One may expect to find instances of phonologic (morpho-phonemic) alternation in almost any language. In English there are changes in consonants, vowels, and stress, insertions, and elisions; and many of these are conditioned only by the phonetic surroundings of the morphemes. An example is the possessive suffix, which is -əz (or - 𝚤z) after a sibilant, -z after a voiced non-sibilant, -s after a voiceless non-sibilant.

As early as 1921 Emil Forrer advanced the hypothesis that Hittite separated from the Pre-Indo-European parent stock before the time of the Primitive Indo-European language that has been partially reconstructed by the comparative method. Forrer did not undertake to prove his point, but I have tried to do so in a series of articles.

It is probably no more than a coincidence that in both OArmenian and O Irish a distinction of voice rather generally observed in other parts of the verbal paradigm is wanting in the imperfect indicative, so that in Armenian the active and the passive of normal thematic verbs have here a common inflection, as do the active and the deponent verbs in Irish.

[Etymologies are offered for: aduggere ‘to irritate, anger’; Sien, asciunare ‘to gather’ and dialectal cognates; attraccare ‘to draw a boat to shore’; bifolco ‘cow-herd’; coccolarsi ‘to squat; to enjoy oneself’; dunque ‘therefore’; elce ‘holm-oak’; incignare ‘to begin’; infolcarsi ‘to interfere, meddle’; ischio ‘oak’; lazzo ‘(stage) trick, gag’; mucca ‘milch-cow’; pozzanghera ‘mud-puddle’; uggio ‘disgust’; also for Span, mozo, muchacho, Port, moço ‘boy’.

These etymologies are divided into two groups: those involving contamination or the like, and simple etymologies.]

The Badagas of the Nilgiri Hills in South India are a community numbering some 40,000 and comprising seven sub-castes. They are known to be immigrants from Mysore state, having appeared on the Nilgiris some five or six centuries ago. Their language is certainly an offshoot of Kannaḍa, as is shown by its having in common with modern Kannaḍa among other features initial h instead of p of other Dravidian languages. A. N. Narasimhia, in The history of p in Kanarese, has demonstrated for Kannaḍa that h- appeared first in inscriptions in the language in the 10th century A.D. and by the 16th century had established itself in inscriptions in almost all words. In the 13th century h- < p- had already begun to disappear and has now entirely disappeared in the uneducated vernaculars; we have no information whether this is the case in all the local vernaculars. In the Badaga language this h- still remains, but in the absence of a dialect atlas for Kannaḍa this survival of h- is of no use in dating the advent of the Badagas in the Nilgiris. The Badaga language is said by speakers of it to have different dialects, but no work has been done except on that spoken in the neighborhood of Kateri. Enough work has been done by the writer on this one dialect of Badaga to permit a short treatment of phonology and especially of the vowels.