The basic tenet of Indo-Europeanists with regard to personal proper names has been, stated briefly, that the typical Indo-European name is a dithematic compound, as exemplified most eminently by many Indic, Greek, and Germanic names. Any name that is not compounded in a like manner, it is asserted, is either actually a shortened hypocoristic form of an original dithematic name, or is some sort of nickname or hybrid—at any rate not a genuine original. If indeed the theory went only so far as to state that the predominant feature of a great number of Indo-European names, particularly in Indic, Greek, and Germanic, is a form of composition resulting largely in the dithematic onomastic type, no serious objection could be raised. But the creators and propagators of this theory were obviously determined to arrive at some more comprehensively valid rule. Of course, this rule was bound to include exceptions; but not even the existence of an abnormally great number of exceptions seemed to arouse any grave doubts as to the universal applicability of the rule, nor was it deemed necessary on their account to investigate the causes of the divergencies.

The indefinite pronoun, ὁ ἡ τὸ δϵῖνα ‘so-and-so’, like other Greek pronominal stems (cf. αὐτvός, ἐκϵῖνος) is difficult to analyze, and even more difficult to etymologize. It has been much discussed in the literature and a number of ingenious suggestions have been made; nonetheless, as discerning a judge as Gustav Meyer noted long ago that the word was ‘in Ursprung and Bildung dunkel’ and concluded his analysis with a non-committal ‘non liquet’.

The following paragraphs attempt to give a partial phonemic analysis of modern standard German. §1 establishes the segmental phonemes that are needed to describe utterances of only one syllable. §2 discusses the role that stress plays in German, describes the effect of stress on the segmental phonemes already established, and adds one new segmental phoneme (ž) which does not occur in monosyllabic utterances. §3 describes the inadequacies of this phonemic analysis, and shows how they may be removed either by giving phonemic status to a number of sounds which are not phonemic in monosyllabic utterances, or by assuming a segmental phoneme of open juncture. §4 gives the reasons why the present writer considers it preferable to assume a phoneme of open juncture. §5 discusses a possible second type of juncture, for which no explanation is offered.

[The theory is proposed that PMP D became Tg. d in initial and postconsonantal positions; that l appears instead of d in initial position as the result of analogy, and in postconsonantal position as the result of some secondary change.]

The most extensive body of evidence that is available on the Annamese lexicon is G. Cordier's Dictionnaire annamite-français à l'usage des élèves des écoles et des annamitisants. Elsewhere I have described the contents of this dictionary as valuable and very full, but only to be used with exactitude if a native speaker is available, since all periods, local dialects, and styles are included without any indication of the ascription of any particular word. During work on the language with two native speakers, a basic vocabulary of just over 2000 words was completely analyzed and has been used for various chapters of grammatical analysis, which it is hoped will be published in the near future.

[Affixes and groups of bases, as well as alternants of these, are numbered. The method for showing the correlation between them is stated (0); all the affixes to be treated are given in a list (1); after an exhaustive listing of bases (2), the distribution of these two sets of morphemes and of their alternants is stated in formulae (3–8). A few major distributional classes emerge (9).]