Paper Balloons, Secret Censorship, and the First Intercontinental Drone Campaign

On May 5, 1945, a group of six civilians gathered near Bly, Oregon, for a picnic. Their encounter with a strange object in the forest would become one of the most unusual and tragic episodes of the Second World War on the American mainland.

The object was a Japanese Fu-Go balloon bomb.

Elsie Mitchell, who was pregnant, and five children were killed when the group encountered the device and disturbed it. Their deaths remain the only fatalities caused by enemy action in the continental United States during the Second World War.

The attack was not delivered by aircraft, submarine, or conventional missile. It came from a paper balloon carried thousands of miles across the Pacific by the jet stream.

Japan's Fu-Go campaign was ultimately a military failure. But the underlying idea was significant: an inexpensive, unmanned weapon could cross an ocean, exploit atmospheric conditions, penetrate an adversary's homeland, and create uncertainty without requiring a conventional delivery platform.

That makes Fu-Go more than an obscure wartime curiosity. It is an early example of a problem that would become increasingly important in modern warfare: how much strategic disruption can be created by a weapon whose cost and technical sophistication are relatively low?

The Fu-Go campaign began in November 1944 and continued until April 1945.

Japan launched more than 9,000 hydrogen-filled balloons, often cited at approximately 9,300. Each balloon was roughly 33 feet in diameter and carried incendiary and antipersonnel explosives.

The objective was straightforward in concept. The balloons would rise to approximately 30,000 feet, enter the prevailing high-altitude winds, and travel eastward across the Pacific toward North America.

The journey could cover more than 6,000 miles and take approximately 70 hours.

The weapon therefore did not need an engine to cross the ocean. The atmosphere itself provided the propulsion.

That was the central innovation.

The balloons were equipped with an automatic altitude-control system. As hydrogen-filled balloons rose or descended, mechanical systems could release ballast or vent hydrogen to keep them within the appropriate altitude range. The system was designed around the assumption that the jet stream could provide the necessary long-distance movement.

The balloons themselves were also products of wartime improvisation. Japanese engineers developed large paper envelopes using traditional materials and techniques, including a konnyaku-based adhesive. Production involved wartime labor, including schoolchildren.

The result was an unusual combination of advanced atmospheric understanding and relatively simple materials.

Fu-Go emerged at a point when Japan's ability to conduct conventional strategic attacks against the continental United States was severely constrained.

The Doolittle Raid of April 1942 had demonstrated that the Japanese homeland could be reached by American forces. Japanese planners subsequently considered ways to retaliate against North America without relying on conventional aircraft capable of making the round trip.

The balloon offered an asymmetric alternative.

It did not require a pilot.

It did not need an engine capable of sustaining an intercontinental flight.

It did not need to carry enough fuel for the journey.

Instead, Japan could manufacture the weapon relatively inexpensively and allow the Pacific's atmospheric circulation to do most of the work.

The intended effects extended beyond physical destruction. Japanese planners hoped the balloons could ignite forest fires, cause casualties, create anxiety among the American population, and force the United States to divert military and public-safety resources.

The campaign therefore depended partly on psychology and uncertainty.

A weapon did not necessarily have to destroy a major target to create a problem.

It only had to make the defender believe that another attack might be coming.

Fu-Go demonstrated an important limitation of long-range unmanned weapons: reaching a continent is not the same as reaching a target.

The balloons had no sophisticated guidance system capable of directing them toward a particular installation, city, or military facility.

Their route depended heavily on atmospheric conditions.

This created an enormous targeting problem.

Japan could launch thousands of balloons and rely on the jet stream to carry them toward North America, but it could not reliably determine where individual balloons would ultimately land.

Some balloons were recovered or reported thousands of miles from their launch points. Documented incidents occurred across the western and central United States and Canada, with reports extending as far east as Michigan.

The campaign therefore achieved something technically remarkable while remaining strategically imprecise.

The difference matters.

A weapon can have extraordinary range and still have limited military utility if its ability to place an effect on a meaningful target is poor.

American authorities initially had to determine what they were dealing with.

Reports of balloon sightings and recoveries generated an unusual intelligence problem. The objects appeared in widely separated locations, and investigators needed to determine whether they were connected.

The physical evidence eventually became important.

