Atlantis was a Real Place
A Discussion by Dan Crisp
Part 2: Refining the Search
In Part I we saw how the potential locations of Atlantis are narrowed very quickly by the description Critias gives. In this part, we will refine our search still further.
The Location of Atlantis (continued)
[Atlantis] was the way to other islands, and from these you might pass to the whole of the opposite continent which surrounded the true ocean; for this sea which is within the Straits of Heracles is only a harbour, having a narrow entrance, but that other is a real sea, and the surrounding land may be most truly called a boundless continent.
The "opposite continent" is clearly the Americas - which was taken to be a boundless land surrounding the whole ocean: very much a forgivable error.
Between the pillars of Heracles and the Americas fit Atlantis and a string of islands of unspecified size, again suggesting that Atlantis was fairly close to the pillars.
Atlantis being "the way" or "on the way" to these other islands and the Americas suggests it could be skirted without much of a detour. We are not told, for example,
be crossed to reach the other side of the ocean, as might have been the case with an island formed by the uplifted mid-Atlantic Ridge.
On the other hand, being "the way" to the other side suggests you
had to pass Atlantis, which would be the case if Atlantis were
immediately outside the pillars, on the European and/or African coast, or an island dominating the straits.
...the island was larger than Libya and Asia put together... [Tim.]
...Atlantis, which, as I was saying, was an island greater in extent than Libya and Asia... [Crit.]
I gather that "Libya" refers to all of North Africa, except Egypt. However, "the parts of Libya as far as Egypt" suggest Egypt is included in Libya - otherwise, Atlantis would be said to have controlled all of Libya. "Asia", meaning Asia Minor, is usually equated with modern Turkey. We do not need to know the exact limits of these ancient territories because they still tell us important things about the scale of Atlantis.
The usual interpretation is that Atlantis was a genuine island of continental proportions. Taking this passage in isolation, South America itself is a plausible candidate. However, if Critias were talking about areas of land, it would make no sense to define it as a large, somewhat indeterminate area (Libya) plus a much smaller, better defined area (Asia). One is something like ten times larger than the other (perhaps more, depending on how far south you think Libya stretched). Furthermore, Atlantis is said to be "larger" or "greater in extent". Critias was clearly talking in approximations and "larger than Libya" would have been sufficient.
The Asia Critias refers to must be of significant size compared with Libya for it to be worth mentioning. Remembering the singular importance already placed on living by the sea, I suggest that Critias was not talking about the areas of Libya and Asia, but the lengths of their coastlines. That is to say, Atlantis was to be understood principally as a coastal region with a length on the order of the distance from the Straits of Gibraltar to the Bosporus via north Africa. (We can use this approximation even if Critias meant to exclude Egypt and Palestine because he said Atlantis was larger than Libya and Asia.)
Taking the edges of the continental shelves as the Ice Age shore lines, this is equivalent to
the European coast from the straits to Scandinavia, passing around the British Isles and the North Sea; or
the coast of Africa from the straits to the Equator, or thereabouts.
All these and their descendants for many generations were the inhabitants and rulers of diverse islands in the open sea;
This phrase is so non-committal that we can not draw any firm conclusions from it. It may mean Atlantis was an archipelago proper. Or that Atlanteans colonised a variety of islands over a wide area, as well as their home territory, wherever that was. "In the open sea" may mean the islands were widely dispersed. Or it may simply mean "in the Atlantic Ocean", as opposed to "in the Mediterranean, which is, by comparison, only a harbour". In this case, the territories described may or may not be confined to the margins of the ocean.
...and there was an island situated in front of the straits which are by you called the Pillars of Heracles; [Tim.]
This might be highly significant, or highly misleading, depending on how firmly we grasp our first impression. As I have already mentioned, "island" need not be taken too literally. Certainly, the tradition of a continent-sized land mass proper that sank to the ocean floor is well and truly dismissed by our knowledge of the Atlantic Ocean floor and the manner of its formation. But if we find many other aspects of Critias' story are perfectly sensible, should we discount the whole matter on the basis of this one word?
It has been pointed out that "in front of" is a rather specific translation and it should simply be read "before" [I have not heard that there is any dispute about this among Greek scholars.]: implying somewhere outside (west) of the straits. But we already knew that. This paragraph does not help us to draw any further conclusions.
"...which are by you called..." does not imply the pillars are some obscure landmark known only to them: only that the Greeks called the well known straits by the name of a Greek hero.
To his twin brother, who was born after him, and obtained as his lot the extremity of the island towards the pillars of Heracles, facing the country which is now called
the region of Gades in that part of the world, he gave the name which in the Hellenic language is Eumelus, in the language of the country which is named after him,
It is accepted that Gades is the region around Cadiz in southern Spain. That is to say, one "end" of Atlantis reached southern Spain, all or most of the way to the straits. We do not need to determine the exact edge of the kingdom, because Atlantean colonies stretched into the Mediterranean from the pillars, so the area close to Cadiz was occupied one way or another.
This is compatible with everything we already know. However, distinguishing Gades (Cadiz) from the pillars, as well as letting us know where the name came from, introduces the matter of scale again. The edge of the Atlantean kingdom must have been very close to Gades and the pillars: therefore, almost certainly on the continent. The pillars are mentioned first, an important geographical landmark, marking the end of the Mediterranean. From any substantial distance, this is as much as needs to be said; facing Gades is the same thing as facing the pillars. From even further afield, the Americas, say, nowhere can be said to face Gades or the pillars: New York and Miami can both be said to face Cadiz, but only in a sense so broad and artificial as to be meaningless.
Gades marks the extremity of Atlantis, which, being on the northern side of the straits, rather suggests that Atlantis stretched from there northwards. It is still possible that it was mostly to the south, on the African coast, with one kingdom across the straits.
The foregoing paragraphs are somewhat compatible with an island formed by an uplifted Mid-Atlantic Ridge. In this case, the island would be long and narrow, stretching from the North Atlantic to the Equator; as long as the Libyan-Asian coast, according to Critias, which would be about right.
However, to expose the areas that are now the continental shelves, we are only talking about hundreds of metres of relative sea level change; to expose the Mid-Atlantic Ridge, we would be talking about thousands of metres. Although we might be tempted to think the extreme geological conditions involved were the very ones that culminated in extraordinary seismism, volcanism and the end of the Ice Age - and the disappearance of Atlantis - there is much more work required to prove the plausibility of this idea, as against the sea level changes we have come to expect from the waxing and waning of the ice caps. I am not aware of any suggested mechanism for the Ridge to well up and become habitable.
Furthermore, such an island would have been so important a feature of the Atlantic Ocean that it would surely have warranted a more vivid description. Critias speaks of the Mediterranean as a harbour on the Atlantic, which is surrounded by a continent on the other side; which is as it we know it today, not as it would have been if such an enormous island divided the ocean in two.
Such an island would stretch far north of Gades and the pillars. Even if the island is at it widest at the appropriate latitude, the point opposite the straits could hardly be called "the extremity of the island".
For these reasons, I suggest we can dismiss the suggestion that Atlantis was a true island formed by the Ridge, because it simply does not fit the clues we are given.
With the Mid-Atlantic Ridge eliminated from our enquiry, there are no possibilities left for a true island of Atlantean proportions. That means
"island" must be interpreted loosely, not literally;
the only places left that fit the clues so far are the Atlantic coasts of Europe and Africa, just outside the pillars of Heracles and extending for thousands of kilometres.
End of Part II
There are only two candidates left that match the details of Critias' description of the location of Atlantis. In Part III we will see that only one of these matches the remaining clues.