Bryn Mawr Classical Review

In 2001, the French geologist and prehistorian Jacques Collina-Girard first broached the idea that a submarine elevation (now about 55 m below sea-level) lying about 15 km northwest of Cape Spartel (on the Moroccan coast, 12 km to the west of Tangier) had until about 10000 BC been an island above sea-level and might be identified with Plato's famous Atlantis.¹ In recent years, Collina-Girard has continued to work on his hypothesis, and now he presents a whole book on the subject.

In a brief "prologue" (pp. 7-11) he tells us how he first came across Atlantis in early 2001, discovering to his great astonishment how Plato's geographical description of the location of Atlantis contained "l'exacte description du détroit de Gibraltar tel qu'il était il y a 9000 ou 10000 ans avant notre ère" (p. 9), but also how the great majority of French classical scholars, led by Pierre Vidal-Naquet, very soon rejected his theory in total, basing this rejection, however, on nothing more but "états d'âme et. . .affirmations dogmatiques" (p. 10).

In his first chapter (pp. 13-39), Collina-Girard then sets out to sketch "what Atlantis is -- and what it is not." Noting that since Plato wrote down his account of Atlantis "400 ans avant notre ère" (p. 13; a rather vague indication -- we can do better than that!) thousands of publications have created their own image of Atlantis, he proceeds with a series of demonstrations of what Atlantis indeed is not: no comic strip like the one produced by the Belgian artist Edgar P. Jacob ("L'Énigme de l'Atlantide") in 1955 (pp. 15-6); no romance like Pierre Benoit's "L'Atlantide" of 1920 (p. 16); no science fiction novel like the ones written by Jules Verne ("Vingt milles lieux sous les mers") or François Bordes alias Francis Carsac ("Celui qui vint de la grande eau"; p. 17); no film like the one by Jacques Feyder ("Atlantide") of 1926, predecessor of many others (pp. 17-8); no enticing subject of what Collina-Girard derisively calls "para-archéologie", the wild and irresponsible efforts of which he sincerely detests (pp. 18-20), or of "géologie fantastique", cultivated by the likes of Immanuel Velikovsky ("Worlds in Collision", 1950) or Otto Muck (pp. 20-22). After evoking the really staggering presence of Atlantis in today's Internet (p. 23) -- with a little excursus on Donovan's famous song "Atlantis" of 1969 -- and even touching on a psychoanalytic perspective (which treats the drowned island as a symbol for the drowned memories of our early childhood, p. 24-26), Collina-Girard then neatly divides the various approaches to Atlantis into two diametrically opposed camps (p. 26): that of the "too enthusiastic", whom he compares to the all-credulous Don Quijote, and that of the "too sceptical", whom he equates with the all-too-narrow-minded Sancho Panza. This is a clever move, and it already gives us a hint that Collina-Girard will surely range himself neither with the one nor with the other camp.

