The UPenn IE Tree

JoatSimeon at JoatSimeon at
Fri Aug 20 04:08:13 UTC 1999

>X99Lynx at writes:

>Does this mean that PIE co-exsts with Anatolian?

-- it's supposed to mean that Anatolian branched off before PIE started its
general breakup.  In the sense that proto-Anatolian became quite different
from PIE while PIE's internal divisions were still relatively slight.  This
can't be definitely proven but seems highly likely.

"PIE" means "the last stage of the language when all the IE dialects were
still mutually intelligible".  The language, like all languages, was always
undergoing change; it ceased to exist when its component dialects stopped
sharing enough innovations, just as Latin ceased to exist as a living
language sometime between 400 CE and 800 CE or so as its widespread dialects
slowly ceased to be mutually comprehensible.

(By way of contrast, Greek remained one language, changing over time -- not
the same language as its ancestor, though.  Rather as if Italian were the
only descendant of Latin.)

"PIE" means something rather like "late proto-Romance".  The big difference
is, of course, that Latin continued to be known in its written form(s) as a
learned secondary language while PIE vanished into the black hole of entropy
and can only be tentatively reconstructed via the comparative method.

>Then where along that left side diagonal does PIE cease to exist?

-- when there's no longer only one mutually intelligible IE language plus
Anatolian, probably.  Since that was way back in the prehistoric period, with
no written records, we can only say that it was sometime before about 2000
BCE and after about 4000 BCE, by triangulation.

Of course, one could argue that the Balto-Slavic protospeech of 2000 BCE
would probably have been more or less comprehensible to a PIE speaker of 3000
BCE, just to complicate things.

>a sampling artifact favoring Sanskrit, Germanic and Lithuanian/Slavic - the
favored sources in many *PIE reconstructions?

-- Sanskrit is 'favored' because it was recorded in a fixed form very early,
mid-2nd-millenium BCE or so.

Balto-Slavic, and particularly Baltic, are 'favored' because they're very

More information about the Indo-european mailing list