2 Nov 2009

A modification of Indo-Aegean, plus some new grammatical ideas on Minoan

I like to explore new ideas and test them as always. One of my ever-evolving ideas is on the idea that Indo-European and Aegean are related to a common Proto-Indo-Aegean ancestor datable to 7000 BCE. Or so I've been thinking up to now but...

I decided to explore a radical new extrapolation that's got a grip on my mind recently. What would be the consequences to my theories if Proto-Indo-Aegean were dated to as much as a thousand years later in 6000 BCE? The first interesting thing about this fresh perspective is that 6000 BCE is just about the time before Proto-Semitic began to affect Mid IE (MIE) according to my currently defined chronology. Another interesting thing is that if we take for granted a more Balkans-positioned MIE vis-à-vis the later Ukraine-positioned PIE proper, then it begs the question: Where would this theoretical Proto-Aegean of mine be sitting at this time? The most obvious answer would be that it would lie somewhere to the west and/or south of the Balkans in the general area that it historically emerged (see graphic above). Yet my theory also positions Old IE (OIE) back in the northerly territory occupied by later Late IE such that the geographical path from OIE to MIE to PIE looks like a meandering vee that points towards the Aegean Sea (see graphic below). This isn't problematic since nothing says that languages have to spread progressively in only one direction over the course of time. However, this pattern, if taken as correct for the sake of argument, teases in me a further idea that Aegean would have been brought to Greece and/or Turkey by that very southerly movement that brought Mid IE into the same trading zone. It's as if to say that what I call "Old IE" circa 7000 BCE is to be revised as a still-evolving Indo-Aegean and the beginning of the Mid IE period should be called "Old IE" at 6000 BCE. It's as if the temporary spread of an early stage of PIE to the Balkans and the spread of a related Aegean branch perfectly coincide to warrant further pondering.

Given the general conceptual arguments in favour of this deviation from standard, I went towards examining all the morphological what-ifs with even more profound consequences. The unfortunate problem with Etruscan, Lemnian and Rhaetic (and probably too with Eteo-Cypriot and Eteo-Cretan) is that no personal endings appear to be attached to verbs in these languages despite the fact that many features like the 1ps and its oblique form (mi and mini), demonstratives and the declensional system (ie. the demonstrative accusative, s-genitive, animate and inanimate plural endings) all find direct connections to PIE. If Aegean is related to PIE then something has happened to these endings and they've disappeared at some unknown point in time motivated perhaps by reasons that are lost in the mists of time.

I refuse to believe the answers aren't recoverable and I don't particularly like mist. I've been poring over Minoan texts recently and while very hesitant at first, I've been rethinking on the published but nonetheless speculative view by some that -SI and -TI are the 3ps and 3pp endings respectively. This is an obviously PIE-inspired interpretation and given the lack of success in translating Minoan with PIE values, we have reason to be skeptical.


It's interesting to observe that if we stick by my values of the Libation Formula such that *una (U-NA) means 'libation' (cf. Etruscan un 'libation') with plural *unar (U-NA-RU), and *kan- in KA-NA-SI/KA-NA-TI is cognate with Etruscan cen- 'to bring', then not only do we have a perfectly sensible phrase "a libation was given"/"libations were given" that coincides with the fact that it's written on several Cretan libation tables, but if we take the variation KA-NA-TI in PK Za 11 to be correctly read and written on purpose by scribes to indicate a different inflection, then what we have here is a language with personal endings that apparently have not been completely lost! It would seem that -TI might indeed correlate with plural subjects while -SI would correlate with singular ones.

If we additionally corroborate this with CR (?) Zf 1 (an inscribed gold pin) where we find a perfectly Etruscoid sentence with the ubiquitous SOV word order and with intriguingly Indo-European-like verbal endings, A-MA-WA-SI KA-NI-JA-MI (*Amawasi kaniami 'I (ie. the pin itself) was brought for Amawa'[1]), then we have a very exciting verbal system that might help crack the language: 1ps *-mi (cf. PIE *-mi), 3ps *-si (cf. PIE *-ti), and 3pp *-ãti (cf. PIE *-énti).

The reasons for this strange hodgepodge grammar, neither fully Etruscan nor fully PIE by any sensible definition, would then relate back to the modified chronology that I suggest above. Speculation? You bet. But worth a look, I think.

[1] Ego-focussed dedicatory inscriptions such as these were plentiful in later Etruria and were also found in the Greek and Faliscan languages as well. Read for example Pallottino, The Etruscans (1955), p.253 (see link) who testifies to the Faliscan inscription eco quto ... enotenosio ... 'I (am) the pitcher of ... Enotenus ...'.


  1. This is a great post, I should have come across it earlier!

    Acccidentally, I pondered over the same question about the origins of PIE tree and the Aegean family years ago. IF (and this is a big if) the Aegean family is in fact a side-branch of a common PIE-Aegean tree, then one can contemplate over their common homeland (This is tempting to assume but we know too few Etruscan or Minoan phrases to answer this with 100% certainty).

