Group think and ‘Yes’ men

The space shuttle Challenger blows up shortly after launch (28 January 1986)

“All that is required for evil to triumph is for good men to do nothing.”

This quote and its many variants are widely attributed to Edmund Burke, although according to many sources he never put his idea into this concise a statement. Others attribute the statement to John Stuart Mill’s inaugural address to the University of St. Andrews in 1867, and even to a translation of Leo Tolstoy’s classic “War and Peace”.

It is often used in the context of some of the great evils the world has seen, like the Nazi Holocaust against the Jews of Europe. Had more people stood up to the Nazis sooner, might the horrors of the Second World War been avoided? I fear that is a question easier to pose in hindsight – the future cannot be predicted easily, if at all.

But what if good people do speak up and nothing changes, or worse, something bad does happen? I’m not necessarily thinking here about war and genocide, but things more mundane, and not necessarily controversial (or seen as risky) when they were decided at the time, but the structure of the business, organisation, government, or indeed wider democracy is incapable of seeing the flashing warning signs that the path they are going down is a REALLY BAD idea?

Whilst it might have been the politician, or the CEO, who took the decision, what if the actual failure was the inability to listen to people saying “don’t do this”; or the inability of the people saying “don’t do this” to articulate properly the warning that something bad might happen; or worse still the ultimate decision maker, despite hearing and understanding the warning from the technical expert, carries on regardless out of some wider imperative that they perceive is more important. It might be political (“the voters won’t like me”), it might be the balance-sheet (“the shareholders won’t like this”), or even an inappropriate weighing of what was thought acceptable risk (“it’s been fine before, we don’t think that will happen.”) It could even be the political opponents have the right arguments against a dangerous course of action, but the voters discount the messenger, or are persuaded otherwise.

The history of the American space program is a useful example of the group think that led to bad policy, and the eventual loss of life, reputations and careers. Despite successfully putting a man on the moon in 1969, the Nixon administration decided to terminate the program the built the Saturn V Moon rocket. The most significant cost of getting into space is the propulsion required to get you from A to B and the Saturn V rocket, as it turned out, had five times greater launch capacity that the later Space Shuttles and cost 25% less to develop and 50% less to operate (adjusting for inflation). NASA’s plans included, by the end of the mid-1980s a lunar base and a semi-permanent Martian base by 1999. Instead by 2021 we’ve only got as far as putting a robot on the Martian surface.

However, the Moon/Mars ambitions and the Saturn V rocket were abandoned, largely for political reasons, and NASA spent the next 25 years in near-Earth orbit with its Space Shuttles and the International Space Station. If you wanted to go into deep space now, with a human crew, the decision to terminate Saturn V meant the closure of production facilities and research and development into other lunar exploration components. Whether you think this was the right approach or not – there are of course other priorities to spend tax money on – the USA lost the technological means for exploration and the knowledge of engineers and scientists trained in human spaceflight. NASA also lost the ability to deliver larger and faster payloads elsewhere in the solar system. For example, a modified Saturn V rocket could have reduced the Cassini probe’s journey to Saturn by three and a half years. Instead Cassini spent two years looping around the inner solar system to pick up speed.

Rather NASA embarked on the Space Shuttle program, since mothballed after two horrible disasters and leaving NASA reliant on other means and States to achieve its space aims. Fair enough, the 1970s were an austere decade – oil prices were under threat and the US was fighting the Vietnam war – hence the political drive to abandon the Saturn V on then grounds of perceived cost. Nixon made it clear that whilst he favoured an on-going space programme, funding on the order of the Apollo program was not possible and the Shuttle became the focus of NASA’s ambitions, and possibly a space station if the Shuttle program was successful. Design and production of the Shuttle took place with the first shuttle, Columbia, being launched in 1981. Design required two recoverable solid rocket boosters (SRBs) – a concept attractive in terms of cost and re-use when up to 60 launches a year were envisaged. However the rationale for solid rocket boosters was questioned from an early stage on grounds of safety and controllability. The company awarded the contract for the boosters, Thiokol, was chosen on the basis that their model offered the lowest cost per flight.

Each of the two solid rocket boosters was made up of seven sections, six of which were permanently joined in pairs at the factory. For each flight, the four resulting segments were then assembled at the NASA space centre. The factory joints were sealed with asbestos-silica insulation applied over the joint, while each field joint was sealed with two rubber O-rings. The seals of all of the SRB joints were required to contain the hot, high-pressure gases produced by the burning solid propellant inside, thus forcing them out of the nozzle at the bottom of each rocket. However, during the launch of Challenger in January 1986, the O-ring joints failed on one of the boosters causing a fuel leak, that burnt through the booster casing, ignited the main external fuel tank, and destroyed the Shuttle and killed all seven crew.