Sand recovered from balloon equipment provided clues about where the devices had originated. U.S. scientists and military investigators analyzed the material and compared it with geological samples from Japan.

This investigation involved organizations and specialists outside conventional military intelligence, including the U.S. Geological Survey's Military Geology Unit.

The lesson was significant: intelligence about a new weapon does not necessarily come from the weapon itself.

It can come from its materials.

The sand beneath a balloon could reveal information that the weapon's designers never intended to disclose.

The most consequential known infrastructure incident occurred in March 1945.

A Fu-Go balloon contacted a high-voltage transmission line serving the Hanford Engineer Works in Washington State, temporarily interrupting electrical power.

Hanford was a critical part of the American wartime effort to produce plutonium for the atomic bomb.

The incident did not cause a catastrophe. Backup safeguards allowed the facility to recover, and full operations were reportedly restored after approximately three days.

But the event illustrates a different dimension of the balloon campaign.

The weapon did not have to deliberately target Hanford to create a serious strategic concern.

The balloon's ability to arrive in the vicinity of critical infrastructure was enough to expose a vulnerability.

This is one of the most important aspects of the Fu-Go story. The threat was partly created by uncertainty.

If a weapon can arrive somewhere it was not specifically designed to reach, every sensitive facility within its possible range becomes part of the risk calculation.

The United States also faced an unusual information-security problem.

By early 1945, American authorities understood that public reporting of balloon incidents could potentially help Japan evaluate the success of the campaign.

If newspapers reported where balloons had been found, Japanese analysts could potentially learn about their geographic reach and effectiveness.

The U.S. Office of Censorship therefore requested that media organizations refrain from publicizing balloon incidents.

This created a deliberate information gap.

American officials wanted to prevent the Japanese government from receiving feedback, while also limiting public alarm.

The strategy had an obvious benefit: it reduced the amount of information available to the adversary.

But it also created a difficult public-communication problem.

People were encountering unexplained objects and incidents without necessarily receiving an explanation of what was happening.

The Fu-Go campaign thus became not only a physical-security problem but also an information-management problem.

Despite the scale of the effort, Fu-Go did not produce the strategic effects Japan had hoped for.

More than 9,000 balloons were launched, but only a few hundred confirmed sightings and recoveries were documented in North America. Estimates suggest that perhaps around 1,000 balloons successfully reached the continent, although the exact number remains uncertain.

The balloons were difficult to control.

Weather affected their movement.

The devices were not capable of precision targeting.

And the conditions needed to produce widespread forest fires did not materialize on the scale Japanese planners hoped.

The wet winter and spring conditions in much of the western United States reduced the likelihood that incendiary devices would produce major fires.

The campaign also suffered from a basic feedback problem.

Japan had limited information about what happened to the balloons after launch. American censorship reduced the amount of useful information that could reach Japanese planners through public reporting.

In April 1945, Japan halted the launches.

The operation had consumed significant effort without producing the intended strategic results.

Yet describing Fu-Go only as a failure misses what makes the campaign historically important.

The balloons were inexpensive compared with conventional strategic bombing systems.

That did not make them harmless.

A defender still had to investigate sightings, recover devices, assess unexploded ordnance, protect infrastructure, and determine whether the objects represented a larger campaign.

The cost of the weapon and the cost of responding to it were not necessarily equivalent.

Fu-Go depended on the jet stream.

The same atmospheric system that made the campaign possible also made it difficult to control.

Weather was simultaneously Japan's delivery mechanism and a source of uncertainty.

Modern security planning similarly has to account for systems whose behavior depends partly on environmental conditions.

The balloons demonstrated extraordinary geographic reach.

But their inability to reliably control their final destination limited their military value.

Long-range delivery without adequate guidance can create impressive technical statistics without producing reliable strategic effects.

The investigation of balloon debris demonstrated the value of interdisciplinary intelligence.

Geology, materials analysis, meteorology, engineering, and military intelligence all contributed to understanding the campaign.

A new threat may reveal itself through physical evidence that initially appears unrelated to conventional intelligence work.

American censorship reduced the information Japan could obtain about the balloons' success.

That mattered because the campaign depended on large-scale deployment without reliable direct observation.

An attacker that cannot determine whether its weapons are reaching their intended effects faces a serious operational problem.

