STOP
Early Journal Content on JSTOR, Free to Anyone in the World
This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in
the world by JSTOR.
Known as the Early Journal Content, this set of works include research articles, news, letters, and other
writings published in more than 200 of the oldest leading academic journals. The works date from the
mid-seventeenth to the early twentieth centuries.
We encourage people to read and share the Early Journal Content openly and to tell others that this
resource exists. People may post this content online or redistribute in any way for non-commercial
purposes.
Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early-
journal-content .
JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people
discover, use, and build upon a wide range of content through a powerful research and teaching
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please
contact support@jstor.org.
V.— SOBRIQUET AND STEM.
i. Composition and Suffixation : dk w - ' eye ', ano- 'face'.
The preconceived idea of suffixation has often made for blind-
ness. But we now generally recognize that Lat. ferox and
atrox are compounds with the posterius -6k w - ' looking, seeing ;
eye ' ; and that, even though the sense of the posterius is quite
in the way of vanishing ; cf. also velox celox (see § 2 ; and cf
Kretschmer, Einl., p. 160). The same posterius is quite trans-
parent in Greek -<ow(o)- in words like olvoxf/ (olvunfr, oivanros) ,
yopy<ail/ (yopyoiTros) , irap6ivo>w6's and, with adverbial prius, in
t^Awjtos ' far-seen > distant '. Like ferus : ferox is 8«vos : Seivdxj/
(8«vw7ros), and the evanescence of the posterius is even greater
in Horn. crTctv«)7rds : aruvo? (cf. tvpo>ir6<s, KoiA«)7rds). Again, many
scholars recognize in wp-avr/s Trpo<rqvr)<s 0.1^^ a posterius -ones- :
Skr. and- 'face V and so they ought, cf. Lat. pronus{ <pro-anos).
Further note Skr. sam-and-s 'similis' (and almost 'aequos')
<'uno-ore' (=sam-anana-) : Lat. (s)m-anos 'aequos> be-
nignus'; imm-ani-s 'atrox' (prius ismo-: Skr. ismln- 'rapi-
dus ; turbulentus ', see IF. 26, 41 ) : aidvr^ ' atrox ' (prius alao- :
Av. aesa-sa- 2 'petens, adoriens ') ; c-aros (used of cloth), from
es(w)-( :iv<s ' bonus ')+ ano- =' bona- facie'. Skr. sahasana-s
(1), with evanescent posterius, amounts to a long form of
sdhas ' violentus ' and argasands 3 (2) ' iniuriosus ' belongs with
'Boisacq's objection that a is "gunated" (he means vriddhied) a is
puerile. What is o in ambages, pray? His defense of the complicated
derivation of -acijs from -anses- has no other purpose than to find an
etymon for Goth, ansts ' gnadengabe '• — which he ought to look for in
Lat. <h>onos (with <h> as in <h>umerus), in the plural = 'gifts
of honor' and honestus; also in ivi-vti-iii (not o-vi-vrjiu) , as Uhlenbeck
has seen.
2 IE. aisos k (h) 0-, in Lat. aeruscator 'mendicus impudens', see CQ.
9, III, where u is wrongly explained as original.
3 Macdonnell, Ved. Gr. § 527, lists twelve other such forms as j-aor.
participles. If we bear in mind that sdhas 'violens' also means (as a
noun) 'vis', we find for over half of our list nouns in as, e. g. (3) ohas
' expectatio ', (4) jrdyas 'extensio, spatium', (5) nomas 'honor', (6)
SOBRIQUET AND STEM. 83
* areas ' iniuria '. Has via a similar long form in vt-avias (pos-
terius *aniya- like Skr. asiya-m: Lat. os 'mouth, face')? cf.
Skr. kalyinl ( f em.) = 'lovely ': kalya-s, koAos (on n see
Wackernagel, ai. gr. I, § 173).
2. There is another group of -ana- compounds in the Rig-
Veda, the analysis of which yields astounding results. Of the
proper names, omitting Apnavana-s with its obscure prius,
Cydv-ana-s (one raised from the dead by the A^vins) and
Cydvat-dna-s seem to mean ' mobili-ore ', and the priora cyo-
and cydvat- are either adjective and participial (Grundriss,
2 - x > §3 r 3 Y) or c y6- 1S an imperative (ibid., 2. 1. 64), and
cydvat- a transitive participle governing the posterius (ibid. 7).
