Divers keep finding stone streets of a sunken civilization off the coast of Cuba

Off the western tip of Cuba, sonar operators once mapped what looked like an organized grid of stone structures resting in total darkness nearly half a mile beneath the surface. More than two decades later, no scientific team has confirmed what those shapes actually are, and the story of the “Cuban underwater city” says as much about the limits of sonar interpretation as it does about any lost civilization.

What the Sonar Actually Showed

In 2001, marine engineers Paulina Zelitsky and Paul Weinzweig, working through a Canadian survey company called Advanced Digital Communications in partnership with the Cuban government, recorded sonar images off the Guanahacabibes Peninsula in Pinar del Rio Province, according to the summary of contemporaneous reporting compiled on the Cuban underwater formation Wikipedia page. The sonar readings appeared to show symmetrical, geometric shapes spread across roughly two square kilometers, at depths ranging between about 600 and 750 meters. On a return expedition, the team deployed an underwater remotely operated vehicle that captured footage interpreted by some observers as pyramid-like and circular formations built from large, smooth blocks resembling hewn granite. Zelitsky herself was careful in her public statements at the time, telling reporters it would be “totally irresponsible” to describe the find as a city before gathering actual physical evidence, even as she called the structures “a really wonderful” formation that resembled an urban center.

Scientists Pushed Back Almost Immediately

The skepticism arrived just as quickly as the initial excitement. National Geographic senior editor John Echave told reporters the formations were “interesting anomalies,” but cautioned that without a firsthand dive to examine them, “it will be difficult to characterize them.” Oceanographer Robert Ballard, famous for locating the wreck of the Titanic, was blunter, saying the depth alone made a human origin implausible and warning that sonar interpretation can resemble reading an inkblot, where “people can sometimes see what they want to see.” Cuban marine geologist Manuel Iturralde raised a more specific technical objection: based on established models of sea-level change, he calculated it would have taken roughly 50,000 years for structures built at the coastline to subside to the depth where the sonar readings were recorded, a timeline that predates any known human civilization capable of quarrying and arranging stone on that scale. A specialist in underwater archaeology at Florida State University summed up the broader scientific reaction bluntly, saying the structures were “out of time and out of place” for anything known from the pre-Columbian Americas.

Why No One Went Back Down

Part of what has kept the site a mystery rather than a settled scientific question is that no confirmed follow-up expedition ever took place. According to the same compiled reporting, subsequent dives to physically examine and sample the formations never materialized, largely because of political and financial obstacles tied to conducting deep-water research in Cuban waters during a period of strained relations with the United States and limited international funding for speculative marine archaeology. Without physical samples, cores, or close-up imaging beyond the original sonar and remotely operated vehicle footage, there has been no way to determine whether the shapes are cut and fitted stone, naturally occurring basalt or limestone formations, or some combination shaped by geological processes like fracturing and erosion along fault lines. Skeptics have specifically pointed to pareidolia, the well-documented human tendency to perceive deliberate patterns and familiar shapes in ambiguous or randomly organized data, as a likely explanation for why grainy sonar returns were interpreted as streets, plazas and pyramids in the first place.

The Guanahacabibes Coastline Itself

The formations, whatever their origin, sit off a peninsula that is itself a recognized natural landmark. Guanahacabibes, at the western end of Cuba, is protected largely for its terrestrial and marine ecology rather than for any confirmed archaeological remains, and the surrounding waters include karst limestone geology known for naturally forming dramatic underwater topography, including sinkholes, ridges and blocky fracture patterns that can appear artificial in low-resolution sonar without ever having been touched by human hands. That geological context matters because it gives skeptics a plausible natural explanation that does not require inventing an unknown ancient culture: block-like limestone terrain, shaped by tectonic fracturing and later reshaped by underwater currents, can visually mimic cut stone architecture on a sonar display even though no comparable dry-land structures of similar scale, materials or arrangement have ever been documented anywhere in the Caribbean from that era.

A Case Study in Unresolved Evidence

More than two decades after the initial sonar survey, the Cuban underwater formation remains exactly what its cautious original discoverers said it might be, an interesting anomaly rather than a confirmed archaeological site. No peer-reviewed excavation, radiocarbon dating or geological core sample has been published to settle the question in either direction, and without a properly funded return expedition capable of retrieving physical samples from those depths, the formation is likely to remain a subject of speculation rather than scientific consensus. The story nonetheless offers a useful lesson in how initial sonar data, however striking, requires direct physical verification before it can support claims about lost civilizations, and how the absence of that follow-up work can leave a genuinely intriguing anomaly permanently suspended between mystery and explanation.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview