Add to favorites Examiner Bio

Phaistos Disk: Greek or Luwian?

In the July 2000 issue of the American Journal of Archaeology, Yves Duhoux reviewed Jean Faucounau’s book Le Dechiffrement du Disque de Phaistos: Preuves et Consequences (The decipherment of the Phaistos Disk: Evidence and Results, 1999, Paris: L’Harmattan). Monsieur Faucounau was kind enough to send me a copy of this book and to write a response to Professor Duhoux’s review, which he felt required a reply. The Word Geek will respond, in turn, to this author's responses, adding her thoughts on this most interesting book.  She apologizes for having taken so long to write this column, but was making a translation of the book for the edification of her husband, who showed some interest.

1)      Duhoux: "Faucounau's book is the latest of many attempts to decipher the Phaistos Disc."

Faucounau: “Actually, the decipherment was published in 1975 in the R.E.A. The book is just a justification of it. Why such a long delay ? Because any decipherment of such a short text HAS TO BE PROVED, and it took me some 25 years to find, one by one, some 30 pieces of evidence. None was examined by the reviewer.” 

Word Geek’s two cents: It will not matter, in the end, what sequence the correct decipherment comes in, whether it is the 100th or the 1,000th attempt, so long as that decipherment is correct. For example, once Linear B writing was shown to be Greek by Ventris and Chadwick, it didn’t matter how many or how few people had previously tried to prove that it couldn’t be Greek. Still, it took others to pin down many of the finer details. It so happens it took a whole committee to make the big breakthroughs in Mesopotamian cuneiform. The point is, when one is standing amid a crowd, especially a not-very-distinguished crowd which includes a great many crackpots who have claimed to have deciphered the Phaistos disk in mostly unconvincing ways, one must be prepared to encounter a certain amount of skepticism.

2)      Duhoux: "On this criterion (of the direction of reading).the attempts by numerous authors are suspect".

Faucounau: “They are not ‘suspect.’  They are plainly wrong.  Ten pages of my book are dedicated to this question.  Not a word about that by the reviewer!” 

Word Geek’s two cents: The world will finally know definitively in which direction to read the Phaistos disk when it has been conclusively deciphered, in the opinion of the Word Geek. Looking closely at the overlaps among signs on the disk can demonstrate pretty clearly in which direction the signs were stamped. But, if an illiterate person was doing the stamping, following a model that was created by a different but literate person, the illiterate might have gone about it in the wrong direction. Faucounau’s book indicates that he thinks this impossible for the disk, due to the fact that the “scribe” of the disk made many errors. An illiterate would not make so many errors and then correct them, he feels.  However, different decipherers come to different conclusions on this point based on different criteria.

3)      Duhoux:  "Any grid that assigns a consonantal value to signs has.. a good chance of being wrong".

Faucounau: “What is ‘a good chance’? Where are Duhoux' calculations supporting this statement ?” 

Word Geek’s two cents: The hypothesis most commonly held among linguists is that the signs on the disk represent syllables, based on the number of distinct signs (around 45, though there is some variation even in this number among would-be decipherers). 

If there were about 26 symbols, people would think it was an alphabet. If there were about 100 different symbols, they’d suspect it of being more purely hieroglyphic (which is to say pictographic). Still, not everybody in the ancient world wrote with a syllabary. The ancient Egyptians did not. The Phoenicians, who came along quite a bit later, did not use a syllabary either. They only wrote consonants. However, the Phoenicians are, as mentioned, much later. In the Bronze Age, there is also cuneiform, which was much like Egyptian in combining syllabic symbols and determinatives (or hieroglyphs) in an large, untidy -- and as she mentioned before, syllabic -- system. Other writing systems of the time were hieroglyphic or syllabic. Considering the preponderance of data on systems in existence during the Bronze Age, then, it is indeed more likely that the symbols on the Phaistos disk represent syllables than that they are hieroglyphs or simple consonants. That has little or nothing to do with statistics. It’s based on the observation that, in the Bronze Age, syllabaries were pretty much what everybody used for writing, with the Egyptians being the main exception (unless you get into the Semitic systems in Canaan and Faucounau isn’t suggesting the disk is Canaanite). M. Faucounau does statistical calculations but not all statements are based on that sort of thing.

4)      Duhoux: "Greek is probably the language most often proposed."

Faucounau: “For a good reason : a long series of statistical calculations shows that it has more than an 80% chance of being correct !” 

Word Geek’s two cents: “Since this disk was found in Crete, and the people of Crete today speak Greek, that’s a good language to assume was spoken by the maker of the disk. Still, that’s a guess, or a hypothesis, not a fact. Besides that, we know that not everybody on Crete spoke Greek in the Bronze Age. The classical Greeks mentioned people they called Eteocretans who did not speak Greek. Further, we know that Linear A, written by the Minoans on Crete before the Mycenean Greeks came, did not represent Greek. Professor Hubert LaMarle considers it to be an early Indo-Iranian language, related to Old Persian and Sanskrit. So it’s also possible that the Phaistos disk does not contain Greek, but this Indo-Iranian language.

