A drill with no missile
March 16, 2022. Newport News, Virginia.
The aircraft carrier USS George Washington sits in dry dock. Below deck, a row of sailors in firefighting gear crouches down, hands over their heads, waiting for a missile to hit.
The missile isn't coming. The ship isn't at sea either. This is an imagined naval battle — they are drilling battle-damage response inside the dry dock.
The sailor who took the photograph was the ship's photographer, Corbin M. Hagman.
Twenty-four days later, the first sailor of that April died. He was not on the ship when it happened.
The ship in the dock
A refueling overhaul is something a carrier goes through once in fifty years of service. Two reactors are refueled, the hull is overhauled, the systems are upgraded.
The Navy's own investigation says, in Finding 14, that the shipyard is a dangerous and noisy industrial environment: the shriek of needle guns, bells, constant announcements over the 1MC, grinding. Noise like this, it says, is common.
The endorsement changed one word. The draft had called that noise "normal." The reviewer changed it to "common."
By July 15, 2022, 370 of the 422 people who had been living aboard had moved ashore.
Two months earlier, on May 11, the House Armed Services Committee held a hearing. Chairman Adam Smith opened by noting that the ship had made the news over its number of suicides, that 200 crew members had been pulled off, that complaints about living conditions were widespread — and that on this, things did not seem to have been handled well. Lives had already been lost.
April
On April 9, Mika'il Sharp, 23, who had enlisted in June 2020, died at a private residence. He was not living aboard at the time; he was away at training.
On April 10, Natasha Huffman died.
On April 15, Mitchell-Sandor was found unconscious aboard the ship and pronounced dead after arriving at a local hospital. He was 19 and had been aboard about three months.
The Navy's report records two things about him. Senior enlisted aboard knew he had been sleeping in his car; the report says they counseled him, but investigators found no one who had asked him why. And the investigating team ran its own estimate: three stretches of more than 24 hours without sleep every week, roughly 130 hours of accumulated sleep debt in the 10 weeks before he died.
The investigation's conclusion about all three was that the deaths were not connected to one another. The three had no social and no working relationship. Each was facing a different set of personal pressures.
The door
Set the ship aside for a moment and consider two people.
Evan Slusser was a Marine Corps fighter pilot. He told his own story publicly in The War Horse. After a suicide attempt, he walked into the flight surgeon's office. He wrote that he finally spoke up, that he was quietly grounded, and that not long after he was out of the Marine Corps — processed, as he put it, fairly quickly.
He did not write what he was diagnosed with, or which administrative route took him out. We don't know that, and we won't guess.
Maj. Jane Marlow flies the C-130J for the Air Force. In an official Air Force release, she said that like many of her peers she kept putting off getting care until she was moved to a non-flying billet.
Note what she did not say. She did not say she hid it. She said she waited until she wasn't flying.
Why wait? Because some billets have it written down.
Under the nuclear Personnel Reliability Program, a suspected suicidal act can trigger a procedure: the person's nuclear duties are suspended until the psychological evaluation is complete. That is nuclear duty eligibility, not military eligibility in general.
Naval aeromedical psychiatry guidance provides that aircrew diagnosed with adjustment disorder accompanied by suicidal ideation, suicidal behavior, or non-suicidal self-injury must complete at least six months of non-flying observation after symptoms have fully resolved. That applies to that diagnosis. It is not a rule that any pilot who has ever had a suicidal thought is automatically grounded for six months.
Six months not flying. For a pilot, no reassuring sentence offsets that.
But the rules are only the half written on paper. The other half is in the berthing compartment.
On a Nimitz-class carrier, some enlisted berthing compartments sleep around sixty people. Racks stacked three high, one small locker each. Some of those working below deck can go weeks without seeing daylight.
In a compartment like that there are no doors and no partitions. You get up, and whichever way you walk, sixty people look up and see you.
Seeing a psychologist means requesting leave and saying where you're going. And everyone in the compartment knows which way that door is.
But this is not a cover-up story
It is easy at this point to slide toward a conclusion: the military is hiding it. The material doesn't support that, and the evidence against it has to be laid out.
First. DoD Instruction 6490.08 establishes that when a service member voluntarily seeks mental health care, the presumption is that the commander is not notified. Notification may be required only in nine enumerated categories — risk of self-harm, risk of harm to others, mission risk, hospitalization, special-duty personnel, and so on. "Go to a psychologist and your unit will find out" is not what the rule says.
Second. The SF-86 asks about mental health in Question 21, provides exemptions for certain kinds of counseling, and states on its face that receiving mental health counseling is not grounds to deny or revoke a security clearance.
Third. Gen. Mike Minihan, then head of Air Mobility Command, put his own mental health appointment openly on his calendar in 2022. At a gathering in Phoenix on April 29, 2024, he told the room that if they wanted to see what a pilot with PTSD looks like, they were looking at one. He was still in command when he said it.
Fourth, and this one matters most. In 2015, Sharp and six co-authors published a systematic review in Epidemiologic Reviews covering military mental health research from 2001 to 2014. Two figures from it get quoted constantly: 44.2 percent worried about being treated differently by leadership, 42.9 percent feared being seen as weak by peers.
The same review contains a finding that almost never gets quoted. Among the studies it analyzed, nine found no statistical association between anticipated stigma and actual help-seeking behavior — and four found a positive association, meaning those who perceived more stigma were subsequently more likely to seek help.