    The PIE 'Urheimat' was variously sought for in southern Ucraine, west Turkestan, Caucasus, the Balkans and even Anatolia by different authors. Yet if the Aegean languages have arisen from the same trunk, than perhaps we do have a clue. What you wrote about a putative Balkans Urheimat for early PIE is pretty much possible, but there is only one logical problem: this is way too far for the proposed Semitic-PIE mutual influences. I do not think that trade relations were so intense between the (northern) Balkans and the Levantine coast.

    So I would suggest some alternative: Do you believe that an Anatolian Urheimat for early PIE is possible? (Because I do think so.) Then we could explain not only the cultural exchange between the (neighbouring) Semitic and PIE tribes, but also the distribution of the Aegean languages. These might have developed as a result of early isolation on the islands north and south from the Anatolian plateau. Later, the PIE speakers might have left Anatolia for the Balkans, whence they migrated into Ucraine. This migration has likely not affected the larger islands of the Aegean, where earlier forms of the language persisted.

    I also believe that the distribution of "Aegean-type words", that is parallel to the spread of Aegean languages, is a result of a secondary expansion: Once the Cretan and Cypriot civilizations have reached the point of urban revolution (thanks to Egyptian cultural influence), a spreading of the (previously isolated and relictual) Aegean languages happened, all across the Aegean Sea, onto the Greek and Anatolian coast. In this sense, the Etruscan-Lemnian group is not a real mainland Anatolian relic language, but instead a result of re-colonization from the islands.
    This process lasted only until the spread of the mainland, genuinely IE people (Greek, Hittite, etc.) have not conquered and replaced the Aegean folks altogether (some, like the Etruscans wandered off to distant lands, thereby surviving assimilation).

    What do you think of this line of thoughts? Is it complete nonsense or not?

  2. Bayndor: "The PIE 'Urheimat' was variously sought for in southern Ucraine, west Turkestan, Caucasus, the Balkans and even Anatolia by different authors."

    Perhaps in the 1950s. Nowdays, despite a few drifters, it's senseless to place PIE anywhere but the NW of the Black Sea. I'm repeatedly asked about the Anatolian Hypothesis and I repeatedly explain that it lacks linguistic, historical and archaeological facts to support it.

    In fact, it *defies* known facts. One sharp problem is that Proto-Indo-Iranian is well-proven to have been in contact with Proto-Finno-Ugric at around 2500 BCE. If one were correct that the PIE had left Anatolia (simultaneously wiping out all traces of itself in prehistory and in all languages supposedly surrounding it, like Semitic, Hattic and Hurrian) then Indo-Iranian has made a ridiculous loop-dee-doop around the Black and Caspian Seas. And what of Tocharian which is surely one of the two earliest branches to separate from the common PIE community? This theory would have us believe that it had veered straight to the east for no apparent reason, deep into Asia.

    Even from a Nostratic point of view, it's nonsense because PIE is likely to be closer to Uralic and Altaic than any other language group identified as 'Nostratic' and yet their respective protolanguages are securely placed to the north and east of the Black Sea. Try to draw a dialect map for PIE based on this wild view and one realizes how out of whack this idea really is.

    Therefore, the Anatolian Hypothesis isn't a hypothesis any more than Intelligent Design is. It has no facts and ignores others. It's a dead horse that people continue to beat.

    Bayndor: "I do not think that trade relations were so intense between the (northern) Balkans and the Levantine coast."

    When you say 'think', do you mean 'assume' perchance? What facts support your view? The contribution of Western Asia to the European Neolithic is clear by all archaeological data overall. There's also a possibility that some dialects of Proto-Semitic were found in Asia Minor at the time, alleviating your perceptions of 'far', although boats makes things less far than by land (nb. Perlès, The early Neolithic in Greece: the first farming communities in Europe (2005), p.62; also note what the author says on p.61).

    Bayndor: "In this sense, the Etruscan-Lemnian group is not a real mainland Anatolian relic language, but instead a result of re-colonization from the islands."

    I'm starting to reconsider things on this, actually. If Etruscan were continuously in contact with the Aegean islands through the 1st millenium BCE, then Lemnian could be an out-of-Italy dialect of Old Etruscan (as is usually claimed afterall).

  3. Oops, correction: The early Neolithic in Greece was published in 2001, not 2005. Mea culpa.

  4. Speaking of Perles's _The Early Neolithic in Greece_, it has often made me wonder whether a Semitic/Afroasiatic language of Asia Minor and/or the northern Levant might have hopped across to Greece with agricultural incomers, and then have spread north towards the regions west of the Black Sea, such that in effect a language associated with the Cucuteni-Trypillian culture might have been its descendent (keeping in mind the couple of millennia between the start of the Greek Neolithic and the earlier Cucuteni-Trypillian phases). With a Mallory/Anthony-style view of PIE, this would provide ample opportunity for PIE-speakers to interface with Semitic/Afroasiatic via such a language in such a location (at such a time). But this is all more or less wild and idle speculation on my part, uninformed by any real engagement with the evidence.