During the Shuttle design process, a report as early as 1971 raised the risk of an especially dangerous fault: a burn-through by hot gases of the rocket’s casing which might prevent safe abort of a mission. The O-ring joints in the SRBs were supposed to close more tightly due to forces generated at ignition, but a 1977 test showed that when pressurized water was used to simulate the effects of booster combustion, the metal parts bent away from each other, opening a gap through which gases could leak. This phenomenon, known as “joint rotation”, caused a momentary drop in air pressure. This made it possible for combustion gases to erode the O-rings. In the event of widespread erosion, a flame path could develop, causing the joint to burst—which would have destroyed the booster and the shuttle.

Engineers wrote to the manager of the Solid Rocket Booster project, George Hardy, on several occasions suggesting that Thiokol’s field joint design was unacceptable. For example, one engineer suggested that joint rotation would render the secondary O-ring useless, but Hardy did not forward these memos to Thiokol, and the field joints were accepted for flight in 1980. Evidence of serious O-ring erosion was present as early as the second space shuttle mission, which was flown by Columbia. Contrary to NASA regulations, Thiokol did not report this problem to senior management at NASA, but opted to keep the problem within their reporting channels with the company. Even after the O-rings were re-designated as “Criticality 1″—meaning that their failure would result in the destruction of the Orbiter, no one at Thiokol suggested that the shuttles be grounded until the flaw could be fixed.

After the 1984 launch of Discovery, the first occurrence of hot gas “blow-by” was discovered beyond the primary O-ring. In the post-flight analysis, Thiokol engineers found that the amount of blow-by was relatively small and had not impinged upon the secondary O-ring, and concluded that for future flights, the damage was an acceptable risk. However, after the Challenger disaster, Thiokol engineer Brian Russell identified this event, in hindsight, as the first “big red flag” regarding O-ring safety.

By 1985, with seven of nine shuttle launches that year using boosters displaying O-ring erosion or hot gas blow-by, Thiokol realized that they had a potentially catastrophic problem on their hands. Perhaps most concerning was the launch in April 1985 of Challenger, in which the worst O-ring damage to date was discovered in post-flight analysis. The primary O-ring of the left nozzle had been eroded so extensively that it had failed to seal, and for the first time hot gases had eroded the secondary O-ring. They began the process of redesigning the joint but did not call for a halt to shuttle flights until the joints could be redesigned, but rather treated the problem as an acceptable flight risk. A flight manager for the SRB project since 1982 issued and waived launch constraints for six consecutive flights. Thiokol even went as far as to persuade NASA to declare the O-ring problem “closed”. A member of the commission that later investigated the Challenger disaster likened this situation to an airline permitting one of its planes to continue to fly despite evidence that one of its wings was about to fall off.

The day of the Challenger launch (Jan 28, 1986) US President Ronald Regan was about to give his annual State of the Union address to Congress that evening. It was thought he would reference (later denied) the launch, especially as one of the astronauts was a teacher, selected as part of special programme to be the first teacher in space. The launch had already been delayed and a further delay on the day of a high profile political speech might be embarrassing to NASA. Earlier delays had led to negative press coverage on the CBS Evening News. The weather in Florida for the launch was unusually cold with temperatures close to −1 °C (30 °F), the minimum temperature permitted for launch. The Shuttle was never certified to operate in temperatures that low. The O-rings, as well as many other critical components, had no test data to support any expectation of a successful launch in such conditions.

However, back in mid-1985, Thiokol engineers had worried that others did not share their concerns about the low temperature effects on the boosters. Engineer Bob Ebeling in October 1985 wrote a memo— entitled “Help!” so others would read it — of concerns regarding low temperatures and O-rings. After the weather forecast, NASA personnel remembered Thiokol’s warnings and contacted the company. When a Thiokol manager asked Ebeling about the possibility of a launch at 18 °F (−8 °C), he answered “[W]e’re only qualified to 40° [40 °F or 4 °C] … ‘what business does anyone even have thinking about 18°, we’re in no-man’s land.'” After his team agreed that a launch risked disaster, Thiokol immediately called NASA recommending a postponement until temperatures rose in the afternoon. NASA manager Jud Lovingood responded that Thiokol could not make the recommendation without providing a safe temperature. The company prepared for a teleconference two hours later during which it would have to justify a no-launch recommendation.

At the teleconference on the evening of January 27, Thiokol engineers and managers discussed the weather conditions with NASA managers from Kennedy Space Center. Several engineers reiterated their concerns about the effect of low temperatures on the resilience of the SRB rubber O-rings, and recommended a launch postponement. They argued that they did not have enough data to determine whether the joints would properly seal if the O-rings were colder than 54 °F (12 °C). This was an important consideration, since the SRB O-rings had been designated as a “Criticality 1” component, meaning that there was no backup if both the primary and secondary O-rings failed, and their failure could destroy the Orbiter and kill its crew.

