Sunday, May 2, 2010

Divine names on Linear A tablets

Parsing through the Linear A tablets several times, one can discover a large number of interesting phrases. The tablets mention places and persons, debths and taxes, a large variety of goods and even the names of gods offerings were given to. Given the intimate ties between religious and political authority in the Minoan and Mycenean era, the frequent references to religious ceremonies and donations should not be surprising. This is exactly what I would like to discuss.

It has already been well-known for several decades, that Linear B archives feature a high number of theonyms. Some tablets exclusively deal with the acts of sacrifice made to please the gods - like the famous Pylos tablet PY Tn316. This is an unusual document and therefore quite renown. It mentions golden chalices and people as a subject of sacrifice; the term PO-RE-NA might translate as "sacrificial victim", implying human sacrifice. Greek legends mention several instances of human sacrifice (remember Iphigeneia), so it must have existed in the bronze age - and now we have direct proof of it. But by any means, it is expected to having been carried out on very rare occasions only. Some have suggested that this tablet was written in the last moments before Pylos was overrun by the enemy. The mentioned sacrifices could have been the last desperate attempt to enlist divine aid, and were perhaps never carried out - as the tablet was found among the burnt ruins of the palace.

The sacrifices at Knossos were of much milder nature than what the former Pylos tablet has shown. Human sacrifice is never mentioned, and even animal sacrifice seems to be lacking. The gifts are exclusively agricultural goods, mostly oil or honey (supposedly given as libations). As an example, I will show one of these tablets below:

The text is roughly the following:
In the month of Lapathos: to Dwikutos(?) 1/18 unit of OIL, to Pipituna 1/18 unit, to Aurimos 4/18 unit of OIL, to all the gods 1/3 unit, to Qerasia 1/3 unit, to the priestess of winds 1 unit of OIL, to the Utanos priestess of winds 1/3 + 3/18 units.

All the personal names on this tablet are expected to be divine entities. Unfortunately we know nothing of Qerasia or Pipituna, except that the former name resembles the name "Thera" and the latter is somwehat similar to goddess Diktynna. The mentioned fractional oil units (the logogram is featured once in each line) are in fact quite large quantities: 1 unit of oil equals approximately 36 litres, so the smallest quantity is still about 2 litres.

Now, as we have seen the examples in Mycenean Linear B, it is time to look at those tablets that likely detail similar offerings in Minoan Linear A. We shall see that several putative theonyms can be discerned on Linear A tablets, among them the names of Maia, Eileithyia and Apollon!

We have already seen - thanks to the Egyptian scribes - that the Minoans had some divinities by the names Razija and Amaja. While for the former "Great God(dess)" there is no evidence on the Cretan clay tablets, there is one clear instance of the latter name. The Khania tablet KH14 features a broken heading, but sure-enough, a name A-MA-JA can be made out between two word-dividers. Given the frequent loss of initial vowels on Minoan words borrowed by Greek i.e. *Adikitu -> Dikte, it is quite possible that this goddess corresponds to the classic Greek Maia, leader of the Pleiades.

Tablet KH14
?-RA • A-MA-JA •CYP6
 *336 (horses)2
?NI (figs)1/4
 VIR+*307 (women?)2

The list of offerings to this deity is interesting, too: Apart from the figs (that were commonly used as food) and the CYP logogram (that might have meant barley or something other instead of Cypress-wood), we have a strange and unique logogram: a horse-head (Lin A *336), with a number '2' following it. Albeit horses were an important commodity in the later Mycenean age, the Linear A tablets never mention horses apart from this single occasion, so we must assume horse-breeding was less commonplace in the Minoan than it was in the Mycenean era. Thus this pair of horses is rather a special gift. In the lowest row of the tablet, a logogram VIR+*307 can be made out (with a number '2'), likely referring to women, but the context cannot be determined, as the rest of the tablet is broken off. But if the relatively little quantity of food FIC 1/4 belongs to them, we should rather expect temple-servants instead of sacrificial victims.