Of the mere adjectives, tdkav-dnas 'velox' serves as a long
iorm to taku-; vdsav-ana-s 'dives' tovdsu- 'benignus' (vdsu-
also = 'opes', cf. optimus) ; bhfgav-ana-s 'splendidus' to
Bhfgu- ' Splendens '. Latin Di-ana, the moon goddess, may be
from dlyo-dno- ' of shining face ' or, if Varro's Divi-dna- is not
an invention, from divi-ydna- ' in caelo iens ', cf. Skr. divi-
gamana-s ' planet/
3. But it is from the proper name" Pfthav-dnas that we can
derive most instruction, and I will make bold to say at the
beginning, but deprecating any charge of insolence, that I do
not write for readers who, on account of looking at the forest,
refuse to look at the trees. To learn, we must seize the indi-
vidual thing that seems to be true and if, tested with and by
other things of its kind, it still rings true, true it must be. So
I will start from the truth that seems to glare from the proper
name,
P'rthav-dna-s. Vocative origin of sobriquet compounds.
4. The vocative P'rthav Ana =(0) Broad Face. Its prius,
prthav, 1 is the legitimate prevocalic samdhi form of p'rtho,
bhiyas 'fear', (7) rabhas 'vis', (8) (avas 'vis'. For the others, -as
nouns, though not of record, are perfectly legitimate formations ; and
rhjas-ana-s ' celer ' is no less apt to contain -ana- ' face ' than Lat. velSx
to contain -Ok™-. For yama-sands ' reined up', I can but think we have
a compound of yama- ' rein '+ sdna-s 'vinctus' (:Vsa, in infin. satum).
1 The objection that in secondary derivation before vowel suffixes u
stems show av in Sanskrit will not lie. If we may not claim the spread
of av from cases like Prthav-anas and tdkav-anas (§ 2) , still in deri-
vatives like hanaviya- 'genulnus' (spelled hanavyd-), the stem hanavi-
84 AMERICAN JOURNAL OF PHILOLOGY.
vocative to prthu-s 'broad' (see Whitney, Gr 2 . § 134, a) ; the
second vocative ana is properly enclitic (ibid. §314, d). The
vocative P'rthaiana, after being used as a nominative (cf. Lat.
Ju-piter and its likes), picked up the nominatival -s; and
thence pfthav- was carried through the paradigm. Starting
from a case like Pfthav-ana-s, where the vocative prius has
the look of being an inexpugnable fact, we may infer that
such compounds as ugrd-bdhu-s ' stout-armed ' will also have
started as ugra-bdho (O) Stout Arm; and will contain in
their a (< 0) an IE. -e, deflected in the post tonic syllable to
IE. o. The final accentuation of ugrdbahu- will be due to un-
compounded ugrdm bdhum (ace), cf. uruh kdksah 'Broad
Shoulder', to be cited presently; Lat. os durum (Terence,
Eun. 806).
5. The notion that the Indo-European proper name repre-
sents a predominant early type of compound has often been
advanced. As soon as we let such sobriquet compounds start
in the vocative, where it seems that they certainly must have
started, the figment of the composition stem or pre-flexional
word, or casus indefinitus (!) in -0 yields to the sane con-
viction that the o-stem, for all its wide subsequent develop-
"ment in composition, originated in a genuine and isolable
word form, viz. the vocative in -e (deflected to -0). As for
the typical name Broad Face (Latine, Bronze Beard), we have
somehow failed 1 adequately to realize its actual vitality in our
own tongue ; in " Grandaddy Long Legs " ; in our Texas so-
briquet of "Big Foot Wallace"; in "Flat Head Mountain";
in " Faint heart ne'er won fair lady " ; in Shakespeare's voca-
tive instance, " O you hard hearts, you cruel men of Rome ".