After this, Collina-Girard finally gets down to where all the stuff about Atlantis originated in the first place: Plato's dialogues Timaeus and Critias (pp. 27-33). Here, Collina-Girard's major weakness soon becomes apparent: his knowledge about Plato, his works and the (historical and cultural) context in which they came into being is a purely derivative one, totally depending on the French translation of the dialogues by Luc Brisson² and on the somewhat eccentric opinions of Jean-Luc Périllié,³ who apparently believes every word of the story by which, as Plato's Critias claims, the Atlantis tale came into Critias' family. A significant number of inaccuracies shows that Collina-Girard treads on rather uncertain ground here. He claims that Plato devotes long sections to "le mode de gouvernement et l'organisation sociale de la cité atlante pervertie" (p. 27); in fact we read next to nothing about the "social organisation" of the metropolis of Atlantis, but only about the ways the Atlantean kings do business with each other and how their once excellent moral constitution declined into reckless pride and arrogance. The time-table depicting "Les grands étapes de la civilisation grecque antique", which Collina-Girard presents on p. 28, is at first sight highly misleading: next to the indication "400 avant J.-C." we read "Perte de l'indépendance grecque à la bataille de Chéronée" and "Platon", and next to "500 avant J.-C." we find "Thucydide", "Guerres [sic] du Péloponnèse", "Hérodote", "Tragédies", "Pindare", "Guerres médiques"; this would all have made more sense if Collina-Girard had written "IVe siècle" and "Ve siècle" instead of "400" and "500". His description of the first part of the Timaeus -- "un débat philosophique lancé par Socrate: il s'agit de s'accorder sur ce que pourrait être le mode de gouvernement de la ?Cité idéale? décrite dans La République" (p. 28) -- is again quite misleading, to say the least: The Timaeus begins with a recapitulation of some parts of a previous discussion about an ideal state (which shows some similarities with the ideal city described in the Republic, but is not identical with it⁴), and then Socrates states that he would very much like to see the citizens of the previously described ideal state involved in warlike action, so that they can really prove their worth. It is this wish that prompts Critias to relate how he came across Athens' most heroic deed in the past, namely its defeat of the overwhelming forces of Atlantis. Nothing of this can be found in Collina-Girard's account: he only mentions a "demande" (p. 29) of Socrates, without telling us what that "demande" is. He also gets the name of the fourth participant of the two dialogues consistently wrong, calling him "Herménocrate" instead of Hermocrates. He also states that Plato "affirme à plusieurs reprises que la tradition orale à l'origine de son récit est absolument authentique" (p. 33), but this is wrong on more than one account. First of all, Plato affirms nothing whatsoever in his own name in the Timaeus and Critias, but describes a conversation by four speakers none of whom may prima facie be regarded as a mere mouthpiece of his author. Secondly, there are in the Timaeus just two passages, where the 'truth' of the story is stressed: in Tim. 20d Critias insists upon it, but he is certainly not to be regarded as Plato's primary (or particularly trustworthy) spokesman; in Tim. 26e Socrates seems to agree with Critias' affirmation, but he has, of course, only Critias' word for it. Thirdly, nowhere does Critias (or any of the other speakers) breath a single word about an "oral tradition" lying behind the tale: the Egyptian priest refers Solon to old written records in Egyptian temples and even insists that no oral tradition could have survived the numerous global catastrophes between the distant past and the present (Tim. 23ab).

Quite remarkably -- after quoting Tim. 26c-d, where Critias states that he wants to treat the Athenians of that primeval city which defeated Atlantis as if they were the citizens of Socrates' ideal state -- Collina-Girard unambiguously declares that "la métropole idéale, décrite en détail dans le Critias, est totalement imaginaire" (p. 32), but he nevertheless wants to regard Plato's imaginary story as developed "sur fond de véracité" (ibid.), claiming to follow in this Brisson's interpretation, who in the introduction to his translation of the Critias toyed with the idea of regarding Plato as the inventor of the "roman historique."⁵ Taking his cue from this, Collina-Girard insists that "historical novels are always founded on a truth transformed by imagination" ("les romans historiques sont toujours fondés sur une vérité transformée") and that to refuse even to consider "la possibilité d'un événement réel au départ du ?roman de l'Atlantide?" is an unacceptably dogmatic stance (p. 33). After having castigated (once more) Vidal-Naquet as the main proponent of this stance, Collina-Girard turns to Périllié's much more credulous position (see above) and tries to divide the elements of Plato's Atlantis tale into two categories: on the one hand "des descripteurs qualificatifs, passionellement surinvestis par Platon" -- among these Collina-Girard reckons all details pertaining to the virtue of the old Athenians and to the depravity of their Atlantean foes --, on the other hand "des descripteurs neutres d'un cadre géographique et géologique" (p. 34); it may, however, be doubtful whether such a neat division is really feasible. He then outlines "trois catégories d'opinions sur l'Atlantide de Platon" (p. 35), which he finds already prefigured in the account of the Neo-Platonist Proclus. Two of these he has already evoked on p. 26 (see above): the "too credulous" (who take everything in Plato's tale as historic) and the "too sceptical" (who regard everything as invented by Plato, like Vidal-Naquet); he now adds a third in-between category which he will champion himself: "le mythe de l'Atlantide serait une fiction construite a partir des événements géologiques bien réels" (p. 36). To make this stance seem more plausible, he cites the examples of two ancient writers who (he claims) were regarded as tellers of unbelievable tales in their own time, but whose credibility is now restored: Herodotus and Pytheas (pp. 36-7). Regarding Pytheas this may be true, but as for Herodotus, Collina-Girard obviously knows nothing about the heated discussion raging about this author since Fehling, in 1971, claimed that he had invented more or less all his sources.

