Saturday, June 18, 2011

Place-names on Cretan sealstones - A key to the decipherment of Minoan Hieroglyphics?

There is one last post I would like to append to my long series on bronze-age Cretan place-names. This one will encompass some fairly new research into the oldest relics of Minoan writing. I am struggling to make it simple, so I hope you will find it interesting even without being a professional in ancient writing systems.

In my previous post, I mentioned that some words found on Cretan hieroglyphic seals can be interpreted as place-names. I also made a number of assumptions when reading toponyms in Linear A. For example, one of the place-names was reconstructed there as *306-KI-TA, from barely two occurrences of the name - both texts were damaged, and they were showing derived cases only. This would have left this toponym highly tentative - until now.

While browsing the database for Cretan hieroglyphic seals, to either confirm or dismiss the idea about reading place-names on sealstones, I came across specimen CHIC302. This single seal presents a word WO-KI-TA - a toponym in its "base" case, just as it was predicted from Linear A! To get this reading, some very simple rules have to be kept in mind: The sealstone is actually a multi-faceted bar. It was drilled in the middle, in order to be worn on a necklace by the owner. When used to "sign" a document, the bar was rotated on a flat layer of clay (by an indefinite number of times), to give a continuous impression. Therefore the sign-groups on each facet are not independent from each other. On the contrary: they give a coherent text from the start until the end, with many of the words "overflowing" from one side to another. There are no word-divisors to help us, just small "start signs" to emphasize the direction of reading. In most cases, the inscriptions turn out to be boustrophedons: The signs are arranged in the most economic way possible, and their direction reverses (alternates) each line.

On the cited sealstone (CHIC302), there is only one facet that has a start sign on it. Therefore it is concieveable that reading has to be started there. While the introductory term is found on certain other seals with longer text, we do not have any hint on its meaning. On the other hand, the second side clearly presents a hapax: a one-time word, suggesting that this is a personal name. The immediately following term consists of three signs, and it is reasonably common on other seals as well. Although the value of the middle sign is uncertain, a potential reading could be JA-RA-RE. In some seals, it returns as JA-RA only (basic case?). I labelled it secondary title, to reflect the fact that it is not an obligatory component of any sealstone, and found only on a fraction of them - but there, it can also substitute a personal name. Next comes the term WO-KI-TA (split between two facets) - this is clearly a place-name, based on Linear A parallels, and could be an early reference to Lyttos. The last two signs make up an incredibly common word - found on most seals. This is what I call primary title. Despite the fancy name, I have no idea of its precise reading or meaning: it could have designated an impersonal entity as well ("polity", "kingdom", "province", etc.)

Perhaps it is useful to make a de-tour from the topic, and examine the Eteocretan material for parallels. Unfortunately, Eteocretan inscriptions are few and far between, and most of them are pretty fragmentary. Yet one of the Praisian stone slabs offers us a particularly interesting insight to the sequence seen on CHIC302. On the second line of the stone, the following sequence can be read (in Ionic letters): ?δο??ιαραλαφραισοιιναι. Unfortunately there is no word separation; yet - if we follow van Effenterre's considerations - we can be almost sure that the word *inai was separate. This phrase is also seen on a bilingual Drerian inscription, where it seems to parallel the Doric Greek verb εϝαδε = "(it) pleased", "(it was) decided", "(it) came to pass that". If so, it is most plausible that it be preceded by a name or a title - or even a series of them. That phrase could have been either *?doph? iarala Phraisoi (this is the most straightforward one) or *?doph? iaral Aphraisoi (this is what Linear A parallels suggest - c.f. SI-DA-TE vs. A-SI-DA-TO-I, both on ARKH2). In either case, the term *iaral(a) could correspond to our "secondary title" JA-RA-RE. Note that there is not a single occurrance with an initial A-, so the J- initial was probably part of the stem, and not an attached prefix particle. That would make it similar to the Greek word ἱερός (='holy'), despite the fact that ἱερός has a good Indo-European etymology: It is thought to stem from PIE *(e)is-əro = 'exalted one', making any connections to the Minoan title *yara-(a)le very dubious.

