoring both by the captain as well as the first officer. All
humans are prone to errors. It is simply human to err. While flying the pilot has a mental map of
situation from time to time. His decisions are based on these mental maps. Once
these mental maps are formed it is difficult to realize that it could be
incorrect and devise alternate scenarios. The fault could have been due to
wrong data or outdated data or simply because brain becomes saturated due to
voluminous information needing attention. Humans are by design terrestrial
animals and some of our organs give out incorrect information in the form of
illusions (Somatogravic, autokinesis, startle reflex, false horizon, black hole
etc.) when it is put up in flying environment which is unnatural for us. Pilots
are prone to disorientation, confusion, startle-reflex effect, fatigue,
fixation etc. We are prone to forgetting. Our memory, vision, hearing is far
from perfect. Our brain suffers from multiple cognitive biases which distorts
the incoming information. Errors and mistakes are bound to happen while flying.
However, errors could easily turn into catastrophe and therefore there is a
need to check/correct them in time. Based on these facts the system where
one person is actively flying and the other is monitoring his actions and
inputs, to catch and rectify any mistake, has been found to be most safe and
efficient way to fly and has been adapted by most of the transport category
aircrafts. The important role of person monitoring is to be recognised and
given the importance it deserves.
Humans
are ill equipped to multitask whereas flying is all about juggling and
monitoring multiple parameters and rapidly analysing vast amount of information
for decision making. This makes the person flying focussed on multiple
instruments and gadgets to monitor the flight path. His overall situational
awareness is markedly reduced due to focussed attention on instruments. Here
the monitoring pilot whose job is to keep track of the overall situation will
have a better Situational Awareness (SA). He will be better placed to suggest a
decision in case of any abnormality. This understanding about human
limitations would also dilute the most influential factor which makes a person
compliant i.e., perceived legitimate authority. In aviation all
captains, Commanders and appointment holders are legitimate authorities. They
are proved experts in the profession of flying and considered as worthy of
being obeyed. We, though, add one more aspect, erroneously and naively, that
experts do not make mistakes. This is a myth which is time and again exposed by
mounting statistics of very senior and experienced captains landing up in
accidents.
It is very simple to blame the FOs in
case the accident had resulted due to him not speaking out, but the fact is
that he had never been prepared to handle ‘strong situations’ like, “what to do
if my instructor decides to do low level aerobatics?” “What to do if the CO/
Flt Cdr/authority is breaking rules and SOPs?” He needs to be trained and given
skills and methods to resolve such strong situations. His rights to question
and check the captain need to be endorsed in SOPs and orders.
In an interesting precursor to the experiment Milgram had conducted a poll asking many renowned scholars including psychologists and psychiatrists to predict the behavior of teachers in his experiment. Most believed that hardly one out of 100 would reach to the final lethal shock. The results showed how wrong they were. This gap between prediction and actual results is referred to as “The Milgram Prediction Error.” As per Eugen Tarnow, “The Milgram prediction error is part of a larger social illusion of the effectiveness of ethical teachings. The force exerted by the moral sense of the individual is less effective than social myth would have us believe.” The illusion exists in cockpit too, where we all believe that captains erroneous decisions, if leading to disaster, will be challenged by the other pilot. The Air India Express Flight-1344 (Boeing 737-800) report on crash (07 Aug 2020) at Kozhikode states that, “In this accident, the first officer had been able to identify the un-stabilized approach conditions, but a steep authority gradient in the cockpit had apparently precluded him from taking over the controls or to enforce any corrective actions despite of being well aware of the grave situation”. The aircraft was destroyed killing 21 people including both pilots. 92 passengers and 2 crew members were critically injured.
Comments
Post a Comment