Why Cases Remain Unresolved

The difference between what is unknown and what is extraordinary

A Report Is Made

On the afternoon of June 24, 1947, a private pilot named Kenneth Arnold was flying a search pattern near Mount Rainier when he saw nine bright shapes moving in echelon formation against the peaks. He had no instruments trained on them — no radar, no camera, nothing but his eyes, his aircraft clock, and the landmarks he used to judge distance. Timing their passage between Rainier and Mount Adams, he calculated a speed that seemed impossible for anything built in 1947.

Arnold told a newspaper reporter in Pendleton, Oregon what he had seen. Within days the story was national news, and the phrase he used to describe the objects' motion — like a saucer skipped across water — gave the phenomenon its name.

Army Air Forces intelligence officers interviewed Arnold and a second nearby witness. They found both credible. They also concluded, without a clearly stated technical basis, that Arnold had seen a mirage.

This is where nearly every case in this collection begins: with a person, a sighting, and a story someone had to decide whether to believe. What happens next — whether that story can ever become more than a story — depends on something Arnold's account never had.

What Investigators Need

A single account, however sincere, is difficult to verify. What moves a report from "someone said" toward "here is what can be shown" is corroboration: independent instruments and independent witnesses recording the same event from different vantage points, at the same time, in ways that can be checked against each other.

The RB-47 encounter of July 17, 1957 shows what that can look like. Flying an electronic-reconnaissance mission over the Gulf Coast, an Air Force crew watched an intensely bright light close on their aircraft and change direction sharply. Their electronic countermeasures gear, tuned to a specific microwave band, intermittently picked up a signal tracking counter to their own flight path. Ground radar at a separate Air Defense Command station near Duncanville, Texas, logged a target in the same position and relayed it back to the crew by radio. Three channels — eyes, receiver, and radar — appeared to register the same thing across ninety minutes and roughly seven hundred miles.

That kind of layered evidence is what lets investigators test explanations rather than merely guess at them. But corroboration has limits of its own: no written report from the crew survives from 1957. The detailed account investigators eventually worked from was reconstructed from memory a decade later — a reminder that how evidence is preserved matters almost as much as how it was gathered.

When Metadata Is Missing

In January 2015, an F/A-18 aircrew off the Florida coast used an infrared targeting pod to record an object that appears to spin in place while holding a steady course. The clip, later called GIMBAL, shows a bright, elongated shape against a flat sky — and little else.

That "little else" is the problem. Metadata is the technical scaffolding around a recording: timestamps, sensor settings, aircraft position, range and altitude data — the information that lets an analyst check a claim against a recording rather than take the recording on faith. GIMBAL's public version carries almost none of it. There is no radar track to compare against the infrared image, no visible-light footage from a second camera, no confirmed distance to the object.

One technical explanation fits what is visible: infrared glare from a distant aircraft, with the apparent spin produced by the pod's own tracking mechanism reaching a limit and "derotating" its image, a known behavior of this hardware. It is plausible. It is not confirmed, because the data that would confirm or rule it out was never released, if it still exists at all.

Missing metadata does not mean something extraordinary happened. It means the record cannot answer the question either way.

Why Apparent Speed Can Mislead

Recorded the same month and from the same deployment as GIMBAL, a second infrared clip shows an object crossing rapidly over the ocean surface. Aircrew can be heard on the recording remarking on its speed. For years after the video became public in 2017, that impression — something small, low, and extremely fast — was the case's defining fact.

In February 2025, the Department of Defense's All-domain Anomaly Resolution Office published a geometric reanalysis. Using historical wind data and a modeled range of the aircraft's possible headings, analysts placed the object near 13,000 feet — not skimming the waves but drifting at altitude, likely carried by wind of roughly 69 miles per hour. What had looked like a fast, low object was consistent with motion parallax: the same illusion that makes a nearby fence post appear to race past a car window while a distant mountain barely seems to move.

The reanalysis worked from a compressed copy of the video — the original file had been lost — and could not draw on complete flight telemetry, real limits that keep this from a full identification. But it shows how a viewer's impression of speed and altitude, formed in the moment, can be an artifact of geometry rather than a measurement of anything.

Insufficient Data, or Corroborated Evidence

Two cases can both remain unresolved and mean entirely different things.

Kenneth Arnold's 1947 sighting is unresolved because there is too little to work with: one witness, no instruments, an official mirage finding that later reviewers have disputed without offering anything sturdier in its place. That is insufficient data — a case where the record simply cannot decide the question.

Contrast that with the night of August 13, 1956, over Suffolk, England. Ground radar at RAF Bentwaters, a separate radar unit at RAF Lakenheath, the onboard radar of a scrambled RAF Venom fighter, and multiple ground and air observers all tracked related targets for several hours, including a target that appeared to reposition itself behind the intercepting aircraft. The Condon Committee's own investigator, no friend of extraordinary claims, wrote that conventional explanations could not be ruled out but judged their probability low. That is corroboration: several independent channels, evaluated and still not explained.

Both cases are labeled unresolved. Only one clears the higher bar this project reserves for corroborated evidence. Unresolved means that the available evidence does not support a reliable identification. It does not imply extraterrestrial or nonhuman origin. The difference between these two cases is not how strange they sound. It is what can actually be checked.