No CrossRef data available.
Article contents
Germanic Influence on Old French Syntax
Published online by Cambridge University Press: 22 April 2026
Extract
In a previous article (Language 7.194-9) U. T. Holmes sought to establish the probability of German influence on the syntax of late Latin as spoken in northern Gaul. He also suggested a method for determining those Germanic constructions which differed most from popular Latin; and, where these German constructions are present in OFrench syntax he assumed that an interconnection was quite possible. Willem L. Graff has observed recently that ‘it is not yet possible to formulate the general laws that are thought to rule the drift of syntactical features.’ This may be true, in part, but it is safe to argue that every language has inherent tendencies which work themselves through to an expected conclusion, once the restrictions of artificial grammar become lax. It is possible to foresee much of late vulgar Latin in primitive and early Latin. The drift is continuous—no radical departures of syntactical construction are apt to develop without a rude shock from without, and, as was argued on good authority in the earlier article, the most vigorous shock which Gallo-Latin had to bear was from the horde of Germanic speakers who eventually took political possession of northern France. Furthermore, whatever we may say of the phonology and spelling of Saint Jerome's Vulgate Bible, its syntax (particularly before the revision by Alcuin) shows the popular concept of the phrase and sentence as held in the fourth century A.D. By confronting the Vulgate text with painfully literal translations made from it into OHG it is not difficult to discover those Germanisms which were essentially different from the late Latin drift. At the same time no construction must be labelled Germanic, even by this careful comparison, until a check has established that it is not present, or only slightly so, in the other Romance tongues. Italian and Provençal suffered less Germanic influence than French; Spanish had still less, and Roumanian almost none at all.
Information
- Type
- Research Article
- Information
- Copyright
- Copyright © 1933 Linguistic Society of America
References
1 Recently (Rev. Ling. Rom. 4.309 ff), as an argument for the unsoundness of historical syntax, Carl de Boer has cited the fact that German and Romance have occasional resemblances in syntax not to be accounted for in classical Latin. Eugen Lerch (ibid., 6.127) answers this with the suggestion that in some individual cases there may be influence from low Latin constructions on both Germanic and French. Surely it is safer to search, as we have done, for influence from OHG to Gallo-Romance. See also G. W. Small in Language 8.314.
2 Language and Languages 291 (Appleton 1932).
3 Histoire de la langue française 434 (Paris : Payot 1930). He offers no proof.
4 Our examples are drawn mostly from the following texts : Eduard Sievers, Tatian (Paderborn: Schöningh 1892), which we designate as T; G. A. Hench, The Monsee Fragments (Strassburg: Trübner 1890), indicated by the letters M-W; Le Roux de Lincy, Les Quatre livre des Reis (Paris: Imprimerie royale 1841), indicated by K; and such OFrench works as the Chanson de Roland, the Life of Saint Alexis, and the lays of Marie de France, which are easily available in a number of editions. Marie's lays are cited by their individual names, Eliduc, Guigemar, Yonec, Equitan, Bisclavret, Milun, etc.
5 Examples from the Lex Salica are given by Holmes in Language 7.196.
6 Petite syntaxe de l'ancien français 301 (Paris: Champion 1930).
7 Gramm. d. rom. Spr. 3.698.
8 An excellent illustration of this is Guardet aval e si guardet amont (Roland 2235) where si certainly has the force of donc.
9 Note that the reinforcing adverb is not the same as a preposition introducing an adverbial phrase. The two must not be confused.
10 ibid. 18
11 It is evident that Brown sees in such a translation as a ‘man’ many varying shades of indefiniteness which he tends to differentiate from the generic usage. This is a philosophical rather than a grammatical problem. We contend that homo is grammatically indefinite only in cases (2) and (3) outlined herewith.
12 ibid. 272.
13 W. Ritchie, The Plays of Terence 189 (London: Bell 1927).
14 H. T. Riley, The Comedies of Platus 1.105 (London: Bell 1894).
15 A Chronology of Vulgar Latin, Z. f. rom. Phil. Beiheft 78.74 (1929).
16 Ausführliche lateinische Grammatik 3.518-23.
17 The Language of the Eight-century Texts in Northern France 241 (New York 1932).
18 The NED is our authority that English anew and afresh go back ultimately to of new and, possibly, of fresh.
19 Althochdeutsche Grammatik 268, n. 3 (Halle : Niemeyer 1911).
20 ibid. 437.
21 Essai sur la structure logique de la phrase 33 (Paris : Champion 1926).
- 1
- Cited by