Nearing the end of this first important chapter, Collina-Girard cites as predecessors of his 'moderate' median position once more Brisson, Périllié and Proclus, whose report of an author called Marcellus he cites with approval, getting, however, (again) a few things a bit wrong: According to Proclus, this Marcellus⁶ wrote about seven rather large islands in the "Outer Sea"; Collina-Girard places them "à la sortie des colonnes d'Hercule" (p. 38), which is not borne out by Proclus' quotations.

I have had to examine this substantial first chapter in considerable detail in order to demonstrate Collina-Girard's conspicuous shortcomings in matters regarding Classical Antiquity. The following chapters can be dealt with more succinctly, as Collina-Girard here is on his own turf presenting matters which may be regarded as largely uncontroversial. Chapter 2 ("Commment reconstituer l'histoire d'un passé lointain?", pp. 41-64) gives a competent survey of the substantial progress made in geology in the 20th century (discovery of the Middle Atlantic submarine ridge, discovery of the rise and fall of water-levels in the oceans throughout the earth's history, discovery of the great periods of glaciation, discovery of the movement of tectonic continental plates: pp. 42-6), demonstrating how earlier theories which seemed to make the assumption of a big island like Atlantis in the Atlantic Ocean at least tenable had to be discarded, thus depriving such assumptions of every right to exist Moreover, since the 19th century the discovery of the great age of the earth and of the various means to measure it has brought great advances in long-term chronology (pp. 46-9), and since the 1970s methods have been developed to measure climate changes of the past (pp. 49-51). Additionally, the development of humans in past epochs has been thoroughly investigated since the middle 19th century (pp. 51-4); in this regard Collina-Girard stresses the fact that around 9000 BC the way of life of Paleolithic hunter-gatherers came rather abruptly to an end and gave way to the "Neolithic Revolution" (pp. 53-4).

At this point, Collina-Girard returns to Atlantis ("Relire l'Atlantide avec le regard de la géologie actuelle", pp. 54-6) and first of all rejects the attempts to connect the famous Santorini explosion in the mid-second millennium BC with the tale of the drowning of Atlantis: "ni le lieu ni la date ne correspondent aux indications de Platon" (p. 55). He then goes on to have a closer look at the last period of glaciation (and its aftermath) in the recent history of the earth with regard to the expansion and migrations of modern humans in Europe and adjacent Africa ("L'Europe sous la glace avant le 'déluge'", pp. 56-64). He describes how the process of deglaciation went hand in hand with a considerable rise of oceanic water-levels (125-135 m), flooding many coastal areas where humans had lived before, and he points out that this flooding chronologically occurs more or less at the same time when Plato posits the flooding of Atlantis (p. 64).

In the next chapter ("Où chercher l'Atlantide", pp. 65-78) he first of all rightly rejects the attempts of the Italian journalist Sergio Frau to move the "Pillars of Hercules" from the Strait of Gibraltar to the Channel of Sicily (between Sicily and Tunisia), underlining the fact that all Classical Greeks known to us (Anaximander, Hecataeus of Miletus, Herodotus) seem to have identified the Pillars with the Strait of Gibraltar (p. 67). He notes that Herodotus already mentions "Atlanteans" in the northwest of Africa (pp. 68-9). To demonstrate the Greeks' familiarity with the Strait of Gibraltar, he also evokes Herodotus' description of the famous circumnavigation of Africa by Phoenicians sent out by the Egyptian Pharaoh Necho (pp. 69-71) and the exploratory voyage by the Carthaginian admiral Hanno along the northwestern coast of Africa down to the Gulf of Guinea (p. 71).

