According to the glottalic theory, Indo-European had ejective stops instead of voiced aspirated ones.

The traditional reconstruction of Indo-European includes the following stops:

b is bracketed because its very rare, and its reconstruction is thus uncertain.

Now, there are several problems with this reconstruction. From a typological point of view, no language is known which has voiced aspirates unless it also has voiceless aspirates. Furthermore, if a voiced stop is missing from a phoneme inventory, it would normally be /g/ that is missing, not /b/; on the other hand, if a voiceless stop is missing, the labial (/p/) would be the most likely candidate. Finally, Proto-Indo-European did not permit a root to begin and end with a voiced stop, i.e., there are no such roots as *deg or *ged in the reconstruction above; this is typologically very odd again.

Thomas Gamkrelidze and Vyacheslav V. Ivanov found a possible solution to the problem but postulating that the voiced unaspirated series was actually ejective (or glottalized), while the voiced aspirated series was just plain voiced. That is, the system looked as follows:

This solves the problems mentioned above. First of all, such a system is very common among the world's languages. Secondly, the absence of /p'/ is again typologically very common. Finally, prohibiting two ejectives in one root (no *t'ek' or *k'et') is a common constraint in such languages.

Opponents of the glottalic theory criticize that ejective consonants are not found in any of the Indoeuropean languages (whereas voiced aspirates are found in the Indian languages).

Sources

• Anthony Fox : Linguistic Reconstruction, Oxford, 1995.