Websites appear and disappear every day. The internet is an amorphous beast, constantly being edited and updated. We accept these changes as simple fact, but crucial information is frequently removed from government websites—and small changes over time can have major downstream consequences.
I made the disturbing discovery recently that, amid a dramatic five-year spike in aviation accidents, the Navy has put aviation safety data that used to be public behind a wall. I bumped into this change while researching for the documentary film Who Killed Lt. Van Dorn?, which uncovers the long history of negligence and institutional failure surrounding the 53E helicopter, the deadliest aircraft in the U.S. military.
I lost a high-school friend in the 2014 training accident that killed Wes Van Dorn, and soon after, I began investigating why he died. The safety data on Navy accidents that is now blocked was public when I began my reporting, and it was indispensable in producing what was first a newspaper series, then an NBC Nightly News investigation, and now a feature documentary. Without that safety data, these stories would have been far more difficult to tell.
The decision to suppress this data from public view came as the Navy, which also oversees Marine aviation, was dealing with headlines pointing out that its safety problems are increasing faster than any other branch: The Navy has seen an 82 percent spike in accidents between the 2013 and 2017 fiscal years, while the overall military increase for that same period of time is 40 percent.
The Navy’s decision may be part of a larger trend. In May 2017, the foreign-policy magazine National Interest obtained emails from top Navy officials discouraging the disclosure of readiness numbers—figures that report how many aircraft are ready to fight in the event of a major conflict. At the time, nearly two-thirds of the Navy and Marine F-18s were unfit to fly. In 2015, the Marines’ CH-53E fleet was in even worse shape, with just 23 percent of aircraft able to fly a mission. A March 2017 email from the Pentagon spokesman Captain Jeff Davis instructed the heads of public affairs across the military to “be cautious about publicly telegraphing readiness shortfalls.” The guidance reportedly came from Secretary Mattis himself, according to National Interest.
“There have been a series of removals of information from the public domain,” said Steven Aftergood, the director of the Federation of American Scientists Project on Government Secrecy. He pointed to the Defense Department’s decision last year to stop publishing active-duty-troop numbers in Syria, Iraq, and Afghanistan; these numbers had been posted online for more than a decade. The Department also removed past troop numbers from its quarterly reports and issued conflicting numbers of troops active around the world.
It’s unclear whether the Navy’s decision to remove the data was a product of Davis’s email guidance. The Naval Safety Center, a Navy command dedicated to working on safety issues, told me that the decision was simply made as part of a website redesign. It was “definitely not directed by Secretary Mattis or part of any directive issued by him,” said April Phillips, the Naval Safety Center’s public-affairs officer.
The value of the removed data sets was made clear in 2016, when this information helped inform my report with The Virginian-Pilot that a potentially catastrophic safety issue had gone unaddressed and led to the grounding of more than 150 Navy and Marine Corps helicopters.
I relied on these data sets to learn the condition of Navy and Marine Corps aviation. How often were aircraft crashing? How did that number compare to prior years? Which aircraft were most accident-prone? Without being able to access these records directly, the public has to rely on an understaffed and overworked team that handles public-records requests. While it’s possible I could have done the same work under current conditions, it’s certain that reporters interested in Navy and Marine Corps aviation safety will now have a significant barrier to overcome, a barrier that none of the other military branches have put up. The public no longer has easy access to detailed summaries of accidents, data on accidents by aircraft types, the comparative information provided by annual safety data, and other safety reports and studies.
The removal of this data from public view is troubling for aviators themselves, too. Their aircraft may now face less scrutiny both from the public and from reporters.
The change in policy was implemented quietly sometime between January 22 and March 18 of this year, according to Internet Archive’s Wayback Machine. The web page where the information used to be available gives the following message: “Note: For further statistical information, visit NAVSAFECEN’s Common Access Card (CAC) site. A CAC and a one-time account registration are required for access.”
So I emailed a Navy representative to ask about the change.
About a week later, I got a reply from April Phillips: “The information you are referring to has not been removed, but it has been placed in a section of our website requiring a DoD Common Access Card (CAC) for access, along with a great deal of other information...”
The Department of Defense (DOD) Common Access Card cited in the email is only available to active-duty military, military reserves, government contractors, and civilian DOD employees. By removing this information, the Navy has obviously reduced its transparency. The Navy made this decision to limit access to the information even though none of the files or information on the old version of the website were marked “For Official Use Only.”
To learn how the decision was made, I submitted a Freedom of Information Act request for any documents or emails about the decision to remove the data. The response reads: “The decision to make the change you refer to was the result of an exclusively verbal discussion … There was no written guidance, no emails, discussions (except for verbal conversations), memos, notes, or policy change documents relating to the withdrawal of Naval Safety Center public information, specifically on the statistics page, to the DOD CAC (Common Access Card) site.”
While I did get confirmation that the information was “withdrawn,” I’m no closer to understanding why the Navy made the decision. Phillips, the Naval Safety Center public-affairs officer, implied that publishing the data online created a security risk. “When you aggregate so much information in one place,” she said, “bits and pieces of that can be compiled by those who wish us harm to make inferences and gain intel.”
Aftergood, from the project on government secrecy, finds this reasoning unpersuasive. “It’s disappointing and, I would say, inappropriate,” Aftergood said. “There is no security reason for limiting public access to statistical summaries of safety mishaps. On the contrary, public disclosure of such information can help to improve safety and security by motivating officials to adopt remedial measures.”
In contrast to the Navy, the Air Force could be considered a model of transparency. The Air Force Safety Center not only discloses accident data year after year; it also breaks the data down by individual aircraft. (The Center even lets you see which bird species have caused the most damage to Air Force aircraft. The Canada Goose has caused more than $91 million in damages from 1995 to 2016.)
The Air Force’s clear handling of its public data raises an obvious question: Are the security concerns of the Department of the Navy really that much greater than those of the Air Force? Even though the Navy’s accident data will hypothetically still be available through a request system, the act of making the data private effectively makes it secret. “Most people, including most reporters and official oversight personnel, are not going to take the trouble to uncover suppressed information of this kind,” Aftergood said.
Small changes to government websites will inevitably go unnoticed by the public, and there’s no telling what information is slowly and irrevocably slipping out of reach.
We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.