What does the Mueller Report have to do with auto safety?

On March 24, newly minted Attorney General William Barr released a four-page summary of the Office of Special Counsel’s 448-page report of a two-year investigation into President Donald Trumps ties to and cultivation and acceptance of Russia’s help in defeating Democratic opponent Hillary Clinton via the use of stolen emails, and of Trump’s efforts to stop an investigation into his actions.

In delivering his much truncated and highly curated version, Barr put this partisan thumb on the scale, opining that Special Counsel Robert Mueller did not find criminal wrong-doing. President Trump used the occasion to stand on an airport tarmac and proclaim in his trademark sing-song: “It was just announced – no collusion with Russia. The most ridiculous thing I ever heard. There was no collusion. No obstruction and it was a complete and total exoneration!”

It would be two weeks before Barr would release a redacted version, in which Congress and any member of the public who so desired could sit down in a wing chair with two fingers of scotch and actually read most of it, to more fully understand the context of Special Counsel Robert Mueller’s prosecutorial conclusions, or feast on dozens of damning details about the Trump presidency that had not already been reported.

Barr’s rollout garnered headlines like these:

  • “Barr's summary of Mueller report sparks GOP celebration, fresh Dem calls for full release”
  • “Trump did not collude with Russia, says Mueller, and is cleared of obstruction by the attorney general” and
  • “Mueller finds no Trump-Russia conspiracy”

Of course, the Mueller report did not vindicate Trump. Special Counsel Mueller explained:

[I]f we had confidence after a thorough investigation of the facts that the President clearly did not commit obstruction of justice, we would so state. Based on the facts and the applicable legal standards, we are unable to reach that judgment. Accordingly, while this report does not conclude that the President committed a crime, it also does not exonerate him.

Over at The Safety Record, the collective hairs on the backs of our collective necks rose in concert. We’ve heard it before.

On August 29, 2009 California Highway Patrol Officer Mark Saylor, his wife, young daughter and brother-in-law died in a horrific crash, when the loaner 2009 Lexus ES 350 Saylor had been driving experienced an Unintended Acceleration (UA) event. Before the Saylor crash, Toyota UA was the subject of eight National Highway Traffic Safety Administration investigations and two agency-influenced, limited recalls. After the Saylor crash, more investigations were launched – among them, one from Congress. Toyota admitted no wrongdoing and embarked on an aggressive public relations campaign to quiet critics and to re-gain its reputation.

In April 2010, the Department of Transportation hired the NASA National Engineering and Safety Center (NESC) to examine Toyota’s electronics to determine if they were to blame for the many incidents of unintended acceleration that began to occur when Toyota transitioned to its Electronic Throttle Control System – Intelligent (ETCS-i) in 2002.  After 10 months of investigation, it issued two reports.

In February 2011, Secretary of Transportation Ray LaHood stood before the cameras to declare:

“The jury is back …he said,  "The verdict is in. There is no electronic-based cause for unintended high-speed acceleration in Toyotas. Period.”

The dense, 177-page highly technical report was released to reporters only one hour before LaHood’s victory lap, so his bold declaration grabbed headlines like these:

  • "Toyota's electronics exonerated in unintended acceleration claims"
  • "Electronic Flaws Did Not Cause Toyota Problems, U.S. Says"
  • "DOT: No Electronic Sudden Acceleration in Toyotas"

Of course, the NESC report did not vindicate Toyota’s electronics; instead, it said

Due to system complexity which will be described and the many possible electronic hardware and software systems interactions, it is not realistic to attempt to "prove" that the ETCS-i cannot cause UAs. Today's vehicles are sufficiently complex that no reasonable amount of analysis or testing can prove electronics and software have no errors. Therefore, absence of proof that the ETCS-I  has caused a UA does not vindicate the system.   

