Image by C Messier

The Phaistos Disk

This article first appeared in The Skeptic, Vol.20, No.3, pp.24-26.

Reproduced with permission

The Phaistos Disk is a flat disk of baked clay, 16 cm in diameter, which was found in 1908 during the excavation of a Minoan palace at Phaistos in Crete dated at around 1700 BC.

It is inscribed on each side with a text apparently running from right to left and spiralling in from the rim to the centre (but see below). Depending on how they are counted, there are about 240 characters in all, representing 45 distinct types, some pictorial and some apparently abstract; they are divided into 61 groups by broken radial lines. Very remarkably given the early date, the signs were impressed into the clay when it was soft by means of a set of cut punches, so that all tokens of a given type are effectively identical. Neither the Disk itself nor the characters resemble any other items yet discovered in the Aegean at all closely, and both the intended use of the artefact and the interpretation of the text remain mysterious.


Since 1984 three very different full or partial ‘decipherments’ of the Disk have been accessibly published, either in print or on the web (two of them in the last couple of years). To misquote Mark Knopfler, two of them must be wrong! Very probably, all three are wrong. At the very least, despite the strong advocacy of the authors in question, there is no good reason to accept any of them.

There had, in fact, been many previous efforts by scholars and amateur decipherers to interpret the Disk, starting with early ideas based on the similarity of some of its symbols to those of a local hieroglyphic script which itself cannot be read. a) In 1931, F. G. Gordon, who interpreted the later and then undeciphered Linear B script as representing a language allied to Basque, read the Disk in the same terms, producing a strange story of dogs, dogfish and circling paths. b) In the same year, Stawell translated the Disk as a prayer in Greek, though her Greek was not archaic enough and the ‘decipherment’ involved a novel acrophonic method. c) Later Ktistopoulos decided that the text was in a Semitic language and dealt with gods, stars, prophecies and the white of eggs. Meanwhile work was progressing on the better represented scripts Linear A and B, and the latter was finally deciphered as early Greek (a surprise!) in 1952, by the brilliant and superbly erudite amateur, Ventris. This gave encouragement both to his academic colleagues and to other amateurs less well versed in the relevant languages and methods, but Linear A, the hieroglyphs and the Disk have resisted decipherment to this day (though the increasingly fringe linguist Cyrus Gordon was confident that he had deciphered Linear A as Semitic). The pre-Linear-B scripts are all thought to belong to stages of the Minoan civilisation, and are generally regarded as representing some unidentified language used in the area before Greek (which would explain why they resist decipherment). For more on all this, see Chadwick’s highly readable The Decipherment Of Linear B.

More recent ‘decipherers’ (or partial ‘decipherers’) of the Disk include: d) the Danish scholar Hagen, who believes that he can identify month names in the text and, like Butler (see below), interprets the Disk as a calendar; e) the American arch-epigrapher Fell, who found Polynesian elements and accordingly proposed early contact over very long distances; f) the Bulgarian Georgiev, who read the Disk as a story in Luwian (Anatolia) about the fabled Cretan King Minos (his account was seized upon by promoters of the Oera Linda Book who sought to link it with their nationalistic Frisian fantasies); g) the Latvian Kaulins, who advances the outrageous hypothesis that his own mother tongue is the oldest in the world but reads this particular text as a geometric proof in rather odd Greek, written in an Egyptian-based syllabary; h) the Russian Rjabchikov, who (more predictably) reads the Disk – and Linear A – as early Slavic, more specifically as instructions for rituals; he also believes that Etruscan was close to early Slavic. There are more. Many of these more recent proposals are rehearsed on websites and some are supported with considerable erudition or at least the appearance of same. But they are all a long way from presenting an overall reading – or even a partial one – that will actually hold up; and naturally they all completely disagree with each other.

The first of the three new more complete and more accessible ‘decipherments’ of the Disk was Fischer’s, in 1984 (i). Fischer, a Polynesian languages expert based at Auckland, has published extensively on his claims, including book-length treatments in 1988 and 1997 (the latter book also deals with his ‘decipherment’ of the Easter Island Tablets). With some but not all qualified commentators, he concluded that the Disk text is clearly acrophonic and tried to arrive at the phonetic values of the symbols by comparison with the other local scripts, with no assumptions as to the language represented. These other scripts are predominantly syllabic in character, so the values adopted here were also syllabic. Fischer’s decisions are by no means all obvious, but his later ‘decipherment’ persuaded him that almost all of his initial identifications of phonetic value were in fact correct.

Like Ventris, Fischer gradually came to the idea that he might be dealing with early Greek or at least Indo-European. But he knew that the earlier date of the Disk would make it even harder to convince scholars of this than in the case of Linear B (for historical reasons) and that the Greek would have to be even more archaic.


