6 Jul 2011

Minoan Asasarame is not a deity??

The transliteration of an inscription on a libation table from the House of the Frescoes (KN Za 10) is written out as ]-TA-NU-MU-TI • YA-SA-SA-RA-MA-NA • DA-WA-[•]-DU-WA-TO • I-YA[ by John Younger to which Bayndor (Andras Zeke) of Minoan Language Blog has plausibly revised with DA-WA-SI. Shedding the ugly brackets, I would thus reconstruct the inscription in full as:

Bayndor's Minoan translation turns "bloody"

In his latest entry Those "bloody" Minoans..., Bayndor commits a number of false deductions towards his favoured value of YA-SA-SA-RA-MA-NA and I'm left disappointed by how he rationalizes this. The status quo maintains that Asasarama refers to a goddess figure (or figures) and despite any perceived difficulties in the comparison with Anatolian epithets, it still fits the context best. It's comparison with Isassaras-mis (see The Song of Ullikummi), with Mycenaean Potniya, with 'My Lady' equivalents in Asia Minor (cf. Kubaba, Cybele, Asherah, Ashtarte, Ishtar, etc.), and Egyptian ones (cf. Hathor, Isis) only adds to its historical plausibility. So I expect that any credible objection to all of this must be next to flawless.

To be brief, Bayndor's alternative translation based on Luwian asḫarmis 'sacrifice' is shockingly incoherent for someone who spends much of his time studying Minoan. The Luwian sequence /-sx-/ can't possibly explain reduplicated -SA-SA- in any meaningful way. This reduplication is so consistent in Linear A that it's absurd to avoid interpreting it as anything other than underlying -sasa-, not *-ssa- or *-sa-. As such, Bayndor has no leeway here. I won't dwell further on something so easily falsifiable.

Asasarama is *not* Minoan

Given the formulation of the original hypothesis, Bayndor errs some more when he states: "Isḫassara- is a compound stem, made up from isḫa- = 'lord' and the feminizing suffix -sara-, thus meaning 'lady'. None of its parts have a particularly good Indo-European etymology." Yet the source of -sara- is already commonly known to be from Proto-Indo-European *-s(o)r-, a suffix present also in Celtic and Indo-Iranian! Therefore Asasarama *can only be* from an Anatolian Indo-European language like Hittite. Even Judith Weingarten, who we may also assume studies Minoan rather extensively, falls into the same false reasoning in her comment further below: "So, I'll stick with Isḫa-ssara as the most likely parallel, also because it seems non-Indo-European in origin." Sigh.

There's a difference between the origins of isḫa- and of isassara-

As I said above, isḫassara- 'lady' is a transparently Anatolian formation so any talk of its possible Minoan origins is off to left field. Nonetheless it's true that the *root* of this Hittite word, isḫa- 'lord', may very likely come from Hattic asaf 'lord, god' (= asapasaw) as per Jaan Puhvel in his Hittite dictionary. This particular non-IE etymology can have little to do with the source of Minoan Asasarama though and we must endeavor to keep these irrelevant side-facts separated in intelligent discussion on the matter.

On the other hand, these facts about the Hittite root suggest a stress accent on its second syllable. Thus Hattic asáf /əs'xaɸ/ would be lent to Hittite nominative isḫás /ɪs'xas/, then extended by the IE feminine suffix to isḫássaras, in turn used to form an epithet which in the vocative case becomes Isḫássara-Mi /ɪs'xassara-mɪ/ 'O My Lady'. If anything, we may best trace underlying Minoan Asásarama /ə'sasaramə/ from this foreign vocative. Searching for a Luwian equivalent to explain initial a- becomes unnecessary.

From this, we have the Minoan locative case form Asásaram-e and the qualifier Asásarama-na 'pertaining to Asasarama'. To respond to Bayndor's objections regarding the nature of the distinctive Aegean suffix -na, I maintain that the semantic distinction between a true genitive form and a qualifier is rather moot. We must note that Anatolian languages too had gone so far as to systematically replace their inherited Indo-European genitive case forms with adjectival formations. Finally, the same case and derivational endings I theorize for Minoan are amply attested in Etruscan with precisely the same usage, lending further weight to my interpretation of the term.

I'll speak more later on the reasoning behind a full translation of KN Za 10 I have cooking in my wok right now.