CHIC302 is not the only seal that features toponyms. There are at least a dozen seals with comparably long inscriptions. But the scarcity of signs with easily identifiable Linear A counterparts severely limits our reading capability. Four other seals exist that feature the term KI-TA-NA or its derivatives. CHIC295 has a fairly similar composition of names and titles as we have seen before. The text is also a boustrophedon - this is highlighted by the "start signs", featured in every line. The only interesting feature of this seal is the presence of not one, but two primary titles. Conversely, the second title seems to be declined - as it possesses both a prefix (the MA- prefix seen on the Phaistos Disc) and a suffix (perhaps -SE or -RI). However, I can offer no guide on whatever these might mean.

Some caution is yet to be exercised; in spite of the plausibility of these readings. I have intentionally selected seals that have a relatively clear composition and direction of reading. Many of the seals are not so easily cracked: they are full of artistic ligatures, complicated circular arrangements of signs, and decorative placeholders - that might look like signs - while they are in fact nullities - fancy decorations only. Sometimes they are even inserted in the middle of a line - making the job of the reader really hard.

This problem also applies to sealstone CHIC260: a nice triangle-based prism with rather clear figures. The reading of the first line is however, dubious: it depends on whether we regard the simple, circular drill as a sign (Hiero *73, probably QE, giving the name JA-QE-RA), or a nullity, in which case the remaining signs form a "secondary title" JA-RA. In the second and the third line, an already familiar term is found: these signs read KI-TA-NA-SI, a declined form of KI-TA-NA, similar to what is seen in Linear A on the pithos PEZb3: KI-TA-NA-SI-JA-SE. Finally, the last two signs give the same "primary title" as seen on all our previous examples. We can see that in this case, reading is linear: this is due to the mathematical impossibility to make an infinite but regular boustrophedon from an odd base number of sides on the prism to be rotated.

No matter how many sealstones exist in the museums of the world (CHIC260 and CHIC302 can be seen in life at the Metropolitan Museum, New York, while CHIC295 resides on Crete, in the Iraklio Museum), the find-spots of specimens are rarely known. This is because they are valuable, and can be looted rather easily from tombs. Therefore we do not know where CHIC302 hails from: but it could have been plundered off Kastelli hill (the site of ancient Lyktos). Out of the four seals which feature KI-TA-NA, only two has a known provenience: CHIC238 comes from Mochlos and CHIC310 was found near Sitia. Looking at the map where these places lie shows a spectacular overlap with Linear A inscriptions containing KI-TA-NA: all these spots concentrate in a well defined area of easternmost Crete. While the vessels tell only little of history and geography (as they are traded freely), the presence of sealstones with the same city-name over a wider area paints a more definite picture on the political landscape of eastern Crete. It could easily imply that the towns at Mochlos and Sitia - and perhaps Palaikastro, Praisos and Makryghialos as well - fell under the same single authority. Given the impressive size of its city-center ("palace"), that central authority could not have been other than the polity of Zakros.

Based upon this realization, I slightly amended the map of Prof. Metaxia Tsipopoulou: the spread of references to KI-TA-NA implies that Eastern Crete was a politically unified entity, not a collection of rivalling city-states as previously assumed. While Zakros was deserted after the LMIb period - making it a ghost-town in the Mycaenean era, it was not completely erased from memory. For example, the Linear B tablet Am821 clearly refers to a person as hailing from KI-TA-NE-TO • SU-RI-MO. The latter name is known to be a place lying on the easternmost end of the island, together with U-TA-NO: Thus *Surimos could have been the same as Palaikastro (it was a powerful settlement in the LMIII era), and *Utanos the neighbouring Itanos. So it could be that Mycaenean Greeks still referred to the Eastern Lasithi province as KI-TA-NE-TO, despite the earlier demise of its name-giving capital city at Zakros.