One problem for both decipherments lies in the poor fit between a rather small syllabary which could represent only open syllables (consonant-vowel) and the sound-system of Greek with its many consonant clusters and syllable-final consonants. This leads in Linear B to frequent homography (eg, pater ‘father’ and pantes ‘everybody’ both appear as pa-te), and Ventris was thus unable to demonstrate that the language really was Greek until he could explicate many of the longer sequences as intelligible wholes. In these sequences the chance of accidental fit was lower. Elsewhere, Ventris’ interpretation was sometimes supported by pictorial logograms/ideograms. Fischer, working with one short text and what appeared to be an equally restricted syllabary, did not have these advantages; he was forced to guess more often.

And, perhaps even more crucially, Fischer’s knowledge of historical linguistics and of Greek did not match that of Ventris (though he certainly tried, and his 1988 book has a very scholarly appearance). His first specific claim involved a vowel shift confined to his (naturally unattested) ‘Minoan’ dialect of Greek, which had created what appeared to be the ‘wrong’ values in many words. But not all of the ancestor forms he posited were archaic enough or otherwise plausible, and he also seems to have misconstrued some of the relevant phonetics and phonology (though it must be said that his 1997 book is more popular in style and in places suggests a higher level of error than actually exists). Elsewhere he posited many blatant exceptions to this and other rules, to suit the requirements of specific cases. Fischer also had to propose a ‘flip-flop’ rule, exchanging two vowels in certain environments; but such changes are rarer and more contentious than he apparently thought. And he had to posit some non-obvious consonantal substitutions (some in the spelling, some in the language itself) which, while not impossible, cannot easily be accepted without independent evidence. In respect of the grammar, his proposed constructions suggest a limited ‘feel’ for Greek (eg, his phthos kros ‘initial sum’, presented as a key piece of evidence in support of his ‘decipherment’, does not seem feasible as Greek) and in places he apparently misinterpreted basic sources (eg, hos ‘to [a person]’ is actually a very rare form; Fischer must have misunderstood his lexicon).

The translation offered involves a published announcement, or the transcription of a speech, by the commander of a Minoan naval force, urging his troops on to battle (apparently near Naxos) against invaders from Anatolia.

Fischer’s Easter Island material, though equally contentious, has met with a modest but not wholly negative reception and has appeared in refereed journals; but his ‘decipherment’ of the Disk has not met with general acceptance among classical philologists in the 15 years since it was proclaimed. This, and his problems with the material as detailed above, have not persuaded him that he might be on the wrong track, and his tone in his more recent book is quite arrogant in places. He adopts a patronising attitude to Ventris’ collaborator Chadwick (a very experienced philologist and decipherer); he is apparently unwilling to consider the possibility that Chadwick’s rejection of his ‘decipherment’ might not have been motivated by bias against novel ideas but rather by genuine doubts about the methodology and/or the results.


In 1999 another book on the Phaistos Disk was published by Butler (j). Butler is an enthusiastic and widely-read amateur who stumbled on the story of the Disk and was reminded of a Mayan calendar disk. He became convinced that the key to the text lay in the numbers of symbols rather than in the symbols themselves, and eventually decided that the primary meaning of the text was mathematical rather than linguistic. He is prepared to allow for a second, perhaps less important, linguistic reading; but arranging the text in this way would presumably have complicated matters enormously for the compiler, and no such text is known (in any event, Butler does not say much more about this second reading).

Specifically, Butler interprets the Disk as an astronomically explicit calendar. However, actual records of Minoan astronomy are virtually non-existent. Butler is thus forced to rely on speculation, deriving much support from Ovendon and Roy’s contentious suggestion that the zodiac must have been first recorded somewhere in the latitude of Crete at around the time to which the Disk is dated. The interpretation of the Disk in these terms goes beyond the basic zodiac to include planetary movements. Butler’s reading involves some individual identifications of symbol-sequences which are not implausible, but also, perhaps inevitably, a good deal of special pleading. This latter is very typical of numerically-based interpretations of ancient artefacts, and arguments of this kind typically prove nothing. Butler may know more mathematics than most who advance such arguments; but, before his interpretation could be accepted, a careful analysis would have to be conducted in order to determine the number of different interpretations of the Disk which could be made to work (given the rest of our knowledge of the Minoans) to the degree that Butler’s interpretation works, when manipulated in the same way and to the same degree as Butler manipulates his numerical reading.

It must also be noted that on the evidence which is actually available the observational astronomy of the relevant period, even in Egypt, was almost certainly too inaccurate to support Butler’s case. His ‘decipherment’, if valid, revolutionises our interpretation of the science of the period, and it must stand or fall in its own terms.

