Many of the categories used in the discussion of the Indo-European verbal paradigm are functional rather than structural, a fact recognized and succinctly stated by Brugmann: ‘Bei der Gruppierung der Formen eines Verbalsystems ist die traditionelle Grammatik von der Bedeutung der Formen ausgegangen, nicht von formantischer Gleichartigkeit. Infolge davon hat sie zum Teil solches, was formal zusammengehört, voneinander getrennt und solches was formal verschieden ist, vereinigt.’ And later in the same paragraph: ‘Zunächst ist zu betonen, dass ein Bildungsunterschied zwischen den Formen des Präsenssystems und den Formen des Aorists ursprünglich nicht vorhanden war.’ Among his examples he points out the structural identity of the ‘imperfects’ Skt. ábhāt, Gk. with the ‘aorists’ Skt. ásthāt, Gk. , and the use of the stem *ĝene- to form an ‘imperfect’ in Skt. ájanat, but an ‘aorist’ in Gk. Of course, it is the coexistence or the absence, as the case may be, of a ‘present’ built from the same stem that determines whether such a form is to be accounted an ‘imperfect’ or an ‘aorist’. All this is in sharp contrast to the procedure of Semitists, who, in their treatment of finite verbs, employ categories that are structural throughout; e.g. in Arabic grammars and dictionaries the various ‘stems’ are merely listed in an arbitrary numerical order. Thus, once past the ‘First’ or Ground-stem, the structural relations of all the other stems (II-XV) are throughout predictable; it suffices for the lexicon merely to mention which of these derivative stems exist, without quoting the actual forms. One can imagine the mild chaos created in Arabic grammar if a strictly functional classification of verb-stems as ‘reflexive’, ‘causative’, etc. were to be introduced.

[Hittite evidence indicates that the relative kwi- kwo- is a development of the indefinite. In Hittite a subordinate relative clause can scarcely be distinguished from a coordinate indefinite clause. A typical sentence can mean either ‘some utensils are there, and these he picks up’ or ‘which utensils are there, these he picks up’. Six features that characterize the indefinite type are found to be present in an overwhelming majority of examples of the relative type, 390 of which have been examined.]

[Sanskrit has five nasal phonemes: velar, retroflex, dental, and labial nasals, and anusvāra. The palatal nasal is an allophone of the dental.]

[It is well known that Middle Indic has an indefinitely productive causative suffix derived from Skt. -āpaya-. But it is commonly assumed that the ‘causative’ meaning of such forms is often evanescent; that they are often used as synonyms of the underlying primary verbs (as is, undoubtedly, true of not a few Sanskrit ‘causatives’ in -aya-). This paper undertakes to refute that opinion, particularly for Buddhist Hybrid Sanskrit, which however is believed to be typical of all Middle Indic.]

Throughout its history the Greek language has been characterized by a remarkably steady trend toward reduction of the number of permitted final consonants, though at all periods the number has been greater in close juncture than in open. The purpose of this paper is to trace and illustrate this trend, by considering the development of the language in this respect first from Indo-European to Ancient Greek and then from Ancient Greek to Modern. Since an approach of this kind entails a comparison of juncture systems, I have found it necessary to include a descriptive statement of the operation of the juncture systems of Ancient Greek (specifically, colloquial Attic Greek of the fifth century B.C.) and Modern (standard colloquial) so far as final consonants are concerned.

[An IH laryngeal with a following semivowel became IE hy and hw, which in turn, under certain accentual conditions here described, became Gmc. k; e.g. NE quick : Lat. vīvus, Skt. jīvas.]

1. Phonetics

1.1. Phonemes:

Consonants: A. ptkcć čsš C. fh

B. bdg z ž v r 11‘ m n n’ j

Vowels: i e a o u r

Pitch-stress: ‘’

Junctures: close (no symbol), open (#:)

The division of consonant phonemes into groups A, B and C is on the basis 0 distribution (see 2.2).

During the century that scholars have devoted to the study of Akkadian their views on the question, seemingly so simple, as to what the proper plural form of the masculine noun is, have strangely changed.

[Malay tiga ‘3', even though it corresponds to forms in other Indonesian languages, is originally a borrowing of Middle Indic tiga ‘a triad'.]

The interplay of sound change and analogy may create patterns so typical as to make it possible to recover from them the process to which they owe their existence. Such internal reconstruction serves to supplement the comparative method. In this paper sound changes are classified with regard to their effect on structure: 1. non-phonemic change; 2. phonemic change without loss of contrast; 3. unconditioned merger with loss of contrast; 4. conditioned merger with loss of contrast; 5. secondary rearrangement induced by a primary loss of contrast; 6. borrowed contrast.