Kenneth Caroli is a researcher from Florida and a regular contributor to Ancient American and Atlantis Rising magazines as well as The New Archaeology Review. He has tackled subjects such as the age of the Bimini Wall (AA vol.1.4), the date of Atlantis’ destruction (AA vol.7.43) and speculates on the possibility of a Celtic Atlantis (NAR June 2006). In Frank Joseph’s book, Atlantis and 2012, a map drawn by Caroli was included, showing the Azores as Atlantis.
Dating Atlantis is one of the most contentious difficulties faced by Atlantology. The critical problem is to identify the time of the Atlantean War and that of the later destruction of Atlantis itself; two events possibly separated by a period not recorded by Plato. This entry is primarily concerned with the date of the war. However, it should be pointed out that Plato also reveals that the Atlantis story has a very long history before the war, back to a time when ships and sailing did not yet exist (Crit.113e), so it is understandable when Plato filled the historical gap with mythological characters, namely five sets of twins sired by Poseidon. Of course, Poseidon being a sea god did not require a boat to get to the island of Atlantis! Plato also informs us that the twins and their descendants lived on the island for ‘many generations’ and extended their rule over many other islands in the sea (Crit.114c).
There are roughly three schools of thought regarding this important detail. The first group persist in accepting at face value Plato’s reference to a period of 9,000 solar years having elapsed since the War with Atlantis up to the time of Solon’s visit around 550 BC. The second group are convinced that the 9,000 refers to periods other than solar years, such as lunar cycles or seasons. The third group seeks to identify the time of Atlantis by linking it to other known historical events.
While these groups offered some level of evidence, however flimsy, to support their claims, some individuals have placed Atlantis up to millions of years in the past based on nothing more than their fertile imaginations or delusions. Arguably the best known was Edgar Cayce, but purveyors of such daft ideas are still lurking among us!(m)
[1.0] 9550 BC is factually correct
This view has a slowly dwindling number of supporters among serious investigators. Massimo Rapisarda is one such promoter, who has offered his reasons for accepting this early date(p). To support this early date proponents usually cite a wide range of evidence to suggest the existence of advanced cultures in the 10th millennium BC. Matters such as an earlier than conventionally accepted date for the Sphinx, early proto-alphabets a la Glozel or the apparently anomalous structures such as the Lixus foundations or the controversial Baalbek megaliths have all been recruited to support an early date for Atlantis, many, if not all, have their dates hotly disputed. Apart from the contentious dates, there is NOTHING to definitively link any of them with Plato’s Atlantis.
In common with most nations, the Egyptians competitively promoted the great antiquity of their own origins. Herodotus reports that while in Egypt he was told of a succession of kings extending over 17,000 years. The priests of Memphis told him firmly that 341 kings and a similar number of high priests had until then, ruled their country. (Herodotus, Book II, 142). Even an average reign of 20 years would give a total of nearly 7000 years whereas a more improbable 26-year average would be required to span the necessary 9000 years.
It is therefore obvious that the 17,000 years related to Herodotus is not credible raising a question regarding the trustworthiness of the 9000 years told to Solon.
In The Laws Plato refers to Egyptian art going back 10,000 years, seemingly, indicating consistency in his belief in the great antiquity of civilisation and fully compatible with his date for Atlantis. However, I have discovered that in Plato’s time ‘ten thousand’ was frequently used simply to express a large but indefinite number.
A Bible study site tells us that “The use of definite numerical expressions in an indefinite sense, that is, as round numbers, which is met with in many languages, seems to have been very prevalent in Western Asia from early times to the present day.”(h)
The acceptance of Plato’s 9,000 years as literally correct defies both commonsense and the archaeological evidence, which demonstrates that neither Athens nor a structured Egypt existed at such an early period. The onus is on those, who accept the prima facie date of 9,600 BC, to explain how Atlantis attacked a non-existent Athens or Egypt.
