Isoglossa Kentum - Satem

The terms come from the words meaning the numeral “hundred” in the representative languages ​​of each group ( Latin centum and Avestan satəm ).

The languages ​​of the Satem group include Indo-Iranian (Indo-Iranian and Nuristan proper), Baltic-Slavic ( Baltic and Slavic ), Albanian , Armenian and, possibly, a number of undocumented dead languages, such as Thracian and Dacian . In this group, the velar and labio-velar sounds of the Proto-Indo-European language developed into the velar, and the chamber-laced sounds hissing and whistling. Although the Albanian language is recognized as the Sateme language, it can be assumed that simple velar and labiovell sounds were not completely combined in the Praalban language.

The Kentum group is often considered to coincide with the non-satem, that is, including all other dialects. More specifically, according to K. Brugmann ’s definition of “languages ​​with labialization,” the Kentum group includes Italian , Celtic , Germanic , Greek, and possibly small and unknown groups of dead languages ​​(such as Venetian and ancient Macedonian and, presumably, Illyrian languages ) In this group, the velar and palatodelar sounds of the Proto-Indo-European language were combined into velar. In Tocharic languages, all rows of sounds were also united in one velar series: the relative chronology of this change is unknown, which is typical for the languages ​​of the Kentum group.

The Anatolian languages apparently did not undergo the changes characteristic of the Satha or Kentum . The velar series remained unchanged in the Luvian language , while the Hittite may have undergone secondary Kentumization, but exact data are not available.

Isoglossa kentum-satem explains the evolution of three rows of dorsal consonants reconstructed for PIE , * kʷ, * gʷ, * gʷʰ ( labio- velar), * k, * g, * gʰ (velar), and * ḱ, * ǵ, * ǵʰ ; (chamber) in child languages. The division into groups of centum and satem only gives a general idea of ​​the parent language, which had a complete set of dorsal sounds. Later changes in individual groups of Indo-European languages: similar to the palatalization of Latin k into s described above in some Romance languages or the merger of * kʷ with * k in Celtic languages did not have an effect of isolation.

August Schleicher, in his Compendium , published in 1871, believed that there was a single velar series, k, g, gh . Karl Brugmann in 1886, in Grundriss (can be translated as Short Course, Essay, or Fundamentals) recognized only two rows, designating them q, g, gh “velar explosive” against k̑, g̑ and g̑h “ palatalized explosive. " Brugmann terms: “languages ​​with labialization” or “u̯-languages” for Kentum languages ​​and “languages ​​without labialization” for satem. In his opinion,

For words and groups of words that do not appear in any language with a labeled velar sound [pure velar], it is currently impossible to determine the presence of u̯-collapse ever in the past.

In the 1897 edition of this work , Brugmann changed his mind by adopting the Kentum / Satem terminology introduced by Bradke in 1890 . Accordingly, he designated the labio-velar ones as q ^u̯ , q ^u̯ h, g ^u̯ , g ^u̯ h (also introducing mute aspirates).

Currently, the presence of three dorsal rows in the parent language is not accepted by all scientists. The reconstructed “middle” series may be an artifact of borrowing between early child languages ​​in the process of satelliteization. For example, Oswald Semereni (in his Introduction), recognizing the use of the separation * kʷ , * k , * разделения as a symbolization of sound correspondences, argued that support for the three phonologically separated rows in the PIE is unsatisfactory and preferred the double notation * kʷ , * k . Other scholars who believe that PIE had two dorsal rows are Kurilovich (1935), Meye (1937), Winfred Lehman (1952), and Woodhouse (1998).

The languages ​​of the “satem” group show a characteristic change in the consonant Proto-Indo-European chamber sounds (* ḱ, * ǵ, * *) into affricates and fricative consonants pronounced in the front of the mouth. For example, * ḱ became in Sanskrit ś [ʃ] , in Avestan , Russian and Armenian languages ​​- s , in Lithuanian š [ʃ] and in Albanian th [ θ ]. At the same time, the velar ( * k, * g, * gʰ ) and labio-velar ( * kʷ, * gʷ, * gʷʰ ), characteristic of the proto-language , merged in these languages ​​and later lost their rounding of the lips.