Our next tablet of interest is the Haghia Triada tablet HT96. Geographically, Haghia Triada is almost next to Phaistos (3 km away on the next hilltop), so we expect them to be the same polity, possibly even the same city. The Phaistos 'palace' must have been an important religious centre, and now we have some proof on our hands: this tablet mentions the term A-PA-RA-NE in the header on both sides. Although *Apalan(e) is a word somewhat different of the classical Greek Apollon (Latin Apollo, Etruscan Apulu, Luwian Appaliunas), but there is one term that makes this identification probable: the word SI-MI-TA, that is similar to a title of Apollon: Smintheus. This epithet refers to a hard-to-understand role of Apollon (Apollon of the mice). But form Hittite sources, we know that in the bronze age, mice played an important role in religion: they were not only given as religious offerings, but also used as a "scrape-mouse" (like a scrape-goat later) in ceremonies. Although SI-MI-TA is more probably read as *Simintha, a reading *Smintha cannot be entirely ruled out, since a weak-weak joining (a cluster of two non-plosive consonants) was somewhat a "scribe's choice" case: it could either have been simplified or resolved. Such ambiguities of writing rules may be caught on the tablet HT103 with the terms DA-KU-NA and DA-KU-SE-NE-TI. (*takusna versus *takusnethi ?)

Tablet HT96
 *323 *3171
 PI-TA-RA •1
 *323 *3441

?-KU-MA-RO • TENI (figs)20
?NI (figs)2

A-PA-RA-NE • QA-*118-RA-RE •*516 • GRA •40+1/2+1/4
 NI (figs)2+1/16

The list of goods is - again - quite special. Apart from relatively small quantities of grain and figs, side A mentions twice a hard-to interpret item with logogram *323 (it is like a slightly modified TI sign). Once it stands with *317, the logogrammatic "double-axe" sign, and the other time with an image of a roughly-drawn tripartite building (*344). Like this unknown object, items by the name RU-SA and PI-TA-RA were also given (they have integer quantities), but we do not know what on earth they refer to. Given the small quantities and possibly a sacrifice, they must have been valuable. On the reverse, the goods are less exotic: There is a large shipment of grain (more than 40 units), some type of oil (OLE+U) and figs. These items are from an owner or place *Qazir (QA-*118 [KH10] in the suggested "pertinentive" case *-(a)le).

For the last part (and to make this 'trinity' complete), let us observe another Khania tablet (KH5), detailing a transaction that might have had something to do with a religious offering. In the header, we find a name A-RA-U-DA, which shows a reasonably high similarity to the Mycenean theonym Eleuthia, corresponding to classic Greek Eileithyia. This identification is further reinforced with another small detail: the second paragraph on the table mentions a place-name WI-NA-DU, that can be identified with the Mycenean WI-NA-TO and the classic Greek town of Einatos (thanks to Miguel Valério) - actually quite close to the cave sacred to Eileithyia.

Tablet KH5
WI-NA-DU • *301-NA • KU-PA-DOCYP+?3+2/3
 NI (figs)2+1/2+1/3

At this time, the list of goods supposedly offered to goddess *Alauta is less exotic: A smaller quantity of wine, some CYP and a package of figs. Apart from the names, the header is difficult to interpret: while A-DA-KI-SI-KA might have been a verb, even a passive preterite one (ending -SI-KA), the word KU-PA-DO cannot be securely identified as the case of KU-PA (the -DO ending would be a hapax). Undoubtedly, we shall need more time, more finds and much more research to find out the truth behind these mysterious documents.

