Fraggle, 100 in Sanskrit is shatam or shata, not satem.
Yes, I've been corrected on that before, but I haven't managed to go back and edit all my old posts. The Eastern Indo-European languages were formerly known as the
Satem Group because
satem is the
Avestan word for hundred, not Sanskrit.
The Western Indo-European languages were called the
Kentum (or
Centum) Group because
centum is the Latin word for hundred. This illustrates what, in the 19th century, was regarded as the fundamental distinction between the two branches of the language family. The original Proto-Indo-European word for "hundred" was
kmtom. When the Proto-Indo-Europeans (supposedly) split into two tribes, the western tribe retained the K, whereas in the language of the Eastern tribe that K was
palatalized into S. The two prominent sub-branches of Eastern Indo-European are Indo-Iranian and Balto-Slavic. The S of
satem in Avestan (one of the earliest Indo-Iranian languages, in which the Zoroastrian scriptures were written) was used to identify that group. It could have been the S of Russian
sto, but in those days the Indo-Iranian languages were studied more intensively than Balto-Slavic because of their much longer written history, and their people's prominent role in the development of civilization.
This paradigm has subsequently broken down in many ways. For one thing,
palatalization is a very strong force in phonetic evolution and it has continued to "distort" the original words for "hundred." The identifying K in the word for "hundred" in the Western languages has been palatalized in myriad ways. The C in Latin
centum is still there in the written word, but the sound has palatalized and also changed in other ways. It's S in French
cent, it's TH in Castilian Spanish
ciento, and it's CH in Italian
cento. It's H in English "hundred" and German
hundert. The Greeks have managed to keep the K alive in
hekaton and the Irish in
céad. If we were to talk about the "Centum Languages" today, virtually everyone but a linguist would read it as "sentum" and not see any difference from the already-palatalized consonant in
satem. Furthermore, as rcswc points out, even the S in
satem has been palatalized in the Indic languages, and has become SH.
For another, the K/S distinction has not proven to be a definitive identifier. The extinct Tocharian language maintained the K, but it is now regarded as a separate branch of Indo-European, not closely related to the Western/Kentum branch (Germanic, Celtic and Italic). Even Greek, which is more closely related to them than Tocharian, is now considered by some linguists to have split off from the Western branch early in the migration from the Pontic Steppe. Armenian was grouped with the Eastern/Satem languages, but today it is also regarded by some as a distant relative. As for the Albanians, from what I have read nobody really knows how they arrived in Europe, but their language is also not closely related to either the Eastern/Satem or Western/Kentum groups.