This style of nomenclature has also been pointed out in the
Rig Veda, as in uruh kdksah . . . gangydh ' Latus Humerus
will lend itself to explanation as a locative (v. Macdonnell, Ved. Gr.,
§296), cf. Lith. danguje-jis 'heavenly' from loc. dangujc (ex. ap. Brug-
mann, Gr. 2. 1, §66, 3). [I now realize that in Homeric ravai-wovs (and
rava-inris ?) we have a vocative prius, while in (voc.) l%6v v may come
from 3u. So ravav- comes from voc. tn-n^u, cf. Skr. nom. tanb-s,
allocated to the feminine; note v in SeUvvtu. Likewise ravafis is from
tn-nsw<o>-; cf. the suffix -ewo-, set up by Brugmann-Thumb, p.
213']
1 But now cf. W. Petersen in IF. 34. 262. § 14. with a splendid list of
examples.
SOBRIQUET AND STEM. 85
. . Gangeticus ', (see Wackernagel, ai. Gram. II, 1, § 112 b, 3),
We need not involve our explanation in the pars pro toto
figure : barefoot, hard heart, and the like are more primitive
than rhetoric; 1 nor limit it to names of persons (cf. Flat
Head, name of a mountain).
6. A word may be added on the role of the nominative in
making compounds. In RV. 1. 32. 6 d, the problematic rujtnas
was long ago correctly interpreted for the thought by Pro-
fessor Bloomfield. Morphologically, I would compare Eng.
hump-back and Germ, spaltfuss, analyzing as rujd ' cleft '
( :p<l>£ 2 < IE. wrg- \ rug-, see Grundriss 1 § 539, 3 ) + nas ( : Lat.
nares) ' nose' (or + anes- ' face ', §1). Leaving out rujttnah,
our pada states that " Indra's enemy was crushed to pieces ",
and we may insert rujtnas parenthetically as "a cleft his
nose ". 3 In urdhva<s>and-s (hapax in a line with arqasana-s,
'An overwhelming number of sobriquet compounds, exhibiting in the
posterius the name of a part of the body, might doubtless be turned up
in the various tongues, such words as Av. darago-basu- long-arm (ed),
-angusta- ' long-thumbed ', -gava- ' longi-manus ' ; parsSu-varah- 'broad-
chested ', -safa- ' broad-hoofed ', -sraoni- ' broad-hipped '. In Greek we
have like compounds in -xetp ( K P aTe po-x e 'p) > -tow (Kparaiirovs) , -<ppriv
(raXal-tppav) . As for the -at of ra\ai</>pov (vocative) and the -a of
raXacfrpoiv ' stout-heart ', they furnish evidence of an Indo-European
vocative to -a stems in -at | -a ; the former due before consonants, the
latter before vowels, but confused. This vocative prius we have in
K\uTai-/ii7<7Tpa=:(0) Famous Contriver; perhaps in iual-$ot>e=z(0) Pol-
lution, (O) murderer | ; in KpaTai-irous=(0) hard foot (Kparat- a voca-
tive to the < ? nom. masc. > adverbial form /copra ?) ; in raKaai-ippav
the prius will be the vocative feminine of a participle *tll-nt-l ' ferens'.
The explanations of the handbooks, as of KpaTat-from Kparatot and the
like, are entirely artificial and have not been made even algebraically
convincing.
* The river ruj&na mentioned by the commentator on this passage
may well have existed, cf. &wo-pp<!>£, of a branch of the Styx.
3 We also have in Sanskrit rju-nas-^= (O) Straight Nose, com-
pounded of the -it vocative *fju (u lengthened in composition to avoid
the rhythmic succession ^ w <->) -J- the vocative nas; cf. in Greek SiXa-
vos (-vo- : Skr. fern, nas- ' nose ' : : Skr. -da- : -das- ; cf. the o-stem
ayytXos : Skr. dhgiras, ej-stem ; and Skr. compounds with posterius in
-stha-s: stha-s; also Wackernagel, ai. gram. II. 1, p. 96 a). Not Solm-
sen in IF. 30, 13, but Fay in CR. 18, 208, was the first to define XiKavos
by Snub-nose. If in the name of this Bacchic deity we are to recognize
a derivative from Thracian fiXa 'wine' (Kretschmer, Glotta, IV. 351
sq.), he must owe his traditional facial character to what appeared to
be the meaning of his name, in short to "Disease of Language."
