13 Mar 2009

A precise chronology of Proto-Indo-European

This diagram from page 100 of The horse, the wheel, and language (2007) by David W. Anthony gives a detailed solution to the exact chronology of Proto-Indo-European (PIE) and its later daughter languages. I quite like this timeline which places Proto-Indo-European within a range of 4500 BCE and 2500 BCE. The author suggests that Anatolian split away from PIE specifically around 4200 BCE.

For kicks, I might be tempted to overlap it with my own chronology of Pre-IE to create a detailed overview of Indo-European diachrony. I'm perfectly fine with cutting off a couple of centuries from my "Late IE" period like this.

Of course, as we all learn more details concerning this fascinating subject, our picture will be prone to modifications but for what it's worth this represents my understanding of PIE's timeline at present.

9 comments:

  1. Google did something to this link, and the diagram isn't available now (14 March).

    ReplyDelete
  2. Nope, I tested it and it works but you need to scroll up from the highlighted text to see the diagram. Sorry about that.

    ReplyDelete
  3. Actually, if ever any of you readers have problems with my links for whatever reason, you can always try searching for the book I'm referencing for yourself and once you find the book, Google Books allows you to do a search within some portions of the book. I love Google Books although I suppose copyright-loving authors might not love it, but alas, changing times, changing times.

    ReplyDelete
  4. Ah yes, that book is wonderful! I'm just re-reading it now.
    I've just passed the part where he describes how the root *Hw{e/o}l-, from whence *Hwlh2neh2 (wool) comes, likely meant "felt". I wonder if it's related to the word for sheep (Howi-), via an -l- extension and a metathesis of of the vowel and the *w. Just a theory.
    And to see a diagram with your Pre-IE chronology added would be most cool.

    ReplyDelete
  5. I don't think that *Hwlh₂neh₂ could be directly related to *h₂ewi-. If before Syncope we had a derived stem like **haway-ála-, it would trigger a-Epenthesis, becoming **ahwyál- and then PIE **oh₂wyól-. However, I could accept the possibility that both *h₂ewi- and *Hwlh₂neh₂ are built on a root like **h₂eu- since an MIE stem **haw-ála- would yield *h₂wol-. It's hard to be certain but it's an interesting conjecture.

    ReplyDelete
  6. Whoops, I goofed up on the laryngeal for "sheep". That should read *h₃ewi-, not *h₂ewi-, and the conjectured Mid IE forms should then contain *hʷ instead. Mea culpa. I've been drinking tea instead of coffee lately.

    ReplyDelete
  7. I goofed up too. In the note I'd scribbled in the margin I'd written *H{e/o}w. Typo on my part. I seem to making a lot of those lately. C'mon, brain-box, work properly!
    Actually, the reconstructions I've read seem to vary between an *H2 and an *H3 onset.
    That kinda gets me to thinking - when dealing with reconstructions like *How-i-s/Hw-l-H-neh2, with *H{e/o} (or *-{e/o}H), how can one tell if a reflex of *o (or *o: in the latter case) comes from *H3e- (or *-eH3), or from an o-grade form (with lengthening) with any of the 3 laryngeals? I have a feeling I should know this answer, but I'm drawing a total blank.

    ReplyDelete
  8. Lubotsky states on page 29 of The system of nominal accentuation in Sanskrit and proto-Indo-European that we know it's *h₃e because of the Armenian reflex hoviw 'shepherd' which appears to preserve the initial laryngeal.

    ReplyDelete
  9. Hey, I read something in the Metro paper a few weeks ago about the domestication of the horse. Looks like the Botai culture had it at least 3500 BCE!
    http://www.sciencemag.org/cgi/content/abstract/323/5919/1332

    ReplyDelete