The first report entitled, Technical Support to the National Highway Traffic Safety Administration (NHTSA) on the Reported Toyota Motor Corporation (TMC) Unintended Acceleration (UA) Investigation, clearly delineated the difficulties in isolating the causes of such a complex multi-root cause problem, and yet it found:

  • Toyota’s main defense in previous unintended acceleration investigations is false. The NESC team found scenarios in which engine speed can be increased, RPMs can surge, and the throttle can be opened to various degrees in contradiction to the driver’s command and not set a Diagnostic Trouble Code (DTC).
  • One way that Toyota’s electronic throttle control system could cause a UA event was tin whiskers in the Accelerator Pedal Position Sensor (APPS) of potentiometer-type pedals. Tin whiskers are hair-like structures which can cause electrical shorts. The team found the presence of this well-known electronics phenomenon in virtually every potentiometer accelerator pedal assembly inspected. They linked it to a real-world complaint of a UA event. This suggests that the presence of tin whiskers is widespread and deserved further study.
  • Numerous design inadequacies exist in Toyota’s electronic architecture, but the report failed to probe them any further to discern how these deficiencies fit into the problem of unintended acceleration. 

The Unreliable Narrator’s Playbook

Both instances involve a powerful entity trying to get out in front of the bad news by deliberately mischaracterizing its contents to an audience that did not have the advantage of the facts. It’s called “controlling the narrative,” and, as a Forbes editorial notes: “Few phrases in common usage are more pervasive than ‘controlling the narrative.’ Regardless of your walk of life, having control over every aspect of your “story” should be a priority, personally and professionally.”

Indeed, the concept is used several times in the Mueller report to describe attempts by Trump and his minions to spread lies, as in:

Also on May 10, 2017, Sessions and Rosenstein each spoke to McGahn and expressed concern that the White House was creating a narrative that Rosenstein had initiated the decision to fire Comey. The White House Counsel's Office agreed that it was factually wrong to say that the Department of Justice had initiated Comey's termination, and McGahn asked attorneys in the White House Counsel's Office to work with the press office to correct the narrative.

We would not argue that the parallels are exact. The actions of a lawless President to encourage and use the intervention of a hostile foreign power to gain advantage in a U.S. election, and his attempts to obstruct justice, are likely to garner more and sustained public interest and news coverage than an automotive safety problem. And the story – complex as it is due to Trump’s many questionable and illegal activities in his decades as a businessman and as a politician – is more readily graspable than a highly technical problem that is invisible and difficult to find, but each narrative traces a similar path:

  • Use the bad acts of the report’s sources to further the narrative of the preferred outcome
  • Close the investigation before all facts are in
  • Set the narrative before the people see the detailed facts
  • Count on the public’s short attention span to accept the government’s word and move on
  • Count on the misimpression that the government is diligently protecting the public’s interest through rigorous regulations and laws and through enforcement.  

Prosecutors in the Special Counsel’s investigation had to rely on unreliable narrators –  individuals implicated in wrongdoing – and had an incentive to minimize or hide their own actions:

The investigation did not always yield admissible information or testimony, or a complete picture of the activities undertaken by subjects of the investigation. Some individuals invoked their Fifth Amendment right against compelled self-incrimination and were not, in the Office's judgment, appropriate candidates for grants of immunity…Even when individuals testified or agreed to be interviewed, they sometimes provided information that was false or incomplete, leading to some of the false-statements charges described above….Further, the Office learned that some of the individuals we interviewed or whose conduct we investigated – including some associated with the Trump Campaign – deleted relevant communications or communicated during the relevant period using applications that feature encryption or that do not provide for long-term retention of data or communications records.

Similarly, the NASA investigators had to rely on the representations of the target, and they did not have access to Toyota’s complete source code or vehicles that had experienced high-speed UA events. NHTSA relied on Exponent, a science-defense firm specifically retained by Toyota’s counsel for the purpose of defending the company against a class-action lawsuit, to perform an analysis of warranty claims without identifying Exponent as the source of the analysis. This analysis was used by NHTSA and NESC investigators to rule out electronic causes of UA and dismiss the significance of tin whiskers – the physical evidence they discovered in potentiometer-type pedals. (Tin whiskers are crystalline structures that emanate from tin and other alloys used as solder on printed circuit boards. These nearly microscopic metal hairs can bridge circuits, leading to electrical shorts and significant malfunctions.)

