The morning of June 19th, 1944, in the western Philippine Sea, Lieutenant Commander Zenji Abbe lifted his binoculars to his eyes from the bridge of the destroyer Shiratsuyu. His hands trembled—not from fear, but from the impossible sight before him. American carriers were everywhere. Their gray hulls lined the horizon like a wall of steel. He tried counting them. Six. Seven. Eight. No—nine carriers within visual range. And behind them, more masts.

Below deck, radio operators were overwhelmed trying to catalog the contacts. “Task Force 58,” they called it—the Americans’ fast carrier task force. Abbe remembered the intelligence briefings. Tokyo believed the Americans had perhaps twenty operational carriers in the entire Pacific—twenty-five at most. Japanese naval intelligence had tracked every American carrier meticulously since Midway. Every ship had a name, every name a file.

But now, in the waters off Saipan, Lieutenant Commander Abbe was staring at nearly half the supposed American carrier force assembled in one place. It felt impossible. It felt like exactly the kind of sight a desperate empire needed: proof that if Japan could win here—win today—perhaps the war was still salvageable.

What Abbe didn’t know—what almost no one in the Imperial Japanese Navy knew—was that he was looking at ghosts. Task Force 58 was real enough, its carriers made of steel, rivets, and aircraft. But the numbers Tokyo believed in did not exist. The intelligence files back home described a fleet that was a phantom. The United States didn’t have twenty carriers. It didn’t have twenty-five. And Task Force 58 wasn’t half their strength. It wasn’t even close. It was barely a third.

Because by the summer of 1944—thanks to a perfect storm of deception, rapid construction, bureaucratic confusion, and Japan’s own desperate need to believe its enemy was weaker than he truly was—Japanese naval intelligence had reached an extraordinary conclusion: the United States Navy possessed more than eighty aircraft carriers. The real number was twenty-seven.

How could an empire with one of the world’s most sophisticated naval intelligence services make an error this large? How could analysts “see” phantoms while staring at reality? And what happens to strategy, morale, and the will to fight when every battle report seems to confirm the Americans have far more carriers than anyone imagined?

This is a story about the fog of war, but also about something more dangerous—a kind of intelligence failure born not from too little information, but from too much. From seeing patterns that weren’t there. From counting the same ship twice, three times, five times, simply because analysts couldn’t believe how fast the Americans were building them.

The Japanese called it Koku Bokan—the Phantom Carrier Problem. American planners called it Operation Intermittent. Neither side fully understood it until after the war, yet it shaped every major naval battle from 1943 onward.

To understand how Japan lost count, we must go back—back before the phantoms. Back to a time when the numbers were small enough to track.

The Simple Beginning — Seven Carriers

December 7th, 1941. Pearl Harbor. Vice Admiral Chuichi Nagumo launched six carriers against the American Pacific Fleet. At that moment, Japanese intelligence knew exactly what they were facing. The United States had seven aircraft carriers total. Three—Enterprise, Lexington, and Saratoga—were in the Pacific. Ranger, Wasp, Yorktown, and Hornet were split between the Atlantic and other duties.

Seven carriers. Japan had eleven. The mathematics seemed clear.

Lieutenant Commander Nakajima Tadashi ran the carrier-tracking desk at Naval General Staff headquarters in Tokyo. His job was simple: know where every American carrier was, its condition, and when it would next be combat-ready. In late 1941, this was possible. Seven ships. Seven files. Seven pins on a map.

Nakajima kept meticulous notes. When Enterprise and Lexington were at sea during Pearl Harbor, he knew. When Saratoga took a torpedo in January 1942 and returned to Puget Sound for repairs, he logged the date.

Even after the war began, the numbers remained manageable. But by the end of 1942, after the Coral Sea, Midway, Guadalcanal, and Santa Cruz, America had lost four of its original carriers. Only Enterprise, Saratoga, and Ranger remained. Nakajima’s map looked nearly empty.

This should have been Japan’s moment—except the intelligence picture was about to get complicated.

New Names, Old Names, and the Rise of Essex

Even as the old carriers were sunk, new names began appearing. Essex, commissioned December 31st, 1942—a giant, 872 feet long, carrying ninety aircraft, larger than Shōkaku or Zuikaku. This was no replacement—it was a leap in capability.

Then in January 1943, Yorktown. The name stunned Nakajima. The original Yorktown had sunk at Midway only months earlier. Was this an error? Deception? Had the Americans raised her? No—this was CV-10. A new ship. New hull. Old name.

Then came Lexington, CV-16. Another resurrection. To Japanese analysts, every new name was a puzzle. When radio intercepts mentioned Lexington, which one was it? How could analysts tell?

By mid-1943, Nakajima’s tidy system was dissolving. Old names. New names. Reused names. And then came the real disaster: the American command shuffle.