  1. I do not want to dispute your theory (after all, it is a plausible one, but not the only one possible). I would just want to give you a warning on the values I transliterated this inscription (KNZa10) with. There is a big ambiguity regarding the reading of the second syllabogram: It can equally be 'TA2', not just 'MU'. Linear A does have an example of a phrase U-TA2 (*utya) on HT103. We can try to compare this with the Eteocretan verbal form *utat that you very nicely translated as a form of a verb 'to share' (or sim.) This could mean, that if that word reads TA-NU-TA2-TI (*tan-utyati) instead, it could mean "this(acc.) [they] share", and U-TA2 = 'share' (as a noun)! Note that the stem of KA-NA also tend to change over to KA-NI-JA- in the verbal forms. Similarly, the word DU-WA-TO also has an inherent ambiguity regarding the last sign: I must confess, it has a quite unusual shape for a TO syllabogram. This may hint that it isn't a 'TO' at all, but an early example of 'JO' (Lin B *36) instead! And if we take DU-WA-JO, that becomes comparable to the Linear B word TU-WE-A = 'perfume' (it is well known that Mycaeneans perfumed wine and oil before offering it to their gods). I just told you about these ambiguities to help you before making a final translation. Actually, I am also thinking about correcting my transcriptions (at least on MU/TA2), especially if it gives a way better reading and meaning.

  2. Let's not lose our path by failing to abide by Occam's Razor. Pleading relativism (ie. "It's not the only theory possible! It could be anything!") whenever one's views are based on nothing more elaborate than desperate eyeballing (as with your comparisons with asḫarmis) is an overused tactic and is unproductive to even debate about. There are plenty of authors before us like Cyrus Gordon, Joseph Greenberg and Zacharia Mayani who've wasted our time this way. Sufficed to say, Minoan remains untranslated no doubt because too many people like this are idling with this disorganized "method".

    I see that you're throwing some more distracting sand in our eyes by suggesting that the very values we're working with may not be real now. If you really want to go that route, why not suggest that since existence can't be proven, Linear A can be anything we want it to be, hmm?

    So please, figure out what your position is and stick to it. I'm not running around an infinite solution space after your ever-changing positions, okay? ;o)

  3. Here is some update for you. I parsed through the entire Linear A corpus (the GORILA book) to supply you probabilities, not just "gut feelings" about the identities of signs on KNZa10. For the second sign (transliterated as 'MU' by John Younger), I found that the reading 'TA2' is not just possible, but the odds clearly favour the latter version. The internal architecture of that sign is unlike any instances of 'MU', but coincide with a variant of 'TA2'. The latter sign comes in two shapes: First, it can show a closed triangle or trapezoid, filled with dots, on a vertical stalk. But there is also a variant, where there aren't any dots present, the trapezoid isn't closed, and the stalk takes a sharp U-turn left once it enters the 'headpiece'. A good example is for the latter type is the 'TA2' found in ?-WO-KI-TA2 on HT122.

    As for the ambiguous sign within the penultimate legible word, the situation is definitely more nasty: This sign was apparently transliterated as 'TO' by Younger, because he couldn't find a better explanation. There are two clear peculiarities here: first, the stalk of this sign is 'wavy', and there isn't any cross-bar present as it is expected in 'TO'. For this purpose, I checked every single 'TO' sign on the photos of the original specimens. Their stalk is generally straight and they all have cross-bars. There is only a single occurrance of a wavy shaft: in the case of a libation vessel from Prassa. However, the latter has a clear reading, as that 'TO' is part of the word SE-TO-I-JA, a place-name found countless times in Linear B. Yet this libation vessel's inscription is very crudely carved (that explains the distortion of signs), and - on a closer look - the 'TO' sign does have a cross-bar! This implies that the sign in question on KNZa10 is either a super-divergent version of 'TO' (that has a rather low probability), or it presents a sign incredibly rarely seen in Linear A: 'JO'! So-far the only sign that can be plausibly read as 'JO' I have seen on KH11. There, a word (name) reads as A-TO-JO-NA-I (or A-TO-JO-TO-I, undecidable due to the damage). The middle sign (uncomfortably labelled as *349 by Bennett) has a straight headpiece, and its stalk is even more wavy than what is seen in that of KNZa10. Unfortunately, due to the low frequency of 'JO', the probability of reading something as a variant of a 'JO' sign, is also low. So we have little more choice than to work with 'gut feelings' in this case. Unless we can find a really good translation for one of the two possible versions, of course.

  4. Interesting. Thanks for the different perspective. This problem is definitely full of aggravating hidden variables.

    So you would now read it TA-NU-TA2-TI • YA-SA-SA-RA-MA-NA • DA-WA-SI • DU-WA-JO • I-YA? Boy, a photo would sure be handy right now to decide on the matter!!! <:o/ But I digress. Look at me, I'm like a broken record already.