Having ascertained the identification of the "Pillars of Hercules" with the Strait of Gibraltar, Collina-Girard then gives geographical and geological details of the Strait (pp. 72-78). The following (fourth) chapter (pp. 79-102) is devoted to the "Histoire du détroit de Gibraltar et de ses archipels engloutis", starting with an overview of the rise and fall of sea-levels in this region during the last 450,000 years (pp. 80-82). After this, we turn to the islands that existed before the last major rise of the sea-level and that considerably facilitated human travel between the European and the African coasts of the Eastern Atlantic (pp. 83-4). After briefly looking at the still existing Canary Islands, Azores and Madeira, which were not settled before the 1st millennium BC (pp. 85-7), Collina-Girard describes the geographical situation of the Strait of Gibraltar and its surrounding waters during the peak time of the last glacial period around 17,000 BC (pp. 87-90): at this time, there was an island about 15 km northwest of Cape Spartel, which together with six other islands then above sea-level formed a sort of archipelago that (together with ample coastal regions of southwestern Spain and northwestern Morocco that have been submerged since) made the waters west of the Strait of Gibraltar a kind of inland sea extending about 77 km east-west and up to 20 km north-south. Collina-Girard then traces the gradual vanishing of this "little Mediterranean" west of Gibraltar and of its islands over the next 8000 years (pp. 90-93): there were two accelerated periods of submergence around 12,200 BC ("meltwater pulse 1A", p. 92) and around 9500 BC ("meltwater pulse 1B", ibid.), and Collina-Girard does not fail to note that the second of these -- which leads to the final sinking of "l'île du Cap Spartel" -- remarkably coincides with "l'engloutissement de l'île Atlantide: 9600 ans avant J.-C." (p. 93). One might object, however, that this is not altogether correct: In Tim. 23e and Criti. 108e it is the war between primeval Athens and Atlantis that is situated 9000 years before Solon (and thus about 9600 BC), not the sinking of Atlantis, which only happened some time afterwards (and possibly considerably later).⁷

After describing the change of sea-levels around other coastal areas of the Iberian Peninsula during these times (pp. 93-96), Collina-Girard turns to a phenomenon that is certainly much more promising with regard to the drowning of Atlantis than the altogether very gradual rises of the sea he has traced so far: earthquakes and tsunamis. Evoking the famous earthquake that destroyed Lisbon in 1755 (p. 96), he points out that in fact there were earlier and no less catastrophic seismic events (followed by destructive tsunamis) in this region (pp. 97-8), and one of them (with an impact six times as destructive as the Lisbon one) happened around 10,050 BC. Again Collina-Girard wants to regard this as simultaneous with the sinking of Atlantis reported in Timaeus and Critias (p. 98); he does not seem to be bothered by the fact that this earthquake and tsunami is almost 500 years earlier than the "meltwater pulse 1B" he described on pp. 92-3. The same discrepancy can be observed in his summary (p. 101-2) of the geological rise and fall of the areas around the Strait of Gibraltar: Collina-Girard states (p. 101) that around 9500 BC (= 11,500 B.P.) water levels around the island northwest of Cape Spartel reached -55 m (compared to the present level), submerging some smaller islands in the vicinity; but just a paragraph later he adds that around 10,000 BC (= 12,000 B.P.) -- i.e. 500 years earlier! -- an earthquake and a tsunami led to the abrupt sinking of the island of Cape Spartel itself by 10 m, which would mean that already then water levels had risen to about -45 m. How these figures are supposed to go together escapes me. Ignoring this discrepancy, Collina-Girard once more insists on the identity of time and place of these events with the sinking of Atlantis referred to in Plato's dialogues (p. 102).