Task Force 58 / Task Force 38 — The Same Fleet, Two Identities

The Americans created two designations for the same fast carrier force. Under Admiral Raymond Spruance’s Fifth Fleet, they were Task Force 58. Under Admiral William Halsey’s Third Fleet, they were Task Force 38. The ships did not change. The aircraft did not change. Only the name changed.

Japanese intelligence, however, interpreted every new task force number as a new fleet. When radio intercepts placed TF-58 in the Marshalls and TF-38 near Truk, analysts assumed these were separate carrier forces. The numbers began to climb.

Mass Production — Too Fast to Track

From 1943 onward, every two months, a new Essex-class carrier entered service. Five shipyards produced them simultaneously—identical ships, identical silhouettes, identical radio profiles.

Submarine photographs caught glimpses of them in port. Recon planes saw them at sea. Signals intelligence heard their names. But analysts couldn’t determine which were operational, which were in refit, which were training. So they counted all of them.

Then came the Independence-class light carriers—nine of them.

Then the escort carriers—dozens upon dozens of small CVEs, churned out like automobiles.

To Japanese recon planes, they all looked alike: flight deck equals carrier.

The numbers ballooned.

Phantoms Take Over

By January 1944, Nakajima estimated fifty-three American carriers. His superiors thought this was too low. By summer 1944, after the Battle of the Philippine Sea—where Japanese pilots claimed they attacked at least eighteen carriers—estimates rose into the seventies. One analyst even suggested eighty.

The real number of fast carriers? Twenty-seven.

But counting became self-reinforcing. Every report that seemed to confirm higher numbers pushed the estimates higher still. And deep down, many in Tokyo needed the Americans to have impossible numbers. It made Japan’s defeats psychologically bearable.

Commander Masatake Okumiya later wrote:

“We needed the enemy to be stronger than he was.
Otherwise we would have to confront the possibility that we were being outthought and outproduced.
That truth was unbearable.”

The phantom carriers served as an excuse.

Why American Carriers “Never Died”

When the Franklin and Bunker Hill took catastrophic damage in 1945 and still survived, Japanese analysts reached a simple, wrong conclusion: the Americans must have more carriers. They never considered that American carriers were simply harder to sink, that damage control training was better, compartmentalization superior, and industrial support unmatched.

American ships kept returning, battle after battle, and Japanese intelligence counted each reappearance as a new vessel. The lists expanded. The ghosts multiplied.

92 Carriers — The Breaking Point

By April 1945, Japanese naval intelligence officially tracked ninety-two American aircraft carriers. The files overflowed from Nakajima’s desk. Maps were covered in pins.

Task Force 38 and Task Force 58 were still being counted as separate fleets. Escort carriers were treated as fleet carriers. Refitted ships were counted as newly built ships.

And then came the kamikaze attacks at Okinawa. Pilots were briefed to expect twenty-five carriers. They launched, died, and returned reports—Wasp hit, Essex hit, Intrepid hit, Enterprise hit. The same ships appeared again and again, and every reappearance was counted as a different carrier.

Commander Okumiya finally grasped the truth. “We have been fighting ghosts,” he wrote. “We see the same ships, and count them as different ships. We believed Task Force 58 and Task Force 38 were separate forces. We believed carriers that refused to die must have replacements behind them. The enemy we created existed more in our estimates than in reality.”

But by then, nothing could change Japan’s fate.

The War Ends — and the Files Burn

On August 15th, 1945, after the Emperor announced surrender, Nakajima walked into the courtyard of the Naval General Staff building and burned his carrier-tracking files—every report, every mistaken count, every ghost.

An American officer approached.
“What are you burning?”
“Mistakes,” Nakajima answered.

He later assisted American investigators, explaining the errors without admitting he had been the man responsible.

In 1968, asked when he finally realized the truth, Nakajima said:

“In my heart, I knew by early 1945.
But I could not speak it.
To admit the estimates were wrong was to admit we were beaten by fewer carriers than we imagined.”

The phantom carriers, he said, “paralyzed Japan.”

The Irony — The Error Was Wrong, but the Conclusion Was Right

Japan believed the U.S. had eighty carriers. The real number was twenty-seven. But here was the irony: twenty-seven American carriers, operating with superior logistics, doctrine, and industrial power, behaved like eighty.

The Essex-class carriers:

operated for months without rest,

absorbed damage that would sink Japanese ships,

launched more aircraft with greater tempo,

returned to battle faster than Japan thought possible.

Japan counted ships when it should have measured capability. And capability—not numbers—won the Pacific.

The phantoms never existed, but the force they represented did.

Twenty-seven American carriers did what no fleet in history had ever done. They destroyed an empire—not with ghosts, not with illusions, but with relentless reality.