Romance Etymologies

Author(s): Carl C. Rice

Source: PMLA, Vol. 20, No. 2 (1905), pp. 339-345
Published by: Modern Language Association
Stable URL: http://www.jstor.org/stable/456629
Accessed: 10-06-2017 05:55 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

Modern Language Association is collaborating with JSTOR to digitize, preserve and extend access to PMLA

This content downloaded from 140.206.154.232 on Sat, 10 Jun 2017 05:55:56 UTC
All use subject to http://about.jstor.org/terms
 IX.-ROMANCE ETYMOLOGIES.

I.

Frenchfl&chir < Old French feschir <fieschier, "to bend,"

< *fJexicdre <Jfeus <fleetere, "to bend."
French flEchir, O. F. fleschir, fleskir has been derived by
Forster, Zeitchrift f. rom. Phil., III, p. 262, from a Latin
*fleskire < *flescus <flexus. The assumption of the shift of
ks to sk is defended by an appeal to alaskir from laxus, seem-
ingly showing the same metathesis. This phonetic step,
which must be assigned to a Latin period, is in both instances
certainly unjustifiable, although it has been admitted by
excellent authorities. In the Dictionnaire gn&Eal we find
French decher derived from a type *lascare < laxare. Here
the assumption of metathesis seems to go back to Diez,
Etymologisehes Wb., pp. 188 f., who cites as analoga Cam-
panian fisquer for fixer and lusque for luxe; but these forms
clearly represent popular deformations of learned words and
are accordingly irrelevant. French ldcher has also been
derived by Grober, who evidently objects to the dubious
metathesis, from Old High German *lsc, a type assumed to
account for Middle High German lasch, "schlaff," and Old
Norse loskr, "schlaff," "lass." Kluge, however (Etymolo-
gisches Wb., 6th ed.), is inclined to derive the Germanic
from the Romance group. Grober's derivation has also been
disputed on phonological grounds by Mackel; cf. Korting,
s. v. *lask, who rejects the Germanic etymon. The correct
etymon for ldcher, namely *laxicare, was first suggested by
Ulrich, Zeitschrift f. rom. Phtil., ix, p. 429; is rejected
by K6rting, who says that the assumption of the type is
unnecessary and seems to consider the derivation of the
339

This content downloaded from 140.206.154.232 on Sat, 10 Jun 2017 05:55:56 UTC
All use subject to http://about.jstor.org/terms
 340 CARL C. RICE.

French word unsettled; but is accepted by Mey

Rom. Gramm., II, p. 608. It will be seen late
care presents a perfect phonetic type for the d
the French form. The cognate Romance forms
lascar, etc.) present no difficulty. Returning t
need only mention the derivation of the word f
adopted by Diez, which is phonetically impo
Rom., vIII, p. 628, has explained flechir as d
the adjective flesche, "bent," and the latter as
from fleschier, which he derives from *flescar
My objections to this etymology are as follows
place, the existence of the adjective flesche is extr
ful. Scheler and Paris (1. c.) thought it occurre
French passage, namely, in the Saint Eloi, 9
fleches, enclin le chief. Here Forster, however (
reads fleches, and the passage is also quoted in
Godefroy. No evidence for the existence of th
appeared in Godefroy's Complfment, and under
stances it should doubtless be regarded as im
the second place, the phonetic step fromflexar
is without support. Finally Grober, Archiv f.
Gr., II, p. 285, explains flechir as a collater
flechier showing a change of conjugation. This
certainly seems to be the correct one. A g
lexicon is sufficient to convince one that verbs
between the -ir and -(i)er conjugations were fa
in Old French: note, e. g., refroidier, refroidir; en
embalsamir; engrossier, engrossir; amplier, am
enter, empoentir; empreignier, empreignir. T
undoubtedly be greatly lengthened.
Now, to explain this earlier form flechier, Gr
cited) sets up a type *flecticare, which is accepted
but which does not account for the Old French fo
Paris and Firster (articles cited) assume that

This content downloaded from 140.206.154.232 on Sat, 10 Jun 2017 05:55:56 UTC
All use subject to http://about.jstor.org/terms
 ROMANCE ETYMOLOGIES. 341

Old French form of both echier and fleehir had an

in view of the spellings with s cited by Forster, this
certainly seems to be correct. Several forms with s
be seen in Godefroy.
The right etymon is *flexicare. This type was
suggested by Grober (article cited), who rejected it
ground that it should have given 0. F. *fleischier
similar reason the Dictionnaire gynral rejects *taxi
the etymon of 0. F. taster, Modern French tdter, a
that this Latin type would have given 0. F. *taist
both authorities are in error regarding the sound-la
in question, which is stated by Schwan-Behrens, Al
Gramm., 4th ed., ? 158, 2, as follows:-
VIllige Assimilation des Palatals an den folgenden Konsonan
... in vortoniger Stellung in der Verbindung ks + Kons. ein:
.... sextaru > sestier, dextraru > destrier, *tax(i)tare > taster, ent
> entoschier, extendere > estendre, extorquere > estordre, satzunbetont
estre und joxta > juste."

To these examples we may now add *flericare >fle

which is perfectly analogous to intoxicare > entoschier,
care > laschier, and *txicare > taschier. Tdcher, the
form of taschier, is derived by the Dictionnaire general
*tascare, a metathesized form of taxare. But, as ha
shown above, the analoga seemingly justifying the a
tion of a metathesis of the group ke in a Latin perio
no value. The etymon *ta=icare is due to Ulrich, Zeit
f. rom. Phil., ix, p. 429. It is put in brackets by K6
but is accepted by Meyer-Liibke, Rom. Gramm., ii,
The fact that a so-called epenthetic i does not ap
developments like that of *taxitare > taster is to be exp

1 The daring etymology tdche < *tasca < *rdoXts < 'rdts, recen
gested by T. Claussen, Romanische Forschungen, xv (1904), p. 847
deserves mention. The Dictionnaire gnral correctly states that t
postverbal from tdcher.
10

This content downloaded from 140.206.154.232 on Sat, 10 Jun 2017 05:55:56 UTC
All use subject to http://about.jstor.org/terms
 342 CARL C. RICE.

by the chronology of the sound-change. It is

that the pretonic vowel in paroxytones and th
vowel in proparoxytones dropped at different
*taxitare, as is evidently assumed by Schwan-B
had been reduced to *tastare in a period when
still trissyllabic. One might of course also exp
of the verb with epenthetic i, preserved from the
tone forms, to survive, and this actually did happ
cases. We need only cite the postverbal test bes
supposing a form *taister beside taster, and th
striking form entoischier beside entoschier < into
If this reasoning is correct, we have establish
sion diametrically opposed to that of the Dictionn
s. v. flechir, which says with regard to the etymo
word: " Origine inconnue. La forme du mot ne
d'y voir un repr6sentant, direct on indirect, du la
qui a cependant le mmee sens."

II.

Spanish rosca, "screw" < *r6sicare < r6sas < rodere, "to
gnaw."
In Monlau's Diccionario etymol6gico we find the following
note: "Rosca: 'Es del vascuence errosca, y se dijo de
erruzca, E fuerza, por la grande que tiene para mover grandes
pesos.' (Larramendi.) Segun Covarrubias viene del latin
ruere, lanzarse, arrojarse, porque gira sobre si misma. Diez
afirma, con m,s acierto, que el origen de rosca es todavia
desconocido." The word is missing in K6rting's index. On
consulting the recent edition of the dictionary of the Spanish
Academy, we are told that rosca is derived from an absurd
Greek etymon.
I derive the word from *rosicare; "to gnaw," the exist-
ence of which in late Latin is rendered certain by Italian

This content downloaded from 140.206.154.232 on Sat, 10 Jun 2017 05:55:56 UTC
All use subject to http://about.jstor.org/terms
 ROMANCE ETYMOLOGIES. 343

9rosicare, Proveneal rosegar, " to gnaw." The e

presents no phonetic irregularity. For the c, cf. r
scratch" < *rasicare, "to scratch." There is no r
doubting that intervocalic c in this position, in Sp
Provencal and French, may either remain a surd o
a sonant, according to the date at which the prece
dropped. The formation of rosca, " a gnawing ins
as a postverbal from *rosicare, "to gnaw," has
parallels, for which I refer to Meyer-Liibke, Rom
II, pp. 444 ff. I need only mention Italian leva
from levare, "to raise." For the sense-developmen
compare English bit, i. e., apparently "a biting ins
and Italian sncchiare, "to bore," generally deri
*suculare, "to suck." *LRosicare may have a direct
ant in the Spanish technical word roscar, "to
which, however, may also be a recent derivative f

III.