  1. Connecting WO-KI-TA to Lyttos is a terribly large stretch unless you're suggesting that the former name is unrelated to the latter.

  2. András, most of what I read about the hieroglyphs comes from you, so please forgive me if this is a fiercely ignorant question. The start sign concept was new to me and has my mind reeling - is this your own theory? I would love to read more, if you can recommend anything!

  3. Actually, the "start signs" were already identified as such, and called Stiktograms in the CHIC book. Unfortunately (due to my ignorance) I do not know if anyone examined this subject before Olivier and Godart did write their magnificent book. Up to date, it is still the best on this matter. You can also check John Younger's Hiero site, but I loathe to say: it is not as good and accurate as his Linear A homepage.

  4. For the rings from Mochlos and Fourni, I've encountered something from Linear B that adds some further and interesting context.

    Both have *171. Are there other signets or seals that also include it? This symbol appears once that I can find in Linear A (ZA6, and it's tentative), but several times in Linear B. It's attested in an ideogram form at Knossos, Mycenae, Pylos and Thebes.

    The persistence of ideograms is a treasure we have in Cretan literacy. Given that it's an ideogram in the latest branch of the syllabary, it likely would've also been treated as such in Linear A and the Cretan hieroglyphs.

    As an ideogram, what do you think it might be?

    Here is the context from Linear B:

    KN Ga 464
    da-*22-to , / o-pe-ro , *171 (4)[ ] PYC 26 T 3 ku-ta-to , / o-pe-ro , *171 (4) PYC 15

    KN Ga 519
    ] ku-pa-ro *171 2[ ]to a-pu-do-si [ ](4) T 5 o-[ ] vacat [

    PY Aq 64
    ]-re-wi-jo-te [ ]-ja , mo-ro-qa , to-to , we-to , o-a-ke-re-se ZE 1 *171 3 ka-do-wo , mo-ro-qa , o-u-qe , a-ke-re-se ZE 1 ru-ro , mo-ro-qa , o-u-qe , a-ke-re-se ZE 1 ku-ru-me-no , mo-ro-qa , i-te-re-wa , ko-re-te , to-to (,) (we)-(to)[](o)-a-ke-re-set *171 (6) pe-ri-mo , ti-mi-ti-ja , ko-re-te , to-to-we-to[](o)-a-ke-re-se ZE 1 * 171 3 o-a-ke-re-se pe-ri-me-de-o i-(*65) , po-so-ri-jo-no , te-ra-ni-ja , a-ke-re-se , to-to-we-to , *171 1(2) po-ki-ro-qo , e-qe-o (,) a-to-mo ZE 1 vacat o-da-a2 , ko-to-na e-ko-te e-ta-wo-ne-u , to-to-we-to , o-a-ke-re-se ZE 1 *171 6 a-qi-zo-we , to-to , to-to , we-to , o-a-ke-re-se ZE 1 [ ] ne-qe-u , e-te-wo-ke-re-we-i-jo , to-to , we-to , o-a-ke-re-se ZE 1 [ ] me-wi , e-ru-ta-ra , me-ta-pa , ki-e-wo , to-to-we-to , o-a-ke-re-se ZE 1 [ ] vacat

    TH Wu 46
    CAPf PA-RA-WO , O-PA *171 30

    TH Wu 56
    CAPm [O-]PE-(RE)-(TA) O-PA *171 30

    TH Wu 59
    *171 36 [ ] SUSx PA-RO , SA-ME-WE RO-WE-WI-(JA)

    TH Wu 76
    BOSf A-E-RI-QO vacat O-PA *171 30

    MY Ge 606
    [SO]-SI-MI-JA A-KE-RE-U-TE KA-NA-KO AROM+KO T 2 KU V 1 RE-U-KA V 1 MA-RA-TU-WO V 1 / SA-SA-MA V 4 E-RU-TA-RA M 3 MI-TA PE 2 KO-I-NO , DE 1 (*171) 11 vacat