Strand I · Candidate sites
For 140 years, people have pinned Atlantis to real places. The honest test is simple: hold each candidate against Plato's own specifications — beyond the Pillars of Heracles, around 9600 BCE, a vast sea power, sunk beneath the waves. Measured that way, the famous candidates don't just fall short. They fail on almost every axis at once.
Each site against Plato's four stated criteria. ✓ match · ≈ partial · ✕ fail.
| Candidate | Beyond Gibraltar | ~9600 BCE | Vast sea power | Sank in the sea | Verdict |
|---|---|---|---|---|---|
| Santorini / Thera Spyridon Marinatos; Galanopoulos & Bacon | ✕ | ✕ | ✕ | ≈ | Speculative |
| Spartel Bank Jacques Collina-Girard | ≈ | ≈ | ✕ | ✕ | Refuted |
| Doñana / Tartessos Rainer Kühne; Richard Freund | ≈ | ✕ | ✕ | ≈ | Refuted |
| Richat Structure YouTube ("Bright Insight") et al. | ✕ | ✕ | ≈ | ✕ | Refuted |
| Antarctica Charles Hapgood; Rand & Rose Flem-Ath | ✕ | ✕ | ✕ | ✕ | Refuted |
Proposed by Spyridon Marinatos; Galanopoulos & Bacon
The Late Bronze Age Minoan eruption was a real VEI-7 catastrophe that devastated a sea-trading culture — the closest thing to a historical kernel. But it is in the Aegean (not beyond Gibraltar), ~1600 BCE (not 9600), and far too small. Proponents must "divide Plato’s numbers by 10" to make it fit.
Proposed by Jacques Collina-Girard
A real submerged paleo-island in the Strait of Gibraltar — at least it is in roughly the right place. But it submerged from slow sea-level rise ~11,600 years ago, not a sudden sinking, and is tiny. The "matches Plato" version was refuted in verification.
Proposed by Rainer Kühne; Richard Freund
Marshland in southern Spain, near (but not beyond) Gibraltar, tied to the lost culture of Tartessos. National Geographic’s 2011 film amplified it — but the actual excavations found material dated to the Muslim period, and the project’s own anthropologist said no Atlantis was found.
Proposed by YouTube ("Bright Insight") et al.
The viral favorite: a circular "Eye of the Sahara." But it is a natural eroded dome of 94–104-million-year-old rock, ~40 km across, far inland in Mauritania, with no archaeology of a city. It is older than complex life on land, let alone civilization.
Proposed by Charles Hapgood; Rand & Rose Flem-Ath
Proposes the whole crust slipped, carrying a civilization to the South Pole. Plate tectonics provides no mechanism for such a rapid pole-shift, and there is no evidence of a sub-ice city. Pure fringe geology.
The Richat Structure is the candidate the internet loves most — and the easiest to settle, because we can simply look at it. This is a real NASA astronaut photograph (public domain). Those concentric rings are a 100-million-year-old eroded rock dome, hundreds of kilometres inland, with no archaeology of a city. It is geology, not masonry.
The scorecard summarizes; these are the sourced, tested claims underneath it.
One of the largest eruptions in human history (~28–41 km³ ejected), comparable to Krakatau 1883, burying the islands under pumice up to 60 m thick. Radiocarbon dates it to ~1610–1510 BCE — not 9600 BCE.
Verified 3-0.
Peer-reviewed dating pins the Minoan eruption to the mid-to-late 2nd millennium BCE — radiocarbon wiggle-matching of an olive branch buried in the ash gives 1627–1600 BCE; the only live dispute is 17th vs 16th century. That is ~8,000 years adrift from Plato’s ~9600 BCE. The "divide by 10" workaround (Galanopoulos) exists precisely because the real numbers don’t fit.
Verified 3-0. Anchors why the strongest candidate still fails.
Marinatos proposed it and the BBC popularized it, but it fails on Plato’s date and location. Proponents Galanopoulos & Bacon conceded “the date of 9600 BC … is both incredible and impossible” and applied an ad hoc “divide by 10” fix — confirming rather than testing the hypothesis. Mainstream scholarship holds every site-identification fails.
Verified 3-0.
The claim that Spartel Bank is a paleo-island submerged ~11,600 years ago matching Plato did not survive verification. What is solid: the Gulf of Cádiz sees M8–9 earthquakes/tsunamis every ~1,500–2,000 years — a real mechanism, but it does not make Spartel into Atlantis.
Atlantis identification refuted 0-3; the underlying geology verified 3-0.
Jacques Collina-Girard (2001) noted Spartel Bank — a paleo-island beyond Gibraltar — submerged ~11,600 years ago, matching Plato’s date and "beyond the Pillars" location better than anything else. Then Marc-André Gutscher’s 2005 high-resolution multibeam survey (published in Geology) settled it: by that date the island was reduced to wave-swept islets under 500 m across (at most ~5×2 km after subsidence correction), with no man-made geometric structures. His verdict: any inhabitants "would have probably been simple fishermen and not a Bronze Age culture as described by Plato."
Verified 3-0 in the candidate-sites pass. The real Gulf of Cádiz earthquake/tsunami record (8 quakes in the turbidites) is genuine — the civilization is simply absent.
Prof. Richard Freund claimed satellite-imaged structures in the Doñana marshes were tied to “Atlantis/Tarshish” (National Geographic’s 2011 Finding Atlantis). His own project’s anthropologist, Juan Villarías-Robles, said nothing exceeded expectations and items dated to the Muslim period — no Tartessos, no Atlantis.
Verified 3-0. Freund’s claims were criticized as overreach within the project.
A natural eroded geological dome ~40 km across, formed by a subsurface igneous intrusion. Its rocks cooled 94–104 million years ago, it sits far inland (contradicting Plato’s maritime power), and there is no archaeological evidence of any city. The YouTube “Richat = Atlantis” thesis is fringe pseudoscience.
Verified 3-0.
The "Eye of the Sahara," made famous as Atlantis by YouTube channels, is an uplifted, deeply eroded geologic dome of sedimentary and igneous rock — its concentric rings are differential erosion, not city walls. NASA Earth Observatory documents its geology, and NASA/ISS satellite photography of it is public domain. The real photograph is the rebuttal.
Geology corroborated by NASA Earth Observatory; image is public domain (shown on this site).
Rejecting the candidates doesn't mean Plato invented in a vacuum. Two real events sit close enough to have seeded the imagery — a city that genuinely sank in his lifetime, and a prehistoric flood. These are the honest "kernels of truth": inspirations, not Atlantis.
Helike was a Greek city in Achaea destroyed and submerged by an earthquake and tsunami in 373 BCE — a single night, beneath the sea, within Plato’s own lifetime and just before he wrote. It is a documented historical event and a plausible real-world seed for Atlantis’s "single day and night" destruction motif. This is inspiration, not identification.
The sinking of Helike is PROVEN history; the link to Atlantis is reasoned but speculative.
Ryan & Pitman (1997) proposed that ~7,600 years ago Mediterranean water breached the Bosporus and catastrophically refilled the Black Sea — a candidate source for ancient flood myths. Newer Danube-delta stratigraphy (Giosan et al.) suggests the pre-reconnection lake stood much higher than the dramatic version requires, so the catastrophe’s scale is genuinely disputed. At most a flood-myth inspiration, not Atlantis.
A real reconnection occurred; its catastrophic magnitude is contested in the literature.