The Hanford incident demonstrates why infrastructure protection cannot depend entirely on preventing every possible attack.

Unexpected events can occur.

Redundant systems, safeguards, recovery procedures, and operational resilience can prevent an isolated incident from becoming a strategic disaster.

Fu-Go should not be treated as a direct technological equivalent of a modern missile or drone.

The differences are enormous.

Modern unmanned systems can use satellite navigation, sensors, digital communications, autonomous control, precision guidance, and sophisticated targeting.

Fu-Go had none of those capabilities.

The historical connection is instead conceptual.

Fu-Go demonstrated that an adversary could use a relatively inexpensive unmanned delivery system to exploit a geographic or environmental pathway that conventional defenses were not designed around.

That underlying logic remains relevant.

Modern drones, loitering munitions, autonomous maritime systems, and other low-cost unmanned platforms can create security challenges that are not necessarily proportional to their individual technical sophistication.

The important question is therefore not simply, "How advanced is the weapon?"

It is also:

What pathway does the weapon exploit, what systems can it reach, and how expensive is it to defend against?

Fu-Go's answer was the atmosphere.

A modern equivalent could involve airspace, maritime routes, communications networks, infrastructure dependencies, or other systems that provide an attacker with an unexpected route toward a target.

The most enduring lesson of Fu-Go may be the problem of uncertainty.

A conventional attack often produces an obvious event: an aircraft appears, a missile is detected, or an installation is struck.

Fu-Go created a different problem.

A balloon might be seen hundreds or thousands of miles from previous incidents.

Nobody could immediately know how many more were coming.

Nobody could know precisely where the next one would land.

Nobody could know whether the next device would cause a forest fire, damage infrastructure, injure civilians, or simply disappear into the landscape.

That uncertainty has strategic consequences.

Security organizations must investigate anomalies.

Infrastructure operators must consider improbable failure modes.

Scientists may be asked to identify unfamiliar materials.

Law-enforcement agencies may need to determine whether an object represents an isolated incident or part of a broader campaign.

And governments must decide how much information to release without unnecessarily assisting an adversary.

The Fu-Go campaign therefore demonstrates a broader principle of security: the difficulty of an attack is not determined solely by the destructive power of the weapon.

Sometimes the real challenge is discovering, identifying, tracking, and responding to something that was not expected to be there.

Japan's balloon campaign did not burn down the American West.

It did not destroy a major military installation.

It did not alter the course of the Second World War.

But it demonstrated that an adversary could construct an inexpensive weapon, place it into a naturally occurring atmospheric system, and send it across an ocean without a pilot or conventional propulsion system.

It also demonstrated that the defender's response could require considerably more than simply destroying the weapon.

Investigators had to determine its origin.

Scientists had to analyze its materials.

Military organizations had to assess the threat.

Authorities had to manage information.

Infrastructure operators had to rely on safeguards.

And civilians had to confront a weapon that could appear in an unexpected place thousands of miles from where it had been launched.

That combination makes Fu-Go an unusually useful historical case study.

It was a tactical failure that nevertheless anticipated a strategic logic that would become increasingly important: low-cost, unmanned, long-range systems can create disproportionate security problems when they exploit routes, vulnerabilities, and uncertainties that defenders are not prepared to manage.

The paper balloons disappeared from the Pacific skies in 1945.

The underlying problem did not.

The following sources provide historical documentation, research, and additional context for the Fu-Go campaign:

Continue reading: Explore the primary and institutional sources above for additional documentation, photographs, research findings, and historical context surrounding the Fu-Go campaign.

If you found this analysis useful, follow The Strategic Current for more historical and strategic analysis of overlooked events, emerging threats, and the systems behind them.

AI-Assisted Content Disclosure: This article was prepared with the assistance of generative artificial intelligence. AI was used to help organize, draft, edit, and refine the article based on the supplied research materials. The underlying historical research and sources were provided for this article, and the final text was reviewed and edited for publication. AI assistance does not constitute an endorsement by any source, institution, or individual referenced in this article.

AI limitations: While care was taken to verify information, AI tools can introduce errors; readers are encouraged to consult original sources. If any discrepancies in this information are identified, please email [email protected] for immediate correction.

U.S. Copyright Office. Copyright and Artificial Intelligence: Part 2—Copyrightability. U.S. Copyright Office