It is easy to illustrate the “shift” with the numeral “100”, which sounded in the pre-Indo-European language as * ḱm̥tóm , which became, for example, satəm in Avestan (which gave the name to the whole group), ( śata m ) in Sanskrit , sad in Persian , šimtas in Lithuanian , one hundred in Russian , etc., as opposed to the Latin centum (pronounced [kentum]), English hund (red) - (c / h / from the early * k, see First Germanic consonant movement ), other Greek- -κατόν , Welsh cant , etc. The Albanian word qind is a borrowing from the Latin centum .

In the languages ​​of the “Kentum” group, the palatine consonants merged with the usual velar consonants (* k , * g , * gʰ ). Most of the Kentum languages ​​kept the Proto-Indo-European labio-velar (* kʷ , * gʷ , * gʷʰ ) or their historical derivatives different from simple velar; for example, the pre-Indo-European * k : * kʷ → Latin c / k / : qu / kʷ / , Greek κ / k / : π / p / (or τ / t / before the first vowel), Gothic / h / : / hʷ / etc.

The name “ centum ” comes from the Latin word centum “hundred” (from the pre-Indo-European * ḱm̥tóm ), illustrating the combination of * k and * ḱ . You can compare with the Sanskrit śata- or Russian hundred , in which * ḱ changed into a fricative sound.

Confirmation of the existence of labio-velar sounds / kʷ / , in contrast to simple double / kW / , is found in Greek ( q- series linear letters B ), Italian (Latin qu ), German ( Gothic hwair ƕ and qairþra q ) and Celtic ( ogamic letter ceirt Q ). Thus, according to the adopted reconstruction of PIE, the labio-viability of this series may also be new for the Kentum group, causally related to the forward movement of chamber-sounds. The main argument in favor of this approach is provided by the Anatolian languages, although their phonetics is unknown in detail. The Hittites (and Luvians) did not use the symbols of the q- series of existing cuneiform writing (which transmitted a mute uvular sound in Akkadian ), but transmitted the corresponding labio-velor sounds of PIE as ku . Opinions on whether this was transmitted by one sound of the Anatolian languages ​​or the group / k + w / are shared. The similarity of the three dorsal series is also disputed for typological reasons, but this argument does not matter much, since there are languages ​​with a similar three-row system, for example, the Yazgulyam language (one of the Iranian languages , but its dorsal system is not comparable with the PIE phonology). It is true that such languages ​​are rare, and one of the three series is usually not used. It can be considered that none of the existing Indo-European languages ​​has completely preserved the three-row system, even if it was in the parent language.

Recently, there have been reports that the Bangani language in India has signs of a Kentum language, but they have been recognized as false.

In the 19th century, it was believed that the isogloss kentum-satem shared in the past various dialects of the Indo-European language. However, already Karl Brugman and partially Johannes Schmidt believed that the division of the centum / satem was a territorial feature .

Incomplete satemization in the Baltic and, to a lesser extent, Slavic languages, was accepted as evidence of the penetration of the satemic change or as a result of borrowing during the early contacts of the Proto-Balts and Proto-Germans. Examples of traces of labial elements of labio-velar sounds in the Balto-Slavic languages ​​include Lithuanian ungurys “eel" < * angʷi- , Lithuanian dygus “acute” < * dʰeigʷ- . There are also examples of incomplete satemization in Indo-Iranian languages , such as the Sanskrit guru “heavy” ← * gʷer- ; kulam "herd" ← * kʷel- ; also kuru “do” ← * kʷer- , but this word appears only in the texts after the writing of the Rigveda .

The systemic group of languages ​​also completely coincides with the group of languages ​​in which the RUKI law is fulfilled ^[1] , but the exact conditions for the flow of this law differ in groups and are also superimposed on older isoglosses: for example, * i <* participate in Indo-Iranian languages ə and * r <* l ^[2] .

• Solta, GR, Palatalisierung und Labialisierung , IF 70 (1965), 276-315.