After the Huffington Post published a well-researched article about the significance of the tin whiskers findings, Mike Michels, Vice President for External Communications of Toyota Motor Sales, U.S.A., responded with an editorial trying to herd the narrative back to Toyota’s preferred “no electronic causes” story that consisted of easily refuted falsehoods. (You can read them here)

Neither the Office of the Special Counsel, nor the NASA scientists were given enough time to fully follow their investigative threads before the record was written.

And the Toyota Unintended Acceleration fiasco is but one example. We have seen a version of these events play out in other safety crises.

Ford Explorers and Firestone Tires

In 2000, the bad combination of Bridgestone-Firestone's Wilderness ATX tires on Ford Explorers, America’s best-selling brand of SUV, had resulted in at least 173 tire-failure rollover deaths. In May 2000, NHTSA had opened an investigation into Firestone ATX and Wilderness tire tread separations in the U.S. based on 90 reported incidents that caused 27 injuries and four fatalities. The agency did not know that Ford had already acknowledged the safety defect in overseas recalls. The fiasco exploded into view that August, after Strategic Safety (predecessor to SRS) announced to safety officials and the press that Ford had surreptitiously recalled Firestone Wilderness AT tires in Venezuela.

The Ford Explorer was part of that new generation of station wagon with a high center of gravity, built on a truck platform. At the time, there were no federal standards for occupant protection in rollovers, nor a minimum stability standard for SUVs. Industry fought off any regulations, even as the rollover death tolls in light trucks rose to epidemic levels. Ford knew that Explorer was prone to rollovers – its internal data showed this – and the company had developed an electronic stability control system, which it offered to its customers as a high-cost option that was invisible to drivers. But after those overseas tire recalls came to light, the automaker launched a series of campaigns to replace the Firestone tires and began to sell the story that the tires bore all the blame.

In testimony to Congress, Ford's then-CEO, Mr. Jacques Nasser asserted: “The data tell us that the problem is with the tires and not the vehicle.”

The broken parts narrative served both parties well – NHTSA eschews going after automakers for design defects that are prohibitively expensive to fix and in areas the agency has failed to set regulations for. Compelling the replacement of tires is a lot easier for the agency than demanding post-market design fixes to improve handling and stability or vehicle buybacks. And in 2002, NHTSA denied a request by Bridgestone-Firestone to open a safety defect investigation into the Explorer's oversteer characteristics following a tread separation of a rear tire.

“The data does not support Firestone's contention that Explorers stand out from other SUVs with respect to its handling characteristics following a tread separation,” said then-NHTSA Administrator Dr. Jeffrey Runge in a news release.

It was one of the largest tire recalls in automotive history – 13 million Firestone tires. But even after all the tires were replaced, Explorer tire-related rollover deaths and injuries did not abate. In fact, a 2008 independent analyses of crash data showed that years post-recall, the Explorer was very much a standout among peer SUVs for tire-related fatal crashes. By 2008, the number of Explorer rollover deaths involving tire failures had risen to significantly higher levels than before the tires were recalled.

Ford Explorers in Fatal, Tire-related Crashes: Before and After the Firestone Recall, written by statisticians Randy and Alice Whitfield of Quality Control Systems Corporation, concluded that:

The recalls and replacement campaigns by Ford and Firestone have not achieved long-term effectiveness in eliminating tire-related deaths in the Ford Explorer fleet. Additional efforts must be made to avoid the placement of problematic tires on these vehicles. Also, it is imperative to investigate effective countermeasures addressed to the vehicles' performance on higher speed roadways to avoid loss of control when a tire failure does occur.