Thiokol management initially supported its engineers’ recommendation to postpone the launch, but NASA staff opposed a delay. During the conference call, Hardy told Thiokol, “I am appalled. I am appalled by your recommendation.” Mulloy said, “My God, Thiokol, when do you want me to launch—next April?” NASA believed that Thiokol’s hastily prepared presentation’s quality was too poor to support such a statement on flight safety. One argument by NASA personnel contesting Thiokol’s concerns was that if the primary O-ring failed, the secondary O-ring would still seal. This was unproven, and was in any case an argument that did not apply to a “Criticality 1” component. NASA later claimed that it did not know of Thiokol’s earlier concerns about the effects of the cold on the O-rings. Factors in the decision included an assessment that the O-ring issue had occurred before and nothing catastrophic had resulted.

According to Ebeling, a second conference call was scheduled with only NASA and Thiokol management, excluding the engineers. For reasons that are unclear, Thiokol management disregarded its own engineers’ warnings and now recommended that the launch proceed as scheduled; NASA did not ask why.

On launch day, temperatures were actually colder than first forecast and lower than the previously coldest launch. Engineers were further concerned about the amount of ice on the orbiter and around the launch site that might be shaken loose and strike the shuttle’s thermal protection tiles. However concerns were voiced in a manner that led Houston-based mission manager Arnold Aldrich to go ahead with the launch. Aldrich decided to postpone the shuttle launch by an hour to give the ice team time to perform another inspection. After that last inspection, during which the ice appeared to be melting, Challenger was cleared to launch at 11:38 am. Having launched it blew up around 75 seconds into its ascent as both the O-ring seals failed and leaked dangerous hot gases. The significant failings referenced above came to light in the official enquiry. Ever since, the Challenger accident has frequently been used as a case study in the study of subjects such as engineering safety, the ethics of whistle-blowing, communications, group decision-making, and the dangers of groupthink.

I use this as an example of why things go wrong and why people get angry:

  • there is a political / corporate / business imperative: e.g. “People aren’t voting for us/me. Find me something (or a product) that will save me money and will also be popular!”; The ‘brand’ could also be a person or leader.
  • a new policy, process or product is designed (without consulting the experts in the field) and is high profile enough, and exciting enough, to convince the decision maker this is a good idea; (“Stop this; hate that; do this/buy this instead and we’ll make your life better!”);
  • problems emerge in the design of the policy, product or process; these are flagged but are ignored, deemed solvable, or low risk; (“Please note this will be really complicated, will be expensive, and take time!”)
  • Potential high risk consequences emerge but the programme is so far advanced with political or corporate capital heavily invested in the outcome; (“The party/shareholders won’t like it; Look at the opinion polls!”)
  • More concerns are raised, but group think dismisses them; (“We can’t change that because….”)
  • Alarm bells start ringing, but whilst not yet catastrophic, cautious voices do not stand their ground, do not speak up, or do not articulate their concerns properly that the decision maker listens and understands; or they are shouted down; (“Don’t bring me problems, I want solutions!”)
  • the decision-maker does not want to understand because there are wider imperatives or ideology at stake and admitting a mistake was made equals weakness; (“People that voted for me will hate me!” “We must implement the will of the people!”)
  • Implementation – everything then goes horribly wrong (not necessarily catastrophically at first, but a drip, drip of bad news and problems); severe damage ensures to reputations, careers, balance sheets, electoral fortunes and in the worst cases people get ill, or die or businesses go bust.
  • Everyone asks, “Why did this happen?” Some people reply, “We told you so!” Some people reply, “It’s someone else’s fault!” Anger and more disillusionment with politicians or corporations ensues.

So replace the policy decision to use the Shuttle instead of developing further the Apollo programme; and the specific circumstances that led to the Challenger disaster with other events, and you will see the similarities of policy, imperatives, group think and warnings not heeded play out in similar ways. What if actually progress and betterment for society is actually achieved by careful but dull processes and change, using experts in their field, and where necessary taking the time to get things right. However, “Steady as she goes” doesn’t make a marketing slogan or simple political message.

Yet, for the high profile American space shuttle programme all information was there. Experts repeatedly told senior leaders there was a big risk of disaster and why. The ‘political’ impetus not to delay launch was great. Senior leaders did not want to say “No” to their masters. “Helpful” data took priority; “unhelpful” was undermined or ignored. Launch. Bang. Disaster. The longer term aim and benefits of policy were also scuppered because they’d gone down the wrong strategic path in the first place.

Sound familiar?


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: