This paper is intended as a short linguistic contribution to the study of an important feature of Indo-European religious history: the sacrifice of the horse. There exists a copious and valuable bibliography on the subject, and I shall refrain from needless repetition of the materials and conclusions presented there. Only features of immediate importance for my comparison will be explicitly mentioned.

In grammars of Germanic the reflexes of the PIE resonants, /y w r l m n/, are presented either inadequately or unsystematically. Prokosch devotes but a few pages to them, in contrast with his thirty-odd pages on the reflexes of the obstruents. Streitberg deals with them somewhat more fully, at various places in his treatment of Germanic phonology, but his chief interest is in the assimilatory and dissimilatory changes in which they are involved. The relative neglect of the resonants is understandable, for until Edgerton described in detail the allophonic distribution of the PIE resonants, it seemed completely adequate to cite an example or two of the correspondences between PIE [y w r 1 m n] and PGmc. [y w r 1 m n], or PIE [i u] and PGmc. [i u], and then to state with examples that a [u] developed in the neighborhood of PIE syllabic . If however we accept Sievers' law in Edgerton's formulation, we cannot pass over the Gmc. developments of the PIE resonants so lightly. Since we now expect grammars to present full and systematic statements of the general lines of development of a language, future treatments of PGmc. will have to describe in some detail the developments of the resonants into Germanic. In this paper the general outlines will be sketched, especially those environments which have been inadequately treated in the past.

In historical English grammar it is regularly assumed that Gmc. initial [f s þ] became voiced before voiced sounds in southern England at some time in the Old or early Middle English period. For the present purpose, Wright's description of the change is sufficiently representative: ‘The initial voiceless spirants f, s, þ became the voiced spirants v, z, ð in late OE. or early ME. in Kentish and the southern, especially the south-western dialects, as vader, vat, vlesch, vrend; zaule zǫule, zinne zenne zünne “sin”, ðat ðet, ðing. The modern dialects show that this voicing of the initial voiceless spirants must have taken place at an early period, because it is almost exclusively confined to native words, hence the change must have taken place before the great influx of Anglo-Norman words into these dialects. The use of the initial voiced for the voiceless spirants is now obsolete in Ken., Sur., Sus., and obsolescent in s. Pem., Hamp., and the I.W., but it is still in general use in east Hrf., parts of Glo., west Brks., Wil., Som., and Dev.‘

The objections of Kuhn and Quirk to various reinterpretations of Old English digraph spellings, including our own, have made it clear that extensive discussion of single structure points in the overall frame of the Old English phonological system is a wasteful procedure. The system and the minimal oppositions which make up the system are coexistent and difficult to discuss intelligibly without going through the circular but internally consistent process of describing the one in terms of the other. Taking our departure from a pattern whose outlines were implicit but largely unstated, we attempted to deal with one point of the structure in detail. This kind of presentation seems to have been a mistake, and to have led to some of Kuhn and Quirk's misunderstanding of OP 4. The correction of our error of judgment about presentation must, however, await further research and publication, undertaken not to prove a thesis but to arrive at the most complete, consistent, and economical interpretation of the total evidence. The present article, therefore, only points out some matters of fact and clarifies the basic disagreements.

In determining Polish inflectional categories it is not enough to consider the differences in shape between paradigmatic forms. On the one hand, some phonemically different endings have the same function; e.g. the endings of list-u letter', kot-a ‘cat’, ćeśl-i ‘carpenter’, seńź-ego ‘judge’ all have the function of the Gen. Sg. On the other hand, some phonemically similar endings have different functions; e.g. dom-u ‘house’ functions as both Gen. and Loc. Sg., miš-i ‘mouse’ functions as Gen., Dat., and Loc. Sg. and as Nom., Gen., and Acc. Pl. It is obvious therefore that the factors determining the function of an inflectional ending must be sought outside the form itself.

This paper discusses methods for presenting syntactic information in the form of a calculus, and for measuring its goodness of fit to a language.

Students of language have in general placed primary emphasis upon the generating mechanisms of speech. For example, it is common to classify the sounds of speech in physiological terms; phonetic tables and charts usually show organic position or mechanism in one dimension and method of production in the other. There has also been considerable interest in the acoustical effects or results of speech production. Von Kempelen was among the first to promote a serious interest in this subject, with a volume published in 1791. The literature on the subject is far more distinguished and extensive than most linguists realize; but it cannot compare with the much more voluminous literature on physiological phonetics and voice production. Within the past ten years, great impetus has been given to the study of acoustic phonetics by developments in electro-acoustical instrumentation and by the new, important, and challenging problems which have come upon the field of communication engineering.