Linguistic Society of America

Germanic Influence on Old French Syntax

Author(s): Urban T. Holmes
Source: Language, Vol. 7, No. 3 (Sep., 1931), pp. 194-199
Published by: Linguistic Society of America
Stable URL: http://www.jstor.org/stable/408810
Accessed: 02-04-2017 05:29 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

Linguistic Society of America is collaborating with JSTOR to digitize, preserve and extend access to
Language

This content downloaded from 140.206.154.236 on Sun, 02 Apr 2017 05:29:34 UTC
All use subject to http://about.jstor.org/terms
 GERMANIC INFLUENCE ON OLD FRENCH SYNTAX

URBAN T. HOLMES

UNIVERSITY OF NORTH CAROLINA

[The author wishes to show that German influence on the syntax of

OFrench may be expected, and that it is to be recognized in the order of
words (verb at end) in dependent clauses.]

Several years ago Josef Brtich listed' the traces of Germanic influence
in the Latin speech of Gaul, which are admitted by the majority of
scholars.
It is his last item, syntactic influence, which now concerns us, but
we may begin with some generalities on the conditions of speech mixture.
Brtich himself, in another place, divides language interrelation into two
classes: quantitative and qualitative.2 The first is a matter of vocabu-
lary, words which are easily exchanged through commerce and casual
association between two peoples; the second is more of a problem: it is a
question of influence from a foreign source on pronunciation, word-
formation, morphology, and syntax. Everyone must admit the exist-
ence of the first type of borrowing, but certain scholars have been rigidly
opposed to the latter. Max Mtiller flatly denied any qualitative influ-
ence, as an axiom.' Whitney was more generous and admitted that
word-formation might suffer from foreign influence, but not morphology
and syntax.4 Meillet is also hesitant to accept anything but vocabulary
borrowings;' Sapir prefers to ascribe resemblances in morphology and
syntax between two adjacent tongues to parallel drift rather than to
interrelation.6 For Karl Vossler, mixture can only arise from a blending

1 Rev. Ling. Rom. 2. 30-45. E. C. Hills quotes a well-known scholar, whom he

does not name, as suggesting that the preservation of cl-, fl-, and pl-, in Gaul,
while weakened in other Romance districts, may have been due to Germanic
influence (Hispanic Studies 213, Hispania Monograph Series, 1929).
2 Der Einfluss der germanischen Sprachen auf das Vulgirlatein 1 (Heidelberg,
1913).
3 Lectures 16. 86.
4TAPA 12. 18, 20.
6 Linguistique historique et linguistique g6nerale 84, 86, 87 (Paris, 1926).
6 Language 219-20 (Harcourt Brace and Co., 1921).
194

This content downloaded from 140.206.154.236 on Sun, 02 Apr 2017 05:29:34 UTC
All use subject to http://about.jstor.org/terms
 GERMANIC INFLUENCE ON OLD FRENCH SYNTAX 195

of the spiritual outlook of national groups;7 Charles Bally

cerned with historical, only with psychological syntax,
attention to a 'mentalit6 europeenne' which would account
resemblances.8 I might add, apropos of Bally, that this Eu
tality could hardly have been existent at the period which
us here, the early Middle Ages.
In contrast to these opinions is the attitude assumed by Briic
ardt,9 Hermann Paul,10 Windisch,"1 Vendryes,1" and Jesper
also heard Leonard Bloomfield admit the possibilities of syn
ence of one language upon another. Jespersen puts it clear
says '. . . those who know a language imperfectly are apt
to it methods of construction from their own language.'
The conditions under which speech mixture may arise h
clearly discussed by Windisch and by Georg Hempl.14 With
upon any detailed account of these we shall add that the s
Gallo-Latin with respect to the Germanic invaders was
called Type One: the conquerors come in a body and if the
are not too small, and if they have been long adjacent to
vaded, their language holds out for a while; the conqueror
quered borrow mutually from each other before, in the end, th
invading language succumbs. 'Thus the influence of the Fr
language of France.' H. F. Muller, in a very recent work, us
analogy in demonstrating the contrary, that the German s
could not have lasted long in Gaul. 'The son of foreigners r
country among natives, inherits absolutely nothing of
linguistic habits.' Muller proves this by conditions in Ame
the condition of the immigrant in America is Hempl's Typ
it has no analogy with the condition of the German invader in
7This thought recurs throughout Vossler's Frankreichs Kultur
seiner Sprachentwicklung (Heidelberg, 1929).
8 Charles Bally, Trait6 de Stylistique Frangaise2 1. 23 (Heidelber
9 Lit. f. germ. rom. Phil. 4. 236.
10 Prinzipien d. Sprachgeschichte3 365 (Halle, 1898).
11 Verhandl. d. k6n. sachs.-Gesellsch. d. Wissensch. 1897. 101-126.
12 Le Langage 341-2 (Paris, 1921).
13 Language 214 (Holt, 1922).
14 For Windisch see note 11.
16 H. F. Muller, 'A Chronology of Vulgar Latin', Z. f. rom. Phil., Beiheft 78. 126
(1929).
16 The present article is a reduction from a larger unpublished study in which I
examined with some detail the actual relations between the Germanic invaders
and the Gallo-Romans.

This content downloaded from 140.206.154.236 on Sun, 02 Apr 2017 05:29:34 UTC
All use subject to http://about.jstor.org/terms
196 URBAN T. HOLMES

If we are ready to admit then wi

there should be some legacy from
lect), in Old French syntax, it rem
du M6ril, in a work that never ha
took no account of the discoveries o
interrelation. He derived quelque c
werlde, Id < da, fort < fasto, trds
Sax. for tham the, en cas que < i
entour < umher, atot < mit all, pa
etc." It is evident that these exam
were considered so in 1852 when D
one has ever repeated them. Diez h
traces of German in Romance syn
To date, only one case of such infl
and that by Meyer-Liibke, though
Meyer-Lilbke admits that the use
se viaut, si l'oie' (Yvain 5. 1447) 's
deutschen Einfluss zu einer gr6sse
will examine carefully the Latin tex
to be more liberal than the master
sic like German so, almost entirely
does not indicate, nor is it obvious,
Neigung'. Here are several exam
factum, si eis convenit, sic accipi at
non fuerit, sic patris soror in here
summary, cited above, makes no ref
LiUbke.
In an attempt to uncover further
Frankish constructions I have made use of the fact that the earliest OHG
texts, mostly translations from the Latin, followed their original with
slavish closeness.20 If the German text varied with some regularity
from certain constructions in the Latin original, it was because the
German language found those constructions absolutely repugnant to

17 E. du M6ril, Essai sur la Formation de la Langue frangaise 235-8 (Paris:

Franck, 1852).
18 Diez, Grammaire des Langues Romanes 1. 65; 3. 121 (Franck, 1874).
19 Meyer-Liibke, Gramm. d. rom. Spr. 3. 698, and thereafter J. KlapperUch in
his Bedingungss. im Altfranz. 3. 64.
20 H. Reis, Z. f. deut. Phil. 33. 214. Reis says that in observing the character-
istics of early German syntax 'k6nnen daher nur solche Stellen beriicksichtigt
werden in denen der Ubersetzer von dem Text der lateinischen Vorlage abweicht'.

This content downloaded from 140.206.154.236 on Sun, 02 Apr 2017 05:29:34 UTC
All use subject to http://about.jstor.org/terms
 GERMANIC INFLUENCE ON OLD FRENCH SYNTAX 197

its own expression. If the Latin construction was at all poss

early translators tended to follow it. Now it so happens that in
instances the OFrench language (1000-1300 A. D.) as we have it,
the Germanic construction and not the Latin original. I am we
that there are other factors to be considered. E. Stimming rem
apropos of H. F. Muller's theses on the construction 'faire quelq
a' that it is not enough for a Romance phenomenon to be simila
Latin for the former to be derived from the latter,2' and I sup
same can be said for a German source. The other factors are: (
normal drift of Latin speech, as found in the progressive tende
Vulgar Latin;22 (2) outside influence from Celtic; (3) psychologi
tors, such as a search for clarity or beauty of expression; (4) co
imitation of classic Latin of the Golden Age (90 B.C.-1 A.D.). F
(3) and (4) are rather stylistic devices and if they existed in the
period they would hardly influence any constructions with regu
Factor (2) has been minimized by Vendryes on the ground tha
was very little distinction between Gaulish and Latin syntax a
the opposite of what Bergayne has indicated between Germ
Latin construction. Vendryes allows only the possibility of
periphrasis as 'C'est moi qui le fais' to Celtic influence.24 It is f
which might bother us in any such study as I have suggested
paragraph, but it too is somewhat reduced when the Latin original o
Frankish translations is the Vulgate Bible. V. S. Clark has writ
is certain that the strong undercurrent of living colloquial sp
flowing in the direction indicated by the language of the Vulg
and D. S. Blondheim speaks of the vulgar tendencies 'qui ont pr
Vetus latina, dont la Vulgate n'est qu'une revision.'25
I have made use of the East Franconian translation of Tatian's

21 H. F. Muller, 85 (Poitiers, 1912).

22 K. Vossler in Geist u. Kultur in der Sprache 56-83 (Heidelberg, 1925)
takes to formulate this 'drift' of Latin into Romance.
23 The importance of the psychological factor, over against historical investiga-
tion, in syntax, is stressed by many investigators of to-day. It is a matter of
opinion; but Lucien Foulet, who is sympathetic to the descriptive school of syn-
tax, gives no psychological explanation for the phenomenon which we examine
shortly.
24 Vendryes, Rev. Ling. Rom. 1. 265-6, M6m. Soc. Ling. de Paris. 17. 338; Ber-
gayne in M6m. Soc. Ling. de Paris 3. 51.
25 V. S. Clark, The Latinity of the Middle Age and of the Renaissance 4-5
(1908); D. S. Blondheim, Les parlers judeo-romans et la Vetus Latina cxxxiv
(1925).

This content downloaded from 140.206.154.236 on Sun, 02 Apr 2017 05:29:34 UTC
All use subject to http://about.jstor.org/terms
198 URBAN T. HOLMES

Diatessaron, which was made in t

that many Germanic scholars now b
ber of hands,2' and it is also true, as W
that there were more than 400 mon
many of them did not use the Fran
It is none the less probable that the
careful supervision and the syntact
been at variance with the usage of
same intent, the Monsee-Wiener f
century Bavarian version of an e
translation into Rhine Frankish of Isidore's De fide catholica contra
Judaeos has survived in an 8th century form, but the original of Isidore
is not as useful for our purpose as the Latin Vulgate.
Space is lacking here for me to give a full account of results obtained in
this search for the Germanic element in OFrench syntax, but there is one
case which I should like to present in brief: the word order in subordinate
clauses, in particular the position of the verb. In OHG there is almost
a fixed rule that in dependent clauses the subject usually comes second,
after the subordinating conjunction or adverb, and the verb almost
always comes at the end.28 This was hardly a rule in Latin where the
position of the verb was dependent upon emphasis within a group; fur-
ther, it was contrary to the progressively prevailing tendency of Vulgar
Latin to use the order Subject + Predicate + Complement.29 I shall give
some parallel passages from Tatian:
qui est in caelis thie in himilon ist (49)
qui fuerunt ante vos thie thar fora iu uuarun (43)
si tu non es Christus oba thu Christ ni bist (35)
quod Dominus ostendit nobis thaz truhtin uns araugta (24).1?
and from the Monsee-Wiener fragments:
Quis erit ex vobis homo qui habeat ovem unam
huuelih iuuuer ist der man der ein scaf habet (Braune, Althoch-
deutsches Lesebuch5 20).
26 Taylor Starck in Collitz Studies 190-202 (1930).
27 Z. f. deut. Altertums 46. 141, 144.
28 Reis in Z. f. deut. Phil. 33. 217; 41. 222, 221. This fixity tended to disappear
in the MHG period (F. Maurer in Behagel Festschrift 183).
29 A great deal has been written on Latin word-order since the days of H. We
(De l'ordre des mots dans les langues anciennes comparees aux langues modern
Paris 1844). For bibliography consult Marouzeau, L'ordre des mots dans la
phrase latine ix-xvi (Paris 1922).
30 The page references are to Sievers' edition of Tatian.

This content downloaded from 140.206.154.236 on Sun, 02 Apr 2017 05:29:34 UTC
All use subject to http://about.jstor.org/terms
 GERMANIC INFLUENCE ON OLD FRENCH SYNTAX 199

There can be no question that OHG, and this includes

regularly used the order Subject + Complement + Verb in
clauses. This stereotyped order was not fixed in Latin,
Vulgar Latin 'drift'. It so happens that OFrench has exactl
fixed order as the German, in subordinate clauses. Lucien
it construction II and he says '. .. il est certain qu'on l'e
une predilection toute particuliere dans les phrases sub
conjonction et surtout le relatif au cas sujet rejettent volon
a la fin de la phrase'.31 An example in prose is "Si com Dix
les amans ainme . . . ' (Aucassin et Nicolette 26. 11-2). M
who is writing a descriptive syntax of OFrench does not off
logical explanation; he hints rather that by the XIIth centu
was an archaism: that is, a trait contrary to the prevailing
of the time but preserved through habit.32
There are several other constructions which might have be
here, had space permitted, chief among them the OFrench
ires (Fierabras 2285, 2295) and issirfors, etc. where the use o
sition as an adverb is strongly reminiscent of the German
prefix as in en ti hevet iz uz (Braune, p. 19) where the Latin ha
eam'.

3' L. Foulet, Petite Syntaxe de l'Ancien Frangais3 316 (1930).

32 Ibid. 351.

This content downloaded from 140.206.154.236 on Sun, 02 Apr 2017 05:29:34 UTC
All use subject to http://about.jstor.org/terms