The next chapter ("Qui vivait dans le détroit de Gibraltar il y a 12000 ans?", pp. 103-24) is devoted to what Prehistory has up to now found out about the presence of human beings in these areas during these early times. Apparently there lived groups of Late Paleolithic hunter-gatherers along the coasts of modern Morocco and Algeria whose remains have been discovered in a number of caves and who exhibit some common traits that prompted the prehistorian Paul Pallary to invent the denomination "culture ibéromaurusienne" for them in 1909 (p. 104). Describing the finds of two of the caves mentioned in some detail (pp. 105-111), Collina-Girard is, of course, again on the look-out for coincidences with Plato's Atlantis story: thus he states that it coincides temporally "avec la fin de l'expansion des Ibéromaurusiens: leur culture va disparaître avec le rechauffement climatique qui met fin à la dernière glaciation vers 8000 ans B.P." (p. 108). Again I cannot but note the rather large difference (of about 3500 years) of this dating versus his earlier statements that the island northwest of Cape Spartel (his "Atlantis") had finally disappeared beneath the waves about 11,500 or even 12,000 years B.P. Collina-Girard develops yet another "coincidence" on p. 111, comparing the perceived advance of the Iberomaurusians "depuis les côtes atlantiques du Maroc jusqu'à la Tunisie" with the progress of Plato's Atlanteans "depuis la Libye jusqu'à l'Asie" -- this, however, does not strike me as a particularly remarkable coincidence, given the clearly large geographical differences of the two sides of this comparison. Collina-Girard goes on to claim that the Iberomaurusians "brutalement" (p. 111 -- how does he know it was "brutal"?) put an end to a preceding human Paleolithic culture (the "Atérien") and then traces the Neolithic developments subsequent to the Iberomaurusians (p. 112-4). After a look at the human cultures of the European coasts around the Strait of Gibraltar during these times (pp. 114-6), he focuses once more on "l'origine encore mystérieuse de l'Ibéromaurusien" around 21000-20000 B.P. (p. 114-8) and then on evidence of prehistoric human sea-voyaging not only around Gibraltar, but also in the Western and Eastern Mediterranean as well as in the Atlantic and the Pacific Ocean (p. 118-23), before returning to the "conditions de navigation préhistorique dans le détroit de Gibraltar" (pp. 123-4). The chapter ends with Collina-Girard insisting once more on the "coïncidences ... troublantes" between the geological and prehistorical "facts" concerning this region (as presented by him so far) and the Atlantis story. Now, of course, he has to prove that these are not just coincidences but that Plato's story rests on a genuine oral tradition deriving from these remote times and places.

The following (sixth) long chapter ("Les mythes, une 'machine à remonter le temps'?", pp. 125-168) sets out to prove that such a connection could indeed have existed. Early on, Collina-Girard states his firm belief that "les dernières traditions orales du monde des chasseurs-cueilleurs. . .ont constitué le fond des premiers textes de l'Antiquité classique égyptiens, grecs ou moyens orientaux" (pp. 125-6). Once more he invokes his favorite classicist, J.-L. Périllié, who takes for granted everything that Plato's Critias tells his listeners about the way his extraordinary story about Atlantis and primeval Athens came from Egypt to Athens (p. 126), and then he takes us on a truly impressive tour around the world, presenting evidence for long-enduring oral traditions of extraordinary events (especially natural catastrophes) from all continents inhabited by humans (pp. 127-165). Surely, one cannot fail to admire the tremendous efforts Collina-Girard has put into the collecting of all these examples of oral transmission; all of them, however, have in common that they commemorate events which happened more or less in the same region in which the stories were told, and thus conserve a certain significance for the later generations of the people living there; but what significance could the fact that a rather small island at the far western end of the Mediterranean submerged six or seven thousand years ago still have for Egyptians living thousands of years later and thousands of kilometers to the east? Nevertheless, Collina-Girard tries to convince us that such memories were preserved, citing a theory according to which the so-called Book of the Dead, making its first appearance around 1550 BC and containing some texts the first specimens of which can be traced back to the late 3rd millennium BC, harks back to "une connaissance préhistorique. . .des pistes vers l'océan Atlantique à travers le Sahara humide" (p. 166). Even if that were a proven fact (which it is not), it is also a fact that no part of the Book of the Dead refers to the drowning of an island; thus the assumption that the memory of this submergence was similarly recorded somewhere in an Egyptian document of which so far not even the least trace has been found, is pure speculation. Undeterred by such pedestrian objections, Collina-Girard presses on and boldly affirms (at the end of this chapter, p. 168): "Sortie du contexte européen et méditerranéen [but is one allowed to disregard this "contexte"?] ... l'histoire de l'Atlantide, et de sa transmission au fil du temps, apparaît assez banale, plausible et vraisemblable."

Thus reassured, Collina-Girard proceeds -- in his seventh chapter ("Atlantide imaginaire et Atlantide géologique", pp. 169-87) -- to a final systematic comparison between what he has presented as geological and prehistoric facts about the region west of the Strait of Gibraltar (his "Atlantide géologique") and Plato's tale, after first indulging in a little game of make-believe, in which he once more sets out his "geological and prehistoric truth" about this region as Plato might have told it, replacing the Egyptian priest by "le Géologue" and his listener Solon by "l'Historien" (pp. 169-71). After this, the comparison begins in earnest. Collina-Girard tries his best to make all elements of it neatly coincide, but in a number of cases he has to resort to special pleading and fails rather conspicuously.

This is already apparent when he tries to adapt the remarks about the geographical situation east and west of the Strait of Gibraltar in Tim. 25a: while to the east the Mediterranean appears like a harbor with a narrow entrance, the other (western) side is "really a sea" ("c'est réellement la mer" is Brisson's translation cited by Collina-Girard) and the land surrounding this sea may truly and very rightly be called a "continent" (Brisson's translation: "la terre qui entoure cette mer. . .mérite véritablement de porter le nom de continent"). But this "real sea" is for Collina-Girard just his rather tiny "mer intérieure" (the measures of which he gives as "d'environ 77 km de long et 10-20 km de large" and which is thus in fact much smaller than the sea called a "harbor" by Plato!), and the "real continent" is for him the additional coastal areas of Northwest Africa and Southwest Africa not covered by the Atlantic at that time -- this will surely not do.

Collina-Girard then cites the immediately preceding sentence (Tim. 24e-25a) which tells how navigators could easily travel from Atlantis to the other islands and from them to the "opposite lying continent around that real sea" (Brisson's translation: "ils pouvaient passer sur tout le continent situé en face, le continent qui entoure complètement cet océan, qui est le véritable océan"). Again, Collina-Girard tries to make us believe that this "véritable océan" is his small "mer intérieure de l'Ouest du détroit" (p. 172), and he does not shy away from calling the above-mentioned coastal regions "les continents situés au Nord et au Sud" (p. 173), ignoring the fact that Plato talks of one "real continent" surrounding all of the "real sea", which is obviously not the case with these coastal regions.

What bothers Collina-Girard most, however (and rightly so), is the totally different size of Plato's Atlantis ("bigger than Libya and Asia taken together") compared to the rather modest measures of his island northwest of Cape Spartel. He tries to get around this difficulty by claiming that Plato himself has contradictorily employed the expressions "grande comme la Libye et l'Asie réunies" and "toute l'Europe et l'Asie réunies" (p. 175). But where is the contradiction? The first of these expressions is twice used to characterize Atlantis' awesome size, the second to characterize its insatiable imperialist ambitions. Collina-Girard takes the second of these expressions as proof that the first also "ne désigne plus ici les dimensions de l'île Atlantide, mais l'étendue du ?règne? ou du ?pouvoir? de ses habitants" (p. 177). This is patently false: Plato's use of both expressions is absolutely consistent and non-contradictory.

Just as false is Collina-Girard's subsequent claim that Plato "parle donc clairement d'une île-capitale, située devant le détroit de Gibraltar" (p. 177). According to Critias 113c and 118ab, the capital of Atlantis is situated in a middle section of the south coast of the large island, within a plain extending several hundred kilometers both in length and breadth and surrounded by mountain-chains, so that its distance from the Strait of Gibraltar must at least be several hundred kilometers, too.

Collina-Girard then once more concentrates on the disappearance of his island, brought about by an earthquake and subsequent tsunami around 10,000 BC (pp. 179-81), calling it "étonnant. . .que Solon confirme cette date" (p. 181). I have already pointed out that 1) there is a difference of four hundred years (10,000 / 9600 BC) between the date of the geological event and Solon's date and that, moreover, Solon's date does not refer to Atlantis' disappearance but to the war between Atlantis and Athens which preceded the disappearance an unknown but possibly considerable amount of time.

After this, Collina-Girard even presents another interpretation of the famous 9000 years which are supposed to lie between the Atlantis story and Solon, namely as the fourth part of a "great astronomical year" of 36,000 years. To confirm this, Collina-Girard once more cites his philological authority Périllié,⁸ who, however, himself has to confess "que cela [i.e. Proclus' testimony that the Egyptians had already taught Plato the movement of the fixed stars] ne constitue pas une preuve" (p. 183). And can we really have it both ways -- a "real" chronological coincidence in prehistory and a purely "theoretical" date based on Egyptian astronomical conceptions allegedly passed on to Plato?

Lastly, Collina-Girard also wants to establish an agreement between Plato's statement that the sinking of Atlantis led to muddy and very shallow waters (Brisson's translation of Tim. 25d: "la mer reste impraticable ... par la boue que, juste sous la surface de l'eau, l'île a déposée an s'abîmant") and some shallow waters one can today find in the sea west of Gibraltar (pp. 183-5). Unfortunately the former island northwest of Cape Spartel, Collina-Girard's Atlantis, is now 52 m below sea-level (see map on p. 180), thus forming no "shallows" at all.

In a short "Épilogue" (pp. 189-194), Collina-Girard on the one hand repeats some of his false arguments of the preceding chapters, but surprisingly also backtracks a little: "Il est certain que l'idée d'une reprise, par Platon, d'une tradition orale concernant l'Ouest du détroit de Gibraltar, n'est qu'une hypothèse" (p. 190). Then again he insists that "le lieu, la date, le paysage et les modalités de la catastrophe coïncident trait pour trait" (p. 191), which in many respects is simply not true. He once again claims that Vidal-Naquet's rejection of any factual basis of the Atlantis story "s'apparente davantage à une position d'autorité qu'à une argumentation" (ibid.), which is not true either. He then compares the Atlantis story with the Trojan War which - so he claims - was turned by Homer from "petits accrochages guerriers entre bourgades minuscules" (p. 192) into a mighty war; this only shows that he is not aware that Bronze Age Troy was a quite important place around which a major war might well have been fought. He insists that in the case of his island of Cape Spartel underwater archeology is not likely to bring tangible results (p. 193) and that "seule la découverte d'un nouveau texte pourrait fournir un argument définitif à ce débat" (p. 192). As there is every likelihood, however, that Plato really invented his story, such a text is probably never going to turn up.

All in all, Collina-Girard has certainly succeeded in throwing some light upon some momentous developments in human prehistory in the area west of Gibraltar. Just as certainly, however, he has not found Plato's Atlantis.

1.   J. Collina-Girard, "L'Atlantide devant le Détroit de Gibraltar? mythe et géologie," Comptes Rendus de l'Académie des Sciences de Paris, Sciences de la Terre et des Planètes, vol. 333 (2001) 233-240.
2.   Luc Brisson, Platon, Timée; Critias, trad. inédite, introd. et notes, Coll. Garnier-Flammarion 618, Paris 1992, 2. ed. 1995, repr. 1999.
3.   Jean-Luc Périllié is "Maître de Conférences en philosophie ancienne" and "Directeur du département de philosophie" at the "Université de Paul Valéry, Montpellier III". It is quite astounding that for all the appeals Collina-Girard makes to this scholar in the course of this book he is apparently uncertain how to spell his name correctly: we find the form "Périllé" on pp. 126, 196, 197, 207 and the correct form "Périllié" on pp. 33, 38-9, 183, 187, 213.
4.   See H.-G. Nesselrath, Platon, Kritias, Übersetzung und Kommentar (Göttingen 2006) 55-7.
5.   Brisson, Platon, Timée; Critias (as in note 2), p. 325.
6.   Collina-Girard calls Marcellus a geographer; others have regarded him as a historian (therefore he can also be found in Jacoby's Fragmente der Griechischen Historiker: FGrHist 671 F 1-2). I have tried to show that he may very well have been the author of a romance similar to that by Antonius Diogenes; see H.-G. Nesselrath, "'Where the Lord of the Sea Grants Passage to Sailors Through the Deep-Blue Mere No More:' The Greeks and the Western Seas," in: Greece & Rome 52 (2005) [153-71] pp. 168-70.
7.   For this, see Nesselrath (as in n. 4) pp. 217-8 and 450.
8.   On p. 187, Collina-Girard clearly misunderstands Périllié's distinction between logos and eikôs mûthos in the Timaeus, because the latter refers to Timaeus' long exposition starting in 27c, which has nothing whatsoever to do with Atlantis.