 Then again, as Jan Best and Fred Woudhuizen point out (in Lost Languages from the Mediterranean, 1989, Leiden, E.J. Brill), the Cretans were in contact with the west coast of Asia Minor, where Luwian was spoken, so it’s also possible that the disk was written in Luwian, or even Hittite. The Cretans also had contact with people on Cyprus, on the coasts of Canaan, and even with Egypt. How can anyone say for certain that it is not one of these languages? The fact that more people hypothesize it’s Greek than any other language isn’t proof that it’s Greek. That may simply prove that the Grecophiles are more likely to try deciphering the disk than the Hittitophiles (who may be preoccupied with puzzling over Hurrian or Hattic bilinguals).

5)      Duhoux: "Faucounau rejects this important parallel (with the Arkalokhori Axe)".

Faucounau: “Wrong. I did not! I have even explained it in a (still unpublished) paper, that Y. Duhoux preferred to ignore.” 

Word Geek’s two cents: “Very few would-be decipherers of the Phaistos disk talk about this famous axe from Arkalokhori which is said to have symbols on it resembling those on the disk. Even fewer use their proposed decipherments to address the symbols that are on this object. The Word Geek thinks that doing so would be a good test of any proposed decipherment. But not many would-be decipherers care very much what she thinks. One can find many excellent pictures of the Phaistos disk here and there on the web and in books (National Geographic once had an article on it, for instance, and their photos are hard to beat). It’s quite hard to get a good picture of that axe, though, and artists’ drawings are not nearly as good as photos.

Nevertheless, the decipherers who identified the disk as a Luwian-inspired letter between Idomeneus and Nestor (Best and Woudhuizen) did address this axe, showed the symbols alongside Luwian parallels, and translated them. The Word Geek is duly impressed, though she can’t say she’s convinced that (a) the axe actually contains the same script as the Phaistos disk, or (b) that Best and Woudhuizen have truly deciphered the Phaistos disk. Still, their translation of the axe seems plausible enough. It basically seems to say what those pots and such from Italy (predating the Roman Empire) usually said in obscure dialects of Italic and Etruscan, “Joe Schmoe made this here object” (or rather, “In Scheria: these double-axes, Akarkis, the son of Chanu, has made {them}” (p. 99)). What else would one expect an axe to say? 

6)      Duhoux:  "Faucounau gives no edition of the disc of his own, but regrettably reproduces Evans' facsimile."

Faucounau: “ One has to add : with some modifications, But why ‘regrettably’?  Are later designs more accurate?  I suggest the reader compare the design of Sign 17 in L. Godart's publication, recommended by Y. Duhoux in his Bibliography (an alleged ‘pot lid’ with a rounded shape) with the original to be fully enlightened!“

Word Geek’s two cents: There are many excellent pictures of the disk on the web. If one doesn’t like a particular representation, one can easily go to another. Evans’ drawings are fairly nice, as drawings go. But – and it’s a big but – a good photograph is definitely better than a drawing when one is going to discuss fine details and claim to see things that nobody else sees, such as monkeys, owls on poles, and lampreys, which M. Faucounau does. Since no one else sees these items on the disk, it behooves him to provide better evidence than a drawing. Any artist, no matter how gifted, may regularize an irregular detail, leave out some tidbit that seemed unimportant, and perhaps see something that really isn’t there. For example, some people believe that one irregular shape is a seashell whereas others believe this same shape is a jug, complete with a handle. Which is it? A drawing indicates what the artist sees, not necessarily the reality. A photograph allows the reader to make up his or her own mind. 

Faucounau’s book makes a great deal of many such apparently minor details, including overlaps, erasures, damage to the edge of the disk, and the small slash mark, which not every artist renders. The illustrations in his book are not particularly good, so these details are not verifiable. Such claims as he makes render excellent closeup photos a necessity. Closeup views are not as easily found on the internet as general views, either. A closeup would reveal whether or not there really is a lamprey or monkey sign hiding behind another symbol, something this reviewer admittedly could not see. There were no monkeys in Greece, after all. This choice for an erased symbol is less than obvious. The owl was popular in Athens much later, but that was not true in the Bronze Age. This it is indeed regrettable that no photographs accompanied this edition.

 

The Word Geek must say, though, that the biggest problem with any decipherment comes from the amount of subjectivity involved in identifying many of the pictures. The running man is pretty obviously a running man, for example. Even so, what word is to be identified with him is not nearly as obvious. Is it the verb “to run”? Best and Woudhuizen seem to think so, since they pair this glyph with a Luwian pictograph of a pair of feet, not a whole man, making it represent the sounds SARU. Or is this picture meant to represent “man,” or even “young man old enough to be off his mama’s lap but not big enough to pick up a spear and shield and go to war,” and so the Greek kouros? Faucounau goes with this, making the symbol stand for the syllable ku, the first syllable of the Greek word. Which of these two answers is best?