In a 2021 article(h) about 10,000 BC Thorwald C. Franke offered the following opinion; “Many scientists seem to live quite comfortably with Graham Hancock and similar authors speculating about 10,000 BC because these hypotheses are so nonsensical that they do not interfere with real science. Sometimes you have the impression that many scientists even prefer such misleading popular errors over more informed hypotheses because they would make the audience ask more serious questions and then the questions could not be dismissed so easily anymore. But this is only an impression. In truth, scientists shy away from the effort to overcome these popular errors. It is much easier to stay silent and to ignore them.”
Alexandros Angelis wrote(q) of how he is “always suspicious of coincidence. Whenever I hear this word, an alarm sets off in my head. In my book ‘Our Unknown Ancient Past: Thoughts and Reflections on the Unexplained Mysteries of Prehistory’ I state that it cannot be a coincidence that Plato’s date of Atlantis’ destruction (9.600 BC) is spot on, coinciding with the abrupt end of the Younger Dryas (9.600 BC).”
When this compiler first encountered this coincidence I was also suspicious. However, as I investigated further I realised that all of Plato’s numbers seemed to be inflated by what was arguably a common factor – another coincidence? I have devoted an entire chapter in Joining the Dots to this problem.
Angelis considers the ‘rapid’ gradual melting of the ice at the end of the Younger Dryas as the cause of Atlantis’ submergence, which might have been true except that Plato tells us that the catastrophe took place over a day and a night and that the event was triggered by an earthquake. He seems unaware that isostatic rebound is a very slow process of readjustment involving centuries and sometimes thousands of years and even when glaciers melt rapidly, sea levels, because of their vastness, rise slowly.
[2.0] 9000 refers to units of time other than solar years
Advocates of this view, understandably point out, that the Atlantis described in such detail by Plato belongs to the Bronze Age and could not have existed at an earlier date. It is worth noting that the technology is coincidental with the most advanced known to Plato and his audience. For those who argue that mankind has been destroyed on one or more occasions and has had to start again from scratch, it is not credible that if this was the case, that the culture and technology described by Plato as existing in 9500 BC is precisely what he would have experienced himself. There is nothing in the Atlantis texts to connect it with a pre-Bronze Age society, nor is there anything to suggest any technology or cultural advance beyond that of the 4th century BC. Plato’s tale tells of the existence of at least three major nations before the destruction of Atlantis: Egypt, Athens and Atlantis itself. There is no archaeological evidence to indicate anything other than Neolithic cultures existing in Egypt or Athens around 9500 BC. The currently accepted date for the beginning of Egyptian civilisation is circa 3100 BC and also for the existence of a primitive culture around Athens at about the same time. This would parallel the time of the western European megalithic builders.
It is noteworthy that researchers who support a 9,600 BC date for the war between Atlantis and Athens cannot explain how this took place millennia before there was any structured society in Greece.
It may be worth noting the comments of Israel Finkelstein and Neil Silberman who have argued for a 7thcentury BC date for the final draft of the Exodus narrative rather than during the 2nd millennium BC as suggested by the text. “In much the same way that European illuminated manuscripts of the Middle Ages depicted Jerusalem as a European city with turrets and battlements to heighten its direct impact on contemporary readers” (p68). Similarly, it is possible that Plato added architectural and technological details of his day to a more ancient tale of a lost civilisation to make a more powerful impression on his audience.
According to Bury’s translation, Plato mentions (Crit. 119e) that iron was used for utensils and weapons in Atlantis and so forcing us to look to a date later than 2000 BC for its destruction. Olaf Rudbeck drew attention to this reference around 1700.
Diaz-Montexano claims that the ‘9000 years’ in Critias has been mistranslated. He refers to the earliest versions of Critias that are available and insists that the texts permit a translation of either ‘9 times in a 1000 years’ or ‘1009’, the first being the more rational! Frank Joseph has also used this 1009 number, quoting private correspondence from Kenneth Caroli, in his 2015 regurgitation of Atlantis and 2012. Diaz-Montexano has also drawn attention to the commentary on Timaeus by Proclus, writing in the 5th century AD, where he treats Plato’s use of 9000 as having symbolic rather than literal meaning. It should also be kept in mind that many cultures, ancient and modern use specific numbers to indicate indefinite values(e). In a more recent paper Diaz-Montexano concluded that “we can place their military and colonizing expansion towards the end of 3.500 BC, at the earliest, and the end of their civilization (with Atlantis sinking) between 2.700 and 1.700 BC.”(o)
[2.2] In June 2017, a forum on the Historum.com website included the following possible explanation for the Atlantean dates:
“ The date 8000 is given as a fraction of 8 since the Greeks commonly used fractional notation. Plato wrote in 400 BC and Solon obtained the account in 570 BC.