That evidence weakens the argument this piece is making. I am putting it in because it is true.
It also sharpens the picture. What keeps people outside that door may not be a diffuse fear of being looked down on, but the hard consequences written on paper: six months not flying, nuclear duties suspended. How it feels and what the rule says are two different things.
The sentence the report wrote itself
Back to the ship.
Between January 2021 and January 2022, the ship's psychologist and behavioral health technician, plus two substance abuse rehabilitation counselors, handled roughly 2,615 clinical encounters.
Encounters — not 2,615 people.
At the time of the deaths the ship was manned with one psychologist and one behavioral health technician. Two more providers were sent after April 25. The report's opinions section describes those two as overwhelmed.
So how long was the wait for an appointment? Three answers have appeared in public, at three different levels of sourcing.
The investigation found that a routine, non-urgent initial evaluation could take one to two months given the backlog. Adm. Michael M. Gilday, then Chief of Naval Operations, said in a written response that urgent referrals — risk of suicide or harm to others — were evaluated the same day, and that before reinforcements arrived a new patient's first evaluation took about four to six weeks.
Four to six weeks, one to two months: those ranges overlap heavily and very likely describe the same backlog. The third figure came from the hearing, where Rep. Salud O. Carbajal said he had heard the wait ran as long as six months — while noting in his written question that this came from public reporting rather than Navy measurement.
So this is not a case of official numbers contradicting each other. It is one situation, measured at different levels and against different things, producing different numbers. And the figure explicitly relayed at the hearing was the lowest-sourced one.
Finding 42 of the same report says that some sailors chose to get mental health care off the ship, through Navy or civilian providers, in and out of the TRICARE network, rather than wait for an appointment aboard.
The Navy wrote that sentence itself.
What it supports: those visits do not count toward the 2,615, because that figure counts what the shipboard providers handled. What it does not support: that those people disappeared. They may well have entered medical records, the referral system, or a higher-level aggregate.
Which means the 2,615 is not the total mental health need aboard. It does not cover the people who bypassed the shipboard appointment and sought help outside. And how those visits are recorded at a higher level, the public text of the investigation does not say.
One more line. In the endorsement to the first-phase investigation — not the body of the report, but text a reviewer added afterward — the last sentence of one opinion was changed to note that a sailor had said he wanted mental health matters kept out of his medical record.
The investigation's overall conclusion reads that although the three sailors shared common stressors, such as the general pressure of the shipyard environment, the investigating team's opinion was that the deaths of Sharp, Huffman, and Mitchell-Sandor were unrelated and not connected.
Note what it says: the three people are unrelated to each other. In the same sentence, it concedes the shipyard environment as a shared stressor.
A year later, a broader investigation into command climate and sailor quality of life found that the infrastructure around Newport News could not support this many ships in overhaul and construction at once. Parking, traffic, food and nutrition, training grounds, gyms, housing — none of it sufficient, producing poor quality of life and poor morale. The fleet command's final endorsement issued dozens of actionable recommendations; public reporting usually cites 48. One of them: when not at sea, apart from duty watchstanders, sailors should not be required to live aboard the ship or the barge.
Those are recommendations. Not things already done.
2026, the same name
Jump forward four years.
This needs saying first: it is the same ship, but it left Newport News long ago, completed the overhaul, and deployed again. The place, the mission, and the people aboard are all different.
On August 19, 2026, the George Washington entered the Arabian Sea. The next day, U.S. Central Command said it had entered its area of responsibility, relieving the Abraham Lincoln after more than nine months deployed.
Aboard the Lincoln, Rep. Mike Levin, Democrat of California, told reporters he had received statements from a dozen families calling for a full investigation. Family members described moldy showers, broken toilets, no hot water, shortages of food and hygiene supplies, and people close to breaking down.
Navy Times and Stars and Stripes published reports on August 10 and 11 based on interviews with family members and sailors, mentioning sailors who had tried to jump overboard.
Two things need to be clear.
First, those two outlets are independent of each other, but they may be drawing on the same families and the same sailors. Independent outlets does not mean independent evidence.
Second, the Navy has not released an independently verifiable case count. What is confirmed: on August 3, one person went into the water and was rescued. The Navy did not determine that it was a jump overboard, or a suicide attempt.
The Navy's response to the reporting was a written statement saying that, based on information available to the command, it had seen "no increase in reported suicidal ideation or attempts" aboard the ship.
Reported. That word is theirs.
The statement's limits are perfectly clear: the scope is this ship, and the measure is what gets reported. How long the statistical window runs, who the denominator is, what counts as an event — the statement says none of it.
The end
Back to the second when Evan Slusser walked into the flight surgeon's office.
Before that second he was a fighter pilot. After it he wasn't.
He wrote it down later, which is why we know about him. But the ones who never told their compartment mates; the ones who bypassed the shipboard appointment and went elsewhere; the ones who, like Maj. Marlow, waited until they had left the flying billet before dialing the phone — they did not necessarily stay out of every record.
They entered different records, at different points, against different thresholds. Blending all of that into one number is what makes people believe they have seen the whole picture.
An organization can say, quite honestly, that its numbers have not gone up.
That sentence can be true. It simply doesn't answer another question: when those sixty people are watching which door you walk toward, did you get up and go.
If you or someone you know is struggling, please contact your local crisis line. In the United States, the Veterans Crisis Line can be reached by dialing 988 and pressing 1.