There is that so-called “pot lid” which is called a “knife” by others, too. If it’s a knife, the handle is certainly in an odd place. Still, even if everyone agrees it’s a knife, and not a pot lid, what word should go with it? Should it be “knife” or “to cut”? Should it be pronounced in Greek or Luwian? Best and Woudhuizen make it tashuwar. But Faucounau makes it lu. Is the choice between these two options perfectly obvious to every reader? It wasn’t to the Word Geek.

As a third and final example, there is the Mohawk, that warrior head supposedly wearing a helmet decked with feathers. Faucounau assigns a phonetic value of ka to this symbol, based on the Ionian Greek kare for “head.” The feathered helmet is an unimportant detail, apparently. Best and Woudhuizen give this same symbol the designation a (with an acute accent), pairing it with a Luwian face symbol that has neither hair nor helmet. But let’s just suppose for a moment that the Indo-Iranian language that lay behind Linear A was behind the Phaistos disk. Couldn’t “head” be something akin to the Sanskrit mastaka, in that case? Then might one not, with equal justification, assign this same symbol the phonetic value ma? Then again, even supposing the language were Greek, perhaps the word inspiring this glyph was really “feather” and not “head”? Then everyone is barking up the wrong tree! The Word Geek sees no sure way of resolving these dilemmas.

7)      Duhoux: "The script of the disc should total about 60 different signs, not Faucounau's 88 or 95."

Faucounau: “Any mathematician will tell Y. Duhoux that, concerning such a provision, it is incorrect to treat variables in the same way following Poisson's Law and variables following Gauss Repartition.  But Y. Duhoux may have some trouble with mathematics, as can be seen when one compares the statistical conclusions of his 1979 paper on the Disk (‘The language must be Indo-European and possibly Greek’) with those of his 1983 study on the same topic (‘the Disk's language cannot be Indo-European’).”

Word Geek’s two cents: One cannot assign either a language or a language family with full confidence until one has fully deciphered the disk, so one is free to change one’s mind from time to time, as Duhoux did. Statistics can be a useful first step toward decipherment, when one has a lot of material.  Ventris and Chadwick demonstrated this with their very famous decipherment of Linear B, which turned out to be Greek. However, statistics only work on large amounts of data and there is only one Phaistos disk (not a lot of data by definition). When applied to samples of less than one hundred items, statistics give misleading results, as any mathematician knows.

The number of specific pictograms shown on the disk is visible to anyone who looks at it. Even so, not everyone agrees on what number is there. Some say there are 45 symbols there and that’s all there were. Some say 45 are there but some others would have existed in the writing system that just didn’t happen to show up in that sample. Based on knowledge of other writing systems, various people estimate that unknown number that might exist but didn’t show up to be some small number, such as 20. Faucounau is alone is assuming such a large number of symbols in the writing system. But until more samples are found – more disks or more objects with similar writing on them – no one can prove these assertions.

As for the language behind the symbols, it isn’t necessarily Indo-European, even though that’s the most popular choice (and most would-be decipherers speak Indo-European languages, so it’s possible that they are biased). It might be Semitic, since the Cretans were in contact with people from Semitic areas in Canaan. The language on the disk could be Hurrian, for all anyone knows, since the Cretans knew people in ancient Syria, where the Hurrians lived. Even if it is an Indo-European language, it isn’t necessarily Greek, as noted earlier. It might even be the ancestor of Albanian. Nobody knows for sure where they came from. Maybe they were all over the Aegean since the “beginning” and everybody else just crowded them out. No, the Word Geek just made that up on the spot, for entertainment.

8)      Duhoux: "Such observable facts prove Faucounau wrong."

Faucounau: “ The so-called ‘facts’ are experimental statistics made with an implicit HYPOTHESIS, which is, in fact, wrong.“

Word Geek’s two cents: Statistics are unlikely to prove someone’s decipherment wrong to the Word Geek’s satisfaction, but neither are they likely to prove someone’s decipherment to be correct, either. Statistics only suggest such that someone’s decipherment is more or less probable. However, photographs that are reasonably clear should prove one way or the other whether there is a monkey or owl on the disc, as claimed by Faucounau.

9)       Duhoux: "The structure of Faucounau's syllabary is aberrant."

Faucounau: “ Compared to the Cretan/Cypriote syllabaries, yes!.. But why should the Disk's script be Minoan ?.. Just because of Y. Duhoux' ‘intimate conviction’?’..