86 AMERICAN JOURNAL OF PHILOLOGY.
and so liable to explanation, <^> and all, by momentaneous
irradiation), the <^> may also imply a parenthesis ("upwards
<was> his face") ; or -ana- may have stood to urdhvas in a
relation of "specification" (in Greek the accusative; the IE.
instrumental, see Grundriss, 2. 2. p. 543!). Out of the in-
strumental of attendant circumstance, as e. g. in Latin mulier
forma honesta, the possessive (Bahuvrlhi) type of compound
might also have developed ; cf . in reversed order Skr. janv-aknd-
( with) ' knee-bent '. The two lines of development would have
converged. The course of the development from the sobriquet
compound can be very simply comprehended by looking at a
few Sanskrit compounds with prius ugrd- : ( 1 ) ugrd-bahu-
StrongArm; (2) ugra-dhanvan- ' strong-bowed ', ugra-yud ha-
' strong- weaponed ' ; (3) ugrd-sena- ' strong-armied ', ugrd-
vlra- ' strong-heroed ', ugrd-putra- ' strong-sonned ' ; (4)
ugratejas- ' strong-energied '.
7. In the Avesta, the prius of similar compounds is often in
the nominative : does that point to the fact that the nominative,
and not the vocative, was the generalized form, at least in
a-stems ? For Sanskrit ugrd-bahus, the vocative ugra bdho
has dictated the form of the prius ; but in the Avesta we have
in dardgo-bazu- the type of the Skr. nominative ugrd bahu-s
(like uruh kaksdh, § 5).
8. The arguments advanced above seem to me to have made
it abundantly probable that the non-isolable stem which serves
as the prius of composition originated in the vocative — or also
to a much less degree in some other case form — of a sobriquet
group. If this be true, while the developed facts of grammar
as regards composition remain untouched, our conception of
the fabric of the word must be profoundly modified. Thus
we may no longer speak of the vocative of the o and a declen-
sions as the unmodified stem, but say contrariwise that the
stem is the vocative. And the source of the vocatival -e seems
a thing we can come upon. As the augment, e- is an almost
isolable word, and meant ' there ' ! ( ? or ' here ' ; cf . Drewitt,
CQ. VI, 44 sq. ; i-diXw). The same exclamatory adverb -e makes
an ideal vocative indicator, as anybody knows who has ever
sung out " you-there!" to the man of unknown name whose
attention is wanted quickly from the next tennis court. This
e (in the long form e) is also used as a preposition ; and when
SOBRIQUET AND STEM. 87
Brugmann (Gr. 2. 2. §634) explains it as instrumental of the
demonstrative stem e he is confounding an original local
adverb (without case) with that much subsequent thing a case
form x : and it really is this -e that becomes the casual suffix
of the Indo-Iranian instrumental in -a. In the pair 6\pe | 6\f/t-
the same -e competes with -i (cf. v<j/i) as an adverb deter-
minant. — For the identification of vocatival e with -e in the
imperative see e. g. AJPh. 15, 413; and on ex 'there'! Bull.
Univ. Texas, no. 263, § 99 sq. On -e x as the instrumental case
ending cf. AJPh. t>7, io 7 2 ; as the locative, 170, § 28.
Edwin W. Fay.
University of Texas, Nov. 8, 1915.
'He who thinks that IE. "e 'there' is a merely glottogonic inference
and instrum. e something less glottogonic — has never thought !
Addenda.
§1. The Latin pair vetus: veter-anus conforms precisely with the
type of Skr. sdhas- : sahas-and-.
§1, fn. 3. Cf. rsi-sana- (RV.), attached to a seer.
P. 85, fn. 1. In the Vedic proper names and epithets Hdray-ana- ,
Tawny Face (quasi Atrox) ; Turvay-ana-, Mighty Face (quasi Ferox) ;
and Ahray-ana-, Unabashed Face, ay in the priora is the samdhi form
of the e (=IE. ei/oi) vocative of hdri- turvi- d-hri-.
§8. Thus Lat. puer-e may be conceived as like our outcry "boy
there I"