No Egyptian Annals ever went back 9000 or even 8000 years. The furthest back the Egyptian annals went at the time of Herodotus was to 3050 BC to the reign of Menes the first Pharaoh who Herodotus knew about. Therefore it is obvious that the number of years has been given as a fraction which was extremely common in Greek numerology.
Thus the war between Atlantis and Athens occurred in 9000/8 + 570 = 1695 BC (+/-63 years) which is pretty close to the date of the war between the Titans and the Gods c.1685-1675 BC. The entire story of Atlantis runs concurrent to the time of the Thera Eruption. You even have 10 kings ruling the land equivalent to the 12 Titans.”
The bible too denotes years as fractions, i.e. seasons, equinoxes/solstices etc. That is why you have biblical patriarchs that lived 800 and 900 years old. The ages to Noah are all counted in Lunar months.”(i)
While I’m aware that the Egyptians also had a different way of dealing with fractions, I really cannot fully understand the suggestion made above.
[2.3] 900, not 9000 years
To address these apparent conflicts, some have suggested that the stated 9000 years, which allegedly elapsed since the catastrophe, are the result of incorrect transcription by someone along what is a very long chain of transmission and that hundreds have somehow been confused with thousands and that the correct figure should be 900 years. Another suggestion is that the Egyptian hieroglyphics for ‘hundred’ and ‘thousand’ are easily confused. This explanation does not hold water, as there is little room for confusion between these hieroglyphics as illustrated below. This idea has been adopted by Don Ingram and incorporated into his own hypothesis.
However, 900 years earlier than Solon would place the conflict with the Atlanteans during the XVIIIth Dynasty and would have been well recorded. More recently Diaz-Montexano put forward the idea that the Egyptian words for ‘100’ and ‘1000’ when spoken sounded similar leading to Solon’s error. This idea has now been taken up by James Nienhuis and in greater detail by R. McQuillen(a).
Another explanation offered by James W. Mavor Jnr. is that the original Egyptian story emanated from Crete where it may have been written in either the Linear A or Linear B script where the symbols for 100 and 1000 are quite similar. In both scripts, the symbol for 100 is a circle whereas the symbol for 1000 is a circle with four equally spaced small spikes or excrescences projecting outward.
Nevertheless, the most potent argument against the ‘factor ten’ solution is that if the priests did not intend to suggest that Egypt was founded 8000 years before Solon’s visit but had meant 800 years, it would place the establishment of Egypt at around 1450 BC, which is clearly at variance with undisputed archaeological evidence. However, I contend that they were referring to the establishment of Sais as a centre of importance, not the foundation of the entire nation of Egypt.
[2.4] 9000 months not years
>The earliest suggestion that I have found which implied that the age of Atlantis noted by Plato referred to months rather than solar years comes from the early fourteenth century.
Thorwald C. Franke has drawn attention to Thomas Bradwardine‘s rejection of Plato’s, or more correctly the Egyptian priest’s, apparent claim of a very early date for Atlantis [1255.242]. It seems that he found such a date conflicted with biblical chronology. In the end, he proposed that Plato’s ‘years’ were lunar cycles.<
Around two and a half centuries later in 1572 Pedro Sarmiento de Gamboa suggested the application of lunar ‘years’ rather than solar years to Plato’s figures. Augustin Zárate expressed the same view in 1577, quoting Eudoxus in support of it.