Word Geek’s two cents: Since the disk was found on Crete, most researchers assume it was Cretan in origin and thus Minoan. That is a reasonable supposition under the circumstances.  Therefore, most researchers compare the symbols to other writing systems used on Crete or derived from them. That means Linear A and B, Cretan hieroglyphs, and the later Cypriot syllabary. That makes good sense too. One must start with what is known (these other scripts) in order to work toward what is unknown (namely, the Phaistos disk). Also, some symbols on the disk itself suggest Minoan culture, in particular the woman with the flounced skirt. There are a number of frescoes on Crete of women wearing flounced skirts. The little ship symbol has an upraised prow and that, too, appears in frescoes of ships, those on seals found at Knossos, and so on. There is a bug which might be a bee on the disk, and now and then there are hints that bees and honey were important to the Minoans. The rosette (referred to in a previous column as a daisy) is also found elsewhere in Minoan art. The head of the Mohawk or “feathered warrior” as most call him is another symbol which may indicate a Cretan origin.

None of these things prove the disk to be Cretan, admittedly. An image of a woman with a flounced skirt was also found on Cyprus, so perhaps the disk was an import from that island. The rosette appears on mainland Greece. Feathered warriors show up in Canaan. Other items can also be connected with other areas around the eastern Mediterranean. Still, all these things together mean that comparisons are most likely with Linear A, Linear B, and Cypriotic writing systems. If one wishes to compare the writing on the Phaistos disk with something further afield, one has to justify this. 

Some have done just this by pointing out that the disk could be an import from somewhere else, noting that the Minoans were great sailors. They had contact with lots of places as noted earlier. They went to the Near East, hence the Masseys compared the script to Byblic script (the writing found at the city of Byblos). The Minoans also made contacts with Egypt. There are reliefs in some Egyptian tombs clearly showing Minoans bringing Minoan products as tribute to one of the pharaohs, so comparisons to Egyptian are not totally off the wall, although this sort of comparison hasn’t been very popular. The Minoans also went to various locations in Anatolia. This seems an especially good locale for comparisons since the classical Greeks later had traditions that there were ties of kinship between the Eteocretans and some people in western Anatolia, including the Carians. That means scholars feel relatively free to compare the writing to lots of writing systems used in the Bronze Age. 

However, if one wants to describe a whole new language (Proto-Ionian, as Faucounau calls the dialect of Greek which he reconstructs in his book), and a whole different type of writing system (which is what Duhoux means by “aberrant” here), and a whole different interpretation of the culture of the Bronze Age from what is generally accepted by archeologists, and a whole different interpretation of the evolution of the Greek language from what is generally accepted by linguists, and a whole different interpretation of the settlement of Greece from what is generally accepted by historians, then I’m afraid one has to expect a certain amount of animosity from the rest of academia. 

10)   Duhoux: "Decipherers should be completely confident that, if their decipherment is correct, it will triumph in the end."

Faucounau: “Surely! But with reviewers like Y. Duhoux, who don't pay attention to the proofs brought in, one just has to wait half a century, like A. Wegener did with his theory!.. As for me, I am certainly waiting with full confidence for the time when Y. Duhoux' review will rejoin A.J. Beattie's paper on Linear B decipherment in J.H.S. 1956.”..

Word Geek’s two cents:  Even those scholars who work on fully deciphered languages – like Egyptian hieroglyphics and Akkadian cuneiform – are not 100% confident of every element of their work. In fact, this is typically the case. They leave question marks in their translations to indicate words whose meanings they are not sure of, square brackets to indicate uncertain transcriptions, parentheses to show words they felt were understood but not actually included by the original writer, and so on. They fill their papers with long footnotes explaining why they chose this particular translation for this first word, ponder long and hard about the possible meaning of some second word that totally defied translation, while arguing that some other scholar has transcribed yet a third word incorrectly or reconstructed it from an abraded portion of the tablet in a faulty manner. All these uncertainties are simply in the nature of dealing with old languages that haven’t been spoken or written for over a thousand years. One small tablet is not simply, quietly translated by a single scholar and then left alone. 

The first scholar makes his translation, as described.  When other scholars read these translations, with their hesitations, their question marks, parentheses, and brackets, they make comments and criticisms.  These are not personal attacks.  This is, quite simply, how scholarship is done.  Little by little the work proceeds.  If A has made a mistake in the reading, then B catches it.  If C has made an error in a statistical calculation, the D notes it and corrects it because C has provided the data and the mathematical formula that enables the whole thing to be repeated.  That is science.  Realiability and validity are not a matter of who can shout the loudest but whether results can be repeated, whether all can follow the same logic.

In the field of decipherments, no one person ever does it alone, or all at once.  Not even Champollion, the man who "cracked" Egyptian hieroglyphics.  Not even Ventris and Chadwick.  No one man will "crack" the Phaistos disk either, I'm afraid.