Abbildungen der Seite

Spanish. Anque is used in literature even in the XVIIth and XVIIIth centuries. In my Studies in New Mexican Spanish I § 34 and notes I have cited numerous examples. Anque is used by Santa Teresa, Torres Naharro, Sanchez de Badajoz, Ramón de la Cruz and others, while onque, which is less frequent is used by Juan del Encina. I have no cases of enque or unque from literary Spanish. The XVIth century writers who use the forms anque, onque use likewise the forms an, on, aón, áun, aún. Aón for aun is apparently a very old form. In Torres Naharro the form occurs in rime many times, and aonque is also found.45 To explain the development of the above dialectic forms of the Cuentos we must start with the simple adverb aun<ad-huc. When accented áun, as the word is generally pronounced today when in the weak position before the word or phrase modified,46 the reduction of áu to a is explained by regressive assimilation to the accented and more sonorous vowel, as in the cases of alante and ande already discussed.46 If the development of áunque>anque is contemporaneous with áun>an the phonetic process is the same. In the case of onque we have progressive assimilation after the accent has shifted in the middle form áon developed by assimilation from áun. The form aón found in Torres Naharro becomes on and hence onque. But in view of the fact that a form aonque occurs the shift of accent may have occurred here at the same time as in the simple aon and hence aónque>onque through the same process of progressive assimilation. The dialectic enque from the Palencia text, found also in Salamanca (Lamano y Beneite) and in New Mexico, may be due to weakening of the initial vowel through proclisis. It may have developed from onque. The form could also develop through the confusion of the prefixes an-, en-, in the dialects due to the tendency of en- to become an- before velar consonants and in other sporadic cases.47

(To be continued)

45 See Pietsch, Mod. Phil. 5. 100. An example of aon is cited from the XIIIth century Castilian version of the Concilio de León of 1020.

46 Navarro Tomás, Manual de pronunciación española, 157-8. The development of áu>a is found in the Cuentos in other cases: va sté<va ústed, 140; guarde a sté<guarde a usted, 241; ¿ qué le pasa a té? 264; ya sté sabe<ya usted sabe, 379. Under certain conditions this phenomenon is found also in Old Spanish and in Vulgar Latin.

47 In the same text from Palencia we find no estante <no obstante, but this change may come under the more general prefix confusions so common in Spanish. See Studies in New Mexican Spanish I §51 and Menéndez Pidal, Gramática histórica española, op. cit., 39.


Die L-, R- und D-Laute in austronesischen Sprachen. By OTTO DEMPWOLFF. (Sonderabdruck aus der "Zeitschrift für Eingeborenensprachen," Band XV, Berlin, 1925.)

This contribution, although it appeared as a series of journal articles, deserves to be mentioned not only because of its size and importance, but also because the Zeitschrift für Eingeborenensprachen is scarcely known in our country. Dempwolff's study amounts to a comparative Austronesian phonology, probably as complete and detailed as human ingenuity could make it with the data that are available. It thus brings to a conclusion, for the time being, such researches as those of H. Kern and C. E. Conant. Further advance will be possible, it would seem, only when more material can be used. Especially in the Philippine group interest has centered upon the grammars of the early missionaries, as though the languages were not directly accessible. There are few contributions of material, like O. Scheerer's Kalinga Texts (Philippine Journal of Science 19. 175) or the late C. W. Seidenadel's Bontoc Grammar (Chicago, 1909). A result of this scarcity of material is, for instance, that Dempwolff is forced to ignore accent.

Outlining the sound-history of one of the world's largest linguistic stocks, Dempwolff's study ranks as a major contribution to our knowledge of the history of speech.


La Science du Mot: traité de sémantique. Pp. vii + 426. By A. CARNOY. Louvain: Editions "Universitas", 1927.

Professor Carnoy seeks to present a complete treatise on the semantic processes, which had not hitherto been done. He devotes the first chapters (pp. 1-82) to general linguistic and psychological considerations, related to his main subject, and then takes up (pp. 83-378) the various kinds of change of meaning. After this come an excellent history of semantics, with a bibliography (pp. 388-400), a synopsis of the classification (pp. 401-7), and indices (408-26).

The author has in mind also to create for semantics a definite technical terminology, to replace the present unsettled condition of the sub

ject. He accordingly invents some 35 or more terms ending in -sémie (Greek onuela, which does not occur in classical Greek), the prior elements being those well known in technical terms. Thus métasémie denotes in general a change of meaning; métendosémie is a change to a new meaning in which the main emphasis is on a different element already inherent in the old meaning (Abcd becomes aBcd or abCd), as when lunch 'repas de midi' becomes 'collation, petit repas'. Métecsémie is a shift to a new meaning in which at least one of the elements present in the old meaning is no longer present, though at least one such element still remains, as when Latin testa 'cruche' gives French tête 'head'. And so on. It is indeed highly desirable that semantics should have a fixed terminology; but whether Carnoy's scheme will receive acceptance or not is doubtful; for there are two objections to it. First, the words are identical in the second part, which makes them more difficult to distinguish; second, they are all new and unfamiliar, and not infrequently replace terms already familiar in rhetoric, such as metaphor (= métecsémie). A judicious mingling of old and new, with distinctions made by attributive adjectives (which he does indeed sometimes offer as definitions of his subdivisions), would seem more likely to gain acceptance.

All the categories are richly illustrated by examples, drawn chiefly from French, Dutch, and English, though German, Latin, and Greek are well represented, and Italian, Spanish, and Slavonic examples are used now and then. Colloquial language and slang, in which Carnoy has made special studies, are cited continually, since they furnish the best basis for semantic demonstration: the old meaning and the new are normally still both present in the speech, and the comparison may be better appreciated than if the old meaning had to be derived by etymological study.

Americanisms are given much attention, quite naturally, since Professor Carnoy spent nearly five years in this country during and after the war, and made some rich collections (rich in two senses). The reviewer does not feel qualified to criticize the examples drawn from idiomatic uses of other languages, but he wishes to examine chiefly those taken from English, and especially from American English; for he feels that the present work illustrates the extreme difficulty confronting the student of idioms of a language which is not his own. If at any point the reviewer errs, he is at least following his own linguistic feeling for his mother tongue.

On page 266, we read: 'En anglais comme en français, on dit to have it

on black and white = avoir noir sur blanc, pour insister sur la réalité de la chose écrite.' But the English idiom is in black and white, and the sur in the French equivalent has misled the writer. Later on the same page, 'Quiconque prend la parole tient le crachoir (ang. takes floor "occupe l'estrade").' The English idiom is unintelligible without the definite article, takes the floor. Again on the same page, 'Celui qui étonne et effraie fait dresser les cheveux sur la tête (ang. raises hair).' It is true that we speak of a hair-raising experience, but we do not say that anything raises hair except a hair tonic, and even that fails not infrequently: nor even does terror raise my hair, rather it makes my hair stand on end-which is the only permissible idiom.

Perhaps however some of the examples have suffered from undue brevity of exposition: English Jerusalem from Italian girasole (p. 29) is quite bewildering, by itself, although it is later (p. 224) repeated with the necessary addition of the word artichoke. So also the Old Lady in Threadneedle (p. 336) should not be deprived of Street, and the preposition is of, not in. 'Ainsi l'anglais not half "pas à moitié" signifie "tout à fait, fort" (p. 373). But this is only in British English; in America, not half is current only in not half bad, which means not 'very bad,' but 'very good.' Brevity makes one think also that the men of Latium applied Urbs to Athens rather than to Rome (p. 140).

To return to the difficulties of the idiom of a foreign language, the following might be noted as samples: Sob-sister 'femme plaignarde' (p. 255) is used also of men, and denotes pessimism rather than tears. A hell of a dinner, glossed 'copieux dîner' (p. 309), means a 'wretched dinner', unless the enthusiasm of the speaker's voice reverses the meaning (I have never heard such a use myself). Hemp is glossed 'pendard' (p. 261), and this meaning is in the dictionary, but the slang use is normally for a hangman's rope. Red tape means not 'bureaucrate', as Carnoy states (p. 261), but 'bureaucratie'.

Not a few words also are incorrectly quoted, especially English words: thus kookedooledoo (p. 22) should be cock-a-doodle-doo; Rose Mary (the plant) (p. 29) should be rosemary; primerose (p. 29) should be primrose; chandeleer and electroleer (p. 60) should end in -lier, and motoneer should be motorneer; buttry (p. 64) should be buttery; lever (p. 110) should be liver; beach (p. 118) should be beech; T. C. (p. 241) and T. B. C. (p. 350) should be T. B.; salone (p. 339) should be saloon; tris (p. 342) should be tries. Other misquoted forms are Spanish salire (p. 169; for salir), Latin coenam (p. 245; for cenam), intelligo (p. 317; for intellego), transferro (p. 318; for transfero), Italian fachino (p. 374; for facchino).

Latin feto, quoted as 'jeune animal' (p. 135), is a hypothetical word only, assumed as the ancestor of French faon.

Finally, a few items of a more general bearing. The account of the union of different roots into one paradigm, as in sum fui, go went (p. 62), might more than merely hint at the phenomenon of temporal aspect, which, though no longer an active factor in languages of Western Europe, is vital in the Balto-Slavonic tongues. That 'a truly complete dictionary of French or of English would have more words of Greek origin than of Latin or Germanic origin' (p. 75) is to the reviewer very dubious. Artificially created trade-words are rarely fabriqués artificiellement de toute pièce' (p. 77); nearly all can be traced to some other usual words. Pianola, cited as example, is an obvious formation from piano, with a familiar suffix; byrrh (a drink) is under the influence of beer (Bier, bière) and myrrh (myrrhe). The transfer of an abstract substantive to an individual person is regularly made through an intermediate stage of a collective concrete, though Professor Carnoy does not hint at this (p. 188); the best example is English youth 'a time of life', the youth 'young men collectively', a youth 'a young man'. Another example, caught as it makes the second change, is Latin custodia 'guardianship', then 'body of guards', finally, in Ovid Met. 8.684, 'a guardian'. The ancient type of compounds with an imperative as first member (p. 264) does not include all the examples given; notably, English breakfast started out as an ordinary indicative phrase (they break fast, they broke fast, etc.).

But Professor Carnoy has given us for the first time a systematic manual of semantics, with a definite terminology, and every student of the subject must reckon with his work; must in fact consider it as the point from which he starts, and the point to which he returns, as an indispensable guide and a treasure-house of examples.


Histoire de Parfait Grec. Pp. x + 268. By PIERRE CHANTRAINE. (Collection Linguistique publiée par la Société de Linguistique de Paris, XXI.) Paris: H. Champion, 1927.

Monsieur Chantraine, following especially the studies of J. Wackernagel and of A. Meillet, gives us in this volume a history of the perfect tense in Greek, from Homer to the Hellenistic period. His conclusions may be thus summarized: The oldest type of perfect, commonest in the Homeric poems, denotes a state which persists in the subject itself, and is essentially of present time; this is true even when the verb has a direct

« ZurückWeiter »