Then there are others, such as Émile Mir Chaouat and Jürgen Hepke who subscribe to the view that the 9000 ‘years’ recorded by Plato referred to months rather than solar years, as the early Egyptians extensively used a lunar calendar and continued to use it throughout their long history, particularly for determining the dates of religious festivals and since Solon received the Atlantis story from Egyptian priests it would be understandable if they used lunar ‘years’ in their conversations. Eudoxus of Cnidos (c.400 BC- c. 350 BC), mathematician and astronomer, who spent a year in Egypt, declared, “The Egyptians reckon a month as a year”. Diodorus Siculus (1st cent. BC) echoes this statement. (see Richard A. Parker) and Manetho (3rd cent.BC) (Aegyptiaca[1373.40])
Olof Rudbeck also proposed that Plato misunderstood the Greek priesthood’s use of lunar cycles rather than solar years to calculate time. This in turn led him to date the Atlantean War to 1350 BC.
This use of months rather than years would give us a total of just 750 years before Solon’s visit and so would place the Atlantis catastrophe around 1300 BC, nearly coinciding with the eruption of Thera and the collapse of the Minoan civilisation.
A similar explanation has been offered by J.Q. Jacobs to rationalise the incredible time spans found in ancient Indian literature, who suggested that numbers referred to days rather than years(b).
[2.5] 5,000 not 9,000 years
A claim was made on Graham Hancock’s website in 2008(c) that Plato did not write 9,000 but instead wrote 5,000, but that the characters for both were quite similar leading to the misunderstanding. This claim was originally made by Livezeanu Mihai. However, my reading of Greek numerals makes this improbable as 9,000 requires five characters ( one for 5,000 and one for each of the other four thousand), while 5,000 needs just the one.
Adrian Bucurescu claims that Plato originally said 5,000, not 9,000 years had elapsed between the Atlantean war and Solon’s visit to Egypt. He bases this claim on the fact that the works of the Greek philosophers were preserved in Arabic translations after the fall of Constantinople and that their numbers ‘5’ and ‘9’ were sufficiently similar to have led to a transcription error!(b) This is difficult to accept as the Arabic character for nine is rather like our ‘9’, while the Arabic five is like our zero!
[3.1] Sometime after 9500 BC.
Jonas Bergman correctly points out that according to the story related by the priests of Sais to Solon, the Egyptian civilisation was founded 1000 years after Athens was first established in 9600 BC. Although this probably just refers to the founding of the city of Sais rather than the early Egyptian state.
Plato describes the original division of the earth between the gods of old, Poseidon got Atlantis and Athena got Greece. The implication is that both were founded at the same time, namely in 9600 BC. Realistically, the 9000-year time span is better treated as an introductory literary cliché such as ‘once upon a time’ or the Irish ‘fado, fado’ (long, long ago). Plato’s text describes the building of Atlantis and informs us that no man could get to the island ‘for ships and voyages were not yet’. Since Atlantis had twelve hundred warships at the time of the conflict with Athens, the war could not have taken place in 9600 BC. The development of seafaring and shipbuilding would have taken considerable time. Bergman concludes that the war with Atlantis took place long after 9600 BC.
Another date was proposed by Otto Muck  in 1976 when he maintained that Atlantis had been situated in the Atlantic and was destroyed by an asteroidal impact in 8498 BC and proposed that the same event also created the Carolina Bays!
[3.2] Peter James as quoted in Francis Hitching’s The World Atlas of Mysteries[307.138] is reported to have accepted the orthodox date of 3100 BC as the start of Egyptian civilisation and considering the priest’s statement that the events outlined took place one thousand years before the creation of Egypt and so added only 500 years to compensate for nationalistic exaggeration and has concluded that 3600 BC is a more realistic date for the destruction of Atlantis.
[3.3] Early in the 20th century, the German scholar Adolf Schulten and the classicist H. Diller from Kiel, both advocated an even more radical date of around 500 BC, having identified the narrative of Plato as paralleling much of the Persian wars (500-449 BC) with the Greeks. This however would be after Solon’s trip to Egypt and have made little sense of Plato’s reference to him.
[3.4] 4015 BC is the precise date offered by Col. Alexander Braghine who credits the destruction of Atlantis to a close encounter with Halley’s Comet on the 7th June in that year. This is close to the date favoured by de Grazia.