ET-Plus Guardrails

In 2015 the Federal Highway Administration twisted crash test data beyond recognition to maintain the fiction that Texas-based Trinity Industries’ ET-Plus energy-absorbing end terminal was safe for states’ use. In 2005, Trinity secretly altered the design of the ET-Plus as a cost-savings measure, without notifying the FHWA, as required. The change caused the guardrail component to jam in the chute and fold in half, spearing the striking vehicles. This design change caused severe injuries and deaths, but the FHWA didn’t penalize Trinity for violating federal regulations, or even order it to conduct new tests when a whistleblower complained to the agency in 2012. It just accepted seven-year old data and permitted Trinity to retain its FHWA approved status, which allows states to be reimbursed by the federal government for their purchase.

In October 2014, a federal jury found that Trinity defrauded the government when it decided not to tell the FHWA about that design change. Although the verdict was overturned by a federal appeals court in September 2017, the FHWA initially responded to the verdict by ordering Trinity to conduct a series of crash tests. In one, the ET-Plus energy-absorbing end terminal caused the guardrail component to jam in the chute and fold in half, spearing the striking Geo Metro at the driver’s side door – and the occupant dummy – just as it did in real-world incidents. The FHWA then issued reports twisting the data to reach a conclusion that the test failure was really a pass, and that the dummy occupant – which was not instrumented – wouldn’t have been hurt too badly.

The Safety Institute's detailed analysis of the crash test data showed that the ET-Plus did not perform as intended even in a very controlled environment, and that the guardrail terminal failed in a similar fashion to the real-world incidents that raised controversy about the safety and efficacy of this device. In order to call the test failure a pass, the FHWA and its contractors ignored the industry standard (NCHRP Report 350) criteria for occupant risk; failed to properly measure the intrusion; used the wrong set of crashes as a basis of comparison; confined its injury inquiry to the lower leg, even though the door directly struck the upper leg; and did not examine the potential for injury to the torso, head or arm

On the issue of potential occupant injury, the FHWA ignored its own independent expert, Clay Gabler, who found: “My conclusion is that the risk of serious injury cannot be discounted simply because the impact is to the legs. An AIS 3 femur fracture could occur as a result of an impact to the upper legs and would be considered a serious injury.”

Instead of using the information conveyed by the test data to address the problem, the FHWA devoted its energy to promoting the narrative that ET-Plus passed muster. The report was an audacious counter-narrative to what the video of the test clearly showed.

Takata Airbag Inflators

In January 2017, the U.S. Department of Justice indicted three Takata executives for wire fraud and conspiracy and accepted a guilty plea and a $1 billion settlement fine from Takata Corporation, then one of the world’s largest component suppliers, for selling defective airbag inflators using volatile chemicals that, over time, could explode, expelling metal shards into the occupant compartment.

As the DOJ told the story:

*For more than a decade, Takata repeatedly and systematically falsified critical test data related to the safety of its products, putting profits and production schedules ahead of safety,” said Fraud Section Chief Weissmann. “This announcement is the latest in the automotive industry enforcement actions the Fraud Section has taken to protect U.S. consumers against fraud.

Conspicuously missing in this accounting was the role of one of Takata’s biggest customers – Honda. Globally, some 60 million vehicles were recalled; 16 deaths were linked to the defective airbags. The OEMs were happy to cast themselves as unwitting dupes, when it’s clear that Honda, with its much lauded and documented supplier quality control process had the ability to throw the brakes on this runaway freight train of disaster much earlier. Myriad management studies on Honda’s corporate philosophy and practices make the case that Honda knew – or should have known – about Takata’s chronic inflator manufacturing problems. Had it only followed its own rules, it would have detected Takata’s systemic problems long ago and taken an active role in quality control and prevention.

Instead, Honda did nothing until 2008, after five field ruptures, a lawsuit, and a presentation from Takata about the inflator problem. Honda launched the first limited recall in November 2008, and for 10 years accepted Takata’s ever-evolving root cause explanations. It continued to issue limited “rolling recalls,” adding new model years and new models each time. Even in 2009, after two fatal inflator explosions, Honda continued to ignore the big picture. When NHTSA and Congress sought an explanation from Honda, it blamed Takata. Only pressure from Congress at the December 2014 hearings finally forced Honda to do a large-scale recall.

