Bryn Mawr Classical Review

Bryn Mawr Classical Review 2008.08.22

Pierre Vidal-Naquet, L'Atlantide. Petite histoire d'un mythe platonicien.   Paris:  Les Belles Lettres, 2005.  Pp. 199.  ISBN 2-251-38071-X.  €18.00.  

Reviewed by Heinz-Guenther Nesselrath, Georg-August-Universität (
Word count: 4652 words

[The reviewer apologizes for the lateness of this review.]

In the first sentence of his "Introduction" (p. 13) Pierre Vidal-Naquet (henceforth VN) claims that the gestation of this book began half a century ago, in the early 1950s, when he prepared a master's thesis on the "Platonic conception of history". It was in the 1960s that VN's first (and already then influential) publications on Plato's Atlantis came out in print: first his article "Athènes et l'Atlantide. Structure et signification d'un mythe platonicien" (first in Revue des Études Grecques 77, 1964, 420-444, then included in all editions, reprints and translations of VN's famous book "Le chasseur noir", Paris 1983 (2nd ed.) = 1991, 335-360), in which his interpretation clearly brought out that the Atlantis story is a Platonic invention pure and simple. In more recent times, he published interesting articles on the afterlife of the Atlantis story in later European history.1 In his present book,2 both themes come together, making it indeed a "little history of a Platonic myth" from its beginnings, which VN combines (sometimes a bit too extensively, perhaps) with reminiscences about his own intellectual and academic history.3

The "Introduction" then covers some of the more recent popular fantasies concerning Atlantis: its identification with Minoan Crete (p. 13f. 19f.4) or Thera/Santorini (p. 20), with a little submerged island to the west of the strait of Gibraltar (p. 18), with Troy (p. 18f.; by Zangger, on whom VN is much too lenient); it also reviews some more serious modern contributions to the subject (p. 16-18), many of which -- perhaps not surprisingly in a French author -- are French.5

The first chapter deals with Plato's invention and handling of the Atlantis story ("Au commencement était Platon", p. 23-43). In his characterization of the two Platonic texts in which the story is presented, VN once more6 expresses his belief (p. 25) that Plato always wanted to end the Critias where it actually does in our transmitted manuscripts, i.e. just before the great war between Atlantis and primeval Athens is about to erupt. The only reason, however, which VN adduces for this belief is not sufficient: The fact that Socrates and his dialogue partners wonder at the beginning of the Timaeus why an expected fourth participant in the conversation is missing is no compelling argument for a premeditated breaking-off of this conversation just when it's about to get really exciting. On the contrary: there are enough arguments to show that the Critias is really an unfinished work and why it is unfinished.7 In VN's presentation of primeval Athens and Atlantis as described by Plato (p. 28-32), there are some irksome inaccuracies which might easily have been avoided: When he states that Athens was in those times governed by Athena and Hephaestus,8 he clearly goes beyond the Platonic text which makes Athena and Hephaestus the founders but not the governors of this Athens. An even more glaring fault concerns the very end of the Critias: At this point, there is no "décision de Zeus de détruire l'Atlantide" (p. 30); Zeus wants to chastise the Atlantean kings to lead them back to their former virtue, not to destroy them (Criti. 121bc). The geological destruction of Atlantis occurred -- according to indications of the Platonic text (Tim. 25cd) -- some time after the lost war against Athens, as part of a totally natural cyclical upheaval of the earth (Tim. 22d-23b).9 It may also be doubted whether one should really characterize Atlantis as "utopie négative" (p. 31), just because she happens to be the (in Plato's tale quite necessary) opponent of those virtuous Athenians; in their state of glory as described in Criti. 120e-121a, the Atlantean kings look like superhuman heroes, not villains. Concerning the nature of Plato's tale (p. 33), VN may certainly be right in calling it "plutôt...un pastiche de l'histoire", but by restating his thesis of 1964 "que Platon avait voulu pasticher et critiquer Hérodote" he probably narrows down Plato's focus too much.10 He may also go too far in interpreting the mysterious oreichalkos found on Atlantis (and only there) as an allusion to the Attic silver mines at Laurium (p. 34). The final pages of the chapter are devoted to other Platonic writings (Republic book VII, p. 36-38; Politicus 268d-274e, p. 39f.; Laws III and IV, p. 40f.) and their connections with the Atlantis story, thus underlining that Plato himself is the best source for this tale.

In chapter 2, VN surveys the history of the Atlantis story in post-Platonic Antiquity ("Atlantides antiques", 45-63). He starts with the historian Theopompus, who -- in one of the numerous excursuses of his Philippika -- produced what might be called a parodic pastiche11 of Plato's Atlantis story and one of the earliest literary reactions to it.12 VN's characterization of Theopompus' piece could be more accurate. First of all, the great continent described by Theopompus is named Meropis and not Meropia (as VN calls it); furthermore, in the summary of Theopompus' account preserved by Aelian (VH III 18), which is more or less our only detailed source for Theopompus' fantasy, the two antagonistic cities, Μάχιμος and Εὐσεβής (not Eusébie, as VN calls the second), are not "séparées par une prodigieuse étendue" (46) -- this chasm only appears towards the end of the summary in one of the outlying regions of the fabulous continent, where also the two rivers Ἡδονή and Λυπή flow. VN briefly mentions another Aelian passage (NA XV 2), in which the kings and queens of Atlantis itself make a brief appearance, because they used to wear headbands taken from the skin of the dreaded "sea-ram",13 and he not implausibly argues that the comparison made here with the tiaras of early Hellenistic monarchs may indicate that the source of this extraordinary tale may go back to High Hellenistic times.

The next author on VN's list is Aristotle (46-48). Plato's greatest pupil obviously did not consider Atlantis as something even remotely connected with reality. VN stresses the fact that Aristotle did not breathe a word about Atlantis in a passage where he might have easily been tempted to do so, namely in De caelo II 14 p. 298a9-15, where Aristotle favorably comments on the idea that the waters beyond the Pillars of Hercules and those around India do really belong to one and the same ocean (for which people advocating this idea even cite the existence of elephants both in North Africa and in India as supporting evidence). When dealing with another Aristotelian text (Meteor. II 1 p. 354a22-3), however, VN gets some things wrong:14 He believes that Aristotle here follows Plato (Tim. 25d) in stating that the waters beyond the Pillars of Hercules are shallow "because of the mud" and that "sur ce point Aristote s'est laissé duper par Platon" (48). Now, if ever a pupil was not duped by his teacher, it was Aristotle -- it was simply commonly received Greek knowledge in Aristotle's (and already Plato's) time that you could not sail beyond the Pillars of Hercules because of the shallows out there.15

After a brief look at a passage in the Pseudo-Aristotelian Mirabilia,16 VN moves on into Hellenistic times and -- after a glance at the philosopher-polymath Posidonius who according to VN believed that Atlantis once had existed17 -- to Diodorus and his depictions of 'Atlantioi' (49-50). Again, however, VN unfortunately gets several things wrong. He claims that "Diodore ne mentionne pas sa source" (49), the mythographer Dionysius Scytobrachion,18 but Diodorus names him twice, in 3.52.3 and in 3.66.5-6. Moreover it is not correct that Diodorus (reproducing Dionysius) "nous parle des Atlantes en disciple d'Hérodote plutôt que de Platon" (49): Dionysius not only used Herodotus and Plato, but also Theopompus' Meropis (see above) for the (in some respects rather surprising) picture of his Atlanteans.19

After mentioning the Elder Pliny's acceptance of the one time existence of Atlantis,20 VN turns to Plutarch (51-2) and claims that he was not sceptic about the Atlantis tale21 and even added more details, providing, e.g., the names of the Egyptian priests from whom Solon had got it. VN then detects "une contradiction interne" between various references to the Atlantis story in Plutarch's Life of Solon (in 26,1 Plutarch mentions "Psenophis of Heliopolis and Sonchis of Saïs" as Solon's informants; in 31,6 he names "the learned men of Saïs" as sources for the Atlantis tale), but he does not consider whether such discrepancies might not be due to internal discussions of Plato's Atlantis story within the Academy which Plutarch may have known; some of the "additional" information presented by Plutarch might, e.g., go back to Crantor, who allegedly went to Egypt around 300 BC and inquired there about a possible historic basis of Plato's tale.22

VN then briefly comments on other authors and their (short) references to Atlantis (52-55, not always in chronological order): Ammianus Marcellinus, Philo of Alexandria, Tertullian ("le premier chrétien à parler de l'Atlantide", 55), Arnobius. More space is devoted to the Neo-Platonist Proclus (56-60) and his comments on Atlantis, which provide a short survey of the various interpretations of the Atlantis story within the Academy, starting with Crantor who saw real history in it (see above) and moving on to symbolic/allegorical interpretations. Proclus apparently accepts both 23, but favors the latter approach, regarding the tale as expressing an "opposition...cosmique entre le monde de l'Un et celui de la Dyade, entre le Même et l'Autre, le Mouvement et le Repos, la Limite et l'Illimité" (59). 24 The latest ancient author to appear in this chapter is Cosmas Indicopleustes (60-62), who took the Atlantis tale as a confirmation of the Great Deluge in the Bible and even as an adulterated version of the Bible's tale.

In the third chapter ("Le retour des Atlantes 1485-1710", 65-82), VN discusses the re-entry of Atlantis into (Western) European thought in early modern times.25 He singles out Marsilio Ficino's Latin translation of Plato's Critias in 1485 and Columbus' discovery of America in 1492 as the starting point of a new Atlantis discussion, which in the 16th century was mainly conducted by Spanish authors, mostly revolved about the identification of Atlantis with the newly discovered American continent (67-70) and also tried to connect this with the myth of the Ten Lost Tribes of Israel (first to be found in IV Esra 13,40-44) and with other "data" found in the Bible itself -- all this with the aim of bolstering the Spanish claim to the new Spanish possessions in Central and South America. On p. 70-72, VN shows that Montaigne, for one, did not accept the identification Atlantis/America, but pointed out the elements irreconcilable with this theory in Plato's tale.

After a brief look (72) at Justus Lipsius, who also situated Atlantis between the continents of Europa (and Africa) and America, VN devotes two pages (72-3) to Francis Bacon's 'New Atlantis', which is, of course, not really an attempt at locating Plato's Atlantis but a scholarly utopia situated somewhere in the South Pacific. VN then turns to efforts to unravel Biblical chronology by using Plato's Atlantis chronology,27 citing as an example Isaac La Peyrère's book "Praeadamitae" (of 1655). After this, he discusses the massive efforts (published between 1679 and 1702) of Olaüs/Olof Rudbeck, rector of the University of Uppsala, to identify Plato's Atlantis (and the home of the descendants of Noah's son Japhet) with his own country of Sweden (75-77), shows how they coincide with the times of Sweden's greatest political power, and that other nationalisms at about the same time claimed similar things for their own countries (e.g. A. Audigier for France). He ends the chapter by pointing to two contemporaries of Rudbeck: the German Jesuit Athanasius Kircher, who even produced a map of Atlantis situated in the central part of the Atlantic (reproduced on p. 82), and the Danish scholar Niels Steensen, who was also prepared to believe that Atlantis' one time existence was no mere fiction (78).

Chapter IV ("Lumières de l'Atlantide 1680-1786", 83-98) is devoted to discussions of Atlantis during the Enlightenment. VN shows how the problem of reconciling Biblical and non-Biblical traditions with each other (raised already by people like La Peyrère, see above) became more acute. Pierre-Daniel Huet, a French bishop of the later 17th century, tried to show (in his "Demonstratio evangelica" of 1680) "que les dieux antiques sont les héritiers de la Révélation" (84); in 1726, the advocate Claude Olivier pointed out that the ten tribes of Israel west of the river Jordan coincided with the ten regions of Plato's Atlantis (85). Both Catholics (like J.-J. Bonnaud) and Protestants (like C.-F. Baer,28 the Protestant pastor of the Swedish Embassy in Paris) tried to show that Plato's Atlantis was after all only a "description travestie de la Judée" (85, quote of Bonnaud), and VN (86) also records how Voltaire ridiculed such efforts. Others followed in the footsteps of Rudbeck and made efforts to show that their own country had once been Atlantis; thus the Italian count Gian Rinaldo Carli (on whom see pp. 86-88) "fit de son pays natal à la fois l'héritière de l'Atlantide et la fontaine de la sagesse antique" (86). Still others looked far beyond their own countries or the ancient homeland of the Bible: Jean-Sylvain Bailly found Atlantis and mankind's origins even "plus au nord et plus à l'est que la Suède de Rudbeck" (89).29 Hereafter, VN turns to the "historians" and presents Nicolas Fréret (1688-1749) as the first who regarded Plato's Atlantis story as "une fiction philosophique" (91), while the great Voltaire was not as radical, preferring to consider that Atlantis could be identified with Madeira. With that opinion, Voltaire approaches those who (starting with Athanasius Kircher, see above) entertained the idea that Madeira or the Canaries or the Azores might be the remnants of the vanished Atlantis (92). VN closes this chapter by looking at Nicolas Boulanger (1722-1759), who like the Egyptian priests in Plato's Timaeus saw the Earth's history periodically interrupted by great floods, after which human civilisation always had to start anew (93-95),30 and at Giuseppe Bartoli (1717-1788; 96-98), who in VN's eyes would deserve wide acclaim, because he "pour la première fois...a compris...depuis le temps de Platon, que l'Atlantide était le masque de l'Athènes impérialiste et maritime" (97; which is, of course, VN's own thesis, too).

Chapter V ("Le grand tournant31 1786-1841", 99-113) covers, as the dates given in the heading show, the period of the French Revolution and the decades afterwards. It starts with Delisle de Sales (1743-1816; 100-102), whose original name was Jean-Baptiste Isoard and who in 1779 started a multi-volume enterprise titled "Histoire nouvelle de tous les peuples du monde ou Histoire des hommes", in which he placed the Atlanteans "au coeur de l'histoire universelle" (101) and located the cradle of human civilisation in a "greater" Caucasus extending "du Turkestan à la mer glaciale" (102). After a brief look at Franc,ois-Auguste de Chateaubriand, who -- though influenced by Delisle de Sales -- only briefly and enigmatically mentioned "les chimères d'une Atlantide" (103) in his "Le Génie du Christianisme", VN devotes the bulk of this chapter to another author originating from Delisle de Sales' circle: Fabre d'Olivet (1767-1825; 103-110), "théosophe impérial" (104), who wrote about Atlantis after 1797 and made it one of the corner-stones of his breathtakingly strange "reconstruction" of universal human history. In his "Lettres à Sophie", Fabre presented a very imaginative amalgamation of Plato's Atlantis story with the book of Genesis (with Adam being the son of Éloïm, high priest of Poseidon, and Eve, priestess of Aphrodite, both living on Atlantis, after the inundation of which they find themselves like Noah in the Caucasus); in his "Histoire du genre humain" he went even further and developed the tableau of peoples already present in his former work into a history of the conflict between human races, the "black" or "southern" one becoming that of the Atlanteans, the "white" that of Celts and Scythians; Egypt and America play a role as well. Compared with "ce grand délire" (110), other efforts (briefly dealt with by VN on pp. 110-113) by authors of these times to place Atlantis within the pre-history of their own people almost look rational: that by the Italian Angelo Mazzoldi, by the English poet William Blake and by his countryman F. Wilford, as well as that by the Irish writer Henry O'Brien. The last author mentioned in this chapter is again a Frenchman, the scholar Jean-Antoine Letronne, who "tient le récit de l'Atlantide pour une fable, mais...pense que cette fable vient authentiquement d'Égypte" (113).32

In chapter VI (115-131, headed by the somewhat enigmatic title "Il faut qu'une nation soit ouverte ou fermée"), VN once more returns to France, briefly discussing why in France there are almost no parallels to Rudbeck, Carli and others (115-6) and then presenting the poem "L'Atlantiade" of Népomucène Lemercier (published in 1812), of which he gives a not altogether clear summary (116-7: apparently the poem presented the origin of our system of sciences, embedded in a narrative, in which an overbearing Atlas was beaten in a war of aggression he had let loose, his island and palace destroyed, and a small group of idealistic survivors made it to the American continent).

VN then discusses two novels by the famous Jules Verne (118-120), first (and briefly) the well-known "Twenty thousand leagues beneath the sea" (published in 1869) in which captain Nemo and a group of passengers of his submarine "Nautilus" visit the ruins of Atlantis on the bottom of the sea, second (and more extensively) "The Begum's Millions" (of 1879), in which Verne depicted the Franco-German antagonism of those times in a story about the rivalry between two very different cities: on one side the ideal city "France-Ville" founded by the French humanist "docteur Sarrasin", on the other the factory-like "Stahlstadt" founded by the German militarist "professeur Schultze" who only wants to produce weapons and ultimately annihilate France-Ville. VN is surely right to see Jules Verne drawing here on the Platonic antagonism between Atlantis and primeval Athens; the aggressive designs of Stahlstadt fail like those of Atlantis, and the pacifist France-Ville prevails.

After that, we turn to Spain (120-21) and to the Catalan epic poem "L'Atlàntida" by Jacint Verdaguer (made public in 1877), which celebrated the Iberian peninsula as a more worthy successor to the degenerate Atlantis and later was converted into music by Manuel de Falla. VN then adds, very briefly (121), two scholarly undertakings to "really" (i.e. historically) connect Atlantis with Spain, that by Victor Bérard in 1929 and that by Adolf Schulten in 1939.

After this we move back into the 19th century and are presented with (mostly French) efforts (connected with the rise of the French colonial empire in Northern and Western Africa) to locate Atlantis in the Sahara or at least connect the imaginary island with this vast desert (122). These efforts sort of culminate in the novel "L'Atlantide" (published in 1919) by Pierre Benoit (123-4). Now it is the turn of Germany to provide new theories revolving around Atlantis (124-128), but they are tainted almost immediately -- the only exception being Gerhard Hauptmann's novel "Atlantis" of 1912 -- by National-Socialist ideology.33 Demonstrating that such tainting went on even after 1945, VN briefly characterizes four books: 1. "Atlantis die Urheimat der Arier" (of 1922) by Karl Georg Zschaetzsch;34 2. "Unsere Ahnen und Atlantis" (of 1934) by Albert Herrmann (a professor of Berlin University whose position VN characterizes as "comme le Führer de la presse allemande", 125), who based his theory of the existence of a "grand empire atlanto-germanique" in the 2nd Millennium BC on the so-called "Chronik von Ura-Linda" which even in Herrmann's time had already been recognized as a fraud; 3. "Der Mythos des 20. Jahrhunderts"35 by A. Rosenberg (who managed to make Jesus Christ a descendant of Atlanteans by claiming that those had once settled in Galilee); and 4. several books by Jürgen Spanuth (of 1953, 1965, and 1976; the 1965 book has the telltale title "Atlantis, Heimat, Reich und Schicksal der Germanen") which try to identify Heligoland as the remnant of a once flourishing Atlantis -- this identification had first been proposed in Heinrich Himmler's famous (or infamous) "Ahnenerbe-Institut". After this instructive tour through some of the more sinister manifestations of the ongoing fascination with Atlantis, it is a bit disconcerting to find VN making a very abrupt "saut en Méditerranée" (128) to a rather recent book by the Italian journalist Sergio Frau ("Le Colonne d'Ercole -- un'inchiesta", 2002), who tries to find Atlantis in his home island of Sardinia. VN is surely right to dismiss this newest claim to rediscover Atlantis in one's backyard, but he might have found another place in his book to discuss it.

Chapter VII ("Interlude: Notes sans musique, 133-139) presents two musical works on the Atlantis theme: 1. The opera "Der Kaiser von Atlantis", the music (by Viktor Ullmann) and the libretto (by Peter Kien) of which were composed in the concentration camp of Theresienstadt. The work (in which Atlantis is the symbol of a totalitarian empire 36) was finished in January 1944; ten months later, both the composer and the librettist died in Auschwitz. 2. The "roman semi-autobiographique" (135) "W ou le souvenir d'enfance" by the French-Jewish author Georges Perec, published in 1975. VN proudly claims to have been the first to discover in it "une nouvelle adaptation du mythe de l'Atlantide", though it may be regarded as a rather strange (and, in my opinion, rather flimsy) one. In one part of the book (which consists of two narratives interlockingly juxtaposed), an island called "W", named after its discoverer Wilson 37 and situated at the tip of Tierra del Fuego in the extreme south-west Atlantic, is decribed. On this island -- which at first seems to be an earthly paradise, but then turns more and more into a hell of human brutality --, three kinds of periodic sports games take place, one of them being the "Atlantiades", and this finally establishes (for VN) the connection with Atlantis: "Comme l'Atlantide de Platon, l'île W est située dans un lointain Occident, elle a une végétation luxuriante, elle est une utopie négative...Elle devient [italics by VN] Auschwitz comme l'Atlantide devient le monde de l'Autre et de la dissemblance" (138).

In the last chapter ("L'Eau, la Terre et les Songes",38 140-148), VN turns (or rather returns) to occultist decipherings of Atlantis, discussing the "Story of Atlantis" by William Scott-Elliott (142-144), published in 1909, and casting a brief glance at Helena Blavatsky ("créateur de l'Atlantide théosophique", 142) and at "occultistes contemporains comme Colin Wilson et Rand Flem-Ath" (143)39 on the way. Hereafter we move back -- again somewhat unexpectedly40 -- into the later 19th century and to an author whom VN rightly claims to have had the greatest and longest-lasting impact on modern Atlantologist thinking, Ignatius Donelly, the 13 main "theses" of whose book "Atlantis: The Antediluvian World" (first published in 1882) he reproduces on pp. 146-7. The chapter closes with a short look at a more modern Atlantologist hailing like Donnelly from Minnesota, F. Joseph, who discovered "Atlantis in Wisconsin" (as already the title of his 1995 book alleges).

There follow (apart from the end notes to all the chapters on pp. 149-176) an index of proper names, ancient and modern (177-186) and an appendix (187-195) containing two articles (in English) from the London Times of Febr. 19, 1909, the first of which, written by an unnamed correspondent (later identified as the archeologist K. T. Frost, who also published an article on "The Critias and Minoan Crete" in JHS 33, 1913, 189-206) and entitled "The lost continent", tries to identify the Minoan civilization of Crete with Plato's Atlantis (see above), while the (shorter) second one, under the simple heading "Atlantis", endorses the hypothesis of the first. These articles provide an interesting flashlight on the status of the Atlantis discussion in the early 20th century; they are, however, disfigured by a multitude of atrocious spelling mistakes,41 which must have got into these texts by the process of scanning them for digitalization and insufficient proof-reading: I counted 33 misspellings on nine pages. Their enumeration would take up too much space here, therefore I restrict myself to the names thus affected: Theseus has become "Thesous" (187), the pharaoh Necho II "Nocho" (189), the region of Tyrrhenia has twice become "Tyrrhenian" (190), the Etruscan king Lars Porsenna "Lars Persona" (190), the pharao Merenptah "Morenptah" (190).

The poor handling of English in the appendix is, however, not the only instance of sloppiness in this book. Others are: On p. 28 (3rd paragraph, 3rd line), "l'Athènes" should be replaced by "l'Attique", because Attica is referred to in the quote here given from Crit. 111b. On p. 69, the date given for the report of Gasparo Contarini (who was Venetian ambassador to the court of the German Emperor -- and Spanish king -- Charles V) before the Venetian senate should be 1525 (not 1625). Note 3 (to p. 46) on p. 155 should refer to Ael. VH III 18, but in fact refers to Ael. NA XV 2, which in the text is mentioned five lines below. In note 31 (to p. 62) on p. 158, the dating "deux siècles et demi plus tard" must be changed to ". . . plus tôt", because Eusebius of Caesarea lived and wrote two and a half centuries earlier and not later than Cosmas Indicopleustes. In note 18 (to p. 70) on p. 161, the name of Gomara has to be replaced by "Acosta", because Acosta is the subject of discussion in the text connected with this note, while Gomara is mentioned on the preceding page. In note 24 (to p. 74) on p. 162, the title of La Peyrère's book ("Praeadamitae, sive exercitatis quibus traducuntur . . .") is gravely distorted; its correct form is "Praeadamitae, sive exercitatio super versibus duodecimo, decimotertio et decimoquarto, capitis quinti Epistolae D. Pauli ad Romanos, quibus inducuntur primi homines ante Adamum conditi". In note 5 (to p. 118) on p. 172 the title of P. Jordan's excellent book "The Atlantis Syndrome" has been ludicrously changed to "The Atlantic Syndrome". And why is there (on p. 79) a reproduction of a page from the French translation (of 1595) of Ortelius' "Theatrum Orbis Terrarum", when there is not a single word on Ortelius in VN's text?42

Even the French language has suffered from this sloppiness: On p. 35, the long sentence starting with "Quand Hérodote décrit..." runs on for more than five lines and then breaks off without a main clause. On p. 69, end of 2nd paragraph, read "ait rendus" (instead of "aient"), as the subject of the sentence is singular ("Dieu"); it's just the other way round on p. 101 (l. 4 from bottom), where "ait" must be changed into "aient" because the subject is plural. On p. 83 (l. 3 from bottom) read "des usages" (instead of "des usage"). On p. 142, the sentence at the end of the 1st paragraph remains unfinished, as the "parce que" sub-clause in vain waits for its predicate).

There are also unnecessary repetitions: Proclus (together with Iamblichus) is twice called a Neo-Pythagorean (p. 56 and 57). Gerhart Hauptmann's novel "Atlantis" is also mentioned twice with the same additional information on p. 124 and in n. 20 on p. 173.43

One puts the book down with mixed feelings: There is no doubt that VN can write very well and that he really knows the things he is talking about, having occupied himself with them for half a century; on the other hand, one can only wonder why he sometimes prefers vague (though often witty) allusions to a clear exposition of things one would like to be informed of,44 and why he not rarely connects topics with each other that clearly do not belong together (e.g. Frau's Atlantis theory and the Nazis, or Donnelly's influential book and the theosophic occultists). All in all, the book could have been better; it looks like something of a missed opportunity.


1.   P. Vidal-Naquet, Hérodote et l'Atlantide: entre le Grecs et les Juifs. Réflexions sur l'historiographie du Siècle des Lumières, in QS VIII 1982 No. 16, 3-76; P. Vidal-Naquet, L'Atlantide et les nations, in id., La démocratie grecque vue d'ailleurs, Paris 1990, 139-159 + 353-361.
2.   Already translated into German (Atlantis: Geschichte eines Traums, München 2006) and English (The Atlantis Story. A Short History of Plato's Myth, Exeter 2007).
3.   On p. 13 he reminisces about taking his "diplôme d'études supérieures" with Henri-Irène Marrou and his years as a teacher at high school; on p. 15 and 16 about being acquainted with Pierre Lévêque, Jean Bollack, Heinz Wismann, Luc Brisson, Marie-Laurence Desclos, and Anissa Castel-Bouchouchi; on p. 18, he draws attention to being asked to provide a preface to a volume of papers given at a colloquium (which he feels compelled to repeat several times: in n. 17 on p. 151, in n. 22 on p. 161, in n. 33 on p. 170, and in n. 7 on p. 172); on p. 39 he mentions discussions with Luc Brisson; on p. 93, an anecdote about his academic teacher in geography; on p.113 we read that his pupil Richard Gordon has given him a book as a present; on p. 122 he hints at his friendship with Hervé Duchêne; on p. 135 he stresses that he was the first to detect an allusion to Plato's Atlantis in Perec's novel "W ou le souvenir d'enfance"; and on p. 141, he lets us know how much he would have liked to be a pupil of Gaston Bachelard. With all this, one might feel tempted to rename the book "L'Atlantide et Moi" or perhaps even "Moi et l'Atlantide".
4.   VN traces the origin of this hypothesis back to an article in the London Times in 1909, but the idea of a possible Cretan background for Atlantis was for the first time(?) voiced more than half a century earlier: Already A. S. von Moroff, Die Atlantis nach griechischen und arabischen Quellen, Petersburg 1854 surmised that Atlantis might have been a large island within the Eastern Mediterranean the remains of which might be still be present in the islands of Crete or Cyprus.
5.   On my Platon und die Erfindung von Atlantis of 2002, VN passes a rather harsh judgment: "Il n'y a malheureusement rien à retenir du petit livre de H. G. Nesselrath . . ., qui se contente d' enfoncer des portes largement ouvertes" (p. 153, n. 7). As basis for this statement, VN in fact seems to assume that the fictitiousness of Atlantis is nowadays totally uncontested; on p. 36 he even -- much too confidently -- asserts: "il est établi depuis des siècles que ce continent a été fabriqué par Platon"; but how can one say so when every year at least two or three new hypotheses concerning the location of this "fictitious continent" are being developed and almost always received with much bally-hoo by the worldwide media? According to a poll cited in TIME Magazine (p. 30, of 6 Nov. 2006), 41% of all Americans believe that Atlantis as a very advanced ancient civilisation really existed. It may, moreover, well be that VN has "retained" more from my "little book" than he would like to admit: e.g. the qualification of the Atlantis story as "science fiction" by VN on p. 41 is already found in Nesselrath 2002, 24. Coincidence? It also has to be said that VN does not take into account other German scholarship on Plato's Atlantis: On p. 35, he draws attention to J. Bidez' demonstration of 1945 that many traits of Plato's Atlantean metropolis probably derive from the big cities of the East (Babylon, Susa, Ecbatana), which the Greeks at least knew from reports and sometimes from having gone there themselves. That is true, but the same point was made already 17 years earlier by Hans Herter in his article 'Platons Atlantis', Bonner Jahrbücher 133, 1928, 28-47.
6.   He stated the same position already in "Athènes et l'Atlantide" (above n. 1; p. 356f. in the 1981=1989 edition).
7.   See H.-G. Nesselrath, Platon, Kritias: Übersetzung und Kommentar, Göttingen 2006, 34-41.
8.   VN p. 30: "Athéna et Hephaïstos règnent sur Athènes." Already on p. 16, he states (wrongly) that primeval Athens "est gouvernée par les dieux, non par elle-même ni par des hommes politiques."
9.   There are other faulty statements as well: On p. 26, VN claims that the conversation reported (by Socrates himself) in Plato's Republic "commenc,ait au Pirée, . . . mais se développait pendant la remontée vers Athènes"; in fact all of this conversation takes place in the Piraeus house of Polemarchus -- there is perhaps confusion here with the dialogue situation of the Laws, which indeed unfolds while the participants are walking across the Cretan countryside.
10.   On p. 34 he gives further alleged Herodotean allusions, not all of them convincing. To conclude from Tim. 25bc (where Athens is depicted as leader of a Greek coalition, which then crumbles and leaves Athens alone in her fight against Atlantis) that Plato reversed the Herodotean sequence of events ("Platées, si je puis dire, y précède Marathon") is surely an exaggeration. The only model Plato probably had in mind for depicting the war between Athens and Atlantis was the battle of Marathon. Recently, a book by a leading contemporary German "Atlantologist" has also taken VN to task for his overstressing Herodotus as Plato's source (Th. C. Franke, Mit Herodot auf den Spuren von Atlantis: Könnte Atlantis doch ein realer Ort gewesen sein?, Norderstedt, 2006, 153-231). In other respects, this book (with its extraordinary claim that studying Herodotus may help us to find out more accurately about Atlantis' real existence) has quite a few deficiencies, but its painstaking rebuttal of VN's Herodotus theory is worth reading.
11.   On p. 45, VN regards it as a "plagiat" on Plato, but this is surely the wrong categorization; on p. 46, he more correctly calls it a "pastiche ironique".
12.   See my "Theopomps Meropis und Platon: Nachahmung und Parodie," in Göttinger Forum für Altertumswissenschaft 1, 1998, 1-8, ignored by VN.
13.   On this, see my article "'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, [153-171] 166-171.
14.   VN also strangely mistranslates Aristotle's sentence, rendering it "les parties de la mer situées au-delà des Colonnes d'Héraclès sont à l'abri des vents, à cause de la vase" -- why should mud protect a sea from winds? He refers to the citation of this Aristotle passage in Proclus' Commentary on the Timaeus of Plato and Festugière's French translation (vol. I p. 245), but this is as follows: "Aristote a rapporté qu'il y a de la vase dans la mer extérieure après le detroit et que ce lieu-là comporte des bas-fonds . . ." Moreover, Proclus does not cite Aristotle very accurately. The correct translation of the original passage gives a quite different sense (translation by P. Louis in the Aristotle Budé edition): "les zones situées au-delà des Colonnes d'Hercule sont peu profondes en raison de la vase, et elles sont à l'abri des vents, car cette mer est située dans une espèce de cuvette."
15.   On this, see again my article cited in n. 13, p. 159-60.
16.   By remarking that the source of the Mirabilia "est souvent Théopompe" (48), VN insinuates that this passage, too, may go back to the author of the already-mentioned Meropis; however, already in 1972, H. Flashar (in his annotated German translation of the Mirabilia, Berlin 1972), convincingly pointed out that this passage (ch. 84 = 836b30-837b6) goes back to the historian Timaeus of Tauromenion (see FGrHist 566 F 164.19 p. 655f. Jacoby).
17.   It would be more correct to say that Posidonius proposed to keep an open mind regarding this question. Strabo records Posidonius' opinion as follows: "He [Posidonius] does well to cite Plato's remark that it may be possible that the story about the island of Atlantis is not a fiction...and he [Posidonius] believes that it is better to say this than that 'he who created it [i.e. Atlantis] also made it disappear'" (Strabon II 3,6 p. 102 C. = Poseidonios FGrHist 87 F 28 = fr. 13 Theiler = fr. 49 Edelstein-Kidd, ll. 297-303).
18.   VN dates him to the 2nd century BC; apparently he has not taken note that because of a papyrus find Dionysius -- whom VN calls "Denys de Mytilene" -- was plausibly re-dated into the 3rd century BC by J. S. Rusten, Dionysius Scytobrachion, Opladen 1982, 85-92.
19.   On this, see my "Atlantes und Atlantioi: Von Platon zu Dionysios Skytobrachion," in Philologus 145, 2001, 34-38. It is certainly wrong to say that ten sons begotten by Zeus (not Poseidon) and born by a daughter of Atlas (an "Atlantid") are "le seul point de contact entre Diodore de Sicile et le récit platonicien" (50). As my article shows, there are many, many more.
20.   In Pliny's addition si Platoni credimus, VN detects "un petit brin de scepticisme" (51); but these words may qualify only the reputedly immense size of Atlantis and thus may not express any doubt to Atlantis' former existence as such (here is the wording in full: In totum abstulit terras [scil. rerum natura] primum omnium ubi Atlanticum mare est, si Platoni credimus, immenso spatio...).
21.   But looking at Plut. Sol. 31.6, where Plutarch in fact hesitates whether to call the Atlantis story a λόγος or a μῦθος, one might think that he did not accept it unquestioningly.
22.   On Crantor, see my "Atlantis auf ägyptischen Stelen? Der Philosoph Krantor als Epigraphiker," in ZPE135, 2001, 33-35.
23.   As proof for Atlantis' real existence at one time, he cites a certain Marcellus, on whose existence VN remains non-committal; for this Marcellus, who may not have been a historian but the author of a romance in which Atlantis played a part, see my article cited above in note 13, p. 168-171.
24.   This notion -- certainly a very appealing one to French structuralists -- is found by VN already in Plato's text itself (59).
25.   On p. 65, he briefly considers whether there might be traces of the Atlantis tale in the Irish legend of St. Brendan, but does not really answer the question.
26.   Though others took part as well: VN (67) draws attention to Gerolamo Fracastoro, who in his poem Syphilis (published in 1530) juxtaposed syphilis and the drowning of Atlantis as two divine punishments afflicting the human population of the New World for having become too rich and insolent. In n. 9 (on p. 159) to p. 67, VN cites only editions and (French) translations of Fracastoro's Syphilis from the 18th and 19th centuries, but there is a bilingual Greek-English edition with introduction and commentary by G. Eatough (Fracastoro's Syphilis, Liverpool 1984) and a bilingual Latin-German one by G. Wöhrle (Girolamo Fracastoro, Lehrgedicht über die Syphilis, Wiesbaden, 2nd ed. 1993).
27.   A more precise account of these efforts can be found in M. Ciardi, Atlantide: Una controversia da Colombo a Darwin, Rome 2002, 44. Ciardi's book (not cited by and apparently unknown to VN) provides better coverage of the authors treated by VN also in other cases (see, e.g., p. 55-61 on Rudbeck).
28.   Baer is another case where Ciardi (p. 75-78; see preceding note) provides much more detailed and precise coverage, while VN just briefly mentions his name.
29.   Namely at Spitzbergen, though VN does not explicitly say so. VN's critique of Bailly's arguments and conclusions is a bit strange: "Il faut admettre, par exemple, qu'Hérodote a identifié la mer Rouge à l'Atlantique, ce qui n'est guère plus sérieux que les contes fantastiques de Rudbeck" (90). But Herodotus indeed says that the "Red Sea" and the "Atlantic Sea" are just one (1.202.4)! VN should have chosen his example more carefully.