[3.5] 3590-1850 BC has been suggested by the Czech writer Radek Brychta who has developed an ingenious idea based on the fact that the Egyptians who were so dependent on the Nile, divided their year into three seasons related to their river, the flooding, the blossom and the harvest periods. Brychta points out that counting time by seasons rather than solar years was common in the Indus civilisation that occupied part of modern Pakistan. Even today Pakistan has three seasons, cool, hot and wet. Brychta contends that the 9000 ‘years’ related to Solon were in fact seasons and should be read by us as 3000 years which when added to the date of Solon’s Egyptian visit would give an outside date of 3590 BC. If Brychta is correct this 9000 year/season corruption could easily have occurred during the transmission and translation of the story during its journey from the Indus to the Nile valley.
[3.6] 3100 BC as a date for the destruction of Atlantis has been proposed by several investigators including, David Furlong, Timo Niroma, and Duncan Steel. Hossam Aboulfotouh has proposed a similar 3070 BC as the date of Atlantis’ demise(f).
[3.7] 2200 BC is the proposed date put forward by Dr Anton Mifsud for the end of Atlantis, located in the vicinity of his native Malta. He arrived at this conclusion after studying the comments of Eumelos of Cyrene who dated the catastrophe to the reign of King Ninus of Assyria. Around the same time, in Egypt, unusually low Nile floods led to the collapse of centralised government and generations of political turmoil(f). According to some commentators(g), the Los Millares culture in Iberia also ended around the same time.
[3.8] 1200 BC is a date favoured by investigators such as Frank Joseph, Eberhard Zangger and Steven Sora. It is dependent on the acceptance of Plato’s 9000 ‘years’ being lunar rather than solar. It is worth noting that this date has also been linked to the suggested close encounter with the Phaëton comet and its destructive effects globally.
[3.9] Stelios Pavlou has taken a different approach, basing his conclusion on a close analysis of the Egyptian King Lists with particular reference to that of Manetho. Pavlou’s paper is well(l) worth studying. In the end, he contends that the time of Atlantis was in or around 4532 BC.
[4.0] More than one Atlantis!
It is not unreasonable to consider Plato’s Atlantis narrative as a literary amalgam of two or more historically based stories or myths. One possibility is that the Egyptian priests related to Solon the tale of the inundation of a powerful and advanced culture in the dim and distant past. Such an event did occur, worldwide, when the Ice Age glaciers melted, resulting, for example, in the eastern Atlantic, the flooding of the North Sea, the Celtic Shelf and dramatically reducing in size the Canaries and the Azores and creating the British Isles. The entire world was affected by this event so there were also major inundations in the West Indies and the South China Sea. However, events off the coasts of Europe and Africa would be more likely to become part of folklore on this side of the Atlantic.
[5.0] My preference is to treat the use of 9000 by Solon/Plato as an expression of a large but indefinite number or an exaggeration by a factor of ten. At the beginning of my research, I strongly favoured the former, but as I proceeded to investigate other aspects of Plato’s Atlantis story, I realised that virtually all other large numbers used by him also appeared to be inflated by a comparable amount. In seeking a solution to this I found myself drawn to Occam’s Razor, which states that where there are competing theories, the simpler is to be preferred.
It is worth noting that the Egyptian hieratic numerals also stopped with the highest value, expressed by a single character, being 9000. However, having studied the matter more closely I am reluctantly drawn to the ‘factor ten’ theory. This I have written about at some length in Joining the Dots.
The 1st millennium BC saw the introduction and gradual development of new writing and numerical systems by the Greeks. Some claim that the Greeks borrowed the Egyptian numbers(k).
At an early stage, 9000 was the highest number expressed by a single character in Greek, which in time came to be used to denote a large but uncertain value. As the needs of commerce and science demanded ever higher numbers a new character ‘M’ for myriad with a value of 10,000 was introduced. It also was used to indicate a large indefinite number, a practice that continues to this very day. Greek numerical notation was still being developed during Plato’s life.