Spanish sesgo, "oblique" < sesgar, "to cut obliquely

*sesecdre, "to cut apart."
To explain Spanish sesgo, "oblique," Baist, Zeitschrif
rom. Phil., vII, p. 122, sets up a type *sesecus, whic
attempts to support by the analogy of circumsecus, extrinse
The formation of *ssec?us is not made sufficiently prob
and the etymon is rightly rejected by K6rting, who fa
the derivation from *subsecare. The latter type, howe
presents insuperable phonetic difficulty, to say nothin
semantic obscurity. Ulrich, Zcitschrift fib' rom. Phil.
p. 383, derived sesgar from *sexicare < *sexus < sectus,
Korting objects to the etymology on the ground that *s
for sectus is a monstrosity. *Sxicare also presents phon
and semantic difficulty.
The right etymon is *ssesecre, "to cut apart," an un

This content downloaded from 140.206.154.232 on Sat, 10 Jun 2017 05:55:56 UTC
All use subject to http://about.jstor.org/terms
 344 CARL C. RICE.

peachable formation presenting no phonetic i

This type was also thought of by Baist (articl
dismissed it on account of the existence of
sesgo. But the derivation of sesgo as a post
sesgar presents no difficulty. On the formati
verbal adjectives I refer to Meyer-Liibke, Rom
p. 448, and to the Dictionnaire genEral,
sense-development also presents no difficulty
apart," "to cut across;" 2) "cut across," " ob

IV.

French ruche, "hive" (beside rouche, "hull of a ship on

the stocks ") < 0. F. rusche, Prov. rusca, Piedmontese and
Lombard rusca, "bark," < Comascan ruscd, "to scale off,"
< *ruspicare < *ruspare, "to scratch."
In the 1ictionnaire general we are told that ruche is of
Celtic origin. Korting's article on the word reads as
follows :-

"RUsca ist das vorauszusetzende, aber beziiglich seines Ursprunges ganlz

dunkle Grundwort zu prov. rusca, Baumrinde (auch piemont. und lomb.
rusca); altfrz. rusche (norm. ruque), neufrz. ruche (aus Baumrinde ge-
fertigter Bienenkorb, Schiffsrumpf). Diez 673 hielt das Wort fiir keltisch,
Thurneysen, p. 111, verneint dies."

To make it clear that the words for "hive " and " bark"
are identical, Diez, 1. c., cites Spanish corcho, meaning both
"bark of the cork-tree" and "bee-hive." I propose to
derive the group from the verbal type *ruspicare, which
seems to explain perfectly all the forms. For the dropping
of the middle vowel in Provengal and French we may com-
pare Latin hospitale > Prov. ostal, 0. F. ostel. In Tuscan,
hospitale becomes ospedale, retaining the pretonic vowel.
But Meyer-Liibke, Italienische Grammatik, p. 71, notes that
the Italian dialects diverge widely from Florentine in their

This content downloaded from 140.206.154.232 on Sat, 10 Jun 2017 05:55:56 UTC
All use subject to http://about.jstor.org/terms
 ROMANCE ETYMOLOGIES. 345

treatment of syncope, and in view of the vagueness of

present knowledge of the whole question I hold that, un
the contrary assumption can be supported by evidence,
should admit the regularity of the development of *rut.sic
into an early Italian *ruscare, surviving in Comasca
ruscd, whence as postverbals Piedmontese and Lomb
rusca. We may perhaps cite Tuscan fastare< *taxit
destare < *de-excitare as showing a development parall
that assumed, though, to be sure, the consonant-group
question are quite different. This verb *ruspicare is der
readily enough, by the elimination of the common suf
-icare, from *ruspare, the existence of which in Latin
the original meaning of " to scratch " is generally adm
by Romance scholars, e. g., by Diez, Korting, and Schu
hardt (Romanische Etymologieen, I, p. 27) on accoun
the existence of Latin ruspari, "to examine," and It
ruspare, " to scratch." The semantic series,-1) " to scra
off," "to peel," 2) "peel," bark," 3) " hive made of bark
4) "hive,"-seems perfectly legitimate, particularly in v
of the fact that the first stage is supported by Comas
ruscd, "to scale off."
CARL C. RICE.

This content downloaded from 140.206.154.232 on Sat, 10 Jun 2017 05:55:56 UTC
All use subject to http://about.jstor.org/terms