Indo-European

[James A. Landau]

In a message dated 02/11/2002 12:16:59 AM Eastern Standard Time,
write at SCN.ORG writes:


>  ...books talk about the ancestor of English (and all western languages?),
>  Indo-European.  It seems (Am I right here?) that the idea of the
>  Indo-European language is a hypothesis.  Since Indo-European was a
>  prehistoric language, there is no written record of this language, and
>  certainly no recording of this language.  How do linguists know anything
>  about the Indo-European language when there is no record of the language?

Look at the Romance languages (Spanish, Portuguese, French, Italian,
Romanian).  You will see they have considerable similarities.  Why?  The
obvious conjecture is that started out long ago as dialects of a single
language, and kept on diverging.  Let's call that original language
"Proto-Romance."

We could reconstruct the vocabulary, phonetics, and grammar of Proto-Romance
by comparing the Romance languages.  (If you have some multi-lingual
students, this might make a good class project.)  Then compare this
reconstructed language to Latin, which is known to be the ancestor of the
Romance languages.  How good is your reconstruction?

Probably pretty good.  There will be some mistakes.  You will probably give
Proto-Romance a definite article, miss the -que suffix for "and", and most
likely come up with only two or three noun cases rather than five.  Still,
you will be close.

Now do the same process with German, Latin, a Celtic language, old Church
Slavonic, Greek, Lithuanian or Lettish, and Sanskrit.  The similarities are
not as blatant as within the Romance languages, but they are easily observed.
 Once again you can reconstruct the vocabulary, phonetics, and grammar of a
postulated comman ancestral language, known as Proto-Indo-European.

We know Proto-Romance exists (it is Latin).  Should not a language
recognizably similar to your reconstructed Proto-Indo-European also have
existed?  And have been the common ancestor of the languages cited above?

Yes, this is a hypothesis, and no direct evidence exists.  However, it is a
convincing hypothesis, as it explains the known data and it is not easy to
come up with a competing hypothesis that works anywhere near as well.

Does this answer your questions?

Some sidebars:  There are five languages spoken in Europe today that are NOT
Indo-European.  (Hence there are non-IE "Western" languages.)  These are
Turkish, Hungarian, Finnish, Estonian, and Basque.  The first two are the
result of invasions within historical times.  Finnish and Estonian (very
similar languages---an Estonian co-worker told me he can understand Finnish
but not speak it) are closely related to languages of the Ural region of
Russia---either the Finns and Estonians moved west from the Urals or they
were squeezed into their present locations by Slavs and Scandinavians moving
into their territories.  Basque is a mystery, but considering that it is
restricted to a single mountainous area,  one speculates that it is the
survivor of an otherwise unknown family of languages spoken in southern
Europe before the Indo-Europeans staked out most of the continent.

Dates:  the splitting off of the Romance languages was a long process, but it
was pretty much complete when the Roman Empire fell in 481 AD.  Call it 1600
years ago.  How long ago was it that Proto-Indo-European split up?  A (in my
opinion somewhat GIGO) statistical technique called "glottochronology" gives
a date of not later than 5,000 years ago, so I would say "5 to 10 thousand
years ago" is probably correct.

Are there other language families similar to Indo-European?  Yes, indeed,
e.g. Hamito-Semitic, also called Afro-Asiatic, which includes Hebrew, Arabic,
Coptic, and Amharic.  There are also some that are the subject of
considerable scholarly debate (so you see some hypotheses about language
families are surer than others).  One long-standing debate is whether Turkish
and Mongol fall into a "Ural-Altaic" family with the Finno-Ugrian languages
such as Finnish and Hungarian.

Is Proto-Indo-European itself a descendant of some older language which split
up earlier?  According to the "Nostratic Hypothesis", yes, Indo-European,
Hamito-Semitic, and several others split off from a common language around 10
to 15 thousand years ago.  However the similarities on which the Nostratic
Hypothesis are based are so few and so debatable that that the hypothesis
remains speculation rather than something generally accepted.

     - Jim Landau

Dialectic = making a fire by rubbing two ideas together.