READER'S BLOG: Volkswagen, Challenger and Feynman

Friday, March 18 marked six months since the Volkswagen ‘Dieselgate’ debacle first broke, and in a scandal that has run deep through the heartland of what is now a morally insolvent company, there seems very little in the way of hard and fast answers to the questions that just keep coming.

Volkswagen has made clear that answers are not what they are prepared -- or allowed -- to give, legitimising its steadfastly denialist attitude by claiming that anything they say at this point stands to jeopardise an internal investigation into the affair that is currently underway.

This report, commissioned by Volkswagen itself, is due for completion in the second half of April. It’s not something anyone should take without a grain of salt. Volkswagen is countering concerns about the report's efficacy, reminding us that investigators are pawing over in excess of 100 terabytes of data; in context, that’s more than 50 million books.

But, judging by the company’s language in Friday’s press release, we don’t need 50 million books to work out how the April report is going to ascribe culpability.

In outlining its defence against legal action by shareholders, Volkswagen makes some powerful assertions about why CEO Martin Winterkorn cannot be held responsible for illegally withholding information about the case until the public announcement of the scandal on September 18, 2015.

The legalese is par for the course, but what is particularly striking is Volkswagen’s fairly explicit allusions to what seems to be a convenient theory for upper management: that the illegal software code was written by a, “group of persons at levels below the [VW] Group’s management Board in the powertrain development division, [who] decided to modify the engine management software using relatively small changes, within the budget that was available for the development of the engine management software, and without the need to involve superior levels.”

Quite simply, it appears Volkswagen is resting its defence on the Rogue Engineer Theory.

If that is the case, if that Rogue Engineer Theory is a key component of VW’s findings, then the company has a problem. You see, if history teaches us anything, it is that by ascribing blame to the engineers, the upper management are inadvertently shifting the axis of blame right back onto themselves. And to explain why, we have to introduce a world-renowned physicist, and we need to talk about Challenger.

The Space Shuttle Challenger disaster on January 28, 1986, was a defining moment of late 20th century history. Carrying seven crew members, including the civilian teacher Christa McAuliffe, the Shuttle exploded and was destroyed 73 seconds into its takeoff sequence, erupting into a catastrophic fireball that tore the craft apart in the skies above the Atlantic Ocean, off the coast of Cape Canaveral, Florida.

Images of the disaster, broadcast live on television, shocked the world. How could NASA, an organisation with such a stellar reputation for safety and a culture of engineering integrity, possibly have been involved in such a tragic catastrophe?

President Reagan ordered the Rogers Commission to investigate the disaster, with findings quickly pointing to a failure of the Right Solid Booster O-Ring, whose performance could easily be compromised by low temperatures such as those experienced during the Challenger launch, which allowed pressurised hot gases to blow-by the seal and contact with the external fuel tank, causing structural failure.

But whilst the design of the O-ring was determined as the mechanical cause of the disaster, deeper concerns were raised in the report about just why NASA engineers decided to go for launch in conditions they knew were too cold, temperatures at which they were told previously O-ring performance had never been tested.

Professor Richard Feynman, a remarkable theoretical physicist, was uniquely qualified as an objective voice on the Commission and he gave a fascinating hypothesis on what he saw as a fatal flaw within the NASA bureaucracy that pushed the engineers to OK the launch of an unsafe and underdeveloped shuttle.

In his investigations, Feynman was made privy to a deal between NASA and Congress, where government funding would be diverted away from the Department of Defence to the shuttle program, whose costs had become unsustainable.

In return, Challenger would make available payload allocations for US Cold War spy satellites, and the Shuttle would become the sole access point for US military into space. NASA convinced Congress they’d be able to launch twice a month, every month; they promised launches at any time, under any conditions.

“It was so to speak impossible from an engineering point of view,” Feynman noted. “In the beginning, all kinds of exaggerations were made about what this shuttle would be able to do. It will fly 60 flights, it will only cost so much, it’ll be recoverable and there’ll be no real problems, they said.

"But the engineers at the bottom were most probably screaming up, ‘No, no, it can’t be this way, it can’t be this way. We can only go 10 flights, we haven’t got enough equipment to train that many crews a year and so forth and so on.’”

But the people at the top of NASA, the people who were talking to Congress, didn’t want to hear it.

Feynman explained: “Each level in a bureaucracy kind of understands what it’s supposed to do: Keep the negative information from moving up; management don’t have to, don’t want to, hear about the engineering problems… because it would be uncomfortable going to Congress and saying, ‘We’re going to do 60 flights a year’ when just that morning the engineers had told them it was impossible.”

What the professor described was a dangerous bureaucratic discord, one that existed in the Challenger program between the reality of engineering and the expectations of management.

It is the same discord that produced a culture in which a small enclave of engineers were forced to cheat the system to meet the unrealistic engineering goals set by Volkswagen management for the EA189 diesel engine in America.

They were asking for the impossible: an engine that would maintain its class-leading performance whilst reducing its NOx emissions to meet restrictions that were six times stricter than those in Europe, all without adding costly urea-injection systems or modifying the engine at great expense.

It was a Sisyphean task. But that engine was to be the keystone of Volkswagen’s mammoth push into diesel-wary America. And so no-one from engineering could tell the bosses that their grand plan was, in fact, impossible.

Hence I believe it is Professor Feynman’s hypothesis on Challenger that offers the clearest historical precedent with which we might better understand the Volkswagen case. Certainly, it’s a piece of history Volkswagen cannot ignore.

Yes, it is convenient for the Board to blame the ‘rogue engineers’ for immoral misconduct in creating the illegal software to cheat the system. But if Volkswagen management are to move forward, if they are to push themselves to reform, then they must accept the blame for creating a culture in which engineers are forced to conjure unethical solutions to questions that have no realistic answer.

As we move through this scandal, then, Volkswagen would do well to take heed from the wise words of Professor Feynman:

“Let us make recommendations,” he concluded, “to ensure that officials deal in a world of reality. For a successful technology, reality must take precedence over public relations, for nature cannot be fooled.”

How right he was.

Guest Blogger: Harrison Boudakin