Many authors dealing with the Minoan religion forget about the essentially polytheistic nature of the Minoan pantheon. We can almost always read about a "Great Goddess", with little or no reference to others. But it is beyond doubt that like the Myceneans, the residents of Minoan Crete worshipped a plethora of divine entities. What is more, most of the classic Greek 'Olympic Gods' are though to be borrowed from earlier civilizations. Now we have seen a few intriguing examples of Greek divinities appearing in a purely Minoan context. Without doubts, this is just the 'tip of the iceberg': We know next to nothing about the portfolio of these gods, or their original attributes. Seems like we are looking forward to many more interesting discoveries in the near future!


  1. "These items are from an owner or place *Qazir (QA-*118 [KH10] in the suggested 'pertinentive' case *-(a)le)."

    Not sure what you're basing this on but if you're basing this on Etruscan grammar, then *-(a)le must be understood as a type-II dative ("to", "for"), not pertinentive.

  2. I am not basing it directly on Etruscan, but rather on the suffix system observed in Linear A contexts. I know it is quite tentative (the corpus is small, too), but I have no better concept at this time. Since I haven't found any overlap between -(a)le endings and the -(a)se/-si class, there is a possibility that these two were complementary to each other, especially the -(a)se and -(a)le endings. But the distributions of -(a)le and -si are different. Personal names quite often end in -RE (also in -SE), while the -SI ending seems to be present in more "special" contexts. (That may imply 'outgoing goods', I don't know how to put it better.) Thus the meanings based on the *-e/*-i alternation aren't in agreement with Etruscan (even if the consonantal bases are). So if we are to produce a consistent 'Aegean' theory, my theory probably needs amendment (I noticed it too).

    I am still considering the possibility of a system like the Mycenean Greek illative / locative / elative cases show. These endings: *-de / *-di / *-den are always problematic when derived from proto-IE, so I do not exclude a direct borrowing from a non-IE language. But even these are in poor agreement with Etruscan or Lemnian suffixes. This is the point we need a great intuition to be able to reconstruct how Aegean case-endings evolved over time. I invite you and anyone else one who would have constructive ideas and insight to reconstruct the Proto-Aegean and Minoan systems.

  3. Bayndor: "These endings: *-de / *-di / *-den are always problematic when derived from proto-IE, [...]"

    Nope, they aren't problematic at all. All I see in Proto-Aegean is a postposition *tʰi 'in' which was originally not a true case ending since it was used in tandem with the locative case ending *-i. This is the state of affairs preserved in Old Etruscan.

    However, Indo-European had the locative postparticle *dʰi. Mystery solved.

  4. Yes, I am aware of the possibility to derive such structures from PIE particles *de, *dhe or *dhi. Apparently these suffixes are limited to Indo-Iranian, Hellenic and Slavonic branches. Nevertheless, Greek is supposed to not only have these endings conserved, but also to have innovated a third case -θεν (that I erroneously wrote with -d- in the previous comment) with elative meaning, to complete the paradigm to a three-directional system. I cannot shake the thought of substratum influence here, especially if that language influencing Greek was an 'Aegean' one, endowed with similar and related suffix-particles in -θ- or -t-. For example, if a nearby Aegean language used endings -θι and -θε(ν) for locative and elative purposes (which is perfectly in-line with the Linear A -TI and -TE endings), that could have helped to reinforce or to complete this paradigm in Greek. (I don't know if this can solve the question of PIE d- / dh- alternation in the cited theory) The ending -δε, on the other hand, might be a straight PIE heritage, like Iranian *-da. Unfortunately, borrowings between distantly related languages are much harder to detect than completely unrelated ones; this would be the case if we supposed an Indo-Aegean protofamily.

  5. "Apparently these suffixes are limited to Indo-Iranian, Hellenic and Slavonic branches."

    No, the comparanda contains also Armenian telwo-ǰ 'in a place' and Latin unde 'whence'. I'm sure this is a PIE particle because it nicely supplies an origin to PIE's 1pp and 2pp mediopassive endings.

    At any rate, Etruscan simply doesn't do what Greek does with its case endings and there isn't anything in the available Minoan corpus suggesting this. And if one wishes to explore areal influence, why not Anatolian then?