30.   Although VN devotes three pages to Boulanger, he does not make clear what exactly Boulanger did say regarding Atlantis, which is a bit annoying. He also remains rather vague on Boulanger's opponent Poinsinet de Sivry, who tried to make sense of the Earth's history not by invoking (as was Boulanger's mantra) the element of water, but that of fire. On Poinsinet de Sivry, he cites a "superbe raillerie" of a certain "Grimm dans sa Correspondance littéraire" (95), but denies us any further knowledge who this man Grimm was -- not one of the famous Brothers Grimm, but Friedrich Melchior Grimm (1723-1807), a German writer and diplomat, who in 1753 founded the journal "Correspondance littéraire", which quickly became one of the most important journals of his time.
31.   The heading "Le grand tournant" is nowhere really explained by VN; perhaps it is to point towards the 'turn' towards the occult in Atlantis lore, which is observable in this chapter.
32.   This is -- although VN fails to mention it -- a position very similar to that of Th.-H. Martin and his impressive "Dissertation sur l'Atlantide" of 1841 (see below), whom VN mentions several times, but each time only rather briefly; he never points out that Martin, apparently like Letronne, shied away from making Plato the sole inventor of the Atlantis story.
33.   This, according to VN (124), is also true of the novel "Die letzte Königin von Atlantis", of which he gives us the publication date (1931), but not the author. It is Edmund Kiss (1886-?), who was recruited by Heinrich Himmler for the "Ahnenerbe" institute and was to lead an expedition to Bolivia in 1940, which was thwarted by the outbreak of the Second World War.
34.   I wonder whether Zschaetzsch's "phrase-clé" ("Ohne arische Grundsätze kann eben kein Staat bestehen") is here cited correctly; if it is, the French translation ("Sans la présence d'une souche aryenne, aucun État ne peut subsister") seems quite inexact.
35.   As date, VN gives only that of the fourth edition of 1932; the first appeared in 1930.
36.   VN regards "l'évocation du caractère totalitaire de l'empire"(134) in this opera as one of the "rare" allusions to the original -- i.e. Plato's -- Atlantis, but can we really say that the Platonic Atlantis was totalitarian? Apart from the almost absolute powers of its rulers over their subjects, we simply do not get enough information to picture Atlantis as a "modern" totalitarian state. With the same right, we might regard primeval Athens as a totalitarian society, because it has so much in common with the ideal state in the Politeia, but this view would be just as much distorted.
37.   Regarding the name of this man, VN indulges in some strange counting. He offers "quatre hypothèses" (136) as to Wilson's identity but after naming two adds the "dernière hypothèse" -- where's no. 3? And after these "quatre hypothèses", VN himself proposes "une sixième" (137) -- where's no. 5? Possibly he regards Perec's own translation of the name "Wilson" ("Je veux être un fils") as the missing fifth, but he does not do much to spare his bewildered readers much useless thinking.
38.   VN explains the heading of this chapter as "un hommage à Gaston Bachelard", whose pupil he would have liked to be (141).
39.   More detailed information on both these authors and their theories can be found in P. Jordan, The Atlantis Syndrome, Stroud 2001.
40.   On p. 147, VN not very successfully tries to establish a connection between Scott-Elliott and Donnelly, apparently attributing to both a "colossal" ambition; but this might be said about the claims of most Atlantologists. He even has to admit: "En un sens, ces deux interprétations sont radicalement opposées."
41.   Most of the spelling mistakes -- not in all the names, however -- have been removed in the English translation of VN's book (see above, n. 2), but some have even been aggravated by more distortion: e.g., on p. 193 of the French original edition, the end of the 1st paragraph reads: "...a solid foundation in fact for what might otherwise be regarded as the more creation of Plato's abounding fancy"; now the English translation has transposed the "more" (which before "creation" is surely wrong) before "regarded", but the correct text is surely ". . . the mere creation of Plato's abounding fancy."
42.   A few (but not all) of these mistakes have been corrected in the German and English versions of the book (see above n. 2); e.g. in the German version the chronological relationship between Eusebius and Cosmas has been corrected, but the Venetian ambassador to Charles V. is still dated to 1625, the Latin title of La Peyrère's book is still distorted, etc.
43.   Typos: p. 51, 3rd paragr., l. 7: read "Sonchis" (instead of San-); p. 56 (l. 3 from bottom) read "autant" (instead of aur-); p. 109 (l. 2 from bottom) read "ont" (instead of "on"); p. 125, 2nd paragraph (l. 3 from bottom) read "bestehen" (instead of gest-); p. 153 n. 10: the name of the author of "Plato's Mathmatical Imagination" is Brumbaugh, not "Brambaugh"; p. 160 (1st line) read "Vielfalt" and "Hiersemann" (instead of "Vielalt" and "Hierseman"); p. 162 n. 27, l. 5 and 6: read "Svennung" and "Wiksell", not "Svenning" and "Wieksell"; p. 164 (l. 3) read "Stralsund" (instead of --sung); p. 172 n. 9 read "Rheinisches Museum" (instead --scher); p. 173 n. 21 read "mit einer Karte" (instead of "eine"); in n. 25 (same page) read "The Atlantis Researches" (instead of --cher); p. 174 n. 31 read "Das enträtselte Atlantis" (instead of "enträselte").
44.   Compare the very pertinent remark of the BMCR reviewer of the English version (Lynn LiDonnici, BMCR 2008.05.29): "The book is written in a casual and conversational tone...If I were not already familiar with the natural philosophers and geographers discussed here, I am not sure I would have been able to follow the progression Vidal-Naquet makes..."

Comment on this review in the BMCR blog
Read Latest
Index for 2008
Change Greek Display
Books Available for Review

HTML generated at 13:33:05, Friday, 03 April 2009