How to Read a Government Report

So how should you read a government report? The government has an absolute right and duty to use its authority to probe everything, from powerful figures to consumer products, to ensure the integrity of everything from the rule of law to the safety of vehicle electronics. But don’t take the first soundbite as the last word. Read the report. If possible, get the source material, and read that. Think of the report, as a report – some information that might be useful, another data point. Think about how its claims fit with everything else you see and know.

President Trump has trailed criminal and dubious behavior his entire adult life. More than 30 years ago, Australian officials rejected his bid to build the first casino in Sydney because of his Mafia ties. In December, a New York judge approved a deal to shutter the President’s personal charity, the Trump Foundation, for a “shocking pattern of illegality.” His cheating the sub-contractors who worked on his failed Atlantic City casino was legendary. In February 2018, a $25 million settlement ended the scam known as Trump University. And those are just a few examples. Does anyone really think that the office of the presidency has transformed 72-year-old  Trump into a different person? Does the narrative of “complete and total exoneration” really ring true?

In the case of Toyota, neither the recalls nor Ray LaHood’s assurances stopped UA events from occurring, even though the issue’s high profile slowly dropped. The automaker added a brake-override system to 2011 models, with the alleged ability to allow a driver to counter an unintended acceleration event. In March 2014, the U.S. Department of Justice fined it $1.2 billion for misleading its customers about unintended acceleration caused by floor mat entrapment or sticking accelerator pedals.

Some of the most important answers the public sought as to the root causes of Toyota UA were delivered – not by the government, but by litigation. In October 2014, an Oklahoma jury determined that Toyota acted with “reckless disregard,” awarding $3 million in damages to the victims of a 2007 crash that seriously injured the driver Jean Bookout and killed her passenger Barbara Schwarz. But before the trial could move to the punitive damages stage, Toyota quickly settled the case.

The case turned, in part, on the testimony of two plaintiff’s experts in software design and the design process, who reviewed Toyota’s software engineering process and the source code for the 2005 Toyota Camry and concluded that the system was defective and dangerous, riddled with bugs and gaps in its failsafes, which led to the root cause of the crash.

Michael Barr, a well-respected embedded software specialist, spent nearly 20 months reviewing Toyota’s source code at one of five cubicles in a hotel-sized room, staffed by security guards, who ensured that entrants brought no paper in or out and wore no belts or watches. Barr testified about the specifics of Toyota’s source code, based on his 800-page report. Phillip Koopman, a Carnegie Mellon University professor in computer engineering, a safety critical embedded systems specialist, authored a textbook, Better Embedded System Software, and performs private industry embedded software design reviews – including in the automotive industry – testified about Toyota’s engineering safety process. Both used a programmer’s derisive term for what they saw: spaghetti code – badly written and badly structured source code.

After the Bookout verdict, Toyota began to quietly and aggressively settle UA death and injury cases in confidential settlements. And the NASA scientists who discovered the tin whiskers in Toyota accelerator pedal potentiometers continued their research into the phenomenon, publishing a series of technical papers on their methods and discoveries.

But narratives are powerful.

The myth that Toyota UA events are exclusively the result of mechanical problems or driver error continues to be peddled – by NHTSA and others who lack a sophisticated understanding of the evidence. Celebrity pseudo-journalist Malcolm Gladwell was just the latest to spin this fable.

The ability to create and spread narratives is uniquely human, says Israeli Historian Yuval Noah Harari. And in his book Sapiens: A Brief History of Humankind he posits that this ability to use “stories to unify and organize groups and ensure cooperation,” is the reason Homo Sapiens came to dominate all other species.

We could be harnessing this evolutionary advantage and telling stories that help us tackle climate change, racism, income inequality, safety problems and any number of complex and thorny issues critical to our survival. Instead, its most ardent practitioners are powerful entities trying to preserve their dominant status and shield their mistakes, their crimes and their amorality from greater view.