jūs 'broth, soup; juice, liquid', Skt. yūşa- 'broth', etc. Cf. author, Post-Cons. w in IE 16.

7. Lith. pietūs 'Mittagessen, Mittagszeit; Süden', pietwỹs 'Südwind', from *pēitu- 'swelling, gushing', applied to the south wind as moisturebringing, melting, and also to the preparation of broth for food. The explanation of this word as meaning 'meal, noon meal, midday, south' is improbable, just as it is equally improbable in No. 6. Compare Lat. pituita 'slime, clammy moisture; a gummy moisture that exudes from trees', Gr. TiTVs 'pine tree', base *pei- 'swell, gush' in Gr. widów 'gush forth', rida 'spring, fountain', i'w 'fat', Tiaivw 'fatten; increase; make wanton', Lith. pieva 'Wiese', Skt. pīvan- 'swelling, fat', páyate 'swell', etc.

OSCAN DEKETASIO

ROLAND G. KENT

UNIVERSITY OF PENNSYLVANIA

With the main points made by Professor Whatmough in LANGUAGE 3.105-9, as to the etymology and meaning of the Oscan word deketāsio-, a title of magistrates at Nola and at Abella, I am in hearty agreement. In two minor details only I should like to supplement or modify what

he says.

It

Professor Whatmough associates the word with the giving of tithes, and therefore explains it as a derivative of Italic *dekento- 'tenth,' with the suffix as in Umbrian sestentasiaru, a name or epithet of a festival or gathering, which goes back ultimately to the numeral seen in Latin sex. But there is no warrant in Italic for the -to- suffix in 'tenth,' all the recorded forms being referable to *dekemo-; in fact, no Italic numeral from 'seventh' to 'tenth' inclusive has the t-suffix. would seem more likely, then, that deketāsio-, or, to give it its earlier form, with the nasal that disappears in Oscan before t, *dekentāsio-, was built up from *dekemā 'tithe' by the influence of some other word of official or religious significance, which ends in -tasio-. As such a word I propose the equivalent of Latin argentārius, suitable both in form and in meaning to produce a *deken-tāsio-.

The second point is the g instead of k in degetasio-, as the word is written in the inscriptions found at Nola. There seems to me no likelihood of contamination with the Oscan equivalent of vi-ginti or with that of digitus, suggested respectively by Buck and by Conway. But the equivalent of argentārius or of argentum might easily suggest the g. It so happens that while *aragetāsio- is not found in Oscan, aragetud 'argento' does occur, twice, and precisely in the inscriptions of Nola which record the activities of the officials bearing the qualifying title *degetāsio-.

Thus the early Oscan *dekentāsio- was, in my opinion, built up from *dekemā "tithe' with the suffix of (the equivalent of) Latin argentārius. It is found at Abella as a magisterial epithet in the form deketāsio-, but at Nola with a still further influence of (the equivalent of) argentārius or of argentum, as degetāsio-.

THE LANGUAGE OF THE CUENTOS POPULARES

ESPAÑOLES

AURELIO M. ESPINOSA

STANFORD UNIVERSITY

In the following notes I shall study the more important linguistic phenomena that are to be found in my Cuentos populares españoles, the most abundant and most important collection of Spanish folk-tales that has ever been collected and published. The three volumes that contain the folk-tales offer us 280 dialectic versions of some 200 different tales representative of 24 of the 49 provinces of Spain and collected from 88 different towns, villages, and localities from the following regions of north, central, and southern Spain: La Montaña, Asturias, León, Castilla la Vieja, Castilla la Nueva, Andalucía, La Mancha, and Estremadura. About one third of the materials are from Old Castile. All the folk-tales, with the exception of those from Asturias, were collected and taken down (frequently in phonetic script) by myself. The Asturian folk-tales (eleven in number) are the only materials in the entire collection that may be called non-Castilian and for that reason they will not enter into our present linguistic studies. The linguistic materials that are to be the object of our study are, therefore, fundamentally and essentially Castilian materials from Old and New Castile and from the southern Spanish regions that have been castilianized since the XIIth century when these regions began to be recaptured from the Arabs, and they do not, in general, represent any of the special

1 Cuentos populares españoles, recogidos de la tradición oral de España y publicados con una introducción y notas comparativas por Aurelio M. Espinosa, Stanford University Publications. Tres tomos, 1923, 1924, 1926. Vol. IV, Notas comparativas, now in press. See Boletín de la Biblioteca Menéndez y Pelayo, 5. 1-25 (1923); F. Krüger, Archiv für das Studium der neueren Sprachen und Literaturen 150, 267-8 (1926).