Today, we use similar expressions such as ‘I have a million things to do’ with no intention of being taken literally, but simply to indicate ‘many’(e). Unfortunately, this interpretation of 9,000 does little to pinpoint the date of the Atlantean war, but it is not unreasonable to attribute a value to it of something above 1,000 and possibly a multiple of it.
However, having said that, I am also attracted to the ’factor ten’ theory after a study of other numbers in the Atlantis narrative which all seem to be consistently exaggerated by a similar amount, which seems to be a factor of ten!
Andrew Collins in his Gateway to Atlantis[072.52] wrote “a gross inconsistency has crept into the account, for although Critias affirms that Athens’ aggressor came from ‘without’ the Pillars of Hercules, the actual war is here said to have taken place ‘nine thousand years’ before the date of the dialogue, c.421 BC. This implies a date in the region of 9421 BC, which is not what was stated in the Timaeus. Here 9000 years is the time that has elapsed between the foundation of Athens and Solon’s visit to Sais c. 570 BC. Since Egypt was said to have been founded a full thousand years later, and the ‘aggressor’ rose against both Athens and Egypt, it provides a date post 8570 BC. These widely differing dates leave us with a glaring anomaly that defies explanation. The only obvious solution is to accuse Plato of a certain amount of sloppiness when compiling the text.”
Collins’ suggestion of ‘sloppiness’ is made somewhat redundant if my suggestion that Plato was using 9,000 as a large but indefinite idiomatic value, could be substantiated.
The late Ulf Richter was quite unwilling to accept Plato’s 9,000 years as reliable after a close study of the relevant texts.
Others have produced evidence to suggest that this period in the Earth’s history saw one or more major catastrophic events that may or may not have been interconnected; (i) a collision or near-miss with an extraterrestrial body, (ii) a pole shift, (iii) the melting of the glaciers of the last Ice Age and the consequent raising of sea levels worldwide. This rise provides a credible mechanism that could account for the ’sinking’ of Atlantis.
Mary Settegast, an archaeological researcher, has defended the early date for Atlantis with a remarkable book that delves extensively into Mediterranean and Middle Eastern prehistory and mythologies.
>Mexico, according to the most recent evidence has pushed back the arrival date of the first people to the southern regions of the country to around 30,000 BC(e).<
Mexico has not been ignored by Atlantis seekers. As early as the 17th century Carlos de Sigüenza y Góngora suggested that the indigenous people of Mexico had come from Atlantis after an earlier migration from Egypt. Louis de Launay, the French geologist, proposed in 1936, the possibility of such a link possible in 1936. More recently Gene Matlock has promoted the idea again with his own twist to the theory involving a link with ancient India. His ideas are available online(d).
An interesting attempt at linking Plato’s Atlantis story with the Valley of Mexico can be found on the Internet(a) in an article by Ed Ziomek that looks at possible links between the Old World and the Americas 5,000 years before Columbus.
Clyde Winters has published his views that the Olmecs were descendants of Atlanteans who came from Libya in North Africa. However, another Afrocentrist Paul Barton claims(c) that the Olmecs came from the Mende people of West Africa who are now one of the principal ethnic groups of Sierra Leone.
A Mexican engineer, Eduardo Robles y Gutierrez, while working in Vera Cruz, discovered the foundations of an ancient city about 30 miles from the coast, in and near what is now the jungle-covered region of San Lorenzo Tenochtitlán. He has identified that ancient city, with its concentric channels lined with high banks, as had Plato’s Atlantis. The site had been pillaged by the Spanish who consigned considerable treasure back to Spain. His investigations were also published in Mexico as La Atlántida Está en México.
The fact that many Mexican placenames begin with ‘atl’ has prompted a number of commentators to erroneously assume a connection with ‘atl’antis.
A Mexican website(b) informs us that there is a village called ‘Atlantis’ in the Municipality of Cacahoatán (State of Chiapas). It has 8 people and is at a height of 550 meters.
(a) https://www.world.mysteries.com/gw_edziom2.htm (offline) see https://www.eupedia.com/forum/archive/index.php/t-9206.html
(b) Mexico.pueblosamerica.com/i/la-Atlantean/ (offline)