In this very context: in order to handle some anomalies in his thesis, Butler argues that the Minoans were especially concerned with measurements of distance. Drawing off the scholarly but controversial work of Thom, he posits cultural links between Minoan Crete and geographically remote cultures of the period, especially that of the builders of Stonehenge III; and some of his figures suggest to him that the Minoans may have known of Thom’s ‘megalithic mile’ and indeed may have known the circumference of Earth (!). Extrapolating further, he incorporates some highly implausible notions, including the possible reality of Atlantis (in the Atlantic) and the possession by the Minoans of extensive knowledge of the outer solar system. Almost inevitably, he compares his ideas on this front with the case of the Dogon (on which see Newbrook and Groves in The Skeptic 19:4) and with Swift’s alleged knowledge of the moons of Mars (discovered in 1877) in 1726. In fact, he goes so far as to suggest (like the Afrocentrist Adams) that some early human cultures, in both the Old World and the New, possessed telescopes; he actually translates the mysterious Quiche word quilpi as ‘telescope’! This shift to the fringe as the book progresses is very disappointing; by the end, the reader may be forgiven for thinking that serious scholars will probably pay rather little attention to Butler’s ideas in the years ahead.


The third (and least full) of the three recent major ‘decipherments’ has so far been presented mainly on the web, on a site set up on 15/12/97 (k). It is attributed to the American brothers Keith and Kevin Massey, who are also fervent advocates of the authenticity of the Kensington Stone (indeed, they believe they have proved its authenticity). Their Disk enterprise is presented as a ‘project’ rather than a finished decipherment; others are invited to join in and help. But the Masseys believe that they already have the basic answer: the script, in their view, resembles Proto-Byblic script (mainly used to write a Semitic language) more closely than it does any Aegean script, and on this basis they have assigned phonetic values (consonantal values, because of the nature of Proto-Byblic) to many of the Disk symbols. This, of course, conflicts sharply with Fischer’s interpretation. The Masseys claim that their method is more objective than those of others, but it is far from clear that this is really the case.

They go on to report that they tried Semitic languages at first, but eventually came to the view that the text is Greek, albeit written in a script not otherwise known to be used for that language or in the relevant area. As for Fischer, but perhaps even more so given the script identified, one might question the plausibility of this thesis and demand strong evidence.

One important difference between the Masseys and other Disk-‘decipherers’ is that they read the text left to right and outwards from the centre (see above). It is interesting that Greek readings of the text have been proposed for readings taken in either direction! The Masseys regard the text as an inventory of goods similar to most of the Linear B tablets (see below on numerals).

The Masseys have made an effort to learn about archaic Greek, but they are clearly not experts and do not feel confident enough to invent a dialect as Fischer did. Some of their comments are rather naïve and unsophisticated; eg, they seem happy to insert w- more or less where it suits them to do so in words beginning with a vowel, on the ground that ‘Archaic Greek as presented in Linear B…insert[s] (sic!) and include[s] the consonant /w/ in places where it is not today (sic; is ‘in classical times’ intended?) present and is not expected’. Some of the other forms posited seem to be related to known Greek forms in rather arbitrary and inconsistent ways. The Masseys take 16 of the sign-groups, those including a symbol resembling a slash, to be numerals, and a great deal of the partial ‘decipherment’ seems to rest on the specific sound values implied by these identifications. But even they admit that they cannot yet read the whole of this short text; and at present there is no good reason to accept their proposal.

Like most amateur workers on the Disk, those responsible for these three ‘decipherments’ do not refer to each others’ works at all; and, while Butler and the Masseys have published only recently, it is alarming that neither of these refers to Fischer, even in an attempt to debunk him. Where they do refer to other authors, it is to the mainstream, although Fischer and especially Butler seem to regard cautious mainstream scholarship as rather hidebound and fit mainly for debunking (a familiar pattern!). In fact, there is little evidence that any of these ‘decipherers’, or the earlier ones, have been aware of each other at all. Being isolated, private workers or small clusters of the like-minded, each with a growing conviction that they alone are right, they do not see any need to talk to those who espouse other views, and so they do not observe that the same unreliable methods ‘work’ more or less equally well for all of their mutually contradictory claims. One can persuade oneself, using such methods, of almost any identification of a mysterious inscription with a known language. But this is not how to arrive at a decipherment that will stand up to scholarly criticism. Fischer is a serious academic in a related field, and it could be said that he, at least, should have known better.

Still undeciphered…

The provisional conclusion must be that no-one has yet shown that they can really read the Phaistos Disk. It remains to be seen whether it can eventually be deciphered. There are cryptological arguments tending to suggest that the text is in fact too short to be deciphered unless a further lucky find, involving similar symbols, forces particular readings of substantial portions. For the moment, the Disk remains one of the more intriguing enigmas of early Europe.

Newbrook, Mark. 2002.

Dr Mark Newbrook is a skeptic, linguist and football hooligan. He managed to combine two of these interests to help pioneer the field of Skeptical Linguistics.