2 Although the distinctive and fundamental traits of Castilian were originally limited to a very small territory north of Burgos (the old Cantabria and neighboring territory) the reconquest of Spain carried the originally unimportant dialect to most of northern, central, and southern Spain, thus breaking up the old linguistic uniformity of Spain and establishing Castilian as the language of all the conquered territories. See Menéndez Pidal, Orígenes del español §§ 100–06.

linguistic characteristics of Asturian, Galician, Leonese, Aragonese, Valencian, Portuguese or Catalonian. Andalusian Spanish shows, to be sure, certain linguistic traits that appear to separate it from the Castilian of the northern and central regions, but many of these traits are after all developments of phenomena that had their beginnings in Castile.

In our study of the linguistic phenomena of my Cuentos I shall enter into problems of modern pronunciation only in certain special cases. The general character of Castilian pronunciation is well known thanks to the epoch-making publication of Tomás Navarro Tomás and to this publication we shall refer constantly in all matters of modern phonetics. I cannot enter, of course, into a detailed discussion of Andalusian phonetics, but certain outstanding linguistic phenomena that are characteristically Andalusian, such as the frequency of l + cons. >r+cons., the fall of intervocalic r and other consonants, the disappearance of most final consonants except in certain cases before a vowel, the absence of the Castilian z and ll sounds, the various special developments of s, etc., will be treated at some length. I shall also try to show that many of the linguistic phenomena commonly believed to be Andalusian are rather widespread in Spain and that even such developments as -ada>-á, -ede>-é, -ado>-ao, -ido>-ío, b (v) + vowel>g+vowel and the apocopated verbal forms quié, quién, tié, tién, puson, quison, etc., are rather general in all parts of Spain. The linguistic materials will be studied under three general divisions: phonology and lexicology; special phonetic phenomena; syntax. The three volumes of the Cuentos are numbered consecutively. All references will be to pages without indicating the volume.

Part I. Phonology and Lexicology

1. alante

This dialectism is a syncopated form of adelante. There are examples in the Cuentos from many parts of Spain, from Old and New Castile, from León, and from Andalucía, a few of which are listed below: Avila5

* The general development of Latin f->h- (modern Spanish j) in Andalusian, for example, is only a retarded development of a phenomenon once general in Castilian. It seems to have originated also in the Cantabrian region and developed quickly to a weak aspirate and then disappeared generally, whereas in Andalusia it remains at the h stage. See Origenes del español § 41.

Manual de pronunciación española, Madrid,1 1918,3 1926.

"In recording the source of the folk-tales I shall indicate the province or region and not the village or locality for the sake of geographical clearness.

80 ('Y ya se echó el azadón al hombro y se fué camino alante.'), 119 ('Bueno, pues los otros se fueron por la senda y él siguió alante por el camino resto."); Valladolid 436 ('Y salieron la zorra y el lobo por el camino alante.'), 490 ('Y más alante, más alante encontraron en el camino a un gallo, .'); Segovia 300 ('Se marchó el mayor por el mundo alante. .'); Madrid 368; Soria 59, 92, 249, 284, 326, 411; Astorga 396; Zamora 56, 154, 181; Toledo 189, 227, 281, 357; Cuenca 122, 204, 236; Granada 46, 48, 52, 113, 145, 331; Sevilla 264, 462, 476; Córdoba 439, 482.

The phonetic development was as follows: adelante >aelante >alante, a gradual weakening of intervocalic d after it has changed from an explosive to a continuant, its complete disappearance, and finally a reduction of áe to a by regressive assimilation to the accented and more sonorous vowel of the group. According to Menéndez Pidal, Cantar II s.v. adelant, a regular Old Spanish form more commonly used than adelante, the modern literary form, the etymology of adelante is ad-de-inante. This would give in Old Spanish *adenante, *adenant, forms that apparently have not been found in any Spanish documents. By dissimilation *adenante became adelant(e). We are certain that *adenante must have been a primitive Old Spanish form in view of the Xth century denante (Glosas Emilianenses 80, Glosas Silenses 2807) and the Classic and modern dialectic enantes, denantes, endenantes. Just why an old form *adenante is not documented while the adverbs denantes, endenantes persist even to modern times is not clear. Cuervo in his Diccionarios cites cases of denantes from Juan del Encina, Luis de León, Cervantes, Lope de Vega and Calderón, of endenantes from Solís, Calderón and Bretón, and states that enantes as well as the other forms are found in the modern dialects of Spain and America. In Apuntaciones § 374 he repeats this last statement but does not indicate exact localities. Enantes, denantes and endenantes are not found in the Cuentos so they are apparently not as common in Spain as Cuervo believes. According

• Cantar de mío Cid, texto, gramática y vocabulario, 3 vols., Madrid, 1908-11. 7 See Menéndez Pidal, Orígenes del español § 78. There are other examples given for the Xth and XIth centuries. In these old documents denante still appears only with its prepositional character and without de, as in Glosas Emilianenses 80: 'Facanos Deus omnipotes tal serbitjo fere ke denante ela sua face gaudioso segamus.'

Diccionario de construcción y régimen de la lengua castellana, 2 vols., Paris,

1886-93.

'Apuntaciones críticas sobre el lenguaje bogotano, 6th ed., Paris, 1914.